1urt7Ls

joshuahagedorn

1urt7Ls

Comparative Effectiveness ReviewNumber 138Medication TherapyManagementInterventions inOutpatient Settings


PrefaceThe Agency for Healthcare Research and Quality (AHRQ), through its Evidence-basedPractice Centers (EPCs), sponsors the development of systematic reviews to assist public- andprivate-sector organizations in their efforts to improve the quality of health care in the UnitedStates. These reviews provide comprehensive, science-based information on common, costlymedical conditions, and new health care technologies and strategies.Systematic reviews are the building blocks underlying evidence-based practice; they focusattention on the strength and limits of evidence from research studies about the effectiveness andsafety of a clinical intervention. In the context of developing recommendations for practice,systematic reviews can help clarify whether assertions about the value of the intervention arebased on strong evidence from clinical studies. For more information about AHRQ EPCsystematic reviews, see www.effectivehealthcare.ahrq.gov/reference/purpose.cfm.AHRQ expects that these systematic reviews will be helpful to health plans, providers,purchasers, government programs, and the health care system as a whole. Transparency andstakeholder input are essential to the Effective Health Care Program. Please visit the Web site(www.effectivehealthcare.ahrq.gov) to see draft research questions and reports or to join anemail list to learn about new program products and opportunities for input.We welcome comments on this systematic review. They may be sent by mail to ChristineChang, M.D., M.P.H., at: Agency for Healthcare Research and Quality, 540 Gaither Road,Rockville, MD 20850, or by email to epc@ahrq.hhs.gov.Richard G. Kronick, Ph.D.DirectorAgency for Healthcare Research and QualityStephanie Chang, M.D., M.P.H.DirectorEPC ProgramCenter for Evidence and Practice ImprovementAgency for Healthcare Research and QualityDavid Meyers, M.D.Acting DirectorCenter for Evidence and PracticeImprovementAgency for Healthcare Research and QualityCarmen Y. Kelly, Pharm.D., M.P.H., R.Ph.Task Order OfficerCenter for Evidence and PracticeImprovementAgency for Healthcare Research and Qualityiii


AcknowledgmentsThe authors gratefully acknowledge the continuing support of our AHRQ Task OrderOfficer, Carmen Y. Kelly, Pharm.D., M.P.H., R.Ph.; the guidance of Christine Chang, M.D.,M.P.H., and Aysegul Gozu, M.D., M.P.H., of AHRQ; and our Associate Editor, MargaretMaglione, M.P.P. We extend our appreciation to our Key Informants and members of ourTechnical Expert Panel (listed below), all of whom provided thoughtful advice and input duringour research process. The investigators deeply appreciate the considerable support, commitment,and contributions of the EPC team staff at RTI International and the University of NorthCarolina at Chapel Hill. We express our gratitude to the following individuals for theircontributions to this project: Carol Woodell, B.S.P.H., our Project Manager; Christine DeLongJones, M.D., M.S., and Justinne E. Guyton, Pharm.D., C.P.P., our Content Experts; ChristianeVoison, M.S.L.S., our EPC Librarian; Mahima Ashok, Ph.D., Sarah Selenich, M.P.P., and CatherineAspden Grodensky, M.P.H., our Research Associates; Stefanie Knauer, M.A., Amy Greenblatt,B.A., and Claire Baker, B.A., our Research Assistants; Sharon Barrel, Jennifer Drolet, and CarolOffen, our EPC Editors; and Loraine Monroe, our EPC Publications Specialist.iv


Key InformantsIn designing the study questions, the EPC consulted several Key Informants who representthe end-users of research. The EPC sought the Key Informant input on the priority areas forresearch and synthesis. Key Informants are not involved in the analysis of the evidence or thewriting of the report. Therefore, in the end, study questions, design, methodological approaches,and/or conclusions do not necessarily represent the views of individual Key Informants.Key Informants must disclose any financial conflicts of interest greater than $10,000 and anyother relevant business or professional conflicts of interest. Because of their role as end-users,individuals with potential conflicts may be retained. The TOO and the EPC work to balance,manage, or mitigate any conflicts of interest.The list of Key Informants who participated in developing this report follows:Jamie C. Barner, Ph.D.College of PharmacyThe University of Texas at AustinAustin, TXSteve Handler, M.D., Ph.D.Department of Biomedical Informatics andDivision of Geriatric MedicineUniversity of PittsburghPittsburgh, PAJoseph Hill, M.A.American Society of Health-SystemPharmacistsBethesda, MDBrian Isetts, Ph.D., B.C.P.S., FAPhACollege of PharmacyUniversity of MinnesotaMinneapolis, MNMichelle Ketcham, Pharm.D., M.B.A.Medicare Drug Benefit GroupCenters for Medicare & Medicaid ServicesBaltimore, MDJulie Kuhle, R.Ph.Senior Director, Performance MeasurementPharmacy Quality Alliance, Inc.Springfield, VAJames Owen, Pharm.D., B.C.P.S.Practice & Science AffairsAmerican Pharmacists AssociationWashington, DCv


Technical Expert PanelIn designing the study questions and methodology at the outset of this report, the EPCconsulted several technical and content experts. Broad expertise and perspectives were sought.Divergent and conflicted opinions are common and perceived as healthy scientific discourse thatresults in a thoughtful, relevant systematic review. Therefore, in the end, study questions, design,methodologic approaches, and/or conclusions do not necessarily represent the views ofindividual technical and content experts.Technical Experts must disclose any financial conflicts of interest greater than $10,000 andany other relevant business or professional conflicts of interest. Because of their unique clinicalor content expertise, individuals with potential conflicts may be retained. The TOO and the EPCwork to balance, manage, or mitigate any potential conflicts of interest identified.The list of Technical Experts who participated in developing this report follows:Jamie C. Barner, Ph.D.College of PharmacyThe University of Texas at AustinAustin, TXSteve Handler, M.D., Ph.D.Department of Biomedical Informatics and Division of Geriatric MedicineUniversity of PittsburghPittsburgh, PAJoseph Hill, M.A.American Society of Health-System PharmacistsBethesda, MDBrian Isetts, Ph.D., B.C.P.S., FAPhACollege of PharmacyUniversity of MinnesotaMinneapolis, MNMichelle Ketcham, Pharm.D., M.B.A.Medicare Drug Benefit GroupCenters for Medicare & Medicaid ServicesBaltimore, MDJulie Kuhle, R.Ph.Senior Director, Performance MeasurementPharmacy Quality Alliance, Inc.Springfield, VAJames Owen, Pharm.D., B.C.P.S.Practice & Science AffairsAmerican Pharmacists AssociationWashington, DCvi


Peer ReviewersPrior to publication of the final evidence report, EPCs sought input from independent PeerReviewers without financial conflicts of interest. However, the conclusions and synthesis of thescientific literature presented in this report do not necessarily represent the views of individualreviewers.Peer Reviewers must disclose any financial conflicts of interest greater than $10,000 and anyother relevant business or professional conflicts of interest. Because of their unique clinical orcontent expertise, individuals with potential nonfinancial conflicts may be retained. The TOOand the EPC work to balance, manage, or mitigate any potential nonfinancial conflicts of interestidentified.The list of Peer Reviewers follows:Rowena Dolor, M.D., M.H.S.Primary Care Research ConsortiumDuke Clinical Research InstituteDuke UniversityDurham, NCWilliam Doucette, Ph.D.College of PharmacyUniversity of IowaIowa City, IAC. Bernie Good, M.D., M.P.H.Center for Health Equity Research and PromotionVA Pittsburgh Healthcare SystemPittsburgh, PASteve Handler, M.D., Ph.D.Department of Biomedical Informaticsand Division of Geriatric MedicineUniversity of PittsburghPittsburgh, PABrian Isetts, Ph.D., B.C.P.S., FAPhACollege of PharmacyUniversity of MinnesotaMinneapolis, MNJulie Kuhle, R.Ph.Senior Director, Performance MeasurementPharmacy Quality Alliance, Inc.Springfield, VAvii


Andrew Masica, M.D., M.S.C.I.Baylor Scott & White HealthDallas, TXJames Owen, Pharm.D., B.C.P.S.Practice & Science AffairsAmerican Pharmacists AssociationWashington, DCAlan Zilich, Pharm.D.Research Scientist, Roudebush VA Medical Center, Center of Health Information andCommunicationIndianapolis, INAssociate Professor, Purdue University College of PharmacyWest Lafayette, INviii


Medication Therapy Management Interventions inOutpatient SettingsStructured AbstractObjectives. To describe intervention components and implementation features (Key Question[KQ]1) for outpatient medication therapy management (MTM) interventions withcomprehensive medication review, followup, education, and care coordination; assess theeffectiveness of these MTM interventions on intermediate, patient-centered, or resourceutilization outcomes (KQ 2); identify intervention features (KQ 3) and patient characteristics(KQ 4) that moderate the effect of an intervention on outcomes; and assess harms associatedwith interventions (KQ 5).Data sources. MEDLINE ® , Cochrane Library, International Pharmaceutical Abstracts, grayliterature, additional studies from reference lists and technical experts.Review methods. Two trained reviewers selected, extracted data from, and rated the risk of biasof relevant trials and cohort studies. We used random-effects models to estimate pooled effectsfor outcomes with three or more similar studies with a low or medium risk of bias. For otheroutcomes, we synthesized the data qualitatively.Results. We included 44 eligible studies (21 randomized controlled trials, 4 controlled clinicaltrials, and 19 cohort studies) reported in 61 articles, described in detail in the report (KQ 1).Evidence was insufficient on the effect of outpatient MTM interventions on most outcomes (KQ2). In a few instances, described below, the evidence led us to conclude benefit or lack of benefit.Specifically, we found evidence that MTM results in improvement when compared with usualcare for some measures of medication adherence and appropriateness; medication dosing; healthplan expenditures on medication costs; and, for patients with diabetes, the proportionhospitalized and costs of hospitalization. Similarly, we conclude, based on a low strength ofevidence, that MTM confers no benefit for patient satisfaction and most measures of healthrelatedquality of life.We found evidence on five intervention components and intervention features (KQ 3). Onestudy provided information on each feature and yielded insufficient evidence for most outcomes,with the following two exceptions. An MTM program with pharmacist access to brief clinicalsummaries from the medical record reduces the mean number of adverse drug events whencompared with a basic MTM program without such access (low strength of evidence).Community pharmacists increase the generic dispensing ratio more than call-center–basedpharmacists (low strength of evidence). We found no relevant studies on patient characteristicsmoderating the effect of MTM interventions (KQ 4). Similarly, the evidence on harms associatedwith MTM was limited to one study on inconvenience and was rated as insufficient (KQ 5).Conclusions. The evidence base offers low evidence of benefit for a limited number ofintermediate and health utilization outcomes. We graded the evidence as insufficient for mostother outcomes because of inconsistency in direction, magnitude, and precision, rather than lackof evidence. Wide variations in populations and interventions, both within and across studies,ix


likely explain these inconsistencies. Given the widespread implementation of MTM and urgentneed for actionable information, optimal investments in new research require a process ofresearch prioritization in which the value of information from each proposed study is carefullyconsidered. Studies designed to identify causal relationships between MTM interventions andtheir outcomes require adequate controls for confounding but may offer limited information onthe factors that explain program success or failure. Studies designed to explore the reasons forprogram success or failure using qualitative or single-arm designs may offer hypothesesgeneratingrather than hypotheses-confirming insights on MTM effectiveness. New research,regardless of specific focus, will likely continue to find inconsistent results until underlyingsources of heterogeneity are accounted for.x


ContentsExecutive Summary ................................................................................................................ ES-1Introduction ................................................................................................................................... 1Context ........................................................................................................................................ 1Populations .............................................................................................................................. 2Interventions and Comparators ............................................................................................... 3Outcomes ................................................................................................................................ 5Settings .................................................................................................................................... 5Timing ..................................................................................................................................... 6Contextual Factors .................................................................................................................. 6Scope and Key Questions ........................................................................................................... 7Scope of the Review ............................................................................................................... 7Relevance of Research Question to Clinical Decisionmaking or Policymaking .................... 7Key Questions ......................................................................................................................... 7Populations, Interventions, Comparators, Outcomes, Timing, and Setting (PICOTS) .............. 9Organization of the Report........................................................................................................ 12Methods ........................................................................................................................................ 13Topic Refinement and Review Protocol ................................................................................... 13Literature Search and Identification Strategy ........................................................................... 13Search Strategy ..................................................................................................................... 13Inclusion/Exclusion Criteria ................................................................................................. 14Study Selection ..................................................................................................................... 15Data Extraction ..................................................................................................................... 15Assessment of Risk of Bias of Individual Studies .................................................................... 17Data Synthesis ........................................................................................................................... 17Grading Strength of Evidence for Individual Comparisons and Outcomes ............................. 18Assessing Applicability ............................................................................................................ 19Peer Review and Public Commentary ...................................................................................... 19Results .......................................................................................................................................... 20Introduction ............................................................................................................................... 20Results of Literature Searches .................................................................................................. 20Key Question 1: Components and Implementation Features of MTM Interventions .............. 22Overall Descriptors of Study Interventions .......................................................................... 23Study-Level Descriptors of Interventions ............................................................................. 25Key Question 2: Effect of Medication Therapy Management Interventions onIntermediate, Patient-Centered, and Resource Utilization Outcomes ...................................... 27Key Points: Intermediate Outcomes ..................................................................................... 27Detailed Synthesis: Intermediate Outcomes ......................................................................... 28Key Points: Patient-Centered Outcomes ............................................................................... 58Detailed Synthesis: Patient-Centered Outcomes .................................................................. 58Resource Utilization .............................................................................................................. 76Detailed Synthesis: Resource Utilization .............................................................................. 77xi


Key Question 3: Outcomes of Medication Therapy Management by InterventionFeatures ................................................................................................................................... 114Key Points ........................................................................................................................... 114Detailed Synthesis: Intervention Features .......................................................................... 114Key Question 4. Outcomes of MTM by Patient Characteristics ............................................ 117Key Question 5. Harms of Medication Therapy Management Interventions ......................... 117Key Points ........................................................................................................................... 117Detailed Synthesis: Inconvenience ..................................................................................... 117Discussion................................................................................................................................... 118Key Findings and Strength of Evidence ................................................................................. 118KQ 1: Intervention Components and Implementation Features ......................................... 118KQ 2: Overall Effectiveness ............................................................................................... 118KQ 3: Effectiveness of MTM by Intervention Features ..................................................... 125KQ 4: Effectiveness of MTM by Patient Characteristics ................................................... 126KQ 5: Harms of MTM Interventions .................................................................................. 126Findings in Relation to What Is Already Known.................................................................... 126Applicability of the Findings .................................................................................................. 127Implications for Clinical Practice and Policymakers .............................................................. 128Limitations of the Comparative Effectiveness Review Process ............................................. 129Limitations of the Evidence .................................................................................................... 131Research Gaps ......................................................................................................................... 131Conclusions ............................................................................................................................. 133References .................................................................................................................................. 134TablesTable A. Populations, interventions, comparators, outcomes, timing, and settings ................. ES-5Table B. Definitions of the grades of overall strength of evidence ........................................ ES-12Table C. Summary of findings and strength of evidence for intermediate outcomes ofMTM interventions ................................................................................................................. ES-15Table D. Summary of findings and strength of evidence for patient-centered outcomesof MTM interventions ............................................................................................................. ES-17Table E. Summary of findings and strength of evidence for resource-utilization outcomesof MTM interventions ............................................................................................................. ES-19Table 1. MTM services ................................................................................................................... 2Table 2. Populations, interventions, comparators, outcomes, timing, and settings ........................ 9Table 3. Literature search terms for medication therapy management studies ............................ 14Table 4. Inclusion/exclusion criteria for medication therapy management studies ...................... 16Table 5. Definitions of the grades of overall strength of evidence ............................................... 19Table 6. Characteristics of included studies ................................................................................. 22Table 7. Characteristics of medication therapy management interventions ................................. 23Table 8. Characteristics of medication therapy management studies by type of patientpopulation (broad focus or narrow focus on conditions or diagnoses) ......................................... 24Table 9. Anticoagulation: Strength of evidence ........................................................................... 28Table 10. Hemoglobin A1c: Summary of results ......................................................................... 29Table 11. Hemoglobin A1c: Strength of evidence........................................................................ 30xii


Table 12. LDL cholesterol: Summary of results ........................................................................... 31Table 13. LDL cholesterol: Strength of evidence ......................................................................... 33Table 14. Blood pressure: Summary of results ............................................................................. 34Table 15. Improvement in blood pressure: Strength of evidence ................................................. 37Table 16. Drug therapy problems identified: Summary of results................................................ 38Table 17. Drug therapy problems identified: Strength of evidence .............................................. 38Table 18. Total number of drug therapy problems resolved: Summary of results ....................... 39Table 19. Specific drug therapy problems resolved: Summary of results .................................... 40Table 20. Drug therapy problems resolved: Strength of evidence ................................................ 41Table 21. Medication adherence: Summary of results grouped by type of adherenceoutcome ......................................................................................................................................... 42Table 22. Adherence outcome Type 1—proportion of patients adherent based on athreshold of percentage of pills taken: Strength of evidence ........................................................ 50Table 23. Adherence outcome Type 2—absolute measure of adherence as percentage ofprescribed doses taken: Strength of evidence ............................................................................... 51Table 24. Adherence outcome Type 3—self-reported scales: Strength of evidence .................... 52Table 25. Adherence outcome miscellaneous: Strength of evidence ........................................... 52Table 26. Medication appropriateness scales: Summary of results .............................................. 53Table 27. Medication appropriateness for individual medications: Summary of results ............. 55Table 28. Medication appropriateness scales: Strength of evidence ............................................ 57Table 29. Medication appropriateness for individual medications: Strength of evidence ............ 57Table 30. Medication dosing: Strength of evidence ..................................................................... 58Table 31. Adverse events: Summary of results ............................................................................ 59Table 32. Adverse drug events: Strength of evidence .................................................................. 61Table 33. Cognitive, affective, and physical function: Summary of findings .............................. 62Table 34. Cognitive and physical function: Strength of evidence ................................................ 63Table 35. Affective function: Strength of evidence ...................................................................... 63Table 36. All-cause mortality: Summary of results ...................................................................... 64Table 37. All-cause mortality: Strength of evidence .................................................................... 64Table 38. Gastrointestinal bleeding events: Strength of evidence ................................................ 65Table 39. Scores on SF-36 measures: Summary of effects from meta-analyses .......................... 65Table 40. SF-36: Strength of evidence ......................................................................................... 68Table 41. Condition-specific quality of life: Summary of results ................................................ 70Table 42. Condition-specific quality of life: Strength of evidence ............................................... 70Table 43. Patient satisfaction: Summary of results ....................................................................... 71Table 44. Patient satisfaction: Strength of evidence ..................................................................... 76Table 45. Use of generic medications: Summary of results ......................................................... 78Table 46. Use of generics for MTM versus usual care: Strength of evidence .............................. 79Table 47. Measures used in studies of costs of medications ......................................................... 80Table 48. Patient copayments: Summary of results ...................................................................... 83Table 49. Patient copayments: Strength of evidence .................................................................... 84Table 50. Total expenditures on medications by insurers: Summary of results ........................... 85Table 51. Health plan expenditures: Strength of evidence .......................................................... 88Table 52. Total outlays on medications: Summary of results ....................................................... 88Table 53. Total outlays on medications: Strength of evidence ..................................................... 91Table 54. Medication and other costs: Summary of results .......................................................... 91xiii


Table 55. Medication and other costs: Strength of evidence ........................................................ 93Table 56. Number of outpatient visits: Summary of results ......................................................... 94Table 57. Number of outpatient visits: Strength of evidence ....................................................... 96Table 58. Costs of outpatient visits: Summary of results ............................................................. 96Table 59. Costs of outpatient resource utilization: Strength of evidence ..................................... 98Table 60. Number of laboratory and diagnostic tests: Summary of results .................................. 99Table 61. Number of laboratory and diagnostic tests: Strength of evidence ................................ 99Table 62. Costs of laboratory tests: Summary of results ............................................................ 100Table 63. Costs of laboratory tests: Strength of evidence .......................................................... 100Table 64. Emergency department visits: Summary of results .................................................... 101Table 65. Emergency department visits: Strength of evidence ................................................... 103Table 66. Costs of emergency department visits: Summary of results ....................................... 104Table 67. Cost of emergency department visits: Strength of evidence....................................... 105Table 68. Hospitalizations: Mean number, risk and rates ........................................................... 105Table 69. Mean number of hospitalizations: Strength of evidence ............................................ 110Table 70. Risk of hospitalization: Strength of evidence ............................................................. 110Table 71. Rate of hospitalization: Strength of evidence ............................................................. 111Table 72. Costs of hospitalization: Summary of results ............................................................. 111Table 73. Cost of hospitalization: Strength of evidence ............................................................. 113Table 74. Length of hospitalization: Summary of results ........................................................... 113Table 75. Length of hospital stay: Strength of evidence ............................................................ 114Table 76. Access to patient records (basic MTM versus enhanced MTM): Strength ofevidence ...................................................................................................................................... 115Table 77. Intensity of care coordination and followup following comprehensivemedication review: Strength of evidence .................................................................................... 115Table 78. Pharmacy intensity of adoption: Strength of evidence ............................................... 116Table 79. Community pharmacy versus call center: Strength of evidence ................................. 116Table 80. Type of payer: Strength of evidence ........................................................................... 117Table 81. Confusion: Strength of evidence................................................................................. 117Table 82. Summary of findings and strength of evidence for intermediate outcomes ofMTM interventions ..................................................................................................................... 119Table 83. Summary of findings and strength of evidence for patient-centered outcomesof MTM interventions ................................................................................................................. 121Table 84. Summary of findings and strength of evidence for resource-utilization outcomesof MTM interventions ................................................................................................................. 123FiguresFigure A. Analytic framework for outpatient medication therapy management ...................... ES-4Figure B. Disposition of articles on medication therapy management (PRISMA figure) ...... ES-13Figure 1. Analytic framework for outpatient medication therapy management ............................. 8Figure 2. Disposition of articles on medication therapy management (PRISMA figure) ............ 21AppendixesAppendix A. Literature Search StrategiesAppendix B. Abstract and Full-Text Review Form TemplatesAppendix C. Studies Excluded After Full-Text Level ReviewAppendix D. Evidence Tablesxiv


Appendix E. Risk of Bias Evaluations and RationaleAppendix F. Meta-Analysesxv


Executive SummaryBackgroundUsed appropriately, medications can alleviate distressing symptoms that compromisephysical and psychological well-being, help prevent the onset of many acute and chronicillnesses, and improve patient health outcomes. Too often, however, medications are not usedappropriately. 1-3 In the United States in 2001, adverse drug events led to an estimated 4.3 millionambulatory visits. 4 In addition to problems involving adverse drug events, many patients do notreceive optimal pharmaceutical prescriptions. Even when optimal therapy is prescribed, patientinability to adhere closely to medication regimens may lead to poor health outcomes. 5Medication-related problems are especially pronounced among older adults. 6 Individuals 65years or older constitute 13 percent of the U.S. population, but they consume more than 30percent of all prescription medications. 6,7 A 2006 report found that nearly 60 percent of people inthis age group were taking 5 or more medications and that nearly 20 percent were taking 10 ormore medications, 8 placing them at increased risk for experiencing adverse drug events.Medication therapy management (MTM) services are intended to address issues ofpolypharmacy, preventable adverse drug events, medication adherence, and medication misuse. 9MTM is the current term that represents a suite of health care services that have evolved out ofthe philosophy and processes described in the early 1990s as “pharmaceutical care.” 9 TheMedicare Prescription Drug, Improvement, and Modernization Act of 2003 (Public Law 108-173) 10 expanded patient access to MTM services and established the requirements that MedicarePart D Prescription Drug Benefit Plan sponsors have to meet with respect to cost and quality andthe requirements for MTM programs sponsored by Part D drug benefit plans. The Centers forMedicare & Medicaid Services (CMS) requirements for Part D MTM programs have evolvedsince their implementation in 2006.Within a year of the passage of Medicare Part D, 11 national pharmacy organizationsestablished a consensus definition of MTM, 11 and in 2008 a subset of national pharmacyorganizations established five core elements for an MTM service model. 12 These elementsinclude a medication therapy review, a personal medication record, a medication action plan,intervention and/or referral, and documentation and followup. 9 Also in 2008, Current ProceduralTerminology (CPT ® ) for MTM services became available and further defined MTM and servicelevelexpectations. 13-15The evolution from isolated research interventions studying the impact of pharmaceuticalcare interventions to large-scale commercial MTM programs or collaborative medicationmanagement within primary care represents a journey along a continuum of practice settings,patient populations, and intervention components and features. Over time, the practice andstandards for these services have evolved, as have standards for describing and conductingresearch studies involving these interventions. A broadly defined scope for this review risksincluding studies that may be too different from each other to allow for meaningful comparisonand synthesis. A narrowly defined scope for this review risks the omission of studies that met thedefinition of MTM but that predated the Part D era, were conducted in other countries, or usedpatient eligibility criteria that are less restrictive than Part D.ES-1


Scope and Key QuestionsMTM is a complex intervention that could have different components depending on the goalsand scope of the MTM program. This review seeks to catalog outpatient-based MTMintervention components, assess the overall effectiveness of outpatient-based MTM incomparison with usual care, examine the factors under which outpatient-based MTM is effectiveand optimally delivered, assess what types of patients are likely to benefit from outpatient-basedMTM services, and clarify what types of patients may be at risk of harms from such programs.This review does not address (1) MTM services provided within inpatient settings or shortly afterhospital discharge, (2) disease management services provided by pharmacists, or (3)interventions designed as a single episode of contact. The rationale for limiting the scope toexclude some types of MTM interventions is to ensure that included studies are reasonablycomparable with respect to intended purpose and design of the MTM intervention.The Key Questions (KQs) addressed in this review are—KQ 1: What are the components and implementation features of MTMinterventions?KQ 2: In adults with one or more chronic diseases who are takingprescription medications, is MTM effective in improving the following:a. Intermediate outcomes, including biometric and laboratory measures,drug therapy problems identified, drug therapy problems resolved,medication adherence, goals of therapy met, and patient engagementin medication management?b. Patient-centered outcomes, such as disease-specific morbidity,disease-specific or all-cause mortality, adverse drug events, healthrelatedquality of life, activities of daily living, patient satisfaction withhealth care, work or school absenteeism, and patient and caregiverparticipation in medical care and decisionmaking?ES-2


c. Resource utilization, such as prescription drug costs, other healthcare costs, and health care utilization?KQ 3: Does the effectiveness of MTM differ by MTM components andimplementation features?KQ 4: Does the effectiveness of MTM differ by patient characteristics,including but not limited to patient demographics and numbers and types ofconditions and medications?KQ 5: Are there harms of MTM, and if so, what are they?Analytic FrameworkThe KQs are placed in relation to one another and the populations, interventions,comparators, outcomes, timing, and setting (PICOTS) in the analytic framework (Figure A).Specific details regarding patient population, intervention components, and outcomes areprovided in the next section.ES-3


Figure A. Analytic framework for outpatient medication therapy managementa The population, intervention, outcomes, and setting are described in detail in the text.KQ = Key Question; MTM = medication therapy management.Populations, Interventions, Comparators, Outcomes, Timing, andSettingTable A lays out the PICOTS for this review. For this review, we took a broad perspective onthe population and interventions evaluated; we did not require CMS Part D MTM eligibilitycriteria. Specifically, we did not require multiple chronic conditions or a minimum number orlevel of expenditures on prescription drugs. We included randomized and controlled clinicaltrials, systematic reviews, and prospective and retrospective cohort studies. We includedobservational studies because we anticipated, from our topic refinement work, that a reviewlimited to trials alone would fail to yield evidence on our wide range of prespecified benefitsand harms for MTM interventions as a whole and for studies evaluating the modifying effects ofspecific intervention and patient characteristics on outcomes of MTM interventions.ES-4


Table A. Populations, interventions, comparators, outcomes, timing, and settingsPICOTS Inclusion and Exclusion Criteria and Relevant Factors for Study AbstractionPopulations Inclusion criteria:• Patients age 18 or older with one or more chronic conditions requiring the use of prescriptionmedication to manage symptoms or prevent progression of chronic diseaseExclusion criteria:• Patients in long-term or acute care settings without access or control over their own medicationadministration.Relevant factors:• Patient characteristics that may influence intervention effectiveness: age, sex, race and ethnicity,socioeconomic status, health insurance status, educational level, health literacy status, cognitiveimpairment, number and types of chronic conditions, social support, and urban/rural statusES-5


Table A. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS Inclusion and Exclusion Criteria and Relevant Factors for Study AbstractionInterventions Inclusion criteria:• A bundle of medication-related services described by the term “MTM,” “pharmaceutical care,”“clinical pharmacy services,” or a similar phrase that include at a minimum the following 3elements:o Comprehensive medication review covering all prescription and nonprescription drugs,herbs, and supplements taken by the patient (i.e., a systematic process of collecting patientspecificinformation, assessing medication therapies to identify medication-related problems,developing a prioritized list of medication-related problems, and creating a plan to resolvethem with the patient, caregiver, and/or prescriber)o Patient-directed education and counseling or other resources to enhance understanding ofthe use of medicationo Coordination of care, including prescriber-directed interventions; documentation of MTMservices for use by the patient’s other providers; and referral to other providers, clinicians, orresources when appropriate16Exclusion criteria:• Medication reconciliation interventions that did not include all 3 elements described above wereexcluded• The following types of interventions may include MTM services, but MTM may represent only 1component of the overall intervention:o Disease-management interventions17o Case- or care-management interventions17o Patient-centered medical home models of careo Fully integrated collaborative care models involving multiple disciplines and specialtiesThese types of interventions were excluded unless studies contained the same level of overallmedical care or services among different study arms such that the effect of MTM could be isolated.For example, a study with 2 arms that has 1 arm with a care-management intervention that includesMTM services and another arm that has the care-management intervention without MTM servicescould be included. In contrast, a study that includes a care-management intervention with MTM in 1arm and usual medical care (no care-management intervention) in the other arm would not beincluded.Relevant factors:• Implementation features that may influence intervention effectiveness include the following:o Mode of delivery: telephone, face to face, virtual (Web/online/Internet), and remote videoo Type of professional providing initial and followup MTM service: pharmacist, nurse,physician, other cliniciano Frequency and interval of followup for MTM serviceso Specific MTM components usedo Fidelity in implementing MTM components: extent to which services were delivered asdesigned or intendedo Establishing and communicating goals of drug therapy to patients and among care providerso Method of identifying patients for enrollment (e.g., population health data, provider referralfor services, enrollment during a transition in care, targeting highly activated patients,targeting patients at time of high risk for event such as when prescribing a new drug)o Level of integration of MTM with usual care, which includes access to real-time clinicalinformation and laboratory values, and regular and consistent communication amongprescribers and others providing MTM serviceso Reimbursement characteristics (e.g., who is paying for cost of MTM services, who isreimbursed for MTM services, whether services are separately reimbursable)o Health system characteristics (e.g., are services being provided within an accountable careorganization, patient-centered medical home, or some other unique system setting, such asthe Veterans Health Administration, the Indian Health Service, non-U.S. single-payersystem)ES-6


Table A. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS Inclusion and Exclusion Criteria and Relevant Factors for Study AbstractionInterventions Inclusion criteria:• A bundle of medication-related services described by the term “MTM,” “pharmaceutical care,”“clinical pharmacy services,” or a similar phrase that include at a minimum the following 3elements:o Comprehensive medication review covering all prescription and nonprescription drugs,herbs, and supplements taken by the patient (i.e., a systematic process of collecting patientspecificinformation, assessing medication therapies to identify medication-related problems,developing a prioritized list of medication-related problems, and creating a plan to resolvethem with the patient, caregiver, and/or prescriber)o Patient-directed education and counseling or other resources to enhance understanding ofthe use of medicationo Coordination of care, including prescriber-directed interventions; documentation of MTMservices for use by the patient’s other providers; and referral to other providers, clinicians, orresources when appropriate 16Exclusion criteria:• Medication reconciliation interventions that did not include all 3 elements described above wereexcluded• The following types of interventions may include MTM services, but MTM may represent only 1component of the overall intervention:o Disease-management interventions 17o Case- or care-management interventions 17o Patient-centered medical home models of careo Fully integrated collaborative care models involving multiple disciplines and specialtiesThese types of interventions were excluded unless studies contained the same level of overallmedical care or services among different study arms such that the effect of MTM could beisolated. For example, a study with 2 arms that has 1 arm with a care-management interventionthat includes MTM services and another arm that has the care-management intervention withoutMTM services could be included. In contrast, a study that includes a care-managementintervention with MTM in 1 arm and usual medical care (no care-management intervention) in theother arm would not be included.Relevant factors:• Implementation features that may influence intervention effectiveness include the following:o Mode of delivery: telephone, face to face, virtual (Web/online/Internet), and remote videoo Type of professional providing initial and followup MTM service: pharmacist, nurse,physician, other cliniciano Frequency and interval of followup for MTM serviceso Specific MTM components usedo Fidelity in implementing MTM components: extent to which services were delivered asdesigned or intendedo Establishing and communicating goals of drug therapy to patients and among care providerso Method of identifying patients for enrollment (e.g., population health data, provider referralfor services, enrollment during a transition in care, targeting highly activated patients,targeting patients at time of high risk for event such as when prescribing a new drug)o Level of integration of MTM with usual care, which includes access to real-time clinicalinformation and laboratory values, and regular and consistent communication amongprescribers and others providing MTM serviceso Reimbursement characteristics (e.g., who is paying for cost of MTM services, who isreimbursed for MTM services, whether services are separately reimbursable)o Health system characteristics (e.g., are services being provided within an accountable careorganization, patient-centered medical home, or some other unique system setting, such asthe Veterans Health Administration, the Indian Health Service, non-U.S. single-payersystem)ES-7


Table A. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS CriteriaComparators Inclusion criteria:• Usual care, as defined by the studies• Different bundles of MTM services (e.g., 5 components vs. 3 components)• Same MTM services provided by different health care professionals (e.g., pharmacist vs.physician or nurse)• Same bundle of MTM services delivered by different modes (e.g., telephone vs. in person)• Same bundle of MTM services provided at different intensities, frequencies, or level of integrationwith prescribersOutcomes Inclusion criteria:• Studies must report at least 1 eligible outcomeo Intermediate outcomes- Disease-specific laboratory or biometric outcomes (e.g., hemoglobin A1c; blood pressure;total, low-density lipoprotein, or high-density lipoprotein cholesterol; pulmonary function; renalfunction; left ventricular ejection fraction; or other laboratory or biometric outcome specific todiseases covered)- Drug therapy problems identified as defined by primary studies but typically including thefollowing: medications being taken but not indicated; medications indicated but notprescribed; patient adherence issues; supratherapeutic doses; subtherapeutic doses; generic,formulary, or therapeutic substitution issue; complex regimen that can be simplified with sametherapeutic benefit; and potential for drug-drug interactions or adverse event- Drug therapy problems that are resolved as defined by primary studies but typically includingthe following: needed drug initiated; unnecessary drug discontinued; change in drug dose,form, or frequency; or generic, formulary, or therapeutic substitution- Medication adherence- Goals of therapy met- Patient engagement (e.g., initial and continuing patient participation in the MTM program)o Patient-centered outcomes- Disease-specific morbidity, including falls and fall-related morbidity, and outcomes specific tothe patient’s underlying chronic conditions (e.g., PHQ9, disease-specific symptoms, reducednumber of disease-specific acute exacerbations or events)- Disease-specific or all-cause mortality, including fall-related mortality- Reduced (actual) adverse drug events (frequency and/or severity)- Health-related quality of life as measured by generally accepted generic health-relatedquality-of-life measures (e.g., short-form questionnaires, EuroQOL) or disease-specificmeasures- Activities of daily living as measured by generally accepted standardized measures of basicand/or instrumental activities of daily living (e.g., Katz, Lawton, or Bristol instruments) or withinstruments that have demonstrated validity and reliability- Patient satisfaction with MTM care- Work or school absenteeism- Patient and caregiver participation in medical care and decisionmakingo Resource utilization- Prescription drug costs and appropriate prescription drug expenditures- Other health care costs- Health care utilization (hospitalizations, emergency department visits, and physician officevisits)o Harms- Care fragmentation- Patient confusion- Patient decisional conflict- Patient anxiety- Increased adverse drug events- Patient dissatisfaction with care- Prescriber confusion- Prescriber dissatisfactionES-8


Table A. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS CriteriaTiming Inclusion criteria:• Interventions should have at least 2 separately identifiable episodes of patient-directed MTMservices, with any interval of time in between episodes• For studies that report outcomes at different points in time, we considered only outcomesmeasured after the second episode of care and used the longest term outcomes reported by thestudyExclusion criteria:• Interventions designed as single-episode interventions without any provision for followup andmonitoringSetting Inclusion criteria:• Ambulatory settings (e.g., outpatient clinics or private physician offices), long-term care (e.g.,assisted living) settings if residents have control over medication self-administration, or retailpharmacy settings• Home setting• Interventions conducted in the United States• Interventions conducted in non-U.S. countries published in EnglishExclusion criteria:• MTM services that are delivered exclusively in inpatient settings• MTM services delivered at the time of hospital discharge or shortly after (e.g., within a few weeks)Relevant factors:• The MTM intervention itself may be delivered by home visits, by telephone, via the Web, or inother non–face-to-face modalities, such as video teleconferencingMTM = medication therapy management; PICOTS = populations, interventions, comparators, outcomes, timing, and setting;PHQ9 = Patient Health Questionnaire 9.MethodsTopic Refinement and Review ProtocolThe topic of this report and preliminary KQs arose through a nomination from the PharmacyQuality Alliance. Key Informants representing several clinical and scientific disciplines providedinput on the initial KQs; we revised them as needed. An initial draft of the revised KQs wasposted for public comment from March 6, 2013, through April 2, 2013, on the Agency forHealthcare Research and Quality (AHRQ) Effective Health Care Program Web site. We receivedcomments from 23 professional organizations and individuals and further revised KQs asappropriate.Literature Search and Identification StrategyTo identify articles relevant to each KQ, we began with a focused MEDLINE ® search forMTM interventions using a combination of medical subject headings and title and abstractkeywords, and limiting the search to English-language and human-only studies (inceptionthrough January 9, 2014). We also searched the Cochrane Library (inception through January 10,2014) and the International Pharmaceutical Abstracts database (inception through January 10,2014) using analogous search terms. We selected these databases based on preliminary searchesand consultation with content experts. We conducted quality checks to ensure that the searchesidentified known studies (i.e., studies identified during topic nomination and refinement). Basedon these quality checks, we revised and ran additional searches (specifically, using keywordssuch as “drug therapy management,” “drug therapy problem,” and “medications management”)to avoid missing articles that might prove eligible for this systematic review.ES-9


In addition, we searched the gray literature for unpublished studies relevant to this reviewand included studies that met all the inclusion criteria and contained enough methodologicalinformation to assess risk of bias. Specifically, sources of gray literature includedClinicalTrials.gov, the World Health Organization’s International Clinical Trials RegistryPlatform, Health Services Research Projects in Progress (HSRProj), the National Institutes ofHealth Research Portfolio Online Reporting Tools, the Database of Promoting HealthEffectiveness Reviews, the New York Academy of Medicine Grey Literature Report, andCMS.gov. In addition, we reviewed the yield from AHRQ’s request for Scientific InformationPackets in the Federal Register, posted for 30 days from September 16, 2013 onward.We reviewed our search strategy with an independent information specialist and theTechnical Expert Panel, and supplemented it according to their recommendations. In addition, toavoid retrieval bias, we manually searched the reference lists of landmark studies andbackground articles on this topic to identify any relevant citations that our electronic searchesmight have missed.Two trained members of the research team independently reviewed each of the titles andabstracts against the inclusion/exclusion criteria listed in Table A. We applied the same criteriato systematic reviews and primary studies. Each article that either or both reviewers chose toinclude based on the abstract review underwent full-text review. Two reviewers reviewed thefull text for eligibility against our inclusion/exclusion criteria. During full-text review, if bothreviewers agreed that a study did not meet the eligibility criteria (including designation of highrisk of bias), we excluded the study. Reviewers resolved conflicts by discussion and consensus orby consulting a third member of the review team.For studies that met our inclusion criteria, a trained reviewer abstracted information intostructured evidence tables; a second senior member of the team reviewed all data abstractions forcompleteness and accuracy. Reviewers resolved conflicts by discussion and consensus or byconsulting a third member of the review team.Assessment of Risk of Bias of Individual StudiesTo assess the risk of bias of individual studies, we used predefined criteria developed byAHRQ. 18 For randomized controlled trials (RCTs), we relied on the risk-of-bias tool developedby the Cochrane Collaboration. 19 We assessed the risk of bias of observational studies using anitem bank developed by RTI International. 20In general terms, results of a study with low risk of bias are considered valid. Studies markedlow risk of bias did not have any major flaws in design or execution. A study with medium riskof bias is susceptible to some bias but probably not sufficient to invalidate its results. A studywith high risk of bias has significant methodological flaws (e.g., stemming from serious errors indesign or analysis) that may invalidate its results. Primary concerns for our review includedselection bias, confounding, performance bias, detection bias, and attrition bias. Very highattrition rates, particularly when coupled with a failure to control for confounding or to conductintention-to-treat analyses, resulted in a rating of high risk of bias for trials and prospectivecohort studies. Likewise, we rated studies with an inherently high risk of confounding in design(e.g., observational studies comparing refusers vs. acceptors of MTM interventions) as high riskof bias if they failed to address confounding through design (e.g., matching) or analysis (e.g.,regression). Specifically, we evaluated trials on the adequacy of randomization, allocationconcealment, similarity of groups at baseline, masking, attrition, whether intention-to-treatanalysis was used, method of handling dropouts and missing data, validity and reliability ofES-10


outcome measures, and treatment fidelity. For observational studies, we did not assess adequacyof randomization or allocation concealment but assessed for confounding. We also evaluatedtrials for confounding due to randomization failure through biased selection or attrition. In otherwords, we evaluated trials with potential randomization failure for the same risks of bias asobservational studies.We excluded studies that we deemed at high risk of bias from our main data synthesis andmain analyses. We included them for sensitivity analyses; in cases when we had no otheravailable or credible evidence, we included in the report a brief synopsis of studies assessed ashigh risk of bias.Data SynthesisWhen we found three or more similar studies for a comparison of interest, we conductedmeta-analysis of the data from those studies using Comprehensive Meta-Analysis software(Biostat, Inc, Englewood, NJ). For all analyses, we used random-effects models to estimatepooled or comparative effects. To determine whether quantitative analyses were appropriate, weassessed the clinical and methodological heterogeneity of the studies under considerationfollowing established guidance; 21 that is, we qualitatively assessed the PICOTS of the includedstudies, looking for similarities and differences. When we conducted quantitative syntheses (i.e.,meta-analysis), we assessed statistical heterogeneity in effects between studies by calculating thechi-squared statistic and the I 2 statistic (the proportion of variation in study estimates attributableto heterogeneity). The importance of the observed value of I 2 depends on the magnitude anddirection of effects and on the strength of evidence for heterogeneity (e.g., the p-value from thechi-squared test or a confidence interval for I 2 ). Where relevant, we examined potential sourcesof heterogeneity using sensitivity analysis.When quantitative analyses were not appropriate (e.g., because of heterogeneity, insufficientnumbers of similar studies, or insufficiency or variation in outcome reporting), we synthesizedthe data qualitatively. Whenever possible, we computed confidence intervals for individualoutcomes.Numerous articles did not provide complete information about findings (e.g., 95%confidence intervals, statistical significance values, or between-group data). In many cases,therefore, we had to calculate odds ratios, mean differences or standardized mean differences, therelevant 95-percent confidence intervals, and p-values.Grading Strength of Evidence for Individual Comparisons andOutcomesWe graded the strength of evidence based on the guidance established for the AHRQEvidence-based Practice Center program. 22 Developed to grade the overall strength of a body ofevidence, this approach incorporates four key domains: study limitations (includes study designand aggregate quality), consistency (similar magnitude and direction of effect), directness(evidence links interventions directly to outcome of interest for the review), and precision of theevidence (degree of certainty surrounding an effect estimate based on sample size and number ofevents). In addition, the evidence may be rated as lower strength for bodies of evidence withsuspected reporting bias from publication, selective outcome reporting, or selective analysisreporting. Regardless of the specific risk of bias of observational studies, this approach tograding the evidence assigns observational studies a grade of high for study limitations, whichES-11


then leads to low strength of evidence. The strength of evidence from observational studies canbe rated as higher for scenarios such as a strong dose-response association, plausibleconfounding that would decrease the observed effect, and a high strength of association(magnitude of effect). We evaluated optimal information size criteria to make judgments aboutprecision based on guidance from Guyatt and colleagues 23 and based our grades on RCTs withlow or medium risk of bias or on observational studies unless none were available, based onguidance from the Grading of Recommendations Assessment, Development and Evaluation(GRADE) Working Group 24 and the AHRQ Evidence-based Practice Center program. 22Table B describes the grades of evidence that can be assigned. 25 Grades reflect the strength ofthe body of evidence to answer the KQs on the overall effectiveness, comparative effectiveness,and harms of the interventions examined in this review. Two reviewers assessed each domain foreach major outcome, and resolved any differences by consensus discussion or referral to a third,senior member of the team. We graded the strength of evidence for the outcomes deemed to beof greatest importance to decisionmakers and those commonly reported in the literature; we didnot grade the strength of evidence for KQ 1 (on components and features of MTM services). Thegrades shown in Table B describe the state of evidence (which may demonstrate benefit, harm,or no effect) and the confidence in the stability of that state. An insufficient grade is not astatement about lack of efficacy or effectiveness; rather it is a statement about the lack ofconvincing evidence on benefit, harm, or lack of effect.Table B. Definitions of the grades of overall strength of evidenceGradeDefinitionHighHigh confidence that the evidence reflects the true effect. Further research is very unlikelyto change our confidence in the estimate of effect.Moderate Moderate confidence that the evidence reflects the true effect. Further research maychange our confidence in the estimate of the effect and may change the estimate.LowLow confidence that the evidence reflects the true effect. Further research is likely tochange our confidence in the estimate of the effect and is likely to change the estimate.Insufficient Evidence either is unavailable or does not permit estimation of an effect.Assessing ApplicabilityWe assessed applicability of the evidence following guidance from the “Methods Guide forEffectiveness and Comparative Effectiveness Reviews.” 26 We used the PICOTS framework toexplore factors that affect applicability. Some factors identified a priori that may limit theapplicability of evidence include the following: age and health status of enrolled populations,health insurance coverage and access to health care, and complexity and intensity of the MTMintervention.ResultsWe provide a summary of results by KQ below. Detailed descriptions of included studies,key points, detailed synthesis, summary tables, and expanded strength-of-evidence tables thatinclude the magnitude of effect can be found in the full report. Our summary of results belowpresents the strength-of-evidence grades.ES-12


Results of Literature SearchesFigure B presents our literature search results through January 9, 2014. We identified 2,516unduplicated citations. In addition, we identified 233 publications through gray literaturesearches, suggestions from Technical Experts or public comments received during topicrefinement, hand searches of included studies, and Scientific Information Packets. After applyingour eligibility and exclusion criteria to titles and abstracts of all 2,749 identified citations, weobtained full-text copies of 419 published articles. We reapplied our inclusion criteria andexcluded 358 of these articles from further review before doing the risk-of-bias assessment. The61 articles included after full-text review represent 44 studies.Figure B. Disposition of articles on medication therapy management (PRISMA figure)Number of records found through databasesearching after duplicates removed2,516MEDLINE ® : 1,709IPA: 508Cochrane Library: 299Number of additional records identified throughother sources233Hand searches of references: 90Gray literature: 7Recommended by TEP or public comment: 5Scientific Information Packets 131Total number of records after duplicatesremoved2,749Number of records screened2,749Number of records excluded2,330Number of full-text articles assessed foreligibility419Number of articles (studies) included inqualitative synthesis of systematic review61 (44)Number of studies included in quantitativesynthesis of systematic review6Number of full-text articlesexcluded, with reasons358Ineligible publication type 75Ineligible population 8Ineligible intervention 169Ineligible design 69Ineligible comparator 8Ineligible outcomes 10Ineligible setting 14Insufficient information to 5determine eligibilityIPA = International Pharmaceutical Abstracts; PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses; TEP = Technical Expert Panel.This evidence base consisted of 44 studies (21 RCTs, 4 controlled clinical trials, and 19cohort studies) reported in 61 articles. Most RCTs compared an MTM intervention with usualES-13


care rather than with a different active intervention; all observational studies were cohort studies.Numerous studies had methods problems that led us to rate them as having a medium or highrisk of bias; only a few studies were of low risk of bias. When possible (enough studies similar inintervention, populations, and outcomes measured), we conducted meta-analyses of data fromRCTs or cohort studies separately; when relevant, we did two sets of meta-analyses, one withand one without the trials that had high risk of bias.Because of the wide variation in types of interventions classified as MTM, we first catalogedintervention components and implementation features of MTM interventions (KQ 1). We thenevaluated the effect of MTM on intermediate, patient-centered, and resource utilization outcomes(KQ 2). We also reviewed the evidence to identify how these effects might vary by specificintervention components and features (KQ 3) and patient characteristics (KQ 4). Finally, wereviewed the evidence on harms associated with MTM (KQ 5).Below, we summarize the main findings and strength of evidence, where applicable. We thendiscuss the findings in relationship to what is already known, applicability of the findings,implications for decisionmaking, limitations, research gaps, and conclusions.Key Findings and Strength of EvidenceKQ 1: Intervention Components and Implementation FeaturesOf the 44 included studies, over three-quarters were broadly focused MTM interventionswith patients who had a wide-ranging collection of conditions; the remaining studies werenarrowly focused MTM interventions with patients who had a specific condition. All studiesused a pharmacist as the interventionist. Services were provided face to face in half of includedstudies. Included studies provided interventions in a variety of clinical settings, includingcommunity pharmacies, centralized pharmacies or pharmacy call centers, and outpatient medicalclinics, and some used home visits. Half of the narrowly focused interventions were deliveredexclusively in an outpatient medical clinic.Whether they used the term “pharmaceutical care” or “MTM,” studies did not describeintervention components and features in a consistent manner or in sufficient detail. Thesedrawbacks were especially prevalent for intervention intensity and frequency of followup,method of patient enrollment for services, level of integration with usual care, andreimbursement characteristics for rendered MTM services. KQ 1 was descriptive in nature, so wedid not grade strength of evidence.KQ 2: Overall Effectiveness of MTMOf the 44 studies included in this review, we rated 16 as high risk of bias overall; that is,concerns about randomization failure, confounding, or overall attrition increased the risk of biasfor all outcomes. In addition, we rated some studies that were otherwise of low or medium riskof bias as high for individual outcomes, chiefly because of measurement or detection bias relatedto the specific outcome. These instances are specified in the relevant results section in the fullreport.We rated the strength of evidence for each outcome from studies with low or medium risk ofbias when available. MTM significantly improved objective measures of medication adherence,medication appropriateness assessed in general, and medication dosing (Table C). However, wedid not find evidence of benefit for any other intermediate outcomes on which we had data. Nostudies addressed either goals of therapy or patient engagement.ES-14


Table C. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventionsStudy Design:Intermediate No. Studies (N Strength of SupportingFindings and Direction of EffectOutcome Patients Evidence JudgmentAnalyzed)Anticoagulation RCT: 1 (10) Insufficient Medium study Therapeutic INR achieved, 100% vs.limitations, consistency 16.7%; p = 0.048.unknown (singlestudy), direct,impreciseHbA1c RCT: 2 (102) Insufficient Medium studylimitations,inconsistent, direct,impreciseCohort: 2 (2,688) Insufficient High study limitations,inconsistent, direct,impreciseLow-densitylipoproteincholesterolOne trial with significantly greaterpercentage of patients with HbA1c


Table C. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventions (continued)Study Design:Intermediate No. Studies (N Strength of SupportingFindings and Direction of EffectOutcomePatients Evidence JudgmentAnalyzed)Medication adherencemeasured asproportion adherent toa thresholdRCT: 1 (69) Insufficient Medium studylimitations,consistencyunknown, direct,preciseCohort: 2 (224 to200,722)Low for benefitHigh studylimitations,inconsistent,direct, precise100% of intervention patients and88.9% of controls were adherent;p = 0.115.Two studies with findings in oppositedirection; larger study showing rangeof ORs for medication-specificadherence depending onmedication.For comparison of PDP vs. controls,ORs ranged from 0.99 to 1.43; 95%CIs ranged from (0.90, 1.08) to(1.26, 1.62).For comparison of MA-PD vs.controls ORs ranged from 1.10 to1.40; 95% CIs ranged from (0.83,1.24) to (1.29, 1.52).Medication adherencemeasured aspercentage ofprescribed dosestakenMedication adherenceusing self-reportmeasuresMedicationadherence,miscellaneousmeasuresMedicationAppropriatenessGeneral Index ScoresCohort: 2 (120 to4,500)Low for benefit foradherence totreatment forhypertension anddyslipidemiaHigh studylimitations,inconsistent,direct, impreciseInsufficient fortreatment ofpatients withdiabetes,depression, andasthmaRCT: 1 (292) Insufficient Medium studylimitations,consistencyunknown, direct,impreciseRCT: 2 (365) Insufficient Medium studylimitations,inconsistent,direct, impreciseRCT: 1 (208) Low for benefit Low studylimitations,consistencyunknown, direct,preciseFor clinic-based MTM vs. usual carefor adherence to aspirin, odds ofadherence ranged from 5.981 (95%CI, 0.284 to 126.030; p = 0.250)during the intervention to 1.17 1 yearafter the intervention (95% CI, 0.072to 18.903; p = 0.912).Calculated mean difference fromsmall study, -0.040; 95% CI, -0.101to 0.021; p = 0.201.Larger study found a small(difference in adherence ~4.6%) butstatistically significant effect of MTMon adherence to medications forsome (2 of 5) conditions but nosignificant effect for the otherconditions.Calculated mean difference, 0.090;95% CI, -0.076 to 0.256; p = 0.289.Two studies with opposite directionof effect, both with nonsignificantdifferences between groups.Improvement in MTM group fromscore of 17.7 to 13.4 at 3 monthsand 12.8 at 12 months.ES-16


Table C. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventions (continued)Study Design:IntermediateStrength of Supporting Findings and Direction ofNo. Studies (NOutcomeEvidence JudgmentEffectPatients Analyzed)Medication-specificappropriatenessRCT: 2(261)Medication dosing RCT: 1(56)InsufficientLow forbenefitMedium study limitations,inconsistent, direct, impreciseMedium study limitations,consistency unknown,indirect, preciseSignificant improvement inappropriateness in the MTM group forsome medications but not others.Mean difference, -2.2 doses; calculated95% CI, -3.738 to-0.662.Goals of therapy 0 NA NA NAPatient engagement 0 NA NA NABP = blood pressure; CI = confidence interval; HbA1c = hemoglobin A1c; INR = International Normalized Ratio; LDL = lowdensitylipoprotein; MA-PD = Medicare Advantage Part D Plan; MTM = medication therapy management; NA = not applicable;OR = odds ratio; PDP = Medicare Part D Plan; RCT = randomized controlled trial.Similarly, we did not have evidence of benefit for most patient-centered outcomes, includingadverse drug events or mortality (Table D). MTM did not improve most measures of healthrelatedquality of life (low strength of evidence for no benefit). We graded the “vitality” and“emotional role functioning” domains of the Medical Outcomes Study Short-Form (SF36)questionnaire as insufficient for this domain. For the SF-36, neither the other six domains nor thetwo component scores (physical health, mental health) showed significant benefit from MTMinterventions. The various patient satisfaction items also showed no impact from MTMprograms (low strength of evidence for no benefit). We found no evidence on activities of dailyliving, work or school absenteeism, and patient and caregiver participation in medical care anddecisionmaking.Table D. Summary of findings and strength of evidence for patient-centered outcomes of MTMinterventionsPatient- Study Design:Strength ofFindings and Direction ofCentered No. Studies (NSupporting JudgmentEvidenceEffectOutcome Patients Analyzed)Adverse drugeventsCognitive andphysical functionRCT: 2 (806) Insufficient Medium study limitations,inconsistent, direct, impreciseRCT: 1 (133) Insufficient Medium study limitations,consistency unknown, direct,impreciseAffective function RCT: 2 (181) Insufficient Medium study limitations,inconsistent, direct, impreciseDirection and magnitude ofeffect differs between the 2trials.No significant differencesbetween arms.One study with no significantcalculated mean differencesin depression or anxietyscores; the other study withsignificant differences inmean depression and anxietyscores, but no significantdifference in percentageachieving a depressionremission.ES-17


Table D. Summary of findings and strength of evidence for patient-centered outcomes of MTMinterventions (continued)Study Design:Patient-No. Studies (N Strength ofFindings and DirectionCenteredSupporting JudgmentPatientsEvidenceof EffectOutcomeAnalyzed)Mortality RCT: 1 (181) Insufficient Medium study limitationsconsistency unknown, direct,impreciseCohort: 2(173,329)InsufficientGastrointestinalbleeding eventsCohort: 1(unclear)InsufficientGeneral healthrelatedRCT: 3 (1,169) Low for noquality of lifebenefitdomains other thanvitality and emotionalrole functioningGeneral healthrelatedquality of lifefor vitality andemotional rolefunctioning domainCondition-specifichealth-related qualityof life (diabetes)Patient satisfactionRCT: 3 (1,169) InsufficientHigh study limitations, inconsistent(magnitude), direct, impreciseHigh study limitations, consistencyunknown, direct, impreciseMedium study limitations;consistent for physical rolefunctioning, general healthperceptions, and social functioningdomains; inconsistent for physicalfunctioning, bodily pain, and mentalhealth domains; direct; preciseMedium study limitations,consistent, direct, imprecise (notcorrected for multiple comparisonsor wide CIs)RCT: 1 (73) Insufficient Medium study limitations,consistency unknown, direct,imprecise,RCT: 3 (1,463) Low for nobenefitMedium study limitations,consistent, direct, preciseOR, 0.59; 95% CI, 0.12 to2.49; p = 0.48.One study: OR, 0.5;95% CI, 0.3 to 0.9. Secondstudy: adjusted HR, 0.92;95% CI, 0.87 to 0.96;p < 0.001.RRR, 60%; p = 0.001.Variable mean differencewith CIs consistentlyspanning the null effect.Vitality: Mean difference of2.797; 95% CI, 0.655 to4.939; p = 0.010.Emotional role functioning:Mean difference of 5.386;95% CI, -7.244 to 18.013.Nonsignificant improvementof 0.1 point on a 5-pointscale in the interventiongroup compared with nochange in the control group.No differences on 17 of 21items of patient satisfaction;4 statistically significantdifferences ranged inmagnitude from -0.15 to -0.36, favoring MTM.Activities of daily 0 NA NA NAlivingWork or school 0 NA NA NAabsenteeismPatient and caregiver 0 NA NA NAparticipation inmedical care anddecisionmakingCI = confidence interval; HR = hazard ratio; MTM = medication therapy management; NA = not applicable; OR = odds ratio;RCT = randomized controlled trial; RRR = relative risk ratio.Outcomes related to using health resources were also not much influenced by MTMinterventions (Table E). Two exceptions may merit attention: (1) health plan expenditures onmedication costs and (2) the proportion and costs of hospitalization for patients with diabetes.MTM trials implemented in settings with a broad range of patients did not show a consistentsignal of reduction in the number of hospitalizations, but a single cohort study that partiallyaddressed confounding inherent in studies of refusers and acceptors found a lower mean numberES-18


of inpatient visits for patients accepting MTM compared with patients refusing MTM. Overall,we judge the strength of evidence for this outcome to be insufficient owing to lack of consistencyacross studies.Table E. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventionsStudy Design:Resource-UtilizationStrength ofStudy Findings andNo. Studies (NSupporting JudgmentOutcomeEvidenceDirection of EffectPatients Analyzed)Use of generics Cohort: 1 (63,198 to Insufficient High study limitations, consistency Odds range from200,722)unknown, direct, imprecise -0.01 to 0.006.Medication costs:patient copaymentsMedication costs:health planexpendituresMedication costs:total outlaysMedication costs:medication costs plusother expendituresNumber of outpatientvisitsRCT: 1 (NR) Insufficient Medium study limitations,consistency unknown, indirect,precision cannot be determinedCohort: 1 (1,606) Insufficient High study limitations, consistencyunknown, indirect, preciseRCT: 3 (965)NRCT and cohort: 5(120 to 200,722)Low forbenefitInsufficientMedium study limitations,consistent, indirect, impreciseHigh study limitations,inconsistent, indirect, impreciseRCT: 6 (2,636) Insufficient Medium study limitations,inconsistent, indirect, impreciseCohort: 2 (177,565) InsufficientHigh study limitations,inconsistent, indirect, impreciseRCT: 2 (996) Insufficient Medium study limitations,inconsistent, indirect, impreciseNRCT and cohort: 3(5,300)InsufficientHigh study limitations,inconsistent, indirect, impreciseRCT: 3 (2,208) Insufficient Medium study limitations,inconsistent, indirect, preciseCohort: 1 (4,500) Insufficient High study limitations, consistencyunknown, indirect, impreciseCalculated meandifference, -64 USD;variance notcalculable.Calculated meandifference for MTMvs. same-countrycontrol, 80.40 USD;95% CI, 10.43 to150.37; p = 0.024.Calculated meandifference for MTMvs. different countrycontrol, 88.60 USD;95% CI, 24.61 to152.59; p = 0.007.Mean differencevaries from -34 CADto -293 USD over 6months.Mean differencevaries from -800 USDover 1 year to 425USD over 2 years.Mean differencevaries from -20.16USD to +5.25 USDper month.Mean differencevaries from -563 USDto +310 USDannually.Differences in meancosts range from -8.1CAD to 1,947 USD.Differences in meancosts range from-1,039 to 1,100 USD.Standardized meandifference, 0.049;95% CI, -0.034 to0.133; p = 0.247; I 2 =0.Calculated meandifference, 2.48; 95%CI, 1.674 to 3.286; p


Table E. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventions (continued)Study Design:Resource-UtilizationStrength ofStudy Findings andNo. Studies (NSupporting JudgmentOutcomeEvidenceDirection of EffectPatients Analyzed)Outpatient costs RCT: 3 (2,050) Insufficient Medium study limitations,inconsistent, indirect, impreciseVariable estimates.Number of laboratorytestsCosts of laboratorytestsNumber ofemergencydepartment visitsCosts of emergencydepartment visitsHospitalization:numberRCT: 2 (1,842) Insufficient Medium study limitations,inconsistent, indirect, impreciseRCT: 3 (2,050) Insufficient Medium study limitations,inconsistent, indirect, impreciseRCT: 3 (1,552) Insufficient Medium study limitations,inconsistent, direct, impreciseObservational: 3(795 to 200,722)InsufficientHigh study limitations,inconsistent, direct, impreciseRCT: 2 (996) Insufficient Medium study limitations,consistent, direct, impreciseCohort: 1 (150,470to 200,722)RCT: 3 (2,208)Cohort: 1 (4,500)InsufficientLow for nobenefitLow forbenefitHigh study limitations, consistencyunknown, direct, impreciseMedium study limitations,consistent, direct, preciseHigh study limitations, consistencyunknown, direct, preciseHospitalization: risk RCT: 1 (556) Insufficient Low study limitations, consistencyunknown, direct, impreciseDifferences rangefrom +0.15 to -1.6tests.Differences rangefrom +15 CAD to -140USD.Mean differenceranges from -0.7 (pnot significant) to -0.03 (95% CI, -0.113to 0.053).Adjusted OR rangesfrom 0.89 (95% CI,0.6 to 1.3) to 1.09;mean difference (1study), 0.04; 95% CI,-0.043 to 0.123; p =0.346.Mean differenceranges from -52 USDto -5.6 CAD.Difference rangesfrom -16 USD to+12.8 USD.Mean difference,0.037; 95% CI,-0.004 to 0.080.Mean difference, -0.21; 95% CI, -0.265to -0.155; p


Table E. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventions (continued)Study Design:Resource-UtilizationStrength ofStudy Findings andNo. Studies (NSupporting JudgmentOutcomeEvidenceDirection of EffectPatients Analyzed)Hospitalization: rate(patients with heartfailure and homemedicine review)Costs ofhospitalizationCohort: 1 (5,717)Low forbenefitHigh study limitations, consistencyunknown, direct, precise3; 2,151 (2,050) Insufficient Medium study limitations,inconsistent, direct, impreciseCHF or COPD: 1(169,099 to200,722)Diabetes: 1(150,470)Insufficient forCHF or COPDLow forbenefit fordiabetesHigh study limitations, consistencyunknown, direct, impreciseHigh study limitations, consistencyunknown, direct, preciseAdjusted HR, 0.55;95% CI, 0.39 to 0.77.Inconsistent directionof effect butconsistent in lack ofsignificant effect.Differences rangefrom -526 USD to 200USD for CHF andCOPD.Differences rangefrom -363 USD to -399 USD for diabetes.Length of hospital RCT: 1 (208) Insufficient Low study limitations, consistency MTM reduced lengthstayunknown, direct, imprecise of stay 1.8 days.CAD = Canadian dollar; CHF = congestive heart failure; CI = confidence interval; COPD = chronic obstructive pulmonarydisease; HR = hazard ratio; MTM = medication therapy management; NR = not reported; NRCT = nonrandomized controlledtrial; OR = odds ratio; RCT = randomized controlled trial; USD = U.S. dollar.Over all three categories of outcomes, each of which had a substantial number of individualmeasures, MTM improved outcomes in only a couple of instances. Study limitations, lack ofconsistency, and lack of precision of the estimates of effects limited the strength of evidenceconsiderably. As discussed later, even the minimal findings of effectiveness are at best onlynarrowly applicable.KQ 3: Effectiveness of MTM by Intervention FeaturesWe found evidence from one study each on five intervention features: (1) access ofpharmacists to patient records, 27 (2) intensity of care coordination and followup aftercomprehensive medication review, 28 (3) community pharmacy versus call center, 29 (4) level ofintensity of intervention, 30 and (5) type of payer (private vs. Medicaid). 31 With the exception ofthe study on pharmacists’ access to patient records, we rated these studies as high risk of bias.Evidence was insufficient for most outcomes for the first two intervention features, with thefollowing two exceptions. First, MTM delivered by community pharmacists increased theweighted generic dispensing ratio when compared with call-center pharmacists (low strength ofevidence). Second, enhanced MTM with pharmacists’ access to patient records reduced the meannumber of adverse drug events; this finding suggested benefit when compared with basic MTM(low strength of evidence). We found insufficient evidence for all outcomes for intensity ofintervention and type of payer.KQ 4: Effectiveness of MTM by Patient CharacteristicsWe did not identify any studies that analyzed outcomes of MTM by patient characteristics.KQ 5: Harms of MTM InterventionsLack of precision and the limitations of a single study with high risk of bias meant thatevidence was insufficient to judge whether MTM resulted in greater inconvenience 32,33 thanES-21


usual care. We found no evidence on other prespecified harms, specifically including carefragmentation, patient decisional conflict, patient anxiety, increased (actual) adverse drug events,prescriber confusion, and prescriber dissatisfaction.DiscussionKey FindingsWe included 44 eligible studies (21 randomized controlled trials, 4 controlled clinical trials,and 19 cohort studies) reported in 61 articles, described in detail in the full report (KQ 1).Evidence was insufficient on the effect of MTM on most outcomes (KQ 2). In a few instances,described below, the evidence led us to conclude with a low strength of evidence either a benefitor lack of benefit. Specifically, we found evidence that MTM results in improvement whencompared with usual care for some measures of medication adherence and appropriateness;medication dosing; health plan expenditures on medication costs; and, for patients with diabetes,the proportion and costs of hospitalization. Similarly, we conclude based on a low strength ofevidence that MTM conferred no benefit for patient satisfaction and most measures of healthrelatedquality of life.We found evidence on five intervention components and intervention features (KQ 3): onestudy provided information on each feature and yielded insufficient evidence for most outcomes,with the two following exceptions. MTM programs with pharmacist access to brief clinicalsummaries from the medical record reduced the mean number of adverse drug events whencompared with basic MTM programs without such access (low strength of evidence).Community pharmacists increased the generic dispensing ratio more than pharmacists based incall centers (low strength of evidence). We found no relevant studies on patient characteristicsmoderating the effect of MTM interventions (KQ 4). Similarly, the evidence on harms associatedwith MTM was limited to one study on inconvenience and was rated as insufficient (KQ 5).Findings in Relation to What Is Already KnownOur findings contrast with conclusions that Chisholm-Burns and colleagues reached in arecent systematic review. 34 In that review, the authors concluded on page 923: “Pharmacistprovideddirect patient care has favorable effects across various patient outcomes, health caresettings, and disease states.” 34 Several differences between the Chisholm-Burns review and thecurrent review may account for the discrepant conclusions. First, the Chisholm-Burns reviewincluded all studies that cited evidence of pharmacist involvement in direct patient care. Theinterventions examined included chronic disease management and prospective and retrospectivedrug utilization review; we excluded these types of efforts because of the clinical heterogeneitythose interventions would have introduced into the review. Notably, the Chisholm-Burns reviewdid not use the term “medication therapy management” to categorize the interventions in thearticles they reviewed. Second, approximately 30 percent of the studies in the Chisholm-Burnsreview were conducted entirely in institutional settings. In contrast, we did not identify anystudies within institutional settings that met our MTM intervention definition criteria. Third, theChisholm-Burns review included a total of 298 articles and did not omit from the analysesstudies with a high risk of bias; in contrast, we based our strength-of-evidence grades in thisreview on only those studies with no more than medium risk of bias. Thus, a direct comparisonof findings between these two reviews would be ill advised.ES-22


The striking differences between the conclusions reached in these two reviews emphasizetwo important needs for efforts to systematically review MTM programs. The first is forresearchers to specify the MTM intervention based on existing definitions, taxonomies, orservice models. The second is to develop consensus guidelines for describing interventionfeatures and fidelity of intervention delivery in publications reporting findings from evaluationstudies. Progress on these two steps would enable systematic reviews to differentiate betterbetween different types of services and avoid the problem of overgeneralizing review results.Applicability of the FindingsThis body of evidence has significant clinical and methodological heterogeneity, which limitsthe ability to make any universal statements about effectiveness. However, the range of studydesigns, which includes RCTs, nonrandomized trials, and cohort studies, enhances theapplicability of findings for real-world settings. 26 Included studies ranged from relatively smallinterventions in single clinics provided by a single interventionist to evaluations of MTMservices delivered on a large scale through integrated health systems or health plans as aMedicare Part D or other drug plan benefit. This diversity of studies enhanced the applicabilityof findings to a wide variety of settings, including outpatient clinics, community pharmacies, andcentralized pharmacy call centers. A few studies conducted outside the United States includedMTM as part of a home visits program; findings from this model may not be directly applicablewithin the United States.The studies in this review are broadly applicable to a range of chronically ill adult patientpopulations. A majority of interventions were directed at populations with multiple and commonchronic conditions, such as diabetes, chronic heart failure, and hypertension. Several specificallytargeted adults age 65 years or older. Few studies reported sociodemographic characteristicsbeyond age and sex; thus, the applicability of findings to specific populations (e.g., rural, lowsocioeconomic status, cognitively impaired, uninsured) is unknown. The nature of the MTMintervention, which includes involving patients as active participants in the process, limits theextent to which findings can be generalized beyond patients who agree to participate in suchinterventions. Patients who agree to participate may be systematically different from those whodecline to be in such a program. For that reason, the impact of such interventions at a populationor health-plan level may be limited by the degree of uptake among interested patients.The intervention used across most studies can be characterized as complex and moderatelyresource intensive. Components involve identifying applicable patients; initially assessingpatients; providing counseling, education, and care coordination; and following patients overtime. These services were provided per protocol in some studies and as needed or ad hoc inothers. Most studies described intervention components in terms of “pharmaceutical care model”components or Medicare Part D MTM program criteria, but few provided detailed descriptionsor measurement of implementation fidelity.The comparator arm in all studies was usual medical or pharmacy care. This does nottypically include distinct MTM services by health care providers other than prescribing providers(not common for the time period covered by most of the studies). Models of collaborative healthcare delivery are evolving, and the changing roles and training of pharmacists increase thepotential applicability of MTM interventions in future models of health care.The broad sets of outcomes evaluated across this body of evidence spanned a substantialrange of both intermediate and health outcomes as well as outcomes related to resource use.Proximal and intermediate outcomes included number of drugs, identification of drug therapyES-23


problems, appropriateness of medication prescribing, and laboratory or biometric markers ofdisease control (e.g., hypertension, hemoglobin A1c, low-density lipoprotein cholesterol).Patient-centered outcomes focused on numerous measures of quality of life as well as adversedrug events. Many studies also reported outcomes involving health care resource use andexpenditures (e.g., number and costs of hospitalizations, emergency department visits, outpatientvisits).Most studies did not, however, clearly indicate the expected, desired, or intended direction ofeffect on most resource use outcomes, making the applicability of using these interventions toreduce drug-related health care costs or expenditures difficult to assess. For example, it is notclear whether one should expect the number of medications prescribed for heart failure toincrease or decrease under the careful scrutiny of an MTM intervention because the desiredimpact will be based on the goal of therapy for each individual.The focus of outcome measurement in many studies was the short-term identification andcharacterization of drug therapy problems and their resolution; these endpoints are thought to bethe outcomes most sensitive to change as a result of receiving MTM services. However, becauseidentification of drug therapy problems is, by design, a part of the MTM intervention itself,differences between the nature of the intervention and that of the control programs mean thatmeasuring these outcomes cannot be as rigorous in a usual-care comparison group as it is in theintervention group. In fact, many studies were able to measure changes in this outcome only inthe intervention group. Hence, many studies failed to demonstrate a direct analytic link betweenthe resolution of drug therapy problems as a result of MTM and impact on intermediateoutcomes, patient-centered outcomes, and resource utilization. Thus, the applicability of studiesthat demonstrate an impact on the resolution of drug therapy problems is limited.Implications for Clinical Practice and PolicymakersAlthough we found the evidence insufficient in general to draw definitive conclusions aboutthe comparative effectiveness of MTM for most outcomes that we evaluated, our findingssuggest some implications for practice and policy. MTM is already in widespread practice and isnow shaped in the United States largely by Medicare Part D policy; this presents both challengesand opportunities. MTM programs sponsored and administered by Part D drug benefit plans areoften centrally administered and delivered primarily by phone, and may be less integrated intoroutine health care than some of the interventions included in our review. MTM programs of thefuture have the potential to be more integrated into routine health care through participation inaccountable care organizations or patient-centered medical home models. We were unable toanswer definitively whether level of integration matters for effectiveness, but policymakers mayneed to consider expectations about the impact that MTM might have on patient-centeredoutcomes and resource use in the context of other health care delivery transformation activitiesor quality improvement initiatives that are also occurring. More integration of MTM serviceswith other activities may be effective; however, the more integrated MTM becomes withinroutine medical care, the more difficult it becomes to isolate it as a discrete intervention forevaluation.Policymakers could thus consider whether MTM services should be positioned as acontributor to overall improvement in processes of care, health status, and costs or positioned asan intervention to which effects can be discretely attributed. As noted earlier, improvements inmedication appropriateness or drug therapy regimens may not always translate intoimprovements in health or costs, and even if they do, secular trends in related qualityES-24


improvement (e.g., medication adherence interventions, regulatory requirements for medicationreconciliation, meaningful use incentives for electronic health records) may make measuringoutcomes attributable to MTM very challenging.Future training of MTM providers would benefit from a better understanding of which MTMcomponents really matter. At the moment, such information is lacking. Policymakers and funderswho wish to understand the comparative effectiveness of different MTM components couldencourage rigorous program evaluation designs that fit within the context of the real-worldimplementation of these programs. For example, positive deviance analyses 35 with rigorousmeasurement of implementation features or stepped wedge trial designs 36 may be usefulapproaches.A typical approach for evaluating complex interventions is to identify the “core” componentsfor standardization, while allowing for flexibility for peripheral components or variations inimplementation. In complex practice-based innovations, such flexibility may reflect desirable (orunavoidable) adaptations to local circumstances. Policy governing MTM programs may warrantmodifications to permit investigators to conduct rigorous and innovative evaluative designs toidentify core components or effectiveness-enhancing modifications. As future research andevaluation elucidate these components or enhancements, policy will need to evolve to keep pacewith best practices.Finally, consideration of both patients’ and prescribers’ perspectives in future design anddelivery of MTM services may be needed. In our current analytic framework, MTMinterventions require a significant element of engagement by both patients and prescribers if theinterventions are to have a reasonable likelihood of improving outcomes. Although “opt in”strategies may increase the reach of such interventions, keeping patients (and their prescribingproviders) engaged in the intervention over a reasonable amount of time may be the key totranslating the potential of MTM interventions into actual improvements. Further refinement ofeligibility criteria based on evidence to provide interventions to those most at risk from drugrelatedproblems, and therefore most likely to benefit, may also be warranted.Limitations of the Comparative Effectiveness Review ProcessThe constraints for populations, interventions, and settings that we imposed on thissystematic review may limit its applicability, as discussed above. During topic refinement andbased on Technical Expert Panel inputs and public comment, we expanded the scope byremoving an exclusion criterion that would have required MTM interventions to have beendirected at a patient population with two or more chronic conditions. As a result, we includedstudies that focused on one chronic condition. Because of the prevalence of certain chronicconditions in the adult population, and particularly among Medicare beneficiaries, we think thisdecision was sensible and permitted us to examine a broader evidence base than would otherwisehave been the case.Although we tried to distinguish MTM from disease or case-management interventions,making this distinction was challenging. We created a threshold for the intervention componentsthat were required to be present for this distinction. Specifically, we elected to emphasizewhether the intervention entailed a comprehensive review of all medications; for that reason, wedid not constrain studies of interest to those that targeted a single medication or drug regimen orthat focused on a single condition such as diabetes or hypertension.When we were unable to determine which medications the interventionist had reviewed, wewrote to the authors for additional information. We chose to pursue authors in an effort to permitES-25


us to use studies that had been designed as MTM but did not describe the comprehensivemedication review component in detail.Our approach may have been overly inclusive because it led us to include studies thataddressed a single disease, as long as the pharmacist reviewed all medications. For example, 10of the 44 studies were relatively narrowly focused; 2 of these addressed patients with chronicheart failure and 2 addressed patients with either hypertension or hypertension and diabetes. Theremaining six studies focused on patients with diabetes, HIV, glucocorticoid-inducedosteoporosis, or hemodialysis. The fact that we did not require patients to have more than oneclinical condition resulted in an approach that was inclusive of these more narrowly focused(albeit often termed “MTM”) studies and may render our results less applicable to MTMinterventions targeted to patients with a wide range of chronic conditions.Also based on feedback during the process of setting out the scope of this review, we choseto include interventions that were broader than the Medicare Part D MTM-defined interventions.Put another way, we broadened our view of patient populations and intervention criteria, and weallowed studies not conducted in the United States into the evidence base. This decision led us toinclude interventions described as “pharmaceutical care,” which were generally based on thepharmaceutical care model principles; 9 it also permitted us to examine investigations withelements of pharmaceutical care or MTM that did not specifically label the intervention as eitherMTM or pharmaceutical care. These studies were often described as “clinical pharmacistinterventions.”Furthermore, all the non-U.S. studies involved interventions within single-payer healthsystems. Hence, the interventions in this review constitute a more heterogeneous group than ifwe had allowed only those labeled as Medicare Part D MTM programs. This is both a limitationand a strength. Although our approach makes results more challenging to interpret, it enhancesour ability not to miss interventions that include MTM components but lack the descriptor term“MTM.”Studies did not often explicitly describe certain MTM components. In cases when we couldnot determine whether investigators had provided certain MTM components (such as patienteducation and counseling or coordination with other health care providers), we contacted theauthors to gain additional information that would allow us to make an informed decision. Wewere fairly permissive in interpreting the presence of the MTM intervention components otherthan comprehensive medication review. The main reason is that we recognized that termsdescribing some components have evolved over time and may have been absent from the lexiconin earlier years or implicitly conveyed by authors by simply using the terms “MTM” or“pharmaceutical care” to describe their intervention.Our approach to categorizing interventions for KQ 1 relied primarily on the shortdescriptions in published manuscripts and those we were able to obtain via email inquiries. Theirsimilarities or differences substituted for any overarching taxonomy, because none that weconsidered seemed to fit our purpose. Thus, we have introduced intervention labels that,admittedly, do not fully describe or account for clinical heterogeneity among interventions. Thisapproach limits our ability to make definitive statements about the effectiveness of variousintervention components. We believe that the clusters and categorizations we used are usefulheuristics, but some may regard them more as hypothesis generating than as reflecting settledprinciples of classification.Finally, our search process was complicated by having to ensure coverage of all terms thatcould be used to describe MTM interventions over time. Adding to this challenge was our effortES-26


to examine the gray literature, where we thought we might find studies tilted towardeffectiveness and real-world program evaluation. As it turned out, studies of these types ofinterventions were not indexed similarly; for that reason, we needed to rely heavily on handsearches of citation lists from key background articles to identify possibly relevant studies forinclusion. Thus, we may have missed some studies that might have qualified for inclusion. Giventhe considerable diversity in the evidence base we did have, however, we do not think that anypotentially missed studies would have changed our conclusions in any material way. No metaanalysesincluded more than five studies; as a result, we did not examine included studies forpublication bias quantitatively.Limitations of the EvidenceAs a body of evidence, the MTM literature evaluated in this review has measured numerousoutcomes. As indicated in previous sections, very few outcomes, with the exception of harms,remain completely unexamined. Of the 44 studies in this review, we rated 28 as having medium,low, or mixed risk of bias. The 44 studies included 21 trials and 4 nonrandomized controlledstudies. In other words, the literature on this topic is not marked by failure to consider importantoutcomes, universally high risk of bias, or pervasively weak designs.Despite these advantages, we were unable to identify sufficient evidence on the majority ofhypothesized outcomes of MTM. In several instances, our inability to rate evidence as higherthan insufficient came from inconsistent and imprecise evidence or from a body of evidence withhigh study limitations. The choice of outcome measures in this body of evidence limited ourability to come to conclusions in some instances. For example, some studies did not focus onchanges that proponents might expect MTM services to produce. Because effective MTM caneither increase or decrease expenditures or use of services based on the needs of the patient,studies that did not prespecify the expected direction of change had no way to interpret theirresults as an appropriate change. Studies that demonstrated inconsistent results in direction ofchange (i.e., some showing an increase in resource use and others showing a decrease) may wellhave been consistent in terms of appropriate change, but because they generally failed toestablish a priori the direction in which they expected to find an effect, we rated such evidence asindirect and inconsistent.Similarly, studies often used nonstandardized or idiosyncratic measures for outcomes such asadverse events, adherence, and expenditures or costs; this tendency limited our ability to metaanalyzeresults. When studies focused on specific outcomes, they were often significantlyunderpowered to detect differences between groups (i.e., they did not meet optimal informationsize criteria). As a result, we rated several studies as imprecise.MTM intervention studies are largely practice based and incorporate substantialheterogeneity in specific intervention elements and in patient populations targeted. Yet theevidence is sharply constrained in its ability to inform questions about the effectiveness ofspecific MTM components or intervention features (KQ 3 in our review) because study designsdid not often capitalize on variants in MTM programs for a prospective evaluation of outcomesby those variants. Neither did they measure fidelity to intended MTM elements for post hocevaluation. Similarly, the relatively untargeted nature of the MTM interventions meant that, inmany studies, only small numbers of patients had any one specific condition, and most studiesdid not measure patient characteristics beyond age and sex, thus limiting our ability to addressKQ 4 in our review. For this reason, the evidence we identified for this review was most relevantfor KQ 2.ES-27


Research GapsIn many bodies of research, questions regarding the comparative effectiveness of specificintervention components or implementation features are best answered after clear evidence of theoverall effectiveness of the intervention relative to usual care has been established. Our reviewlargely indicates insufficient evidence on the primary question of effectiveness relative to usualcare. By definition, this limited what we could say about comparative effectiveness.Nonetheless, the widespread implementation of MTM coexists with the urgent need foractionable information for policy, program policies, and training. This clinical and policyenvironment means that new research cannot afford to address causal claims relative to usualcare first, followed by comparative effectiveness of the intervention elements in a relativelycontrolled environment, and finally, program evaluation of real-world implementation, all insequential order.In prioritizing among various research goals, therefore, funders may wish to consider therelative value of new evidence on overall effectiveness, effectiveness of implementation features,and program implementation and accountability. Trial research in narrow clinical settings canaddress questions of effectiveness but may lack applicability to real-world implementation.Likewise, evaluations of real-world programs with variable fidelity to interventions can answerquestions about process and implementation, but they offer limited information on effectiveness.Research prioritization exercises will also need to account for already commissioned MTMintervention studies.For new studies focusing on causal claims, a critical gap relates to the failure to specify theexpected direction of effect. New research requires a strong theoretical foundation to helpspecify causal mechanisms and hypothesized effects. Without such an edifice, future researchwill continue to produce inconsistent and uninterpretable results.Heightened attention to causal mechanisms will also help researchers convey theirunderstanding of what outcomes these types of interventions are likely to influence. For instance,how should researchers wishing to establish direct causal links between MTM programs andoutcomes evaluate distal outcomes such as patient-centered outcomes and resource utilization?This effort requires a better understanding of the relationship between proximal outcomes suchas “drug therapy problems identified and resolved” and distal outcomes. For instance, MTM mayreduce outpatient visits to address side effects. MTM may also result in the need for furthertesting and evaluation for some patients, which could, in turn, result in more rather than feweroutpatient visits. Unless the nature of change resulting from MTM is specified in relation togoals of drug therapy, studies cannot assert benefit or harm. Further, drug therapy problems arediverse and may not all have the same causal relationship to health, quality of life, patientsatisfaction, or resource use outcomes. Furthermore, a causal model of these distal outcomes mayneed to take into account the competing or complementary contributions of MTM, new modelsof health care delivery (e.g., patient-centered medical homes), and other quality improvementinterventions.Investigators embarking on new studies focusing on causal links between MTM andoutcomes may wish to consider the limitations of studies based on secondary data from existingMTM programs that use opt-in/opt-out patient enrollment mechanisms. Although these studiesmay provide invaluable information on process measures such as patient engagement, underlyingissues of confounding severely limit the validity of causal claims from such studies.Regardless of the goal of their future research, investigators should consider issues of samplesize to ensure precision of their results. This issue is particularly relevant when evaluatingES-28


outcomes likely to occur in smaller subgroups defined by patient risk, context, or highly adaptedintervention features. Innovative designs (e.g., stepped wedge trials, statistical process control,time-series analysis, simulations, and factorial experiments) may permit both rigor and adequatesample size within the context of real-world implementation. With careful attention to fidelity,new studies may also inform questions about the effectiveness of intervention components andimplementation features. Mixed-methods approaches may allow more information on variationsin context and implementation. Such designs may also help inform our understanding of criticaltraining elements for MTM service providers.Regarding research gaps for specific outcomes such as patient satisfaction, measures specificto the types of services provided through MTM (e.g., patient education about medications) or tothe proximal outcomes that MTM is intended to achieve (e.g., reduced medication side effects,improved disease control) may offer better insights into the effects of MTM. Similarly, amedication-related instrument may better measure patients’ concerns that are directly related tomedication use (e.g., experience of side effects, intrusiveness of the medication regimen) thangeneric tools do.ConclusionsThe evidence base offers low evidence of benefit for a limited number of intermediate andhealth utilization outcomes.We graded the evidence as insufficient for most outcomes because ofinconsistency in direction, magnitude, and precision, rather than lack of evidence. Widevariations in populations and interventions, both within and across studies, likely explain theseinconsistencies. Given the widespread implementation of MTM and urgent need for actionableinformation, optimal investments in new research require a process of research prioritization inwhich the value of information from each proposed study is carefully considered. Studiesdesigned to identify causal relationships between MTM interventions and their outcomes requireadequate controls for confounding but may offer limited information on which factors explainprogram success or failure. Studies designed to explore the reasons for program success orfailure using qualitative or single-arm designs may offer hypothesis-generating rather thanhypothesis-confirming insights on MTM effectiveness. New research, regardless of specificfocus, will likely continue to find inconsistent results until underlying sources of heterogeneityare accounted for.ES-29


References1. Adams K, Corrigan E. Priority areas for qualityimprovement. Washington, DC: Institute ofMedicine of the National Academies, TheNational Academies Press; 2003.2. Kohn LT, Corrigan J, Donaldson MS. To Err IsHuman: Building A Safer Health System.Washington, DC: National Academy Press;2000.3. McGlynn EA, Asch SM, Adams J, et al. Thequality of health care delivered to adults in theUnited States. N Engl J Med. 2003 Jun26;348(26):2635-45. Epub: 2003/06/27. PMID:12826639.4. Zhan C, Arispe I, Kelley E, et al. Ambulatorycare visits for treating adverse drug effects in theUnited States, 1995-2001. Jt Comm J QualPatient Saf. 2005 Jul;31(7):372-8. Epub:2005/09/01. PMID: 16130980.5. Warholak TL, Nau DP. Quality and Safety inPharmacy Practice. New York, NY: McGraw-Hill Medical; 2010.6. Gurwitz JH, Rochon P, Food and DrugAdministration. Improving the quality ofmedication use in elderly patients: a not-sosimpleprescription. Arch Intern Med. 2002 Aug12-26;162(15):1670-2. Epub: 2002/08/03.PMID: 12153368.7. Mahan D. Out-of-Bounds: Rising PrescriptionDrug Prices for Seniors. Families USAFoundation Publication No. 03-106. Washington,DC: July 2003.8. Kaufman DW, Kelly JP, Rosenberg L, et al.Recent patterns of medication use in theambulatory adult population of the United States:the Slone survey. JAMA. 2002 Jan16;287(3):337-44. Epub: 2002/01/16. PMID:11790213.9. Pellegrino AN, Martin MT, Tilton JJ, et al.Medication therapy management services:definitions and outcomes. Drugs.2009;69(4):393-406. Epub: 2009/03/28. PMID:19323584.10. 108th U.S. Congress. Medicare PrescriptionDrug, Improvement, and Modernization Act of2003. 2003.11. Bluml BM. Definition of medication therapymanagement: development of professionwideconsensus. J Am Pharm Assoc (2003). 2005 Sep-Oct;45(5):566-72. Epub: 2005/11/22. PMID:16295641.12. Association of Chain Drug Stores Foundation.Medication therapy management in pharmacypractice: core elements of an MTM servicemodel (version 2.0). J Am Pharm Assoc (2003).2008 May-Jun;48(3):341-53. PMID: 18595820.13. Isetts BJ, Buffington DE. CPT code-changeproposal: national data on pharmacists'medication therapy management services.Consult Pharm. 2007 Aug;22(8):684-9. Epub:2008/01/22. PMID: 18203409.14. American Medical Association. CPT changes2006: An Insider's View. In. Chicago, IL:American Medical Association; 2005:309-12.15. Pharmacist Services Technical AdvisoryCoalition. Medication Therapy ManagementService Codes. 2010www.pstac.org/services/mtms-codes.html.Accessed March 5, 2013.16. Tudor CG. CY 2014 Medication TherapyManagement Program Guidance and SubmissionInstructions. Baltimore, MD: Centers forMedicare & Medicaid Services; 2013www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/Memo-Contract-Year-2014-Medication-Therapy-Management-MTM-Program-Submission-v040513.pdf. Accessed May 2,2014.17. Bodenheimer T, Berry-Millett R. CareManagement of Patients With Complex HealthCare Needs. Robert Wood Johnson ResearchSynthesis Report No. 19. San Francisco, CA:December 2009.www.rwjf.org/content/dam/farm/reports/issue_briefs/2009/rwjf49853/subassets/rwjf49853_1.ES-30


18. Viswanathan M, Ansari MT, Berkman ND, et al.Assessing the Risk of Bias of Individual Studiesin Systematic Reviews of Health CareInterventions. AHRQ Publication No. 12-EHC047-EF. March 2012. In: Methods Guidefor Effectiveness and Comparative EffectivenessReviews. January 2014 Update. Rockville, MD:Agency for Healthcare Research and Quality.Chapters available atwww.effectivehealthcare.ahrq.gov/.19. Higgins JP, Altman DG, Gotzsche PC, et al. TheCochrane Collaboration's tool for assessing riskof bias in randomised trials. BMJ.2011;343:d5928. Epub: 2011/10/20. PMID:22008217.20. Viswanathan M, Berkman ND. Development ofthe RTI item bank on risk of bias and precisionof observational studies. J Clin Epidemiol. 2012Feb;65(2):163-78. Epub: 2011/10/01. PMID:21959223.21. West SL, Gartlehner G, Mansfield AJ, et al.Comparative Effectiveness Review Methods:Clinical Heterogeneity. Methods ResearchPaper. AHRQ Publication No. 10-EHC070-EF.Rockville, MD: Agency for Healthcare Researchand Quality; September 2010.www.effectivehealthcare.ahrq.gov/.22. Berkman ND, Lohr KN, Ansari M, et al. Gradingthe Strength of a Body of Evidence WhenAssessing Health Care Interventions for theEffective Health Care Program of the Agency forHealthcare Research and Quality: An Update.November 2013. In: Methods Guide forComparative Effectiveness Reviews (Preparedby the RTI-UNC Evidence-based Practice Centerunder Contract No. 290-2007-10056-I).Rockville, MD: Agency for Healthcare Researchand Quality. Chapters available at:www.effectivehealthcare.ahrq.gov/reports/final.cfm.23. Guyatt GH, Oxman AD, Kunz R, et al. GRADEguidelines 6. Rating the quality of evidence--imprecision. J Clin Epidemiol. 2011Dec;64(12):1283-93. Epub: 2011/08/16. PMID:21839614.24. Guyatt GH, Oxman AD, Vist G, et al. GRADEguidelines: 4. Rating the quality of evidence--study limitations (risk of bias). J Clin Epidemiol.2011 Apr;64(4):407-15. Epub: 2011/01/21.PMID: 21247734.25. Owens DK, Lohr KN, Atkins D, et al. AHRQseries paper 5: grading the strength of a body ofevidence when comparing medical interventions--Agency for Healthcare Research and Qualityand the Effective Health-Care Program. J ClinEpidemiol. 2010 May;63(5):513-23. Epub:2009/07/15. PMID: 19595577.26. Atkins D, Chang SM, Gartlehner G, et al.Assessing applicability when comparing medicalinterventions: AHRQ and the Effective HealthCare Program. J Clin Epidemiol. 2011Nov;64(11):1198-207. Epub: 2011/04/06. PMID:21463926.27. Touchette DR, Masica AL, Dolor RJ, et al.Safety-focused medication therapy management:a randomized controlled trial. J Am Pharm Assoc(2003). 2012 Sep-Oct;52(5):603-12. Epub:2012/10/02. PMID: 23023840.28. Grymonpre RE, Williamson DA, MontgomeryPR. Impact of a pharmaceutical care model fornon-institutionalised elderly: results of arandomised, controlled trial. Int J Pharm Pract.2001(4):235-41. PMID: CN-00425461.29. Chrischilles EA, Carter BL, Lund BC, et al.Evaluation of the Iowa Medicaid pharmaceuticalcase management program. J Am Pharm Assoc(2003). 2004 May-Jun;44(3):337-49. Epub:2004/06/12. PMID: 15191244.30. Winston S, Lin YS. Impact on drug cost and useof Medicare part D of medication therapymanagement services delivered in 2007. J AmPharm Assoc (2003). 2009 Nov-Dec;49(6):813-20. Epub: 2009/10/28. PMID: 19858047.31. Witry MJ, Doucette WR, Gainer KL. Evaluationof the pharmaceutical case management programimplemented in a private sector health plan. JAm Pharm Assoc (2003). 2011 Sep-Oct;51(5):631-5. Epub: 2011/09/08. PMID:21896463.32. Carter BL, Barnette DJ, Chrischilles E, et al.Evaluation of hypertensive patients after careprovided by community pharmacists in a ruralsetting. Pharmacotherapy. 1997 Nov-Dec;17(6):1274-85. Epub: 1997/12/17. PMID:9399611.33. Barnette DJ, Murphy CM, Carter BL. Clinicalskill development for community pharmacists. JAm Pharm Assoc (Wash). 1996Sep;NS36(9):573-80. Epub: 1996/09/01. PMID:8824077.ES-31


34. Chisholm-Burns MA, Kim Lee J, Spivey CA, etal. US pharmacists' effect as team members onpatient care: systematic review and metaanalyses.Med Care. 2010 Oct;48(10):923-33.Epub: 2010/08/20. PMID: 20720510.35. Bradley EH, Curry LA, Ramanadhan S, et al.Research in action: using positive deviance toimprove quality of health care. Implement Sci.2009;4:25. Epub: 2009/05/12. PMID: 19426507.36. Brown CA, Lilford RJ. The stepped wedge trialdesign: a systematic review. BMC Med ResMethodol. 2006;6:54. Epub: 2006/11/10. PMID:17092344.ES-32


IntroductionContextUsed appropriately, medications can alleviate distressing symptoms that compromisephysical and psychological well-being, help prevent the onset of many acute and chronic healthillnesses, and improve patient health outcomes. Too often, however, medications are not usedappropriately. The Institute of Medicine and other prominent organizations have recognized thatmedication-related problems plague our health care system. 1-3 In the United States in 2001, anestimated 4.3 million ambulatory visits were for adverse drug events. 4 A cohort study ofMedicare enrollees estimated the overall rate of adverse drug events at 50.1 per 1,000 personyears.5 The study rated more than one-third of the adverse drug events as serious, lifethreatening,or fatal; more than 40 percent of these more severe adverse drug events wereclassified as preventable. Another study found that more than 12 percent of hospitalized patientsexperienced an adverse drug event within 3 weeks following hospital discharge. 6In addition to problems involving adverse drug events, many patients are not prescribedoptimal treatment for chronic conditions such as high blood pressure and hyperlipidemia thatincrease their risk of cardiovascular disease and its complications. Moreover, even when optimaltherapy is prescribed, patient inability to adhere closely to medication regimens may lead to poorhealth outcomes. 7Medication-related problems are especially pronounced among older adults. 5 Individuals 65years or older constitute 13 percent of the U.S. population, but they consume more than 30percent of all prescription medications. 5,8 A 2006 report found that nearly 60 percent of people inthis age group were taking five or more medications and that nearly 20 percent were taking 10 ormore medications, 9 placing them at increased risk for experiencing adverse drug events.Moreover, these figures reflect a substantial increase in the prevalence of polypharmacy since1998. 9Medication therapy management (MTM) services are intended to address issues ofpolypharmacy, preventable adverse drug events, medication adherence, and medication misuse. 10MTM services are designed to be distinct from medication-dispensing services; in particular,they employ a patient-centric and comprehensive approach, rather than an individual product orepisodic perspective. 11MTM is the current term that represents a suite of health care services that have evolved outof the philosophy and processes described in the early 1990s as “pharmaceutical care.” 10 Similarto the concept of medical care or nursing care, pharmaceutical care is a term that describesprofessional pharmacy practice, not a discrete intervention. The Medicare Prescription Drug,Improvement, and Modernization Act of 2003 (Public Law 108-173) 12 expanded patient accessto MTM services and identified the following as goals for MTM services within Medicare PartD: (1) educate and counsel to improve patient understanding of their medications, (2) improveadherence, and (3) detect adverse drug reactions and improper medication use. This law alsoestablished the requirements that Medicare Part D Prescription Drug Benefit Plan sponsors haveto meet with respect to cost, quality, and the requirements for MTM programs sponsored by PartD drug benefit plans. The law established oversight for Part D MTM programs by the Centers forMedicare & Medicaid Services (CMS) and provided a general framework for MTM programsbut allowed Part D Plan sponsors flexibility in design, including the patient eligibility criteria forservices. The CMS requirements for Part D MTM programs have evolved since theirimplementation in 2006.1


Within a year of the passage of Medicare Part D, 11 national pharmacy organizationsestablished a consensus definition of MTM. 13 “MTM is a distinct service or group of servicesthat optimize therapeutic outcomes for individual patietns.” 13, p. 572 This definition furtherdescribes MTM as “a broad range of professional activities and responsibilities within thelicensed pharmacist’s, or other qualified healthcare provider’s scope of practice.” 13, p. 72 Table 1lists the types of services that can be considered under the umbrella MTM definition.Table 1. MTM servicesServices defined as MTM 14Medication therapy reviewsPharmacotherapy consultsDisease management coach/supportPharmacogenomic applicationsAnticoagulation managementImmunizationsHealth, wellness, and public healthMedication safety surveillanceOther clinical servicesThe pharmacy profession has developed or supportedadditional efforts to standardize and establish theinfrastructure for MTM service delivery. In 2008, a subsetof national pharmacy organizations published the secondversion of core elements for an MTM service model. Thismodel established five core elements for use in practice,including a medication therapy review, a personalmedication record, a medication-related action plan,intervention and/or referral, and documentation andfollowup. 10,15 Also in 2008, Current Procedural Terminology (CPT ® ) codes were established toprovide a mechanism for reimbursement for services related to medication management. TheseCPT codes define MTM as “services provided by a pharmacist to optimize the response tomedications for the management of treatment-related medication problems or complications.”More recently, the Patient-centered Primary Care Collaborative established its definition ofcomprehensive medication management to describe MTM in the context of a patient-centeredmedical home, which includes elements of the CPT definition. 16,17 The evolution from isolatedresearch interventions studying the impact of pharmaceutical care interventions to large-scale,commercial MTM programs or collaborative medication management within primary carerepresents a journey along a continuum of practice settings, patient populations, and interventioncomponents and features. Over time, the practice and standards for these services have evolved,as have standards for describing and conducting research studies involving these interventions.Thus, establishing the scope of this review was very challenging. A broadly defined scope (allclinical pharmacist interventions regardless of setting or patient population) risks includingstudies that may be too different from each other to allow for meaningful comparison andsynthesis. A narrowly defined scope (e.g., a focus exclusively on Medicare Part D-defined MTMprograms) risks the omission of studies that met the definition of MTM, but that predated thePart D era, were conducted in other countries, or used patient eligibility criteria that are lessrestrictive than Part D. In the next section, we describe background related to population,intervention, comparison, outcomes, timing, and setting (“PICOTS”) that we relied on toestablish the scope of this review. Throughout this review, we will use the term MTM to describethe general class of intervention. However, when describing individual studies included in thisreview, we will defer to the terms used by the study author to describe the intervention they wereevaluating (e.g., pharmaceutical care, clinical pharmacy services, or MTM).PopulationsAdult patients with multiple chronic conditions who take many different prescription ornonprescription medications, herbal products, or diet supplements (and combinations of these)are the target population for most outpatient-based MTM services. 11 Because older adults are2


more likely to take multiple medications, MTM services generally target them. However, MTMinterventions may also target patients taking a single high-risk medication (e.g., Coumadin) ormay target patients at high risk for an adverse drug event, for example, during a transition in carefrom a hospital to home setting. Although some children with complex medication regimens maybenefit from MTM, these program are typically designed and delivered to adults.As part of Medicare Part D implementation, CMS required that MTM programs targetMedicare Part D enrollees, who have multiple chronic diseases, are taking multiple Part D drugs,and are likely to incur annual costs for covered Part D drugs that exceed a predetermined level(“annual cost threshold”). To be eligible for CMS reimbursement, MTM programs originally hadto offer services for at least four of seven core chronic diseases: hypertension, chronic heartfailure, diabetes, dyslipidemia, respiratory disease (e.g., asthma, chronic obstructive pulmonarydisease), bone disease (e.g., osteoporosis, osteoarthritis, rheumatoid arthritis), and mental healthdiseases. As of January 2013, this criterion specifies at least five of nine core chronicconditions—Alzheimer’s disease and end-stage renal disease were the added conditions.Programs may require no more than eight Part D drugs, although they may set the maximum atany number between two and eight. CMS set the annual cost threshold at $4,000 in 2006,lowered it to $3,000 in 2010, and increased it by an annual percentage each year beginning in2012. The cost threshold for 2013 is $3,144. CMS reimburses for MTM services for bothcommunity-dwelling beneficiaries and beneficiaries in long-term care settings. Although initialPart D MTM programs were designed as “opt-in,” more recently, MTM programs must enrolleligible beneficiaries using an “opt-out” approach.Health care systems, pharmacy benefit management organizations, large self-insuredemployers, community pharmacies, or individual medical practices may also provide MTMservices to beneficiaries who do not have Medicare Part D or who do not meet the CMS Part Dcriteria. For example, the Veterans Health Administration (VHA) includes MTM as one ofseveral clinical activities provided to VHA health beneficiaries by VHA pharmacy services. 18The VHA does not specify patient eligibility criteria for MTM services. Non-Part D MTMprograms and research studies of MTM interventions may define their own patient eligibilitycriteria, which may or may not be similar to current CMS criteria, for example, requiring onlyone chronic condition to be eligible for services.Interventions and ComparatorsAs discussed, several pharmacy organizations have proposed core elements for an MTMservice model. 10,11 These features can be summarized as follows:• A comprehensive medication review (CMR) to identify and resolve medication-relatedproblems.• The generation of a personal medication report, which is a written list of the patient’sprescription and nonprescription drugs, herbal products, and dietary supplements.• A patient-directed medication action developed in collaboration with the patient.• Education, counseling, and resources to enhance patients’ understanding about using themedication and to improve adherence.• Coordination of care, including documenting MTM services and providing thatdocumentation to the patient’s other providers and referring patients to other providers asneeded.3


CMS requires that each beneficiary enrolled in a Part D MTM program be offered aminimum level of MTM services. These include:• Interventions for both beneficiaries and prescribers;• An annual CMR with written summaries in CMS’s standardized format:o The beneficiary’s CMR must include an interactive, person-to-person, or telehealthconsultation that is performed by a pharmacist or other qualified provider (e.g., anurse or a physician) and may result in a recommended medication action plan.o If a beneficiary is offered the annual CMR and is unable to accept the offer toparticipate, the pharmacist or other qualified provider may perform the CMR with thebeneficiary’s prescriber, caregiver, or other authorized individual; and• Quarterly targeted medication reviews with followup interventions when necessary.CMS expects the CMR to meet the following professional service definition: “a systematicprocess of collecting patient-specific information, assessing medication therapies to identifymedication-related problems, and developing a prioritized list of medication-related problems,and creating a plan to resolve them with the patient, caregiver, and/or prescriber.” 19, p. 6 Inaddition, CMS expects the CMR to be “an interactive person-to-person or telehealth medicationreview and consultation conducted in real time between the patient and/or other authorizedindividual, such as [a] prescriber or caregiver, and the pharmacist or other qualified provider. Itis designed to improve patients’ knowledge of their prescriptions, over-the-counter medications,herbal therapies, and dietary supplements; identify and address problems or concerns thatpatients may have; and empower patients to self-manage their medications and their healthconditions.” 19, p.6 Written summaries of the CMR are to be provided in CMS’s standardizedwritten format that includes a beneficiary cover letter, medication action plan, and personalmedication list. 19The service-level expectations of a CMR align closely with the definition of MTMscontained in the official health-reporting nomenclature of CPT. ®20,21 The CPT MTM codesdefine three components that may each vary in complexity or time required to complete. Thesecomponents are (1) assessment of drug-related needs, (2) identification of drug therapyproblems, and (3) complexity of care planning and followup evaluation. Recently, transitionalmanagement CPT codes have been established for use within the first 29 days of patientdischarge from an acute care facility. These codes also include elements of medicationmanagement, specifically medication reconciliation. MTM CPT codes cannot be used by thesame professional in the same time frame as transitional care management codes, suggesting thatthe medication management activities during transitions of care are a distinct category of MTMservices. 22Disease-management, case-management, and self-management interventions havecomponents that overlap with MTM components—for example, provision of education andcounseling to increase medication adherence or coordination of care. Our preliminary literaturesearch yielded many pharmacist-led interventions that were termed as one of these three types ofinterventions (e.g., a pharmacist-led diabetes disease management intervention). We relied on thedescriptions in the Robert Wood Johnson Foundation Research Synthesis Report “Caremanagement of patients with complex health care needs” for guidance to make distinctionsbetween MTM and care management, case management, and disease managementinterventions. 23 We determined that our inclusion and exclusion criteria related to theintervention needed to define specific MTM intervention components, such that we would4


identify relevant studies whether they were called “MTM” or not. We also considered the topicnominator’s original request, which was to consider different models for assisting patients inmanaging their medications for chronic disease among patients with multiple conditions. Thus,we synthesized our findings from the preliminary literature search; our exploration of casemanagement, care management, and disease management definitions; and the topic nominator’soriginal request to determine that our intervention criteria needed to narrowly define multipleintervention components related to medication management, but that these components needed tobe applied broadly to patients across their entire medication regimen. As a result, MTM servicessuch as pharmacist-led single-disease management programs or anticoagulation clinics would notbe considered for inclusion in our review. By bounding the review in this way, we end up with amore homogenous set of studies to synthesize.OutcomesMTM is thought to influence a wide variety of outcomes. Two of the most commonoutcomes measured in MTM studies are drug therapy problems identified and drug therapyproblems resolved. Taxonomies to describe drug therapy problems exist, but our preliminaryliterature search revealed many different approaches to measuring and reporting these outcomes.Other MTM outcomes relate to intermediate health outcomes measured typically by laboratoryor other biometric tests for common chronic conditions; these may include hemoglobin A 1c ,blood pressure, cholesterol (e.g., total, low-density lipoprotein, and high-density lipoproteincholesterol), and cardiac or pulmonary function (e.g., left ventricular ejection fraction,spirometry). Finally, still other MTM services relate to patient-centered outcomes (e.g.,morbidity, mortality, adverse drug events, missed days of work or school, patient satisfactionwith care, health-related quality of life). 24 The impact of MTM on health care utilization,intermediate health outcomes, and patient-centered outcomes may derive from identification andresolution of drug therapy problems, including improved medication adherence, fewer drugrelatedadverse events, and more efficient coordination of care.SettingsMTM services can be delivered in a variety of settings. These include inpatient facilities,ambulatory care settings (e.g., outpatient clinics, physician practices), retail pharmacies in thecommunity, and long-term care settings such as assisted living or skilled nursing facilities. Inaddition, telephone-based MTM services may be provided to community-dwelling adults byprofessional staff (often pharmacists) employed by pharmacy benefits management companies orother commercial health care companies that have centralized call centers. The setting in whichMTM is delivered depends on the type of provider delivering the service and the goals and scopeof the MTM program. Because MTM refers to a wide variety of services, a review of suchinterventions needs to be bounded to ensure that the interventions synthesized in the review arereasonably comparable. For example, studies focused on MTM services provided during andshortly after an acute hospital stay may not be comparable to MTM services provided tooutpatients because the goals of therapy and the acuity of the patient’s status are very different.Based on our preliminary literature search, we found that most studies in inpatient settings werefocused either on single-medication reconciliation interventions during or at discharge or focusedon integrated clinical pharmacy management in acute settings. We also considered the topicnominator’s proposed research questions, which were decidedly focused on MTM provided tooutpatients.5


TimingBecause MTM is used to define a broad range of services, MTM services can be provided asone-time interventions or longitudinally during multiple episodes of care depending on thespecific type of MTM service and care setting. For example, medication reconciliation orimmunization is a type of MTM service that is typically done during a single episode of care.The pharmacy profession consensus definition for MTM includes monitoring and evaluation of apatient’s response to therapy, and the MTM Core Service Model includes followup as acomponent. Similarly, CPT codes for MTM services include a component involving complexityof care planning and followup. Requirements for Medicare Part D MTM programs include afollowup component at least quarterly following an initial comprehensive medication review.Thus, we determined that we needed to establish inclusion criteria to distinguish interventionsdesigned to support longitudinal medication management as opposed to studies of one-timeinterventions.Contextual FactorsOur preliminary literature search identified pharmacists as the typical interventionist forproviding MTM services. CMS guidelines require that MTM be delivered by a pharmacist orother qualified health care provider. Professional pharmacy organizations have been activelyinvolved in proposing delivery models, standards, and recommendations for MTM services.Pharmacist training varies considerably. Before the 1990s, individuals could become registeredpharmacists with a bachelor of science (B.S.) degree that required a minimum of 5 years ofstudy. Current regulations require that individuals have a doctor of pharmacy (Pharm.D.) degree,which requires a minimum of 6 years of study and provides more clinical training than B.S.programs. In addition, many Pharm.D. graduates pursue advanced training through residency,fellowship, and certificate programs. Some of these programs focus on areas such as MTM. Theinfluence that interventionist type (e.g., physician, nurse pharmacist), education, and MTMspecifictraining have on MTM effectiveness is unknown.Numerous factors other than clinical specialty may affect the quality of MTM services.Mode, frequency, and interval of delivery may influence MTM success, as may specific MTMcomponents and the fidelity of their implementation. One key factor is how well an MTMprovider understands the patient-specific goals of medication therapy. Integrating MTM serviceswith usual care may help ensure that the goals of MTM are achieved. Integration of services andusual care refers to the ability of the MTM provider to bidirectionally communicate with patientsand multiple prescribers and ease of MTM interventionist access to patients’ medical records.Health care reimbursement systems may also influence the delivery of MTM services. Not allprivate insurers cover MTM services. The degree to which MTM component services differ forMedicare beneficiaries when compared with non-Medicare beneficiaries is not known.Finally, certain patient populations may have considerable difficulty accessing orparticipating in MTM services. Examples include individuals who are homebound, individualswho have physical or cognitive disabilities, patients without health insurance, and patients livingin rural areas.6


Scope and Key QuestionsScope of the ReviewMTM is a complex intervention, which could have different components depending on thegoals and scope of the MTM program. This review seeks to catalog outpatient-based MTMinterventions, assess the overall effectiveness of outpatient-based MTM in comparison withusual care, examine the factors under which outpatient-based MTM is effective and optimallydelivered, assess what types of patients are likely to benefit from outpatient-based MTMservices, and assess what types of patients may be at risk of harms from such programs. Thisreview does not address (1) MTM services provided within inpatient settings or shortly afterhospital discharge, (2) disease management services provided by pharmacists, or (3)interventions designed as a single episode of contact. The rationale for limiting the scope toexclude some types of MTM interventions is to ensure that included studies are reasonablycomparable with respect to intended goals and purpose of the MTM intervention.Relevance of Research Question to Clinical Decisionmaking orPolicymakingThe Key Questions (KQs) we address are highly relevant to both clinical decisionmaking andpolicies regarding MTM services. Identifying demonstrably effective models and components ofMTM services will help patients and their health care providers achieve important intermediateand long-term health-related outcomes. Our findings will help providers of MTM services,particularly pharmacists and pharmacy benefit managers, understand what works well in whichsettings and with which patients; the findings will have the potential to improve the efficiency ofdelivery and thus improve the value of MTM services. Lastly, a better understanding of thecomparative effectiveness of MTM services will assist CMS with future revisions orenhancements to the policies governing coverage for MTM services.Key QuestionsThe KQs are listed below and placed in relation to another and the PICOs in the analyticframework (Figure 1). Specific details regarding patient population, intervention components,and outcomes are provided in the section that follows the analytic framework.7


Analytic FrameworkFigure 1. Analytic framework for outpatient medication therapy managementa The population, intervention, outcomes, and setting are described in detail in the text.Abbreviations: KQ = Key Question; MTM = medication therapy management.KQ 1: What are the components and implementation features of outpatientMTM interventions?KQ 2: In adults with one or more chronic diseases who are takingprescription medications, is MTM effective in improving the following:a. Intermediate outcomes, including biometric and laboratory measures,drug therapy problems identified, drug therapy problems resolved,medication adherence, goals of therapy met, and patient engagementin medication management?b. Patient-centered outcomes, such as disease-specific morbidity,disease-specific or all-cause mortality, adverse drug events, healthrelatedquality of life, activities of daily living, patient satisfaction with8


health care, work or school absenteeism, and patient and caregiverparticipation in medical care and decisionmaking?c. Resource utilization, such as prescription drug costs, other healthcare costs, and health care utilization?KQ 3: Does the effectiveness of MTM differ by MTM components andimplementation features?KQ 4: Does the effectiveness of MTM differ by patient characteristics,including but not limited to patient demographics and numbers and typesof conditions and medications?KQ 5: Are there harms of MTM, and if so, what are they?Populations, Interventions, Comparators, Outcomes, Timing,and Setting (PICOTS)Table 2 lays out the PICOTS for this review. For this review, we take a broad perspective onthe population evaluated; we do not limit the review to populations meeting CMS Part D MTMeligibility criteria. Specifically, we did not require multiple chronic conditions or a minimumnumber or level of expenditures on prescription drugs. For the intervention, we required acomprehensive medication review, patient-directed education or counseling, an element ofprovider followup, and care coordination as the minimum intervention criteria. We establishedseveral eligible comparators and detailed intermediate process or health outcomes, patientcenteredoutcomes, and health care utilization outcomes. We also identified some potential harmoutcomes, including patient confusion or decision conflict. Lastly, we established exclusioncriteria related to studies conducted in inpatient settings or shortly after an inpatient stay andrequired the intervention to be designed such that followup to patients was available, regardlessof whether the patient actually received any followup.Table 2. Populations, interventions, comparators, outcomes, timing, and settingsPICOTS Inclusion and Exclusion Criteria and Relevant Factors for Study AbstractionPopulations Inclusion criteria:• Patients ages 18 or older with one or more chronic conditions requiring the use of prescriptionmedication to manage symptoms or prevent progression of chronic diseaseExclusion criteria:• Patients in long-term or acute care settings without access or control over their own medicationadministration.Relevant factors:• Patient characteristics that may influence intervention effectiveness:o Age, sex, race and ethnicity, socioeconomic status, health insurance status, education level,health literacy status, cognitive impairment, number and types of chronic conditions, socialsupport, and urban/rural status9


Table 2. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS CriteriaInterventions Inclusion criteria:• A bundle of medication-related services described by the term “MTM” or “pharmaceutical care” or“clinical pharmacy services” or a similar phrase, that include at a minimum the following threeelements:o Comprehensive medication review covering all prescription and nonprescription drugs, herbs,and supplements taken by the patiento Patient-directed education and counseling or other resources to enhance understanding of theuse of medicationo Coordination of care, including prescriber-directed interventions; documentation of MTMservices for use by the patient’s other providers; and referral to other providers, clinicians, orresources when appropriate 19Exclusion criteria:• The following types of interventions will be excluded:o Medication reconciliation interventions that did not include all three elements as describedabove• The following types of interventions may include MTM services, but MTM may represent only onecomponent of the overall intervention:o Disease-management interventions 23o Case- or care-management interventions 23o Patient-centered medical home models of careo Fully integrated, collaborative care models involving multiple disciplines and specialties• These types of interventions will be excluded unless studies contain the same level of overallmedical care or services among different study arms such that the effect of MTM can be isolated.For example, a study with two arms that has one arm with a care management intervention thatincludes MTM services and the other arm that has the care management intervention withoutMTM services could be included. By contrast, a study that includes a care managementintervention with MTM in one arm and usual medical care (no care management intervention) inthe other arm would not be included.Relevant factors:• Implementation features that may influence intervention effectiveness include the following:o Mode of delivery: telephone, face-to-face, virtual (Web/online/Internet), and remote videoo Type of professional providing initial and followup MTM service: pharmacist, nurse, physician,other cliniciano Frequency and interval of followup for MTM serviceso Specific MTM components usedo Fidelity in implementing MTM components: to what extent were services delivered as designedor intendedo Establishing and communicating goals of drug therapy to patients and among care providerso Method of identifying patients for enrollment (e.g., population health data, provider referral forservices, enrollment during a transition in care, targeting highly activated patients, targetingpatients at time of high risk for event [e.g., when prescribing a new drug])o Level of integration of MTM with usual care, which includes access to real-time clinicalinformation and laboratory values, and regular and consistent communication amongprescribers and persons providing MTM serviceso Reimbursement characteristics (e.g., who is paying for cost of MTM services, who is reimbursedfor MTM services, whether services are separately reimbursable)o Health system characteristics (e.g., are services being provided within an accountable careorganization, patient-centered medical home, or some other unique system setting, such as theVeterans Health Administration, the Indian Health Service, non–U.S. single-payer system)10


Table 2. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS CriteriaComparators Inclusion criteria:• Usual care, as defined by the studies• Different bundles of MTM services (e.g., five components vs. three components)• Same MTM services provided by different health care professionals (e.g., pharmacist vs.,physician or nurse)• Same bundle of MTM services delivered by different modes (e.g., telephone vs. in person)• Same bundle of MTM services provided at different intensities, frequencies, or level of integrationwith prescribersOutcomes Inclusion criteria:• Studies must report at least one eligible outcome:o Intermediate outcomes:- Disease-specific laboratory or biometric outcomes (e.g., hemoglobin A 1c; blood pressure;total, low-density lipoprotein, or high-density lipoprotein cholesterol; pulmonary function; renalfunction; left ventricular ejection fraction; or other laboratory or biometric outcome specific todiseases covered)- Drug therapy problems identified as defined by primary studies but typically include thefollowing: medications being taken but not indicated; medications indicated but notprescribed; patient adherence issues; supratherapeutic doses; subtherapeutic doses; generic,formulary, or therapeutic substitution issue; complex regimen that can be simplified with sametherapeutic benefit; and potential for drug-drug interactions or adverse event.- Drug therapy problems that are resolved as defined by primary studies but typically includethe following: needed drug initiated; unnecessary drug discontinued; change in drug dose,form, or frequency; or generic, formulary, or therapeutic substitution- Medication adherence- Goals of therapy met- Patient engagement (e.g., initial and continuing patient participation in the MTM program)o Patient-centered outcomes- Disease-specific morbidity, including falls and fall-related morbidity, and outcomes specific tothe patient’s underlying chronic conditions (e.g., Patient Health Questionnaire 9 [PHQ9],disease-specific symptoms, reduced number of disease-specific acute exacerbations orevents)- Disease-specific or all-cause mortality, including fall-related mortality- Reduced (actual) adverse drug events (frequency and/or severity)- Health-related quality of life as measured by generally accepted generic health-relatedquality-of-life measures (e.g., short-form questionnaires, EuroQOL) or disease-specificmeasures- Activities of daily living as measured by generally accepted standardized measures of basicand/or instrumental activities of daily living (e.g., Katz, Lawton, or Bristol instruments) or withinstruments that have demonstrated validity and reliability- Patient satisfaction with MTM care- Work or school absenteeism- Patient and caregiver participation in medical care and decisionmakingo Resource utilization- Prescription drug costs and appropriate prescription drug expenditures- Other health care costs- Health care utilization (hospitalizations, emergency department visits, and physician officevisits)o Harms- Care fragmentation- Patient confusion- Patient decisional conflict- Patient anxiety- Increased (actual) adverse drug events- Patient dissatisfaction with care- Prescriber confusion- Prescriber dissatisfaction11


Table 2. Populations, interventions, comparators, outcomes, timing, and settings (continued)PICOTS CriteriaTiming Inclusion criteria:• Interventions should have at least two separately identifiable episodes of patient-directed MTMservices, with any interval of time in between episodes.• For studies that report outcomes at different points in time, we only considered outcomesmeasured after the second episode of care and will use the longest-term outcomes reported bythe study.Exclusion criteria:• Interventions designed as single-episode interventions without any provision for followup andmonitoring.Setting Inclusion criteria:• Ambulatory settings (e.g., outpatient clinics or private physician offices), long-term-care (e.g.,assisted living) settings if residents have control over medication self-administration, or retailpharmacy settings• Home setting• Interventions conducted in the United States• Interventions conducted in non-U.S. countries published in EnglishExclusion criteria:• MTM services that are delivered exclusively in inpatient settings.• MTM services delivered at the time of hospital discharge or shortly after (e.g., within a few weeks)Relevant factors:• The MTM intervention itself may be delivered by home visits, by telephone, via the Web, or inother non–face-to-face modalities, such as video teleconferencing.Abbreviations: MTM = medication therapy management; PICOTS = populations, interventions, comparators, outcomes, timing,and setting; PHQ9 = Patient Health Questionnaire 9.Organization of This ReportThe remainder of this report describes our methods, presents the results of our synthesis ofthe literature, discusses our conclusions, and provides other information relevant to theinterpretation of this work. The Methods section describes our scientific approach for thissystematic review in detail. The Results section presents our findings for all five of the KQs andincludes summary and strength-of-evidence tables. In the Discussion section, we summarize thefindings and discuss the implications for clinical practice and further research. A complete list ofreferences, acronyms, and abbreviations follows the Discussion section.This report contains the following appendixes: Appendix A contains the exact search stringswe used in our literature searches. Appendix B documents the title and abstract and full-textreview forms. Studies excluded at the stage of reviewing full-text articles with reasons forexclusion are presented in Appendix C. Evidence tables appear in Appendix D. Appendix E listsstudies rated high risk of bias and reasons for excluding them from relevant KQ analyses.Quantitative analyses are presented in Appendix F.12


MethodsThe methods for this comparative effectiveness review (CER) on medication therapymanagement (MTM) follow the methods suggested in the Agency for Healthcare Research andQuality (AHRQ) “Methods Guide for Effectiveness and Comparative Effectiveness Reviews”(available at http://www.effectivehealthcare.ahrq.gov/methodsguide.cfm). We specified methodsand analyses a priori in a protocol posted on the AHRQ website, 25 following a standardframework for specifying population, interventions, comparators, outcomes, and settings(PICOTS). The main sections in this chapter reflect the elements of the protocol established forthe CER; certain methods map to the PRISMA (Preferred Reporting Items for SystematicReviews and Meta-Analyses) checklist. 26 We describe below instances in which our a priorimethods required further specification during the project.Topic Refinement and Review ProtocolThe topic of this report and preliminary Key Questions (KQs) arose through a nominationfrom the Pharmacy Quality Alliance. Key Informants representing several clinical and scientificdisciplines provided input on the initial KQs; we revised them as needed. An initial draft of therevised KQs was posted for public comment from March 6, 2013, through April 2, 2013, on theAHRQ Effective Health Care program Web site. We received comments from 23 professionalorganizations and individuals and further revised KQs as appropriate. Specifically, we1. added a new KQ (KQ 1) to describe the components and implementation features ofMTM interventions,2. included additional intermediate outcomes in KQ 2,3. reworded KQ 3 to include MTM components,4. specified MTM components and implementation features for KQ 3 in the PICOTS,5. specified additional patient characteristics for KQ 4 in the PICOTS, and6. rephrased KQ 5 to make the response conditional on identifying whether any harms ofMTM exist.Literature Search and Identification StrategySearch StrategyTo identify articles relevant to each KQ, we began with a focused MEDLINE ® search forMTM interventions using a combination of medical subject headings (MeSH ® ) and title andabstract keywords and limiting the search to English-language and human-only studies (Table 3)(inception through January 9, 2014). We also searched the Cochrane Library (inception throughJanuary 10, 2014).and the International Pharmaceutical Abstracts database (inception throughJanuary 10, 2014) using analogous search terms.(Appendix A). We selected these databasesbased on preliminary searches and consultation with content experts. We conducted qualitychecks to ensure that the searches identified known studies (i.e., studies identified during topicnomination and refinement). Based on these quality checks, we revised and ran additionalsearches (specifically, drug therapy management, drug therapy problem, and medicationsmanagement) to avoid missing articles that might prove eligible for this CER.13


In addition, we searched the gray literature forunpublished studies relevant to this review andincluded studies that met all the inclusion criteriaand contained enough methodological informationto assess risk of bias. Specifically, sources of grayliterature included ClinicalTrials.gov, the WorldHealth Organization’s International Clinical TrialsRegistry Platform, Health Services ResearchProjects in Progress (HSRProj), the NationalInstitutes of Health Research Portfolio OnlineReporting Tools, the Database of PromotingHealth Effectiveness Reviews, the New YorkAcademy of Medicine Grey Literature Report,and CMS.gov. AHRQ’s Scientific ResourceCenter managed the process of submittingrequests for scientific information packets, whichcontain information about MTM programs andservices of interest from relevant providers.We reviewed our search strategy with anindependent information specialist and theTechnical Expert Panel and supplemented itaccording to their recommendations. In addition,to avoid retrieval bias, we manually searched thereference lists of landmark studies andbackground articles on this topic to identify anyrelevant citations that our electronic searchesmight have missed.Table 3. Literature search terms formedication therapy management studiesPopulations None; no population terms were usedto avoid restricting the search yieldInterventions (“Medication TherapyManagement”[Mesh] OR “medicationtherapy management” OR“comprehensive medication review”OR “personal medication record” OR(“medication” AND “action plan”) OR“medication therapy review” OR“Medication Reconciliation”[Mesh]OR (med* AND reconciliation) OR“medication-related problems” ORMTMP OR prescriber intervention*OR “drug utilization management”OR “chronic care improvement” OR“drug therapy services” OR(“utilization management strategies”OR “utilization managementstrategy”) OR “medicationcounseling” OR “pharmaceutical casemanagement” OR “drug therapymanagement”Outcomes “optimized treatment outcomes” OR(patient OR patients) AND“medication understanding”) OR(“drug therapy outcome” OR “drugtherapy outcomes”)Study designs None; no study design terms wereused to avoid restricting the searchyieldLimits Humans; English languageWe conducted an updated literature search (of the same databases searched initially)concurrent with the peer review process. We also investigated any literature the peer reviewersor the public suggest and, if appropriate, incorporated additional studies into the final review.The appropriateness of those studies was determined using the methods and criteria describedabove.We planned to include pooled estimates of effect or other relevant results from systematicreviews hat meet our inclusion/exclusion criteria and to evaluate the quality of includedsystematic reviews using the AMSTAR tool. 27 If appropriate and feasible, we had planned toupdate the results of these reviews quantitatively or qualitatively. We also planned to reviewreference lists of systematic reviews that used exclusion and exclusion criteria that differed fromours to ensure that we include all relevant studies.Inclusion/Exclusion CriteriaWe specified our inclusion and exclusion criteria based on the population, intervention,outcome, timing, and settings identified through the topic refinement exercise. We excludedstudies published in languages other than English. We excluded study designs without controlgroups to ensure that our pool of included studies can inform the causal link between theintervention and outcomes.14


In conducting the review, we found that we needed to define the intervention with greaterspecificity than originally thought so that we could include MTM interventions but excludedisease management interventions. Specifically, we required that included studies had conducteda comprehensive, rather than condition-specific, medication review, as required in our PICOTScriteria. Although we had not planned to contact study authors routinely for additionalinformation, the lack of clarity regarding intervention elements in numerous published studiesnecessitated our contacting authors. For these studies, we based our decisions on inclusion orexclusion based on email communication. (Appendix D specifies the studies or publications forwhich we sought such information but received no response from authors as of the time the draftreport was submitted for peer review.)Study SelectionPairs of trained members of the research team reviewed each title and abstract independentlyagainst our inclusion/exclusion criteria. Studies marked for possible inclusion by either reviewerunderwent a full-text review. For studies that lack adequate information to determine inclusion orexclusion, we retrieved the full text and then made the determination.We retrieved and reviewed the full text of all included titles during the title/abstract reviewphase. Two trained members of the team independently reviewed each full-text article forinclusion or exclusion based on the eligibility criteria specified in Table 4. If both reviewersagreed that a study did not meet the eligibility criteria, they excluded the study. If the reviewersdisagreed, they discussed differences to achieve a consensus. If they could not reach consensus, athird senior member of the review team resolved the conflict. We tracked all results in anEndNote ® (Thomson Reuters, New York, NY) database. We recorded the reason that eachexcluded full-text publication did not satisfy the eligibility criteria. Appendix C lists all studiesexcluded at this stage together with the reason(s) for exclusion.Data ExtractionFor studies that met our inclusion criteria, we abstracted relevant information into evidencetables (Appendix D). We piloted our approach with a sample of studies and revised the formthereafter. We designed data abstraction forms to gather pertinent information from each article,including the characteristics of the study populations, interventions, comparators, outcomes,timing, settings, study designs, methods, and results. A second member of the team reviewed alldata abstractions for completeness and accuracy. (Relevant forms can be found in Appendix B.)15


Table 4. Inclusion/exclusion criteria for medication therapy management studiesCategory Inclusion ExclusionPopulation Patients aged 18 or older with one or moreconditions requiring the regular use of prescriptionmedication to manage symptoms or preventprogression of chronic disease• Children under age 18• Adults with acute conditionsInterventionsControlinterventionsOutcomesTiming ofintervention andfollowupSettings• Those specified in the PICOTS criteria listed inTable 1 (Introduction)• More complex interventions with an MTMcomponent that are compared with identicalinterventions without an MTM component(including care management and diseasemanagement)• Those specified in the PICOTS criteria listed inTable 1 (Introduction)• Those specified in the PICOTS criteria listed inTable 1 (Introduction)• Interventions should have at least twoseparately identifiable episodes of MTMservices (either patient directed or providerdirected or both) with or without specifying anycertain amount of time between those episodes• For studies that report outcomes at differentpoints in time, we considered only outcomesmeasured after the second episode of care.• Ambulatory (e.g., outpatient clinics, privatephysician offices, or retail pharmacy settings)and long-term care settings• May be delivered by telephone, via the Web, orin other non–face-to-face modalities, such asvideo teleconferencing• Interventions conducted in the United Statesand other countries will be included• Drug therapy services for a single drug(e.g., warfarin clinics, statin clinics)• Interventions in which the effect of theMTM component cannot be isolated (e.g.,case management or diseasemanagement with an MTM component)• Self-management programs• Isolated medication reconciliationinterventions• Integrated pharmacy services withininpatient settings• One-time corrective interventions relatedto medication management• Studies that do not include at least one ofthe outcomes listed under the inclusioncriteria• Studies that measure outcomes only afterone episode of MTM care• Inpatient settings, if delivery of MTMservices occurs almost exclusively in theinpatient settingGeography • No limits • Not applicableDates of search • No limits; searches will be updated while the • Not applicabledraft report is out for peer reviewStudy designs • Original research• Eligible study designs include:- Randomized controlled trials- Nonrandomized controlled trials• Case series• Case reports• Nonsystematic reviews• Studies without a control group- Prospective controlled cohort studies- Retrospective controlled cohort studies- Case-control studies- Systematic reviews and meta-analysesStudy duration No limits Not applicablePublication EnglishAll other languageslanguagePublication type Any publication reporting primary data Publications not reporting primary dataAbbreviations: MTM = medication therapy management; PICOTS = populations, interventions, comparators, outcomes, timing,and setting.16


Assessment of Risk of Bias of Individual StudiesTo assess the risk of bias of individual studies, we used predefined criteria developed byAHRQ. 28 For randomized controlled trials (RCTs), we relied on the risk-of-bias tool developedby the Cochrane Collaboration. 29 We assessed the risk of bias of observational studies using anitem bank developed by RTI International. 30In general terms, results of a study with low risk of bias are considered valid. Studies markedlow risk of bias did not have any major flaws in design or execution. A study with medium riskof bias is susceptible to some bias but probably not sufficient to invalidate its results. A studywith high risk of bias has significant methodological flaws (e.g., stemming from serious errors indesign or analysis) that may invalidate its results. Primary concerns for our review includedselection bias, confounding, performance bias, detection bias, and attrition bias. Very highattrition rates, particularly when coupled with a failure to control for confounding or conductintention-to-treat analyses, resulted in a rating of high risk of bias for trials and prospectivecohort studies. Likewise, we rated studies with an inherently high risk of confounding in design(e.g., observational studies comparing refusers versus acceptors of MTM interventions) as highrisk of bias if they failed to address confounding through design (e.g., matching) or analysis(e.g., regression). Specifically, we evaluated trials on the adequacy of randomization, allocationconcealment, similarity of groups at baseline, masking, attrition, whether intention-to-treatanalysis was used, method of handling dropouts and missing data, validity and reliability ofoutcome measures, and treatment fidelity. For observational studies, we did not assess adequacyof randomization or allocation concealment but did assess for confounding. We also evaluatedtrials for confounding due to randomization failure through biased selection or attrition. In otherwords, we evaluated trials with potential randomization failure for the same risks of bias asobservational studies.We excluded studies that we deemed at high risk of bias from our main data synthesis andmain analyses. We included them for sensitivity analyses; in cases when we had no otheravailable or credible evidence, we included in the report a brief synopsis of studies assessed ashigh risk of bias.Data SynthesisWhen we found three or more similar studies for a comparison of interest, we conductedmeta-analysis of the data from those studies using Comprehensive Meta-Analysis software. Forall analyses, we used random-effects models to estimate pooled or comparative effects. Todetermine whether quantitative analyses were appropriate, we assessed the clinical andmethodological heterogeneity of the studies under consideration following establishedguidance; 31 that is, we qualitatively assessed the PICOTS of the included studies, looking forsimilarities and differences. When we conducted quantitative syntheses (i.e., meta-analysis), weassessed statistical heterogeneity in effects between studies by calculating the chi-squaredstatistic and the I 2 statistic (the proportion of variation in study estimates attributable toheterogeneity). The importance of the observed value of I 2 depends on the magnitude anddirection of effects and on the strength of evidence for heterogeneity (e.g., the p-value from thechi-squared test or a confidence interval for I 2 ). Where relevant, we examined potential sourcesof heterogeneity using sensitivity analysis.When quantitative analyses were not appropriate (e.g., because of heterogeneity, insufficientnumbers of similar studies, or insufficiency or variation in outcome reporting), we synthesized17


the data qualitatively. Whenever possible, we computed confidence intervals for individualoutcomes.Numerous articles did not provide complete information about findings (e.g., 95 percentconfidence intervals; statistical significance values, or between-group data). In many cases,therefore, we had to calculate odds ratios, mean differences, or standardized mean differences,the relevant 95 percent confidence intervals, and p-values. In all such cases in which wecalculated data, we specify this in the Results chapter; information not specifically called out as“calculated” is taken from the original articles.Grading Strength of Evidence for Individual Comparisonsand OutcomesWe graded the strength of evidence based on the guidance established for the AHRQEvidence-based Practice Center program. 32 Developed to grade the overall strength of a body ofevidence, this approach incorporates four key domains: study limitations (includes study designand aggregate quality), consistency (similar magnitude and direction of effect), directness(evidence links interventions directly to outcome of interest for the review), and precision of theevidence (degree of certainty surrounding an effect estimate based on sample size and number ofevents). In addition, the evidence may be rated as lower strength for bodies of evidence withsuspected reporting bias from publication, selective outcome reporting, or selective analysisreporting. Regardless of the specific risk of bias of observational studies, this approach tograding the evidence assigns observational studies a grade of high study limitations, which thenleads to low strength of evidence. The strength of evidence from observational studies can berated as higher for observational studies for scenarios such as a strong dose-response association,plausible confounding that would decrease the observed effect, and a high strength of association(magnitude of effect). We evaluated optimal information size criteria to make judgments aboutprecision based on guidance from Guyatt and colleagues 33 and based our grades on low ormedium risk-of-bias RCTs or observational studies unless none were available.Our approach is consistent with current strength of evidence guidance developed by GRADEand AHRQ EPCs. The GRADE guidance explicitly discourages the inclusion and averaging ofrisk of bias across studies with different underlying risk-of-bias criteria. Rather, it suggestsconsidering including only studies with a lower risk of bias. 34 Likewise, the AHRQ EPCguidance notes that reviewers may focus “strength of evidence on the subset of studies that providethe least limited, most direct, and most reliable evidence for an outcome or comparison, after analysis32, p. 20of all the evidence.”Table 5 describes the grades of evidence that can be assigned. 35 Grades reflect the strength ofthe body of evidence to answer the KQs on the comparative effectiveness, efficacy, and harms ofthe interventions examined in this review. Two reviewers assessed each domain for each keyoutcome resolved any differences by consensus discussion or referral to a third, senior memberof the team. We graded the strength of evidence for the outcomes deemed to be of greatestimportance to decisionmakers and those commonly reported in the literature; we did not gradethe strength of evidence for KQ 1 (on components and features of MTM services). The gradesdescribed in Table 5 describe the state of evidence (which may demonstrate benefit, harm, or noeffect) and the confidence in the stability of that state. An insufficient grade is not a statementabout lack of efficacy or effectiveness; rather it is a statement about the lack of convincingevidence on benefit, harm, or lack of effect.18


Table 5. Definitions of the grades of overall strength of evidenceGradeDefinitionHighHigh confidence that the evidence reflects the true effect. Further research is very unlikelyto change our confidence in the estimate of effect.Moderate Moderate confidence that the evidence reflects the true effect. Further research maychange our confidence in the estimate of the effect and may change the estimate.LowLow confidence that the evidence reflects the true effect. Further research is likely tochange our confidence in the estimate of the effect and is likely to change the estimate.Insufficient Evidence either is unavailable or does not permit estimation of an effect.Assessing ApplicabilityWe assessed applicability of the evidence following guidance from the “Methods Guide forEffectiveness and Comparative Effectiveness Reviews.” 36 We used the PICOTS framework toexplore factors that affect applicability. Some factors identified a priori that may limit theapplicability of evidence include the following: age and health status of enrolled populations,health insurance coverage and access to health care, and complexity and intensity of the MTMintervention.Peer Review and Public CommentaryThis report received extensive external peer review and was posted for public commentDecember 2, 2013, to January 6, 2014. Comments were received from five peer reviewers andfour TEP members. In addition, we received public comments from eight individuals andprofessional organizations. We addressed all comments in the final report, making revisions asneeded; a disposition of comments report will be publicly posted 3 months after release of thefinal report.19


ResultsIntroductionThis section of this comparative effectiveness review (CER) on medication therapymanagement (MTM) first presents the results of the literature searches. We then document theresults for each Key Question (KQ). KQ 1 describes MTM intervention characteristics. KQ 2presents evidence on the effectiveness of MTM interventions, focusing on intermediateoutcomes, then patient-centered (health) outcomes, and then use of health care resources or costs.The presentation of KQ 3 summarizes the evidence by intervention components andimplementation features; KQ 4 summarizes evidence by patient characteristics. KQ 5 examinesthe evidence on harms of MTM programs. Appendix E has two parts pertaining to these KQs:the first part has the lengthy descriptions of the design of all included studies (for KQ 1); thesecond presents the evidence tables, organized by outcome, for the remaining KQs.Generally, for KQs 2 through 5, the text gives key points and the related strength of evidencegrades, followed by a detailed synthesis of the relevant studies. We also present pairs of tablesfor each outcome. One gives basic summary information about the results of included studies,indicating whether the quantitative data had been what the investigators reported or werecalculated by us. The other table in these sets documents the strength of evidence grades formajor outcomes (showing the ratings for required domains and, in a small number of cases, anyratings for optional domains). Appendix F contains the tables documenting how we arrived atrisk-of-bias assessments for individual studies.Most data can be found in tables and are not repeated in text. As noted in Methods, we focuson studies of low or medium risk of bias; when we need to summarize information for studies ofhigh risk of bias, we note the principal problems leading to that rating.Finally, our inclusion criteria for study designs were expansive and included randomizedcontrolled trials (RCTs) and a variety of observational studies (nonrandomized controlled trials,cohort studies and the like). We use “studies” to refer to all types of investigations; we specifyRCTs (or non-RCTs) as appropriate.Results of Literature SearchesFigure 2 presents our literature search results. Literature searches through January 9, 2014,for the final report, identified 2,516 unduplicated citations. Appendix A provides a list of allsearch terms used and the results of each literature search. In addition, we identified 233publications through grey literature searches, suggestions from technical experts or publiccomments received during topic refinement, or hand searches of included studies, or ScientificInformation Packets (SIPs). After applying our eligibility and exclusion criteria to titles andabstracts of all 2,749 identified citations, we obtained full-text copies of 419 published articles.We reapplied our inclusion criteria and excluded 358 of these articles from further review beforedoing the risk-of-bias assessment. Appendix C provides a list of excluded studies and reasons forexclusion at the full-text stage. The 61 articles included after full-text review represent 44studies. Evidence tables for these 44 studies are provided in Appendix D.20


Figure 2. Disposition of articles on medication therapy management (PRISMA figure)Number of records found through databasesearching after duplicates removed2,516MEDLINE ® : 1,709IPA: 508Cochrane Library: 299Number of additional records identified throughother sources233Hand searches of references: 90Gray literature: 7Recommended by TEP or public comment: 5Scientific Information Packets 131Total number of records after duplicatesremoved2,749Number of records screened2,749Number of records excluded2,330Number of full-text articles assessed foreligibility419Number of articles (studies) included inqualitative synthesis of systematic review61 (44)Number of studies included in quantitativesynthesis of systematic review6Number of full-text articlesexcluded, with reasons358Ineligible publication type 75Ineligible population 8Ineligible intervention 169Ineligible design 69Ineligible comparator 8Ineligible outcomes 10Ineligible setting 14Insufficient information to 5determine eligibilityAbbreviations: IPA = International Pharmaceutical Abstracts; PICOTS = populations, interventions, comparators, outcomes,timing, settings; TEP = technical expert panel.The Agency for Healthcare Research and Quality (AHRQ) Effective Health Care ProgramScientific Resource Center placed the request for scientific information packets (SIPs) in theFederal Register on September 16, 2013; it posted them for 30 days. We included relevant datafrom responses to this request.Table 6 summarizes study characteristics of included studies. Overall, 28 (77.8 percent) ofincluded studies were conducted in the United States, and 16 (44.4 percent) were conducted priorto the 2003 Medicare Modernization Act, which established the framework for Medicare Part DMTM programs. Just over half of included studies used an RCT design (either parallel or clustergroup), 3 (8.3 percent) used a nonrandomized controlled trial design, and the remaining studies(38.9 percent) used a cohort study design. Only 3 studies used an active treatment comparisongroup. Intermediate outcomes were the most commonly reported outcomes. Of the 36 studies, 1was considered low risk of bias (2.3 percent), 19 were considered medium risk of bias (43.221


percent), 16 (36.4 percent) were considered high risk of bias, and 8 (18.2 percent) had mixed riskof bias ratings, depending on outcome.Table 6. Characteristics of included studiesStudy characteristic N (%)CountryU.S. 35 (79.5)Non-U.S. 9 (20.5)Multiple 1 (2.3)Conducted prior to 2003 Medicare Modernization Act 19 (43.2)Study designRCT-parallel group 16 (36.4)RCT-cluster group 5 (11.4)Non-RCT 4 (9.1)Cohort study 19 (43.2)Used an active treatment comparison arm 5 (11.4)Outcomes measuredIntermediate outcomes (e.g., disease specific lab values, medication adherence,32 (72.7)drug therapy problems)Patient-centered outcomes (e.g., health outcomes, quality of life, patient satisfaction) 21 (47.7)Utilization and economic outcomes 33 (75.0)Risk of biasLow 1 (2.3)Medium 19 (43.2)High 16 (36.4)Mixed 8 (18.2)Abbreviations: N = number; RCT = randomized controlled trial; U.S. = United States.Key Question 1: Components and Implementation Featuresof MTM InterventionsKQ 1 was designed to synthesize descriptive findings regarding MTM interventioncomponents and implementation features, which have been identified as important factors relatedto effectiveness of these interventions. Because this report is a CER, our study inclusion criteriaincluded a requirement for a control or comparison arm. For that reason, our synthesis ofdescriptive findings related to MTM components and implementation features is limited toinvestigations that comparatively evaluated MTM; that is, it does not include all studies of MTMinterventions, many of which we had excluded because of the lack of a comparison arm. Thus,our findings represent a somewhat circumscribed lens for the descriptive part of this review.Synthesizing intervention components and implementation features across this body ofevidence was challenging. Mainly, studies did not consistently describe the interventioncharacteristics or implementation features in sufficient detail to allow us to determine the extentto which certain components were used, at which intervals, and at what intensity. Even studiespublished after the 2003 Medicare Modernization Act, which formalized some aspects ofpharmaceutical care, lacked sufficient reporting detail in many cases.22


Overall Descriptors of Study InterventionsTable 7 specifies the components and implementation features from our analytic framework(Figure 1 in Introduction). It also gives our assessment of the suitability or feasibility ofsynthesis, based on information available in the included studies across the entire evidence base.Table 7. Characteristics of medication therapy management interventionsCharacteristic of the MTM Intervention(Specified in Analytic Framework inSummarize in Tables Summarize in Tableand Synthesize With but Not SynthesizeIntroduction)CountsWith CountsMode of delivery Yes NA NAType of professional providing services Yes NA NAFrequency and interval of followup Yes NA NASpecific MTM components NA NA YesFidelity of implementation NA NA YesGoals of therapy established andNA NA YescommunicatedType of setting Yes NA NAMethod of patient enrollment NA Yes NALevel of integration with usual care NA Yes NAReimbursement characteristics Yes NA NAHealth system characteristics Yes NA NAAbbreviation: MTM =medication therapy management, NA = not applicable.Neither Summarize inTables nor SynthesizeWith CountsIn the best case, we can summarize data in tables and synthesize the information with actualcounts across the body of evidence. This is true for mode of delivery, type of professional givingthe services, details about followup, settings, modes of reimbursement, and characteristics ofhealth systems. Somewhat less can be done with methods for enrolling patients and level ofintegrating MTM with usual care, so information is just included in study-level summary tables(but not synthesized with actual counts across the body of evidence). Finally, information onspecific MTM components, fidelity of implementation, and MTM goals was so inconsistent orsparse that we could not either synthesize or include information in summary tables.Table 8 summarizes the intervention characteristics and features that were reportedconsistently enough to be synthesized with counts and frequencies—namely, those in Table 7with an X in the first column. It also notes whether the investigators used the phrase“pharmaceutical care” or the phrase “medication therapy management” to refer to the programtested. For details about intervention frequency and interval of followup, the information inTable 8 is “as designed” (i.e., however, the investigators described their initial intentions).23


Table 8. Characteristics of medication therapy management studies by type of patient population(broad focus or narrow focus on conditions or diagnoses)Broad NarrowOverallMTMFocus FocusCharacteristic of the Intervention(N =44)Intervention(N=34) (N=10)N (%)N (%) N (%)Phrase used todescribeinterventionMedication therapy management 20 (46) 18(53) 2 (20)Pharmaceutical care 14 (32) 9 (27) 5 (50)Other 10 (23) 7 (21) 3 (30)Mode of delivery Face-to-face only 22 (50) 14 (41) 8 (80)Telephone only 9 (21) 9 (27) 0Mixture of face-to-face and telephone 11 (25) 10 (29) 1 (10)Not reported 2 (5) 1 (3) 1 (10)Professional Pharmacist as interventionist 44 (100) 34 (100) 10 (100)Frequency offollowup asdesignedOne time with followup as needed 8 (18) 8 (24) 0Two times 7 (16) 7 (21) 0Three times 7 (16) 5 (15) 2 (20)Every 4 to 8 weeks for between 4 and 24 months 6 (14) 1 (3) 5 (50)Varied based on trigger (e.g., refill, physician visit, continuous 3 (7) 2 (6) 1 (10)enrollment for certain duration)Not reported 13 (30) 11 (32) 2 (20)Clinical settings Community pharmacy 7 (16) 3 (9) 4 (40)Centralized pharmacy 10(23) 10 (29) 0Outpatient medical clinic 16 (36) 11 (32) 5 (50)Home visits 1 (2) 1 (3) 0Integrated health system 2 (5) 2 (6) 0Multiple settings 8 (18) 7(21) 1 (10)ReimbursementcharacteristicsHealth systemcharacteristicsServices provided through Medicare Part D benefit 10 (23) 10 (29) 0Services provided through some other health plan benefit 5 (11) 3 (9) 2 (20)Services provided through study-related funding 3 (7) 3 (9) 0Reimbursement details not reported 26 (59) 18 (53) 8 (80)Single payer system (outside U.S.) 8 (18) 5 (15) 3 (30)Academic medical center 5 (11) 3 (9) 2 (20)Integrated health system 13 (30) 13 (38) 0Health plan 9 (21) 7 (21) 2 (20)Pharmacies independent of medical care system or health 3 (7) 0 3 (30)plansOther 6 (14) 6 (18) 0Abbreviations: N = number; U.S. = United States.During our abstraction process, we identified two distinct categories of interventions. Onecategory, of 34 studies, used a broad pharmaceutical care approach or MTM intervention inserving their patient populations; that is, they were not designed to focus specifically on any onedisease or clinical condition as part of the intervention. We refer to these studies in the reviewand Table 8 as “broadly focused.” Many of the studies in this category used retrospective designsof existing MTM programs. In these studies, the focus of the MTM intervention may have beenbroad, but the study may have restricted the evaluation of the MTM program to a patientpopulation with a specific condition or disease. The other category, with 10 studies, involvedinterventions evaluated in the context of a single chronic condition (e.g., chronic heart failure,diabetes) or provided in a highly specialized setting (e.g., specialized HIV/AIDS communitypharmacies). In these studies, the investigators implemented a pharmaceutical care approach orMTM intervention that attended to the patient’s complete drug therapy regimen, but the focus ofcomponent interventions (e.g., education, counseling, care coordination) and outcomes measured24


may have been specific to certain diseases or conditions. We refer to these studies as “narrowlyfocused.”In many cases, to distinguish narrowly focused MTM studies from case- or diseasemanagementinterventions, we had to contact study authors to clarify that their interventionincluded a comprehensive drug therapy assessment and drug therapy intervention beyond thesingle target condition of interest. The distinction between these broad-focus and narrow-focuscategories may be important for interpretation of the effectiveness of these types of interventions.Studies included in this review used “medication therapy management” to describe theintervention (Table 8) in only 18 of the 44 studies. With respect to mode of delivery (Table 8), 9broadly focused studies used only telephone contact; 37-48 by contrast, no narrowly focusedstudies used only telephone contact. Eleven studies (10 broad, 1 narrow) used a mixture of faceto-faceand telephone contact. 49-62 The studies using a mixture of modes often used face-to-facedelivery for the initial consultation and did followup contacts by telephone. Except for the 2studies that did not report mode of delivery, 63-65 the remaining studies used only face-to-facedelivery in pharmacies, clinics, or homes.All included studies used a pharmacist as the interventionist (Table 8). In some studies,however, the interventionist was described as a community pharmacy resident or ambulatorycare pharmacy resident, and in a few studies nonpharmacist staff performed initial interventioncomponents, such as interviewing patient or reviewing records to compile drug history for thepharmacist.Table 8 also summarizes the intervention frequency and interval of followup as designed, notas may have actually occurred, and these features also differed across studies. Of the 44 includedstudies, however, 13 (30 percent) did not report the designed frequency of contact and interval offollowup. Only 5 studies reported on the actual frequency and interval of followup. 49,51-54,58,66-69Studies evaluating real-world experience with these types of interventions often included aminimum contact threshold for inclusion of patients in the data analysis, but the interventionduration and interval of followup was open-ended and determined by clinical need, as is typicalin real-world practice.Included studies provided interventions in a variety of clinical settings including communitypharmacies, centralized pharmacies or call centers, outpatient medical clinics, and some usedhome visits (Table 8). Half of the narrowly focused interventions were delivered exclusively inan outpatient medical clinic. 59,70-75Concerning reimbursement, of the 44 studies in the evidence base, 26 (59 percent) did notreport on reimbursement for MTM services at all. Of the remaining studies, 15 reported thatpharmaceutical care or MTM was a covered benefit to patients; pharmacist services werereimbursed through an existing mechanism (e.g., Medicare Part D or other health care benefit). 37-44,46-48,51-54,57,62,76-80 Three studies clearly indicated that pharmacist services were reimbursedthrough pilot, grant, health system, or study-related funding. 49,58,81Finally, the context of the MTM services also varied in terms of features of the health systemor organization in which they were provided. Academic medical centers, integrated health caredelivery systems, health plans, and single payer health care systems outside the United Stateswere all represented in this evidence base.Study-Level Descriptors of InterventionsIn Appendix D, we have provided study-level summaries to describe the includedinterventions. Those tables (Table D-3 and D-4) document: interventions and the amount of25


integration with usual practice; method of identifying patients for receipt of MTM services;setting, mode of delivery, frequency and interval of followup; and health care system andreimbursement context. Table D-3 describes the 34 broad-focus studies; Table D-4 describes theremaining 10 narrow-focus studies (and additionally specifies the particular focus). Wesummarize the main elements in text below.Of the 10 narrow-focus studies, 2 addressed chronic heart failure and 3 addressedhypertension or hypertension and diabetes. The remaining studies focused on patients with:depression and anxiety, diabetes alone, glucocorticoid-induced osteoporosis, HIV/AIDS, andend-stage renal disease on hemodialysis.The 14 studies described as pharmaceutical care were generally based on the pharmaceuticalcare model as initially described by Strand and associates and further refined by the profession ofpharmacy practice. 51-54,63-65,70-74,79,82-89 Interventions termed medication therapy management(i.e., MTM) were often based on criteria defined for the Medicare Part D program, whichincludes elements of the pharmaceutical care model. 37-49,57,58,60,62,66-69,76,77,80,81,90 The remaininginterventions included elements of pharmaceutical care or MTM but did not specifically label theintervention as either one or the other. 50,55,56,59,61,75,78,91-93 These studies were often described as“clinical pharmacist interventions.”We defined the level with which pharmaceutical care or MTM services were integrated withusual care as having two main elements: (1) the degree of access that the interventionist had toclinical information in the patient’s medical record, such as laboratory results, diagnoses, andprogress notes and (2) the ease of access and method and process of communication between theinterventionist and prescribers. Providing MTM services within an outpatient medical clinic,presumably where the patient is also receiving medical care, is one such marker of integration,particularly when the study indicated that the pharmacist was part of a multidisciplinary careteam. Some studies, however, described the pharmacy or pharmacist simply as co-located in amedical clinic. In these instances, we do not know whether the level of integration with medicalcare would be any higher than if the pharmacist had been located in a community pharmacy.Thus, we could not rely solely on clinical setting as a marker of integration with usual care.Because many studies did not provide sufficient details regarding specific components of theintervention, whether termed pharmaceutical care, MTM, or clinical pharmacist intervention, wewere unable to synthesize the use of specific intervention components beyond the componentswe required for study inclusion.Only four studies used an active treatment comparator group. 57,58,69,88 All other studies(regardless of focus) compared pharmaceutical care or MTM with usual medical or pharmacycare or both. This factor also impeded our assessing the effectiveness of individual interventioncomponents. Furthermore, almost no study reported on the fidelity with which interventioncomponents were delivered (relative to the original design or intention), including whether goalsof drug therapy were established and communicated.The methods by which patients were identified and offered pharmaceutical care or MTMservices has been proposed as a moderator of effectiveness; the aim is to target patients mostlikely to benefit. These factors may include, for example, patients using drugs with narrowtherapeutic windows, complex drug regimens, or patient characteristics such as age, cognitivestatus, or social situation. With respect to data sources that studies used to identify and thenenroll patients for services, pharmacy prescription records (at a community pharmacy, clinic, orhealth plan) were the most common source. Except for the studies evaluating Medicare Part DMTM programs, few studies used the same criteria for identifying patients for enrollment. Most26


equired either some degree of regimen complexity, such as the number of drugs taken or use ofone or more drugs considered high risk for adverse events. Most studies using pharmacy data orclaims mailed or telephoned eligible patients to provide information about enrollment in anMTM program. For Medicare Part D MTM programs, “opt out” is another variation ofenrollment for these services. Patients meeting eligibility criteria are enrolled for services unlessthey specifically “opt-out.” Some studies relied solely on provider referral, patient self-referral,or routine medical record screening at time of a provider visit to identify patients for services.Tables E-1 and E-2 also provide study-level detail on intervention setting, mode of delivery,frequency and interval of followup and health care system and reimbursement characteristics,which were summarized overall in Table 8 and in the preceding section.Key Question 2: Effect of Medication Therapy ManagementInterventions on Intermediate, Patient-Centered, andResource Utilization OutcomesWe present below key findings and a detailed synthesis of intermediate, patient-centered, andresource utilization outcomes separately. (These outcomes were specified in Table 1 of theIntroduction.) When possible (a minimum of three reasonably similar studies for a givenintervention or outcome), we pooled study results and document those findings below. Whenstudies were too heterogeneous to pool, we present effect sizes for individual studies wheneverpossible in summary tables for each outcome that was reported in two or more studies. We alsoprovide strength of evidence tables to support our findings.Because in many cases the investigators did not report a full set of findings that comparedchanges over time between intervention and comparisons groups or other details that wouldpermit full analysis, we calculated various statistics ourselves. In these cases, we present in thetables below only these calculated findings and related statistical levels, and we note thisexplicitly in the tables or text (as “calculated”). The underlying data from the study article(s) canbe found in the evidence tables in Appendix D.Key Points: Intermediate Outcomes• Evidence was insufficient to evaluate the effect of MTM on anticoagulation after 12months due to an imprecise, single RCT body of evidence with medium limitations.• Evidence was insufficient to evaluate the effect of MTM on hemoglobin A1C after 6 to12 months due to an inconsistent and imprecise body of evidence from two RCTs withmedium limitations and two observational studies with high study limitations.• Evidence was insufficient to evaluate the effect of MTM for decreasing low-densitylipoprotein (LDL) cholesterol after 6 to 24 months due to an imprecise, single RCT bodyof evidence with medium limitations and an imprecise observational body of evidence oftwo studies with high limitations.• Evidence was insufficient to evaluate the effect of MTM for reducing blood pressure(BP) after 4 to 12 months based on direct, but inconsistent and imprecise, findings from asingle RCT and two cohort studies with medium limitations.• Several studies did not report outcomes such as drug therapy problems identified andresolved for both intervention and control groups. As a result, limited evidence addressesthe effectiveness of MTM compared with usual care in improving these importantintermediate outcomes. Study limitations, inconsistency, and lack of precision led us to27


conclude that the evidence is insufficient to judge the effectiveness of MTM in improvingthese outcomes when compared with usual care.• We found low strength of evidence that MTM had an effect on the percentage of peopleadherent to at least 80 percent of prescribed doses and on the absolute percentage ofprescribed doses taken. Although these conclusions are based on inconsistent evidencewith primarily nonsignificant findings of effects and high study limitations, two largecohort studies showed consistent effects of MTM on adherence although with high studylimitations.• Evidence was insufficient to evaluate the effect of MTM on medication adherence (asmeasured by self-report) as a result of inconsistent and imprecise evidence. The numberof trials, consistency, and study limitations varied by specific adherence measure.• MTM increases the appropriate use of medications as measured by overall scores onappropriateness indices (low strength of evidence).• Evidence was low for benefit of MTM on medication dosing as a result of indirect, andprecise evidence from one trial with medium study limitations.Detailed Synthesis: Intermediate OutcomesAnticoagulationOne RCT (medium risk of bias) reported on the effects of a pharmaceutical care interventionon anticoagulation among patients in family medicine clinics in a rural community after 12months of followup. 85 This intervention was conducted with 81 patients at high risk formedication-related problems; however, this outcome was reported only for the four patients inthe intervention arm and the six patients in the control arm who were taking anticoagulants. Thepercentage of subjects who achieved a therapeutic international normalized ratio (INR) differedsignificantly between the intervention and control arms (100 percent versus 16.7 percent(p=0.048); calculated odds ratio [OR], 32.94; 95% confidence interval [CI], 1.06 to 1,021.35).Because of imprecision (wide confidence intervals) and unknown consistency, we graded theevidence as insufficient to evaluate the effectiveness of MTM on improving therapeuticanticoagulation (Table 9).Table 9. Anticoagulation: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency Directness Precision(Analyzed)RCT 1; 81(10)MediumConsistencyunknown-singlestudyFindings andDirection Strength of[Magnitude] of EvidenceEffectDirect Imprecise TherapeuticINR achieved:100% vs.16.7%,p=0.048Abbreviations: INR = international normalized ratio; RCT = randomized controlled trial; vs. = versus.InsufficientHemoglobin A1cTwo RCTs and three cohort studies reported on outcomes related to hemoglobin A1c(HbA1c) among patients with diabetes (Table 10). One RCT (medium risk of bias) reported nosignificant difference in mean HbA1c between intervention (pharmaceutical care) and controlpatients in an Australian outpatient hospital diabetes clinic at 6 months. 72 The other RCT28


(medium risk of bias) reported on changes in the percentage of patients with diabetes whoachieved a HbA1c of less than or equal to 7.5 percent at 12 months among patients at high riskfor medication-related problems seen in family medicine practices in a rural community. 85 Thepercentage of patients at goal did not differ significantly between intervention and control armsat baseline (23.1 versus 56. 3, calculated p=0.08) but was significantly different at followup (100versus 26.7, calculated OR, 56.455; 95% CI, 2.811 to 1,133.912. p=0.008).Table 10. Hemoglobin A1c: Summary of resultsStudyOutcome ReportedDesign/Risk of Study Arms N Analyzed a by Study and Time ResultsBiasPeriodClifford et al.2002 72RCT/MediumG1: Pharmaceutical careG2: Standard careG1: 48G2: 25Mean HbA1c at 6months.Calculated mean difference:-0.2095% CI: -0.927 to 0.527Taylor et al.,2003 85RCT/MediumBrummel et al.,2013 66 ;Soliman, er al.,2013 67 ;Ramalho deOliveira et al.,2010 68Cohortstudy/MediumJeong et al.,2009 38Cohort/MediumPindolia et al.,2009 42Cohortstudy/HighG1: Pharmaceutical careG2: Standard careG1: Opted into clinic-basedMTM programG2: Usual medical care(opted out of MTM program)G1: Participants in Part DMedicare MTM program(opted into MTM program)G2: Control subjects withoutPart D Medicare as theirprimary drug benefit butotherwise similar tointervention subjectsG1: Opted into a telephonebasedMTM programG2: Usual medical care(opted out of MTM program)G1: 13 a Percentage withG2: 16 a HbA1c at goal(defined as lessthan or equal to7.5%) at baselineand at 12 months.G1: 121G2: 103Percentage withHbA1c at goal(defined as lessthan 7%) after 12months ofdemonstrationPercentage withHbA1c at goal(defined as lessthan 7%) after 24months ofdemonstration (i.e.,12 months after endof demonstration)G1: 1,323 a Mean change (SD)G2: 1,141 a in HbA1c at 6monthsG1: NRG2: NRp=0.590Calculated OR: 56.455 95% CI:2.811 to 1,133.912, p=0.008At 12 months unadjusted:Calculated OR: 1.038 95% CI0.574 to 1.879, p=0.901Adjusted difference-indifferencecoefficient:2.44 95% CI 1.22 to 4.86,p=0.01At 24 months unadjusted:Calculated OR: 0.502 95% CI0.294 to 0.855, p=0.011Adjusted difference-indifferencecoefficient: NRCalculated mean difference:-0.020 (0.041)95% CI: -0.101 to 0.061p=0.628Percentage with Calculated OR:HbA1c less than 7% 1.142 95% CI: 0.969 to 1.347at 6 months p=0.114Change in G1: + 3percentage of G2: + 7patients with HbA1c Between-group p: inferred toless than 7% at 6 be NS, exact p: NRmonthsWithin-group p: NRa The study included more subjects than the number analyzed and reported in this column, but the investigators assessed thisoutcome only among patients with diabetes within each study arm.Abbreviations: CI = confidence interval; G = group; HbA1c= hemoglobin A1C or glycosolated hemoglobin, MTM = medicationtherapy management; NR = not reported; NS = not sufficient; OR = odds ratio; RCT= randomized controlled trial; SD = standarddeviation.29


One medium risk of bias and one high risk of bias (because of self-selection of participantsinto the intervention arm and control arms) cohort study were conducted primarily by telephonewithin large, integrated U.S. health care systems. 38,42 The remaining cohort study (medium riskof bias, described in three included publications) involved pharmacists providing MTM servicesto patients with diabetes within medical clinics that were part of a large, integrated U.S. healthcare system. 66-68 This study found a larger difference in the percentage of subjects achievingHbA1C less than 7 percent between baseline and the 12-month followup for the interventiongroup (adjusted difference-in-difference coefficient 2.44; 95% CI, 1.22 to 4.86), but thisdifference was not maintained 12 months after the end of the study intervention. 66-68 Thetelephone-based medium risk-of-bias cohort study reported no significant change in meanHbA1c or percentage of subjects achieving a HbA1C less than 7 percent at 6 months for theintervention group compared with the control group. 38 The high risk-of-bias telephone-basedcohort study found similar findings.Based on direct, but inconsistent and imprecise, evidence from two RCTs and twoobservational studies, all with medium limitations (Table 11), we concluded that the strength ofevidence is insufficient to evaluate the effectiveness of MTM interventions to improve meanHbA1c levels or increase the percent of patients achieving a goal HbA1c level.Table 11. Hemoglobin A1c: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency Directness Precision(Analyzed)Findings andStrength ofDirectionEvidence[Magnitude] of EffectRCT 2; 154 Medium Inconsistent Direct Imprecise One trial: no change in Insufficient(102)mean HbA1c at 6months. One trial:significantly greaterpercentage of patientswith HbA1c


481.915). These findings are quite imprecise, largely because of a sample size of only 19subjects in each group for this outcome.Table 12. LDL cholesterol: Summary of resultsStudyN of SubjectsDesign/Risk of Study ArmsAnalyzedBiasTaylor et al.,2003 85RCT/MediumBrummel et al.2013 66 ;Soliman et al.,2013 67 Ramalhode Oliveira et al.,2010 68Cohortstudy/MediumJeong et al.,2009 38Cohort/MediumG1: Pharmaceutical careG2: Standard careG1: Opted into clinic-basedMTM programG2: Usual medical care(opted out of MTMprogram)G1: Participants in Part DMedicare MTM program(opted into program)G2: Control subjectswithout Part D Medicare astheir primary drug benefitbut otherwise similar tointervention subjectsIsetts et al., G1: MTM services provided2008 81by health plan in existingCohort study/High medical care clinics incollaboration with primarycare providers.G2: Usual medical carewithout MTMPindolia et al., G1: Opted in to a2009 42telephone-based MTMCohort study/High programG2: Usual medical care(opted out of MTMprogram)Outcome Reportedby Study and TimePeriodFollowup (N Percentage ofinferred frompercentage inresults)G1: 18 aG2: 17 aG1: 121G2: 103patients at LDL-C goalbased on ATPIIIcriteria at 12 months.Percentage with LDL-C at goal (defined asless than 100 mg/dl)after 12 months ofdemonstrationPercentage with LDL-C at goal (defined asless than 100 mg/dl)after 24 months ofdemonstration (i.e., 12months after end ofdemonstration)G1: 1,515 a Mean change (SD) inG2: 1,323 a LDL cholesterol at 6monthsG1: 128G2: 126G1: NR aG2: NR a(outcomeassessed onlyamong patientswith coronaryartery disease)Percentage with LDLcholesterol at goal(


Table 12. LDL cholesterol: Summary of results (continued)StudyN of SubjectsDesign/Risk of Study ArmsAnalyzedBiasFox et al. 2009 37Cohort study/HighG1: MTM program providedthrough a health planG2: Usual medical care(eligible but opt-out fromMTM program)G1: 255G2: 56G1: 215G2: 46Outcome Reportedby Study and TimePeriodResultsPercentage of Calculated OR: 2.228,patients with diabetes 95% CI: 1.238 to 4.008;with LDL-C less p=0.008than100 mg/dl at 12 to24 monthsMean (SD) LDL-C at12 to 24 monthsCalculated meandifference: -7.495% CI: -17.297 to 2.497p= 0.33 as reported bystudy authors, p=0.143 ascalculateda The investigators assessed this outcome only among patients with hyperlipidemia, diabetes, or coronary artery disease withineach study arm but did not report the specific number analyzed.Abbreviations: ATPIII = Adult Treatment Panel III (Expert Panel on Detection, Evaluation, and Treatment of High BloodCholesterol); CI = confidence interval; G = group; HEDIS = Healthcare Effectiveness Data and Information Set; LDL-C = lowdensity lipoprotein cholesterol; MTM = medication therapy management; N = number; NR = not reported; OR = odds ratio; RCT= randomized controlled trial.We also evaluated the findings from the five cohort studies. One of the medium risk-of-biascohort studies involved pharmacists providing MTM services to patients with diabetes withinmedical clinics that were part of a large, integrated U.S. health care system in Minnesota. 66-68This study found no significant difference in the percentage of patients achieving an LDLcholesterol goal (defined as less than 100 mg/dl) at 12 months (adjusted difference-in-differencecoefficient 1.95; 95% CI, 0.81 to 4.84; p=0.13). The other medium risk-of-bias cohort studyinvolved a telephone-based MTM program delivered within a large, integrated U.S. health caresystem in California; the analysis presented was limited to patients with a diagnosis ofhyperlipidemia, diabetes, or coronary artery disease. 38 This study found a small but significantmean decrease in LDL cholesterol levels in the intervention group compared with controls at 6months (calculated mean difference, -4.1; 95% CI, -6.019 to -2.181; p< 0.001) and also found asignificant increase in the percentage of patients achieving an LDL goal, defined as LDL lessthan 100 mg/dl (calculated OR, 1.392; 95% CI, 1.160 to 1.670; p


Table 13. LDL cholesterol: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency Directness Precision(Analyzed)RCT 1; 81(38)Observational2; 3,062(3,062)MediumConsistencyUnknown—single studyFindings andDirection Strength of[Magnitude] of EvidenceEffectDirect Imprecise Significantlygreaterpercentage ofpatients at LDL-C goal in MTMgroup at 12months (77.8%vs. 5.9%,p


Table 14. Blood pressure: Summary of resultsStudyOutcome Reported byDesign/Risk of Study Arms N Analyzed a Study and TimeBiasPeriodTaylor et al.,2003 85RCT/MediumPark et al., 1996 89RCT/HighPlanas et al.,2009 90RCT/HighBrummel et al.,2013, 66Soliman et al.,2013, 67 Ramalhode Oliveira et al.,2010 68Cohortstudy/MediumG1: Pharmaceutical careG2: Standard careG1: Communitypharmacypharmaceutical careprogramG2: Usual careG1: 24 a Percentage of patientsG2: 29 a with SBP and DBP atgoal at 12 months.G1:23G2:26G1: Community G1: 25pharmacy hypertension G2: 15MTM program forpatients with diabetesG2: Control group (BPrecorded, informed of BPgoals at 3 times duringstudy)G1: Opted into clinicbasedMTM programG2: Usual medical care(opted out of MTMprogram)G1: 121G2: 103ResultsCalculated OR: 28.875,95% CI: 5.486 to 151.993,p


Table 14. Blood pressure: Summary of results (continued)StudyOutcome Reported byDesign/Risk of Study Arms N Analyzed a Study and TimeBiasPeriodJeong et al.,2009 38Cohort/MediumCarter et al.,997 70Barnette et al.1996 71Cohort study/HighIsetts et al.,2008 81Cohort study/HighG1: Participants in Part DMedicare MTM programG2: Control subjectswithout Part D Medicareas their primary drugbenefit but otherwisesimilar to interventionsubjectsG1: Pharmacy-basedpharmaceutical careG2: Usual medical careG1: MTM servicesprovided by health planin existing medical careclinics in collaborationwith primary careproviders.G2: Usual medical carewithout MTMG1: 1301G2: 982(Studyincluded moresubjects butthis outcomewas assessedamong onlypatients withdiabetes andHTN withineach studyarm)Percentage with BPcontrol (defined as


Blood Pressure Goal AttainmentOne RCT (medium risk of bias), conducted among of a small number of patients at high riskof medication-related problems receiving pharmaceutical care through family medicine clinics ina rural Alabama community, reported a significant difference in the number of patients at bloodpressure goal (SBP ≤ 140 mm Hg and DBP ≤ 90 mm Hg) at 12 months (91.7 percent versus 27.6percent, calculated OR 28.875, 95% CI 5.486 to 151.993, p< 0.001). 85 The other RCT (high riskof bias due to high attrition, lack of intention to treat analysis, and no adjustment for baselinedifferences) provided MTM services through community pharmacies to managed careorganization enrollees with diabetes and hypertension in Oklahoma. 90 This trial also reported afavorable effect of MTM on the achievement of blood pressure goals (OR 12.9, 95% CI, 1.5 to113.8; p=0.021). The last trial (high risk of bias for unclear randomization methods, importantdifferences in baseline with no adjustment in analysis, and other factors related to studyexecution) reported a direction of effect favoring the MTM group, but it was not statisticallysignificant. 89The two medium risk-of-bias cohort studies included a telephone-based MTM programwithin a large, integrated U.S. health care system in California. 38 The other study involved MTMprovided by pharmacists in medical clinics within a large, integrated U.S. health care system inMinnesota. 66-68 Both of these studies showed a directional effect favoring the control groups onthe percentage of subjects who achieved blood pressure control at 6 and 12 months, but thesefindings were not statistically significant (calculated OR 0.953, 95% CI, 0.808 to 1.125;p=0.571, adjusted difference in difference coefficient, 0.73; 95% CI, 0.32 to 1.65; p=0.45). Thetwo other cohort studies (both high risk of bias) reported findings that were directionallyconsistent with the trials, but findings were statistically significant in only one of thestudies. 70,71,81Systolic and Diastolic Blood Pressure LevelsThree studies reported on systolic blood pressure outcomes, and all were rated as high risk ofbias. One RCT provided MTM services through community pharmacies to managed careorganization enrollees with diabetes and hypertension in Oklahoma. 90 The MTM group in thistrial had a mean decrease of 20.0 mmHg (95% CI, -32.7 to -7.4; p: 0.003) in systolic bloodpressure compared with the control group. The other RCT provided pharmaceutical care topatients with hypertension through community pharmacies in Illinois and Wisconsin. 89 Thisstudy found a mean decrease of 13.0 mmHg (95% CI, -23.739 to -2.261; p=0.018) comparedwith controls. 89 The cohort study found directionally similar results. 70,71 We found similar resultsfor diastolic blood pressure from the two high risk-of-bias studies that reported thisoutcome. 70,71,89 Overall, we concluded that the strength of evidence is insufficient for theeffectiveness of MTM interventions to increase the percentage of patients achieving a bloodpressure goal or decrease systolic blood pressure or diastolic blood pressure levels based ondirect but imprecise evidence from one RCT with medium limitations and two observationalstudies with high limitations (Table 15). In addition, the direction of effect was not consistentbetween the RCT and observational evidence (Table 15).36


Table 15. Improvement in blood pressure: Strength of evidenceNumber ofFindings andStudies; StudyStudy DesignSubjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 1; 81 Medium(53)Consistencyunknown—single studyDirect Imprecise OR 28.88 (95%CI 5.49 to151.99, p


Table 16. Drug therapy problems identified: Summary of resultsStudyDesign/Risk ofN Analyzed Outcome and TimeBiasStudy ArmsPeriodKrska et al.,G1: 1682001 91G2: 164RCT/HighG1: Pharmacist-ledmedication reviewG2: Usual care includingidentification ofpharmaceutical careissues, but no planNumber of drug therapyproblems identified foreach study arm at 3monthsResultsG1: 1,206G2: 1,380Welch et al., G1: MTM program G1: 459 Percentage with at least G1: 89.8%2009 44Cohort/Highprovided to home-based G2: 123beneficiariesG2: No MTM controlgroup (voluntary opt-out)1 potential drug therapy G2: 83.7%problem during MTMprocess (timing unclear) Risk difference=6.1%, calculatedp=0.062Abbreviations: G = group; MTM = medication therapy management; N = number; RCT = randomized controlled trial.Table 17. Drug therapy problems identified: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency Directness Precision(Analyzed)Cohort 1; 904 (582) High Consistencyunknown—single studyFindings andDirection[Magnitude] ofEffectIndirect Imprecise Riskdifference=6.1%,calculatedp=0.062Strength ofEvidenceInsufficientDrug Therapy Problems ResolvedIn all, we identified nine studies that attempted to report on whether MTM programs resolveddrug therapy problems that were identified. Of these, six studies provided information only fromthe intervention arm. 42,49,51,54,73,74,81,89 Thus, as with drug therapy problems identified, they cannotinform the question of the comparative effectiveness of MTM interventions. Three other studies(two RCTs, one cohort study) provided information on the effectiveness of MTM for resolvingdrug therapy problems when compared with usual care (Table 18). The cohort study (mediumrisk of bias) found a significant effect of MTM on the difference in drug therapy problemsidentified between baseline and a 12-month followup; the investigators interpreted the change innumber of drug therapy problems identified over time as drug therapy problems resolvedbetween baseline and followup. 39-41One RCT had a high risk of performance bias because of issues concerning site and countryspecificvariation, coupled with failure to control for differences at baseline and a high overallattrition. 64,65 The other trial shows higher total numbers of all drug therapy problems resolved inthe intervention arm, but without measures of variance that account for variation among patients,these results cannot be interpreted with confidence. 9138


Table 18. Total number of drug therapy problems resolved: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodMoczygemba et al.,2011 39 ;Moczygemba et al.,2011 40 ;Moczygemba et al.,2008 41Cohort/MediumKrska et al., 2001 91RCT/HighBernsten et al.,2001 64,65RCT/HighG1: Opt-in telephoneMTM programG2: No-MTM controlgroupG1: 60G2: 60G1: Pharmacist-led G1: 168medication review G2: 164G2: Usual care,including identificationof pharmaceutical careissues, but no planG1: Structuredpharmaceutical careprogram in communitypharmacyG2: Usual communitypharmacy servicesBaselineG1: 1,290G2: 1,1646 monthsG1: 1,024G2: 95312 monthsG1: 863G2: 76418 monthsG1: 704G2: 636Medication andhealth-relatedproblems identified atbaseline and 12monthsDrug therapyproblems wholly orpartially resolved at 3monthsNumber of changes intherapy at baselineNumber of changes intherapy at 6 monthsNumber of changes intherapy at 12 monthsNumber of changes intherapy at 18 monthsResultsParameter estimate forintervention group on predictingchange in medication healthrelatedproblems from multipleregression: 0.81Adjusted p: 0.01NOTE: Regression modeladjusted for the followingpredisposing and need factors:(1) age, (2) sex, (3) race, (4) Nof medications, (5) N of chronicdiseases, and 6) medicationregimen complexityG1: 998G2: 569BaselineCalculated mean difference:0.2,95% CI: 0.101 to 0.299,p


Medicare Part D plan outperforming usual care and Medicare Advantage Part D planunderperforming usual care. The authors note that the timing of measurement (one year afterMTM enrollment for intervention arms) may have allowed prescribers to add back problematicdrugs over the course of the year.Table 19. Specific drug therapy problems resolved: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodPerlroth et al.,2013 62aCohort/MediumCongestive heartfailureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD,receiving MTM withCMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD,receiving MTM withCMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD, receiving MTMwith CMRComparison –congestive heartfailureG13: enrolled in PDP,usual careG14: enrolled in MA-PD, usual careComparison - Chronicobstructive pulmonarydiseaseG15: enrolled in PDP,usual careG16: enrolled in MA-PD, usual careComparison - DiabetesG17: enrolled in PDP,usual careG18: enrolled in MA-PD, usual careG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Remove drug-druginteraction within 365days after date of MTMenrollment (forinterventions) orrandomly-assigneddate in 2010 (forcomparators)Discontinue use of highrisk medications within365 days after date ofMTM enrollment (forinterventions) orrandomly-assigneddate in 2010 (forcomparators)ResultsOdds (95% CI)Congestive heart failureG1 vs. G13: 0.87 (0.76,1.00), p>0.05G3 vs. G14: 1.05 (0.88,1.26), p>0.05Chronic obstructivepulmonary diseaseG5 vs. G15: 0.92 (0.79,1.07), p>0.05G7 vs G16: 1.11 (0.89,1.38), p>0.05Odds (95% CI)Congestive heart failureG1 vs. G13: 1.04 (.97,1.11), p>0.05G3 vs. G14: 0.93 (0.86,1.01), p>0.05Chronic obstructivepulmonary diseaseG5 vs. G15: 1.06 (0.99,1.13), p>0.05G7 vs. G16: 1.00 (0.92,1.09), p>0.05Discontinue use of Odds ratio (95% CI)medicationG1 vs. G13: 0.63 (0.58,contraindicated for 0.67), p


Table 19. Specific drug therapy problems resolved: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodChrischilles et al.,2004 79Cohort/HighG1: PCM-eligiblepatients who receivedPCM servicesG2: PCM-eligiblepatients who did notreceive PCM servicesG1: 524G2: 1,687Change in prevalenceof high-risk medicationuse 9 months afterbecoming eligible forPCMResultsG1: -10.8 percentagepoints;p


three different adherence outcomes: (1) the proportion of patients who, based on a threshold ofbetween 75 percent and 80 percent of prescribed doses taken, were deemed to be adherent 42,62,66-68,85 ; (2) the percentage of prescribed doses taken 40,45,89,90 ; and (3) the scores from an adherencescale score (such as the Morisky Scale). 50,64,86 Two studies assessed miscellaneous aspects ofmedication-taking behavior 56,84 these included “remembering to take medication,” a medicationtakingbehavior subscore, and or determining the number of medications (not pills) for which theparticipant’s reported manner of taking (number of pills and frequency per day) exactly matchedthe prescribed directions. When studies did not report statistical significance, we calculated thestandard difference in means, standard errors, and 95 percent confidence intervals based on rawdata.Table 21. Medication adherence: Summary of results grouped by type of adherence outcomeStudyOutcome and TimeOutcome Type Design/Risk of Study Arms N AnalyzedResultsPeriodBiasOutcome Type1: Proportion ofpatientsadherent basedon a thresholdof percentage ofpills takenTaylor et al.,2003 85RCT/MediumPindolia et al.,2009 42Cohort study/HighBrummel et al.2013 66 , Soliman,2013 et al. 67Ramalho deOliveira et al.,2010 68Cohortstudy/MediumG1:Pharmaceutical careG2: StandardcareG1: TelephonebasedMTMprogramG2: Patientseligible for MTMprogram whodeclinedenrollmentG1: Opted intoclinic-basedMTM programG2: Usualmedical care(opted out ofMTM program)G1:33G2:36G1: 292G2: 1,081(study year 1)G1: 121G2: 103Percentage of patientsadherent defined as selfreportedtaking 80% ormore of medications 12months after baselinePercentage of CHFpatients who were adherentto at least 75% ofACE/ARB medicationsbased on 2006 claims data:Measured during 6 monthspost-MTM enrollmentCalculated OR:9.27795% CI: 0.480 to179.263; p= 0.140Calculated OR:1.08895% CI: 0.834 to1.417p=0.533Percentage of CHF Calculated OR:patients who were adherent 1.174to at least 75% of beta 95% CI 0.89 toblocker medications based 1.54on 2006 claims data: p=0.252Measured during 6 monthspost-MTM enrollment cPercentage of patientsadherent to aspirin (frompharmacy claims data) atbaseline (before MTM), 12-month (during MTMdemonstration project), and24-month (1 year postdemonstration)BaselineCalculated OR2.828, 95% CI0.710 to 11.259,p=0.14)12 MonthCalculated OR5.981 (95% CI0.284 to 126.030,p=0.250)24 MonthCalculated OR1.17 (95% CI0.072 to 18.903,p=0.912)42


Table 21. Medication adherence: Summary of results grouped by type of adherence outcome(continued)StudyOutcome and TimeOutcome Type Design/Risk of Study Arms N AnalyzedResultsPeriodBiasOutcome Type Perlroth et al.,1: Proportion of 2013 62 apatientsadherent basedon a threshold Cohort/Mediumof percentage ofpills taken(continued)Congestive heartfailureG1: enrolled inPDP receivingMTM with CMRG3: enrolled inMA-PD,receiving MTMwith CMRChronicobstructivepulmonarydiseaseG5: enrolled inPDP receivingMTM with CMRG7: enrolled inMA-PD,receiving MTMwith CMRDiabetesG9: enrolled inPDP receivingMTM with CMRG11: enrolled inMA-PD,receiving MTMwith CMRComparison—congestive heartfailureG13: enrolled inPDP, usual careG14: enrolled inMA-PD, usualcareComparison—ChronicobstructivepulmonarydiseaseG15: enrolled inPDP, usual careG16: enrolled inMA-PD, usualcareG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Percentage of patientsachieving adherence (>80% of prescribed pillstaken) to various specifiedmedications365 days after date of MTMenrollment (forinterventions) or randomlyassigneddate in 2010 (forcomparators)CHFAdherent to anyevidence-basedmedicine (EBM)for CHFG1 vs. G13: 1.28 a(1.19, 1.37)G3 vs. G14: 1.40 a(1.29, 1.52)p


Table 21. Medication adherence: Summary of results grouped by type of adherence outcome(continued)StudyOutcome and TimeOutcome Type Design/Risk of Study Arms N AnalyzedResultsPeriodBiasOutcome Type1: Proportion ofpatientsadherent basedon a thresholdof percentage ofpills taken(continued)Perlroth et al.,2013 62 aCohort/Medium(continued)Comparison—DiabetesG17: enrolled inPDP, usual careG18: enrolled inMA-PD, usualcareG9: 1.27 a (1.19,1.36)G11: 1.20 a (1.12,1.29p


Table 21. Medication adherence: Summary of results grouped by type of adherence outcome(continued)StudyOutcome and TimeOutcome Type Design/Risk of Study Arms N AnalyzedResultsPeriodBiasOutcome Type2: Absolutemeasure ofadherence aspercentage ofprescribeddoses takenOutcome Type2: Absolutemeasure ofadherence aspercentage ofprescribeddoses taken(continued)Moczygemba etal., 2011 39Moczygemba etal., 2011 40Moczygemba etal., 2008 41RetrospectiveCohort/MediumPlanas et al.,2009 90RCT/HighPark, 1996 89RCT/highG1: Opt-intelephone Scott& White HealthPlan MTMprogramG2: No-MTMcontrol groupG1:Collaborativehome-basedmedicationreviewG2: Nomedicationreview receivedG1:ComprehensivepharmaceuticalcareG2: Usual careG1: 60G2: 60G1: 25G2: 15Visit 1G1: 7G2: 5Visit 2G1: 21G2: 23Visit 3G1: 23G2: 20Visit 4G1: 21G2: 22Percentage prescribed Calculated meandoses taken: Overall difference: -0.040average MPR across all 95% CI: -0.101 tomedications measured at 6 0.021months before MTM p=0.201participation (i.e., baseline)and 12 months after MTMusing pharmacy dataPercentage meanadherence (percentage ofprescribed doses taken) toantihypertensivemedicationMeasured twice (9 monthsbefore and 9 months afterbaseline visit) andcontinuously usingmedication acquisitionmethod, in which days’supply of medication iscompared with datesmedication was filled usingpharmacy refill data.Mean percentagecompliance (percentage ofprescribed pills taken) frompharmacist report of pillcounts4-month time frameCalculated meandifference frombaseline to 9months: 0.07795% CI: -0.127 to0.281p=0.46Calculated meandifference forchange frombaseline to Visit 4:-0.02395% CI: -0.175 to0.129 p=0.76745


Table 21. Medication adherence: Summary of results by type of adherence outcome (continued)StudyNOutcome Type Design/Risk Study ArmsOutcome and Time Period ResultsAnalyzedof BiasOutcome Type2: Absolutemeasure ofadherence aspercentage ofprescribeddoses taken(continued)Park, 1996 89RCT/HighG1:ComprehensivepharmaceuticalcareG2: Usual careVisit 1G1: 7G2: 5Visit 2G1: 21G2: 23Visit 3G1: 23G2: 20Mean percentagecompliance (percentage ofprescribed pills taken) frompharmacist report of pillcounts4-month time frameCalculated meandifference for changefrom baseline to Visit 4:-0.02395% CI: -0.175 to 0.129p=0.767Moore, 2013 45Cohort/MediumG1: MTMprogram (opt-in)G2: controlgroup (refusers)Visit 4G1: 21G2: 22G1: 2,250G2: 2,250Medication possession ratioby medication type frompharmacy data and medicalclaims data from 365 dayspreceding the patient’sprogram invitation date tos365 days following patient’sprogram invitation dateAsthma MPR (%)Calculated meandifference = 0.62; 95%CI: -2.988 to 4.228; p =0.736Depression MPR (%)Calculated meandifference = 1.16; 95%CI: -1.667 to 3.987; p =0.421Diabetes MPR (%)Calculated meandifference = 2.37; 95%CI: -0.549 to 5.289; p =0.112Dyslipidemia MPR (%)Calculated meandifference = 4.71; 95%CI: 2.747 to 6.673; p


Table 21. Medication adherence: Summary of results by type of adherence outcome (continued)StudyOutcome Type Design/Risk Study Arms N Analyzed Outcome and Time Period Resultsof BiasOutcome Type3: Self-reportedAdherenceusing MoriskyScaleBernsten,2001 64 ;Sturgess,2003 65Pooled sample(percentage adherent)OR at baseline: 0.82,calculated 95% CI:0.666 to 1.0, p = 0.050RCT/High(pooled data)Volume et al.2001 86 andKassam 87RCT/MediumG1: Structuredcommunitypharmacybasedpharmaceuticalcare programG2: UsualcommunitypharmacyservicesG1:Comprehensivepharmaceuticalcare servicesG2: Traditionalpharmacy carePooledsample(excludingTheNetherlandsbecause nobaselineadherencedatacollected)BaselineG1: 867G2: 74818 monthsG1: 792G2: 758Total: 292G1: NRG2: NRMedication adherence: selfreportedas assessed byMorisky Scale(Note: Percentage ofparticipants who wereadherent defined aspatients responded thatthey “never” experiencedany aspects ofnoncompliance on the 4-item scale with a 4-pointresponse option per item)Self-reported adherenceusing the Morisky Scalemade up of fourdichotomous items wheresummary score is 0–4 withlower scores being betteradherenceCalculated OR at 18months: 1.084, 95% CI:0.883 to 1.332, p=0.440Calculated meandifference0.09095% CI: -0.076 to 0.256p=0.289Outcome Type3: Self-reportedAdherence usingMorisky Scale(continued)Jameson et al.,1995 50RCT/High(Medium forstudy overallbut high foradherencebecause ofpoor outcomemeasure)G1:G1: 27Consultation G2: 29with a clinicalpharmacistwithin a primarycare office.G2: Standardmedical care atthe primary careoffice12 to 13 months afterinterventionChange in self-reported G1: -1.6composite scoreG2: -0.2“understanding and 95% CI: NRcompliance” over 6 months p: NSwith negative scorerepresenting improvement(no further information onmeasure used)47


Table 21. Medication adherence: Summary of results by type of adherence outcome (continued)StudyOutcome Type Design/Risk Study Arms N Analyzed Outcome and Time Period Resultsof BiasMiscellaneousAdherenceOutcomesHanlon et al.,1996 84G1: 86G2: 83Calculated OR: 1.076,95% CI: 0.527 to 2.197,p: 0.84RCT/Medium(Low for studyoverall butmedium foradherencebecause oflack ofinformationabout andprecision ofadherencemeasure)Sidel, 1990 56RCT/MediumG1: Usual care,plus clinicalpharmacist care.G2: Usual carein the GeneralMedicine ClinicG1: Received at G1: 92least 2G2: 104pharmacist visitsinvolvingmedicationreview, patientspecificeducation andcounseling;followup patienttelephone callsand contacting ofphysicians bypharmacists asneededG2: Contactedonly to completethe surveySelf-reported medicationcompliance with 12-monthtime frame, assessed bydetermining whether theway patients said they tooktheir medicine (in terms ofnumber of pills and dailyfrequency) matched howthe medication wasprescribed. Compliancewas defined as theproportion of medicationsfor which the patients’response agreed with thedirections.Change from baseline to 6- G1: -3.47month followup in G2: -4.38medication-taking behavior 95% CI: NRSubscore (negative scores P < 0.001 for withingroupdifferencesindicate improvement,which means decreased p = 0.52 for betweenrisk)group differencesChange at6 months in normativescore for remembering totake medicineG1: 0.09G2: -0.1995% CI: NRp=0.52a Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement forCMR. We have excluded groups 2, 4, 6, 8, 10, and 12 from our summary tables.Abbreviations: ACE = angiotension converting enzyme inhibitor; ARB = angiotension receptor blocker; CHF = congestive heartfailure; CI = confidence interval; CMR = comprehensive medication review; EBM = evidence-based medicine; G = group;HbA1C = hemoglobin A1C or glycosolated hemoglobin; LAAC = long-acting anticholinergic; LABA = long-acting beta agonist;MA-PD = Medicare Advantage Part D Plan; MPR = medication possession ratio; MTM = medication therapy management; N =number; NR = not reported; OR = odds ratio; PDP = Medicare Part D Plan; RCT= randomized controlled trial; SE = standarderror; T = time; vs. = versus.Of the 4 studies assessing the proportion of patients who achieved threshold adherencelevels, one was a small RCT (medium risk of bias); 85 the others were cohort studies: 1 small 66-68(medium risk of bias) 1 relatively large (high risk of bias), 42 and 1 very large medium risk ofbias). 62 Only 1 of these studies found statistically significant positive effects of MTM onadherence 62 and did so for adherence to some but not comparisons for multiple medications. Ofthe 4 studies that assessed MTM effects on percentage of prescribed doses taken, 2 were smallRCTs (both high risk of bias); 89,90 the other 2 were cohort studies, one small, one large (bothmedium risk of bias). 40,45 Only 1 of these studies found a statistically significant positive effectof MTM on adherence. All 3 studies that assessed adherence using self-reported adherence scales48


were small RCTs (1 medium risk of bias 86 ; 2 high risk of bias 50,64 ). None found a statisticallysignificant effect of MTM on adherence, although 1 high risk-of-bias study 64 did not account forthe marked baseline differences and, hence, may have missed a statistically signficant differencein change in adherence. This same study (high risk of bias) reported a statistically significantincrease in the percentage of individuals who changed from nonadherent to adherent over 18months (15.25 percent in the intervention group and 12.2 percent in the control group;p=0.028); 64 however, this assessment did not take into account the percentage in each group thatchanged from adherent to nonadherent. Finally, the 2 RCTs (both medium risk of bias) thatassessed miscellaneous aspects of adherence found no statistically significant differencesbetween groups in adherence outcomes assessed. 56,84 Hence, of the 13 studies that assessedeffects of MTM on adherence, 2 large cohort studies that used an objective adherence measurefound a statistically significant positive effect on some aspects of adherence to some medicationsbut not others.Overall, we concluded that evidence is insufficient to draw conclusions about theeffectiveness of MTM for improving the proportion of patients who, based on a threshold of 80percent of prescribed doses taken, were adherent at 6 to 12 months based on direct, impreciseevidence from one small RCT (Table 22) and direct, and precise but inconsistent (regardingeffect direction and magnitude) evidence from one small and one large cohort study that togetherhad high study limitations.Overall, we concluded that there was low strength of evidence about the effectiveness ofMTM for improving the proportion of patients who, based on a threshold of 80 percent ofprescribed doses taken, were adherent at 6 to 12 months based on direct, imprecise evidencefrom one small RCT (Table 22) and direct, and precise but inconsistent (regarding effectdirection and magnitude) evidence from one small and one large cohort study that had high studylimitations. Strength of evidence is also low for improving the absolute percentage of prescribeddoses taken at 6 months and 12 months (mean adherence) for hypertension and dyslipidemiatreatment, based on inconsistent direct, imprecise evidence from two cohort studies, one smalland one large with medium study limitations (Table 23). However, we found insufficientevidence to draw conclusions about the effect of MTM on absolute percentage of prescribeddoses taken for other conditions based on these same observational studies. This conclusion isconsistent with findings from two small high risk-of-bias RCTs that provided direct, impreciseevidence of these effects at 4 to 9 months. Of note, these two trials had a high level of studylimitations and reported opposite directions of effect on absolute percentage of prescribed dosestaken, both with nonsignificant differences between groups.49


Table 22. Adherence outcome Type 1—proportion of patients adherent based on a threshold ofpercentage of pills taken: Strength of evidenceNumber ofFindings andStrengthStudy Studies; StudyDirectionConsistency Directness PrecisionofDesign Subjects Limitations[Magnitude] ofEvidence(Analyzed)EffectRCT 1; 81 (69) Medium Unknown(single study)Observational(Cohort)Cohort: 2; 224to 200,722;(224 to200,722)Direct Imprecise 100% of interventionpatients and 88.9% ofcontrols wereadherent (took > 80%of medicine); p=0.115InsufficientHigh Inconsistent Direct Precise Two studies with Low forfindings in opposite benefitdirection; larger studyshowing range ofORs for medicationspecificadherencedepending onmedicationFor comparison ofPDP versus controlsORs ranged from0.99 to 1.43 [95% CIsranged from(0.90,1.08) to (1.26,1.62)]For comparison ofMA-PD versuscontrols ORs rangedfrom 1.10 to 1.40[95% CIs rangedfrom (0.83, 1.24) to(1.29, 1.52)For clinic-based MTMversus usual care foradherence to aspirin,odds of adherencerange from 5.981(95% CI 0.284 to126.030, p=0.250)during theintervention to 1.17one year after theintervention (95% CI0.072 to 18.903,p=0.912)Abbreviations: CI = confidence interval; MA-PD = Medicare Advantage Part D Plan; MTM = medication therapy management;OR = odds ratio; PDP = Medicare Part D Plan; RCT= randomized controlled trial.50


Table 23. Adherence outcome Type 2—absolute measure of adherence as percentage ofprescribed doses taken: Strength of evidenceNumber ofFindings andStrengthStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Directionof[Magnitude] ofEvidence(Analyzed)EffectObservational2; 132-4,500(120-4,500)High Inconsistent Direct Imprecise One small studyfound no differencewith small meandifferenceCalculated meandifference: -0.04095% CI: -0.101 to0.021p=0.201Larger study found asmall (difference inadherence ~4.6%)but statisticallysignficant effect ofMTM on adherenceto medications forsome (two of five)conditions but nosignficant effect forthe other conditions.Low foradherencetotreatmentforhypertension anddyslipidemiaInsufficientfortreatmentof patientswithdiabetes,depressionandasthmaAbbreviations: CI = confidence interval; MTM = medication therapy management; SE = standard error; SMD = standardizedmean difference.Evidence is also insufficient about improving medication adherence as measured by selfreportedscales from one medium risk-of-bias trial (Table 24). Finally (Table 25), regardingmiscellaneous medications taking behaviors, such as remembering to take medication, amedication-taking behavior subscore, and the proportion of medications matched withinstructions, we concluded that evidence was insufficient for the effect of MTM on theseoutcomes, based on evidence from two RCTs that was direct but imprecise and inconsistent.Although the significant degree of heterogeneity across adherence measures precluded ourability to assess strength of evidence across all adherence studies, we note that considering thebody of evidence for the effect of MTM on adherence, taken together, results from all studieswere inconsistently significant with small magnitudes of effect. Across studies, the direction ofeffect was inconsistent, however, for the two outcomes, “proportion of adherent patients,” and“percentage of prescribed pills taken,” which in all studies were measured objectively usingclaims and pharmacy data, we found low strength of evidence that MTM had an effect onmedication adherence, particularly for certain chronic conditions . Hence, considering theadherence studies as a whole, there appears to be low strength ofevidence regarding an effect ofMTM on adherence.51


Table 24. Adherence outcome Type 3—self-reported scales: Strength of evidenceNumber ofFindings andStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 1; 363 (292) Medium Consistency Direct Imprecise Calculated meanunknown,difference: 0.090,single study95% CI: -0.076 to0.256, p=0.289Abbreviations: MTM = medication therapy management; RCT= randomized controlled trial.Strength ofEvidenceInsufficientTable 25. Adherence outcome miscellaneous: Strength of evidenceNumber ofFindings andStudy Studies; StudyDirectionConsistency Directness PrecisionDesign Subjects Limitations[Magnitude] of(Analyzed)EffectStrength ofEvidenceRCT 2; 492 Medium Inconsistent Direct Imprecise Two studies with Insufficient(365)opposite direction ofeffect, both with nonsignificantdifferencesbetween groupsAbbreviations: RCT= randomized controlled trial.Medication AppropriatenessFive studies (four RCTs, 59,63,84,85 one cohort study 70 ) reported on the effects of MTMinterventions on medication appropriateness (Table 26 and Table 27). Of these studies, threeassessed medication appropriateness across a broad spectrum of regimens; 70,84,85 the other twotrials assessed appropriateness for specific medications. 59,63 In addition, two studies evaluatedsingle aspects of medication appropriateness, across a range of medications. 62,79For the three broader studies, two trials used the Medication Appropriateness Index(MAI). 84,85 One of these reported results for the full scale and for each item of the index (eachitem asks about a different aspect of medication appropriateness) individually; 84 the other trialreported results only for each of the individual items. 85 The cohort study of broad regimens useda panel of three pharmacists to rate the appropriateness of the various antihypertensive regimenson a visual analogue scale. 70As shown in Table 26, one RCT (low risk of bias) 84 found a statistically significantimprovement in the MAI Scale at 3 and 12 months’ followup. The small cohort study (high riskof bias) reported no statistically significant improvement in the three appropriateness scoresassessed for blood pressure regimens (appropriateness of regimens, of dosing intervals, and ofdosages) although it was very underpowered. 70 52


Table 26. Medication appropriateness scales: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodHanlon et al., 1996 84RCT/LowG1: Usual care, plusclinical pharmacistcare.G2: Usual care in theGeneral MedicineClinicG1: 105G2: 103ResultsCovariate-adjusted BaselineMedicationAppropriateness Indexassessed at baseline, 3,12 months by blindedresearch pharmacistG1: 17.7 (0.6)G2: 17.6 (0.6)3 monthsG1: 13.4 (0.6)G2: 16.5 (0.6)95% CI: NRp


Table 26. Medication appropriateness scales: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodCarter et al., G1: Pharmacy-based1997 70pharmaceutical careBarnette et al. 1996 71 G2: Usual medicalCohort study/High careG1: 25G2: 26Appropriateness of BPregimenA blinded review panelof three pharmacistsevaluated cases inrandom order on avisual analog scale,using medical records.The investigatorsaveraged andconverted scores to anumerical value bymeasuring the distancefrom the best option.Score arranged from 0to 16.2. Higher scoresare better.ResultsBP regimenBaselineG1: 8.7 (4.7)G2: 10.3 (4.8)FollowupG1: 10.9 (4.5)G2: 10.1 (5.2)p for change scores NRAppropriateness of dailydosageAppropriateness of Daily dosagedosing interval BaselineG1: 11.6 (4.5)G2: 12.6 (4.5)FollowupG1: 13.4 (3.7)G2: 13.2 (4.1)p for change scores NRDosing intervalBaselineG1: 13.8 (4.3)G2: 13.4 (4.6)FollowupG1: 15.1 (2.3)G2: 13.8 (4.1)p for change scores NRAbbreviations: BP = blood pressure; G = group; N = number; NR = not reported; RCT = randomized controlled trial.Of note, one 84 (low risk of bias) of the two trials reporting the effect of MTM on generalmedication appropriateness scales, also provided descriptive data, by intervention group,regarding the proportion of inappropriate prescriptions for each of 10 items on the MAI (whichaddress different aspects of appropriateness). These findings are reported in Appendix E. Whileone is unable to draw conclusions regarding the findings because they report percentages withprescriptions (rather than “per patient”) as the unit of analysis, they do suggest that some itemsare likely driving the improvements in MAI in the MTM group more than others. Specifically,six aspects of medication prescription appropriateness: drug indication; dosage; practicality ofdirections; drug-drug interactions; duplication; duration of therapy seem to show greaterimprovement in inappropriate prescriptions than do those for four other aspects: effectivemedication; correctness of directions; drug-disease interactions; expense of medication.Similarly, another study 85 which did not report on the full MAI scale, also reported dataregarding the effect of MTM on individual MAI items (Appendix E). The ability to interpret54


these descriptive findings is not only, like the other study, 84 hampered by the use of prescriptionsrather than patients as the unit of analysis, but also is limited by the marked baseline differencesthat existed between intervention groups.Two RCTs (both medium risk of bias) assessed the appropriateness of regimens for specificmedications for specific conditions (Table 27). One assessed, among patients at risk forglucocorticoid-induced osteoporosis, the percentages of patients receiving each of three indicatedregimens; 63 the investigators found, at 9-month followup, a statistically significant improvementin the percentage appropriately prescribed calcium supplements among MTM recipientscompared with controls but not for bisphosphonate or estrogen drug therapy. The other trialassessed the use of angiotensin-conversion enzyme (ACE) inhibitors among heart failurepatients. 59 The pharmaceutical care program had a significant effect on the mean percentage oftarget dose achieved and on the proportion receiving an appropriate alternative medicine amongthe subsample; such services did not produce a significant effect on the percentage of patientswho received an ACE inhibitor.Table 27. Medication appropriateness for individual medications: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodMcDonough, 2005 63cluster-randomizedRCT/MediumG1: pharmaceuticalcare provided bypharmacist in acommunity pharmacyG2: usual carePatients at risk forglucocorticoid-inducedosteoporosis)BaselineG1: 70G2: 26FollowupG1:61G2:199-month followupPercentage of patientstaking calciumsupplementsResultsBaselineG1: 38.6G2: 38.5p for between-groupdifferences at baselinepresumed not significant aFollowupG1:55.7 (p


Table 27. Medication appropriateness for individual medications: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodMcDonough, 2005 63cluster-randomizedRCT/Medium(continued)Gattis, 1999 59RCT/MediumG1: Clinical pharmacistinterventionG2: Usual medicalcarePatients with heartfailure.G1: 90G2: 91G1: 12G2: 19Percentage of patientson estrogen drugtherapy6-month followupFraction of target ACEIdose at followupmedian (25 th and 75 thpercentile values)Of those not on anACEI at followup,percentage receivingalternative drug therapyResultsBaselineG1: 12.9G2: 0P NS for between-groupdifference at baselineFollowupG1: 16.4 (p


Table 28. Medication appropriateness scales: Strength of evidenceNumber ofStudies;Study SubjectsDesign (Analyzed)RCT 1; 208(208)StudyLimitations Consistency Directness PrecisionLowUnknown(single study)Findings andDirection[Magnitude]of EffectDirect Precise Improvementin MTM groupfrom score of17.7 to 13.4and to 12.8 in3, 12 months,respectively,p


Table 30. Medication dosing: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency Directness Precision(Analyzed)Findings andDirection[Magnitude]of EffectRCT 1; 90 (56) Medium Unknown Indirect Precise One RCT Lowshowingsignificantdecrease indaily doses ofmedications(Meandifference: -2.2doses,calculated95% CI: -3.738to -0.662)Abbreviations: RCT= randomized controlled trial.Key Points: Patient-Centered OutcomesStrength ofEvidence• Evidence was insufficient to draw conclusions about the effect of MTM on adverse drugevents based on inconsistent and imprecise findings from two RCTs: one with low studylimitations and one with medium limitations.• Evidence was insufficient to draw conclusions about the effect of MTM on cognitive andphysical function based on direct but imprecise findings from one RCT with mediumstudy limitations and on affective function based on direct but inconsistent and imprecisefindings from two RCTs, both with medium study limitations.• Evidence was insufficient to draw conclusions about the effect of MTM ongastrointestinal bleeding based on direct but imprecise findings from one observationalstudy with high study limitations.• Evidence was insufficient to draw conclusions about the effect of MTM on all-causemortality based on one RCT with medium study limitations and two observationalstudies, with high study limitations.• With two exceptions, MTM interventions had no benefit on SF-36 measures (lowstrength of evidence of no benefit); evidence was insufficient for the SF-36 domain ofvitality or emotional role functioning because of imprecision.• Evidence was insufficient to determine whether MTM interventions improved patientreportedmeasures for patients with diabetes (one imprecise medium risk-of-bias trial).• MTM interventions did not improve measures of patient satisfaction (low strength ofevidence of no benefit).Detailed Synthesis: Patient-Centered OutcomesAdverse Drug EventsFour RCTs 50,69,84,85 and one nonrandomized trial 83 reported on prevalence of adverse drugevents (ADEs) following MTM or pharmaceutical care interventions (Table 31). The methodsfor measuring adverse events differed substantially among included studies. Further, although weassumed that the beneficial direction of effect would be for MTM to decrease ADEs, thenonrandomized trial suggested that MTM services may heighten awareness of potential adverse58


outcomes by patients and, thus, increase reporting of ADEs by those receiving the intervention. 83For this outcome, we rated the risk of bias for some studies 50,83,85 as higher than the overall riskof bias because of measurement and detection bias with respect to the measures and methodsused to ascertain this outcome.Table 31. Adverse events: Summary of resultsStudyDesign/Risk of BiasTouchette et al., 2012 69RCT/LowStudy ArmsG1: Basic MTMservices (withmedicationinformation frompatient interview)G2: Enhanced MTMservices (pharmacistprovided with 2-pageclinical summary frompatient medicalrecord)G3: Usual pharmacycareN AnalyzedG1: 211G2: 218G3: 208Outcomes Reportedby Study and TimePeriodPercentage of patientswith an ADE between0 and 3 months andORPercentage of patientswith an ADE between3 and 6 months andORMean number (SD) ofADEs per patientbetween 0 and 3monthsMean number (SD) ofADEs per patientbetween 3 and 6monthsResultsG1 vs. G3: OR: 1.629(p = 0.078)G2 vs. G3: OR: 0.726(p = 0.278)G1 vs. G3: OR: 1.107(p = 0.717)G2 vs. G3: OR: 0.889(p = 0.672)G1 vs. G3:Calculated mean difference,0.191;95% CI, -0.031 to 0.413p=0.091G2 vs. G3:Calculated mean difference,-0.012;95% CI, -0.239 to 0.215p=0.917G1 vs. G3: Calculated meandifference, 0.284;95% CI, 0.056 to 0.512p=0.014Hanlon et al., 1996 84RCT/MediumG1: Clinicalpharmacist carewithin a generalmedicine clinicG1:86G2:83Percentage with anADE at 12 monthsG2 vs. G3: Calculated meandifference,-0.062; 95% CI, -0.225 to0.101p=0.455Calculated OR: 0.649,95% CI: 0.366 to 1.152,p=0.14Taylor et al., 2003 85RCT/High aG2: Usual careG1: PharmaceuticalcareG2: Standard careG1: 33G2: 36Percentage of patientswith at least onemedicationmisadventure at 12monthsG1: 2.8 b (N=4)G2: 3.0 b (N=3)Calculated OR based onreported percent: 0.93, 95%CI, 0.056 to 15.603, p=0.0961Calculated OR based onreported N: 1.515 (95% CI,0.312 to 7.344), p= 0.60659


Table 31. Adverse events: Summary of results (continued)StudyDesign/Risk of BiasJameson et al., 1995 50RCT/High aFischer et al., 2000 83NRCT/High aStudy ArmsN AnalyzedG1: Consultation with G1: 27a clinical pharmacist G2: 29in a primary careofficeG2: Standard medicalcare in a primary careofficeG1: Comprehensive G1: 201drug therapy G2: 368management programG2: Standardcommunity pharmacypracticeOutcomes Reportedby Study and TimePeriodChange in meanmedication side effectscore at 6 months.ResultsG1: -3.7G2: -1.9p: NS and unable to calculate.OR for likelihood of 1.81 (1.16 to 2.83)reporting side effectsor problems fromprescription medication(95% CI)a This study was rated medium risk of bias overall, but due to measurement bias with this specific outcome, it was consideredhigh risk of bias for this outcome.b The percentage reported by authors cannot be generated based on the reported N and the reported number of events.Abbreviations: ADE = adverse drug event; CI = confidence interval; G = group; N = number; NRCT = nonrandomized controlledtrial; NS = not significant; OR = odds ratio; RCT = randomized controlled trial; SD = standard deviation; SMD = standardizedmean difference; vs. = versus.One RCT (low risk of bias) compared usual care with the provision of basic MTM servicesdesigned to mimic conditions similar to a community pharmacy with another study arm thatincluded an enhanced intervention that provided clinical information about the patient to thepharmacist. 69 It reported on outcomes for the period between 0 and 3 months and for the periodbetween 3 and 6 months. The enhanced MTM intervention was superior to the basic MTMintervention at 3 months in the percentage of subjects reporting an ADE; however, the enhancedintervention and usual care at 3 months and three study arms at 6 months did not differsignificantly. In addition, the mean number of ADEs per patient was not statistically differentbetween 0 and 3 months across study arms, but both the enhanced MTM and usual care studyarms had significantly fewer ADEs per patient than the basic MTM study arm between 3 and 6months. This RCT found no statistical difference in mean ADEs per patient between theenhanced MTM study arm and usual care between 3 and 6 months. Another RCT (medium riskof bias) compared clinical pharmacy care within a VA general medicine clinic to usual care. 84This study found that intervention group subjects were less likely to have one or more ADE over12 months, but this finding spanned the null effect (calculated OR, 0.649; 95% CI, 0.366 to1.152, p=0.14).The other two RCTs and the NRCT were considered high risk of bias for the ADE outcome.One RCT provided pharmaceutical care to patients at high risk for medication-related problemsseen in family medicine practices in a rural Alabama community; 85 the intervention and controlarms did not have significantly different findings with respect to “medication misadventures” at12 months. We rated this trial as high risk of bias because it used a nonstandard measure(medication misadventure was not defined) and because the control event rates differed by afactor of 10 relative to the low and medium risk-of-bias RCTs. The other RCT compared MTMintervention provided by a pharmacist within a family health center setting in Michigan withusual medical care and reported no significant difference between change in medication side60


effect scores using a scale that the study authors had developed for use in the study, but novalidity or reliability data for this scale were provided. 50 The nonrandomized trial (high risk ofbias for this outcome)compared participants who agreed to participate in a pharmaceutical careprogram at one of six participating community pharmacies with a group of control patients whoreceived medications at pharmacies that did not provide pharmaceutical care services; 83 studyparticipants were significantly more likely (OR, 1.81; 95% CI, 1.16 to 2.83) to reportexperiencing symptoms or problems related to prescription medication than control participants,an effect the authors attributed to increased awareness of medication side effects in theintervention group. Without a clear understanding of the hypothesized mechanism of action ineach study for influencing ADEs and the lack of study methods for minimized detection bias, wecannot interpret the conflicting results presented by the nonrandomized trial relative to thefindings from the RCTs.Overall, we concluded that evidence is insufficient to draw conclusions about the efficacy ofMTM for reducing adverse drug events based on direct, but inconsistent and imprecise, evidencefrom one low and one medium risk of bias RCT (Table 32).Table 32. Adverse drug events: Strength of evidenceStudy DesignNumber ofStudies;Subjects(Analyzed)RCT 2; 845(806)Abbreviation: RCT = randomized controlled trial.FindingsandStudyLimitations Consistency Directness PrecisionDirection[Magnitude]of EffectMedium Inconsistent Direct Imprecise Direction andmagnitude ofeffect differsbetween thetwo trialsStrengthofEvidenceInsufficientCognitive, Affective, and Physical FunctionTwo RCTs (both medium risk of bias) reported on changes in cognitive, affective, orphysical function at 6 weeks and at 3 months. 75,92 In one RCT, the intervention was provided in ageneral medicine outpatient clinic in North Carolina to simplify medication regimens amongcognitively intact patients ages 65 or older at high risk for medication-related adverse events(Table 33). The investigators measured cognitive function using three different tests. Theymeasured affective function using the Center for Epidemiological Studies Depression Scale andthe Self-Rating Anxiety Scale and physical functioning using three different tests. Patients in theintervention arm experienced no significant changes in any of these measures when comparedwith patients in the control arm.The other RCT evaluated the use of the Dader method for pharmaceutical care amongwomen receiving treatment for anxiety or depression in specialty clinics at a university hospitalin Brazil. This study found significant mean changes in both the Beck Depression Inventory andthe Beck Anxiety Inventory compared with the control arm. In addition, the percentage ofpatients achieving a depression remission was higher in the intervention arm, but these findingswere not significant (calculated OR, 2.406; 95% CI, 0.601 to 9.632; p=0.215). Overall, weconclude that the strength of evidence is insufficient for the effect of MTM on cognitive orphysical function based on a single, imprecise RCT with medium study limitations (Table 34).We also conclude that the strength of evidence is insufficient for the effect of MTM on affective61


function based on inconsistent and imprecise findings from two RCTs with medium studylimitations (Table 35).Table 33. Cognitive, affective, and physical function: Summary of findingsStudyOutcomes ReportedDesign/Risk of Study ArmsN Analyzed by Study and TimeBiasPeriodWilliams, 2004 92RCT/MediumG1: Modification of G1: 57patient’s medication G2: 76regimen by aninterdisciplinarymedication adjustmentteamG2: Usual medical careAt 6 weeks:Timed manualperformancePhysical performancetestFunctional reachResultsCalculated mean differences(unadjusted for baselinedifferences):0.300 95% CI, -1.093 to 1.693;p=0.673-3.300 95% CI, -13.370 to6.770, p=0.520.00 95% CI, -1.076 to 1.076;p=1.0Digit Span (WAIS)Digit Symbol (WAIS)Randt Memory Test0.200 95% CI, -1.277 to 1.677;p=0.7910.00 95% CI, -5.134 to 5.134;p=1.00.00 95% CI, -1.182 to 1.182;p-1.00Marques et al.2013 75RCT/MediumG1: Dader methodpharmacotherapyinterventionG2: Usual careG1: 22G2: 26CES-D scoreSelf-rating AnxietyScale scoreAt 3 months:Mean change (SD) inBeck DepressionInventory-1.10 95% CI, -3.813 to 1.613;p=0.427-0.100 95% CI, -2.392 to 2.192;p=0.932G1: -13.5 (NR)G2: -2.5 (NR)95% CI: NRp: 0.0275Mean change (SD) inBeck Anxiety InventoryG1:-13.0 (NR)G2: -3.5 (NR)95% CI: NRp: 0.0194G1: 5G2: 5G1:13G2:13Percentage withDepression remission(defined as BDI < 11)Percentage with severedepressionimprovementPercentage withmoderate depressionimprovementCalculated OR, 2.406; 95% CI,0.601 to 9.632; p=0.215Calculated OR 2.667 95% CI,0.158 to 45.141. p=0.497Calculated OR 14.00 95% CI,1.385 to 141.485. p=0.025Abbreviations: BDI = Beck Depression Inventory; CES-D = Center for Epidemiological Studies-Depression Scale; CI =confidence interval; G = group; OR = odds ratio; RCT = randomized controlled trial; SD = standard deviation; WAIS = WechslerAdult Intelligence Scale.62


Table 34. Cognitive and physical function: Strength of evidenceNumber ofFindings andStudies; StudyStudy DesignSubjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 1; 140 MediumDirect(133)Abbreviation: RCT= randomized controlled trial.ConsistencyunknownsinglestudyImprecise One study with nosignificantdifferencesbetween armsStrength ofEvidenceInsufficientTable 35. Affective function: Strength of evidenceNumber ofFindings andStudies; StudyStudy DesignSubjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 2; 198(181)Abbreviation: RCT= randomized controlled trial.Medium Inconsistent Direct Imprecise One study with nosignificantcalculated meandifferences indepression oranxiety scores, theother study withsignificantdifferences in meandepression andanxiety scores, butno significantdifference inpercentageachieving adepressionremissionStrength ofEvidenceInsufficientMortalityOne RCT 59 and two cohort studies 44,48 reported all-cause mortality outcomes following MTMinterventions at 6 months to 4 years (Table 36). The RCT (medium risk of bias) conducted in auniversity general cardiology clinic compared a study arm that included a clinical pharmacistintervention for heart failure patients with usual medical care 59 and found decreased mortalitywithin 6 months, but this finding spanned the null effect (OR, 0.59; 95% CI, 0.12 to 2.49,p=0.48).Both cohort studies (Table 34) (both medium risk of bias) measured mortality outcomes forbeneficiaries who met MTM program eligibility and opted in to a telephone-based MTMprogram provided through an integrated health care system compared with eligible beneficiarieswho opted out of the MTM program. 44,48 One study reported a statistically significant reductionin all-cause mortality at 6 months in the intervention arm, when adjusted for age, sex, andbaseline disease, and health care utilization levels (adjusted OR, 0.5; 95% CI, 0.3 to 0.9;p=0.044). The other study reported a similar direction of effect (adjusted HR, 0.92; 95% CI, 0.87to 0.96; p< 0.001). The RCT reporting mortality outcome also reported a composite measure thatcombined all-cause mortality with nonfatal heart failure events and found the intervention armexperienced a significant benefit from the program (OR, 0.22l; 95% CI, 0.07 to 0.65; p=0.005). 5963


Table 36. All-cause mortality: Summary of resultsStudyDesign/Risk ofBiasStudy ArmsN AnalyzedGattis et al., 1992 59 G1: Clinical pharmacist G1: 90intervention in addition to G2: 91usual medical careRCT/MediumG2: Usual medical careWelch et al.,2009 44Cohortstudy/MediumYamada et al.,2012 48Cohortstudy/MediumG1: MTM program providedto home-based beneficiariesG2: No-MTM control group(voluntary opt-out)G1: MTM enrolled patientsG2: Eligible MTM patientsnot enrolled but matched onage, gender, region andDCG riskG1: 459G2: 336G1: 34,352G2: 138,182Outcome Reported byStudy and Time PeriodOR for all-cause mortalitywithin 6 monthsAdjusted OR for all-causemortality, within 6 months(adjusted for age, sex,chronic disease score,specific baselineutilization)Adjusted HR for all-causemortality within 1 to 4years (adjusted for age,sex, Charlson comorbidityindex, CHF, ESRD)ResultsOR: 0.5995% CI, 0.12 to 2.49p=0.48Adjusted OR: 0.595% CI, 0.3 to 0.9p=0.044Adjusted HR: 0.9295% CI 0.87 to 0.96p< 0.001Abbreviations: CHF = congestive heart failure; CI = confidence interval; DCG = diagnostic cost group (a measure of health careuse and comorbidity); ESRD = end-stage renal disease; G = group; HR = hazard ratio; MTM = medication therapy management;OR = odds ratio; RCT = randomized controlled trial.Overall, we concluded that evidence is insufficient for the efficacy of MTM for reducing allcausemortality at 6 months to 4 years based on direct evidence from a single, imprecise RCTwith medium study limitations and two inconsistent observational studies with high studylimitations (Table 37). We relied more heavily on the RCT evidence for our overall SOE ratingbecause of the inability of observational study designs to adequately mitigate for selection biasrelative to a mortality outcome compared with RCT designs.Table 37. All-cause mortality: Strength of evidenceNumber ofFindings andStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectStrength ofEvidenceRCT 1; 181 (181) Medium Consistency Direct Imprecise OR: 0.59Insufficientunknown,single body ofevidence95% CI: 0.12 to 2.49p=0.48Observational 2; 173,329 High Inconsistent Direct Precise One study OR 0.5 Insufficient(173,329)95% CI, 0.3 to 0.9;Second study:adjusted HR: 0.9295% CI 0.87 to 0.96p


nonenrolled patients had no change in gastrointestinal bleeds (p=0.001 for between-groupdifference in change in gastrointestinal bleeds).Overall, we concluded that evidence is insufficient for the efficacy of MTM for reducinggastrointestinal bleeding events based on direct but imprecise evidence from one cohort studywith high study limitations (Table 38).Table 38. Gastrointestinal bleeding events: Strength of evidenceNumber ofFindingsStudies; StudyStudy DesignConsistency Directness PrecisionSubjects Limitations(Analyzed)Observational 1; 1,388(unclear)HighAbbreviation: RRR = relative risk reduction.Consistencyunknown—single studyandDirection[Magnitude] ofEffectDirect Imprecise RRR 60%(p=0.001)Strength ofEvidenceInsufficientSelf-Reported Health Status: SF-36 MeasuresSF-36 Measures: OverviewEight RCTs 54,55,64,84-86,89,91 and one cohort study 70 reported health status outcomes using theMedical Outcomes Study Short-Form questionnaire (SF-36) (Table 39). The eight SF-36domains, which combine into two components, are as follows—physical health: physicalfunctioning, physical role functioning, bodily pain, and general health perceptions; and mentalhealth: vitality, emotional role functioning, social role functioning, and mental health. Seventrials 54,55,64,84,85,89,91 and the cohort study 70 reported scores for all eight domains. One trialreported only its two component scores (i.e., physical health; mental health). 86 Finally, one trialreported both component and domain scores. 55 The trials differed by overall risk of bias (one,low; four, medium, and three, high); the cohort study was high risk of bias.Table 39. Scores on SF-36 measures: Summary of effects from meta-analysesTotal Mean DifferenceSF-36 Time Periods andNumber 95% CIQ-valueNumber ofComponents Risk of Bias forWith/ Lower Limit to (df for Q)Studiesand Domains Included TrialsWithout Upper Limit p-valueMTM p-valuePhysicalfunctioningdomainPhysical rolefunctioningdomainBodily paindomainAll time periods, lowor medium risk ofbiasAll time periods, allrisk of biasAll time periods, lowor medium risk ofbiasAll time periods, allrisk of biasAll time periods, lowor medium risk ofbiasAll time periods, allrisk of bias3 54,84,85 566/603 1.171CI: -3.871 to 6.214p=0.6495 54,55,84,85,89 968/1,038 -0.438CI: -2.641 to 1.765p=0.6973 54,84,85 566/603 3.392CI: -1.223 to 8.007p=0.1505 54,55,84,85,89 968/1,038 0.733CI: -3.429 to 4.895p=0.7303 54,84,85 566/603 3.320CI: -0.792 to 7.433p=0.1145 54,55,84,85,89 968/1,038 1.459CI: -2.793 to 5.711p=0.5013.873(2)p=0.1444.478(4)p=0.3450.988(2)p=0.6107.238(4)p=0.1242.765(2)p=0.25121.061(4)p


Table 39. Scores on SF-36 measures: Summary of effects from meta-analyses (continued)Total Mean DifferenceSF-36 Time Periods andNumber 95% CIQ-valueNumber ofComponents Risk of Bias forWith/ Lower Limit to (df for Q) I-squaredStudiesand Domains Included TrialsWithout Upper Limit p-valueMTM p-valueGeneral healthperceptionsdomainAll time periods, lowor medium risk ofbiasAll time periods, allrisk of biasVitality domain All time periods, lowor medium risk ofbiasAll time periods, allrisk of biasSocialfunctioningdomainEmotional rolefunctioningdomainMental healthdomainAll time periods, lowor medium risk ofbiasAll time periods, allrisk of biasAll time periods, lowor medium risk ofbiasAll time periods, allrisk of biasAll time periods, lowor medium risk ofbiasAll time periods, allrisk of bias3 54,84,85 566/603 1.916CI: -0.007 to 3.839p=0.0515 54,55,84,85,89 968/1,308 2.476CI: 2.123 to 2.829p


(e.g., group mean not contained within 95% CI, group mean not centered within 95% CI); 55 itreported no statistically significant between-group differences on any SF-36 elements.SF-36 Measures: Meta-AnalysesOur analysis focuses on the three trials rated either low or medium risk of bias that providedsufficient data to calculate mean differences for the eight SF-36 domain scores. 1-3 We alsoconducted sensitivity analyses that included the two high risk-of-bias trials in addition. 55,89 Weomitted one trial from the meta-analyses altogether because it reported only that none of the SF-36 domains differed significantly but did not give any precise values. 91 Similarly, we excludedone trial 64 and the cohort study 70 in the meta-analyses because they did not report standarddeviations, standard errors, or exact p-values for any of the between-group comparisons; bothstudies reported that MTM did not produce any significant differences in anySF-36 domain.Finally, we omitted one trial from the domain-specific meta-analyses because it reported onlycomponent scores. 86 To correct for the potential inflation of Type I error attributable to multiplecomparisons, we used a threshold of α/number of tests (i.e., domains; 0.05/8=0.006) whenevaluating statistical significance. Below, we describe our findings for each SF-36 domain,focusing on the meta-analyses of just the low to medium risk-of-bias trials (i.e., the smaller metaanalysis).We did not conduct a meta-analysis for the SF-36 component scores because only onetrial was rated as low to medium risk of bias for these outcomes.SF-36 Domain ScoresPhysical functioning. Results from the low and medium risk-of-bias analysis showed nobenefit for the MTM interventions (mean difference: 1.17; 95% CI, -3.87 to 6.21; p=0.65;I 2 =48.36). Adding the two high risk-of-bias studies did not alter this conclusion (meandifference:-0.44; 95% CI, -2.64 to 1.77; p=0.70; I 2 =10.67) (Appendix F-1).Physical role functioning. Results from the low and medium risk-of-bias analysis showedno benefit for the MTM interventions (mean difference: 3.39; 95% CI, -0.79 to 7.43; p=0.11;I 2 =27.66). Adding the two other studies did not alter this conclusion (mean difference: 0.73; 95%CI, -3.43 to 4.90; p=0.73; I 2 =44.73) (Appendix F-2).Bodily pain. Results from the low and medium risk-of-bias analysis showed no benefit forthe MTM interventions (mean difference: 3.32; 95% CI, -1.22 to 8.01, p=0.15; I 2 =0). Adding thetwo other studies did not alter this conclusion (mean difference: 1.46; 95% CI, -2.79 to 5.71;p=0.50; I 2 =81.01) (Appendix F-3).General health perceptions. Results from the low and medium risk-of-bias analysis showedno benefit for the MTM interventions (mean difference: 1.92; 95% CI, -0.02 to 3.84, p=0.051;I 2 =0). With the additional studies, however, results suggested a beneficial effect of MTMinterventions on general health perceptions (mean difference: 2.48; 95% CI, 2.12 to 2.83,p


Adding the other two studies did not alter this conclusion (mean difference: 3.44; 95% CI, -4.00to 10.88; p=0.37; I 2 =78.66). However, the high I 2 statistic for both these meta-analyses suggestedconsiderable heterogeneity among the studies for this particular domain (Appendix F-6).Social role functioning. Results from the low and medium risk-of-bias analysis showed nobenefits from MTM interventions (mean difference: 2.93; 95% CI, -0.09 to 5.95; p=0.057; I 2 =0).With the additional studies, however, results suggested a beneficial effect of MTM interventions(mean difference: 0.63; 95% CI, 0.29 to 0.97; p


Table 40. SF-36: Strength of evidence (continued)Number ofFindingsSF-36 Study Studies; StudyConsistency Directness PrecisionDomain Design Subjects Limitations(Analyzed)GeneralhealthperceptionsdomainVitalitydomainSocialfunctioningdomainEmotionalrolefunctioningdomainMentalhealthdomainRCTs 3; 1,343(1.169)RCTs 3; 1,343(1.169)RCTs 3; 1,343(1.169)RCTs 3; 1,343(1.169)RCTs 3; 1,343(1.169)andStrengthDirectionof[Magnitude] ofEvidenceEffectMedium Consistent Direct Precise Meandifference:1.916CI: -0.007 to3.839p=0.051Medium Consistent Direct Imprecise(results notcorrected ormultiplecomparison)Meandifference:2.797CI: 0.655 to4.939p=0.010Medium Consistent Direct Precise Meandifference:2.932CI: -0.085 to5.949p=0.057Medium Inconsistent Direct Imprecise(wideconfidenceintervals)5.386CI: -7.244 to18.016p=0.403Medium Inconsistent Direct Precise Meandifference:1.615CI: -0.362 to3.593p=0.109Abbreviations: CI = confidence interval; RCT= randomized controlled trial; SF-36 = 36-Item Short Form Health Survey.Low for nobenefitInsufficientLow for nobenefitInsufficientLow for nobenefitCondition-Specific Quality of LifeTwo small RCTs 72,74 reported condition-specific quality-of-life outcomes (Table 41). OneRCT (medium risk of bias) of just patients with diabetes compared patients in a study arm thatincluded a clinical pharmacist intervention delivered in an ambulatory care clinic with thosereceiving usual medical care. 72 The investigators reported no significant difference in diabetesspecificquality-of-life between the intervention and control arms at the end of 6 months. Theother RCT 73,74 (high risk of bias) of patients with renal disease reported a significant differenceat 1 year favoring the pharmaceutical care program We graded the strength of evidence, usingonly the medium risk-of-bias trial, as insufficient (single study, direct, but imprecise) (Table 42).69


Table 41. Condition-specific quality of life: Summary of resultsStudyDesign/Risk of BiasClifford et al., 2002 72RCT/MediumPai et al., 2009 73 ;Pai et al., 2009 74RCT/HighStudy ArmsG1: Collaborativepharmaceutical careprogramG2: Standardoutpatient care fordiabetesG1: Pharmaceuticalcare, consisting ofone-on-one care, within-depth drug therapyreviews conducted bya clinical pharmacistG2: Standard of care,consisting of brieftherapy reviewsconducted by a nurseN analyzedG1: 48G2: 25BaselineG1: 61G2: 44Year 1:G1: 44G2: 36Year 2:G1: 24G2: 32Outcome and TimePeriodDiabetes Quality of LifeinstrumentScale of 1 to 5Higher scores indicategreater dissatisfaction,worry, or impact ofdiabetesRenal Quality of LifeProfileMaximum score = 172Higher scores indicateworsening of HRQOLResultsBaselineG1: 2.0 (0.6)G2: 1.9 (0.5)p: NS6-month followupG1: 1.9 (0.5)G2: 1.9 (0.4)p>0.15Total ScoreBaselineG1: 71.9 (40)G2: 74.5 (33.5)Year 1G1: 71.4 (33.6)G2: 87.5 (30.4)p


satisfaction measures (i.e., satisfaction with general health care and satisfaction with pharmacyrelatedcare).Table 43. Patient satisfaction: Summary of resultsStudyDesign/Risk of BiasHanlon et al., 1996 84RCT/LowMalone et al., 2000 51 ;Ellis et al., 2000 52 ;Malone et al, 2001 54 ;Ellis et al., 2000 53RCT/MediumStudy Arms N Analyzed Outcome and Time Period ResultsG1: Usual care atoutpatient clinic, plusclinical pharmacistcare.G2: Usual care atoutpatient clinicG1: Pharmaceuticalcare provided byclinical pharmacistswithin ambulatory VAclinicsG2: Usual care (i.e.no pharmaceuticalcare)G1: 86G2: 83G1: 447G2: 484Bernsten et al.,2001 64 ;G1: StructuredcommunityBaselineG1: 1,290Sturgess et al., 2003 65 pharmacy-based G2: 1,164RCT, Cluster-Randomized/Highpharmaceutical careprogram6 monthsG2: Usual community G1: 1,024pharmacy services G2: 95312 monthsG1: 863G2: 76418 monthsG1: 704G2: 636General health careG1: 1.5 (0.7)satisfaction at 12-month G2: 1.6 (0.8)followup(Higher scores indicate greater p=0.70dissatisfaction)Pharmacy-related health care G1: 5.2 (1.5)satisfaction at 12-month G2: 5.4 (1.7)followup(Higher scores indicate greater p=0.52dissatisfaction)Patient satisfaction withprimary health care provider(Higher scores indicate greatersatisfaction)Percentage rating pharmacyservices provided as“excellent”Time 1G1: 51.9 (7.5)G2:51.9 (7.5)Time 2G1: 51.7 (7.3)G2: 51.9 (7.5)p=NSBaselineG1: 66.2G2: 68.2p NR6 monthsG1: 72.8G2: 63.7p


Table 43. Patient satisfaction: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsPercentage agreeing withstatement “I am satisfied withthe services provided by thepharmacy that I regularly visit.”BaselineG1: 92.0G2: NR6 monthsG1: 95.1G2: NR12 monthsG1: 93.9G2: NR18 monthsG1: NRG2: NRCarter et al.,1997 70 ,Barnette et al., 1996 71Cohort/HighG1: PharmaceuticalcareG2: Usual care withpatients seen bypharmacists who didnot participate in theintensive skillsdevelopmentprogramG1: 25G2: 26All results are percentage ofpatients agreeing or stronglyagreeing with a specificstatement, as measured at 6-month followup“I am very satisfied with thepharmacy services I receive,”“Overall, the program provideda valuable service to me”“The quality of informationprovided to me by thepharmacist was excellent”“My participation in thisprogram helped me tounderstand high bloodpressure better”“The area was private enoughfor me to feel comfortabletalking about my high bloodpressure”“I felt comfortable talking withthe pharmacist about myhealth problems”“I am confident the pharmacistis able to help me control myhigh blood pressure”“I am confident the informationprovided by the pharmacist tothe physician improved myhealth care.”“There are things about thehigh blood pressure programthat could be better.”p=NS for all betweengroupdifferencesG1: 100G2: 96p=0.065G1: 100G2: 80p=0.0018G1: 100G2: 88p=0.012G1: 100G2: 83p= 0.011G1: 96G2: 96p=0.036G1: 100G2: 96p=0.052G1: 100G2: 92p=0.340G1: 87G2: 83p=0.325G1: 9G2: 0p=0.15772


Table 43. Patient satisfaction: Summary of results (continued)StudyDesign/Risk of BiasCarter et al.,1997 70 ,Barnette et al., 1996 71Cohort/HighVolume et al., 2001 86 ;Kassam et al., 2001 87RCT-ClusterRandomized/MediumStudy Arms N Analyzed Outcome and Time Period Results(continued (continued_ “I am very willing to continueto see the pharmacist for helpwith my high blood pressurecontrol.”“I think the pharmacist shouldprovide this type of service foreveryone.”“I think the pharmacist shouldbe paid for this type ofservice.”G1: Comprehensivepharmaceutical careservicesG2: Traditionalpharmacy careBaselineN=363G1: 159G2: 204Time 2:N=317G1: NRG2: NRTime 3:N=292G1: NRG2: NRGeneral satisfaction(Higher numbers reflectgreater dissatisfaction)Interpersonal skills(Higher numbers reflectgreater dissatisfaction)G1: 95G2: 88p=0.459G1: 77G2: 75p=0.890G1: 91G2: 82p=0.379BaselineG1: 1.59 (0.77)G2: 1.56 (0.736-MonthG1: 1.51 (0.84)G2: 1.57 (0.72)12-MonthG1: 1.53 (0.77)G2: 1.62 (0.88)p=NS for all betweengroupdifferencesBaselineG1: 1.36 (0.48)G2: 1.37 (0.53)6-MonthG1: 1.37 (0.59)G2: 1.35 (0.57)12-MonthG1: 1.31 (0.50)G2: 1.45 (0.72)Evaluation and goal setting(Higher numbers reflectgreater dissatisfaction)p=NS for all betweengroupdifferencesBaselineG1: 2.58 (1.12)G2: 2.74 (1.09)6-MonthG1: 2.46 (0.98)G2: 2.98 (1.24)12-MonthG1: 2.49 (1.10)G2: 2.90 (1.08)p


Table 43. Patient satisfaction: Summary of results (continued)StudyDesign/Risk of BiasVolume et al., 2001 86 ;Kassam et al., 2001 87RCT-ClusterRandomized/Medium(continued)Study Arms N Analyzed Outcome and Time Period ResultsTrust(Higher numbers reflectgreater dissatisfaction)BaselineG1: 1.62 (0.66)G2: 1.46 (0.57)6-MonthG1: 1.40 (0.54)G2: 1.39 (0.58)12-MonthG1: 1.43 (0.58)G2: 1.51 (0.75)p


Table 43. Patient satisfaction: Summary of results (continued)StudyDesign/Risk of BiasVolume et al., 2001 86 ;Kassam et al., 2001 87RCT-ClusterRandomized/Medium(continued)Study Arms N Analyzed Outcome and Time Period ResultsPharmacy finances(Higher numbers reflectgreater dissatisfaction)BaselineG1: 3.08 (1.82)G2: 2.85 (1.80)6-MonthG1: 2.89 (1.89)G2: 2.86 (1.75)12-MonthG1: 3.08 (1.80)G2: 3.16 (1.88)Drug plan finances(Higher numbers reflectgreater dissatisfaction)p=NS for all betweengroupdifferencesBaselineG1: 3.31 (1.70)G2: 3.41 (1.75)6-MonthG1: 3.45 (1.96)G2: 3.39 (1.83)12-MonthG1: 3.65 (1.67)G2: 3.56 (1.83)Communicates with doctor(Higher numbers reflectgreater dissatisfaction)p=NS for all betweengroupdifferencesBaselineG1: 1.50 (0.77)G2: 1.60 (0.89)6-MonthG1: 1.36 (0.63)G2: 1.72 (1.00)12-MonthG1: 1.36 (0.65)G2: 1.74 (0.97)p


following baseline, and at 12 to 13 months following baseline. This study reported statisticallysignificant between-group change in a measure labeled, Evaluation and Goal Setting. Thismeasure included six items assessing the extent to which the pharmacist involved the patient insetting therapeutic goals. However, none of the items asked directly about patient satisfactionwith the goal setting process. This study also reported a statistically significant between-groupchange from baseline to the 12- 13-month followup on a measure labeled, Communicates withDoctor. This measure included two items asking about whether the patient’s pharmacist anddoctor work together to determine the most appropriate therapy for the patient. Neither itemasked directly about patient satisfaction with the level of pharmacist-doctor communication.Finally, this study reported a statistically significant between-group change in a measure labeled,Trust. At baseline, patients in the intervention group reported lower trust in their pharmacist.Over the course of the study, their level of trust improved to the level reported by patients in thecontrol group at baseline, accounting for the between group differences reported. The studyreported no statistically significant between-group changes on the remaining six satisfactionmeasures, including a measure that directly assessed overall satisfaction with pharmacy services.When grading strength of evidence, we did not consider the results from the remainingcluster trial RCT 64 and the cohort study 70 because they were rated as high risk of bias. We alsodid not consider findings from three other studies (one RCT, 72 one nonrandomized clinicaltrial, 49 and one cohort study 42 ) because they assessed only changes in satisfaction over time inthe intervention arm and did not make any between-group comparisons. Overall, we concludedthat the strength of evidence for MTM interventions with respect to patient satisfaction was lowfor no benefit (Table 44).Table 44. Patient satisfaction: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Findings and Direction[Magnitude] of Effect(Analyzed)RCT 3; 1,625(1,463)Medium Consistent Direct Precise 17 of 21 between-groupdifferences small and notstatistically significant; 4statistically significantdifferences ranged inmagnitude from -0.15 to -0.36, favoring MTMAbbreviations: MTM = medication therapy management; RCT = randomized controlled trial.Strength ofEvidenceLow for nobenefitResource UtilizationKey Points: Resource Utilization• Effective MTM interventions might plausibly lead to either an increase or a decrease inresource utilization, depending on the baseline status of the patient and intended goals ofthe intervention. When studies did not present a clear hypothesis or expected direction ofeffect, we were unable to interpret changes in resource utilization outcomes as either abenefit or a harm of MTM interventions.• Evidence was insufficient to assess the effectiveness of MTM in changing numerousmeasures of use of health care resources. These included use of generic medications;several measures of medication costs (costs of patient copays for medication, overalloutlays on medications, medications and other medical costs); outpatient visits and costs;76


laboratory tests and costs; emergency department visits and costs; and and length ofhospital stay.• MTM intervention reduced medication costs for health plans (3 trials, medium studylimitations, consistent, indirect, imprecise).• MTM interventions among patients with a variety of clinical conditions from trials didnot demonstrate a consistent change in the number of hospitalizations when comparedwith usual care, but one cohort study that partially addressed confounding found evidenceof reduced hospitalization in the intervention arm (high study limitations, unknownconsistency, direct, precise). Together, the lack of consistency across studies suggestsinsufficient evidence on the number of hospitalizations.• In one large cohort, MTM interventions appeared to reduce the risk and costs ofhospitalization for patients with diabetes (high study limitations, unknown consistency,direct, precise), but results for patients with unspecified or other clinical conditions didnot support these results.• MTM interventions in the home reduce the rate of hospitalizations for patients with heartfailure (one cohort, high study limitations, direct, precise, low strength of evidence ofbenefit).Detailed Synthesis: Resource UtilizationUse of Generic MedicationsUnderstanding whether a change in the number of medications taken following an MTMintervention is a measure of appropriate resource utilization requires knowledge of the goal ofdrug therapy. A decrease in the number of medications can represent regimen simplification andresolution of therapeutic duplication; thus, it can be interpreted as a measure of appropriateresource utilization. The converse—that is, an increase in number of medications—cannot,however, be interpreted as a measure of inappropriate resource use. An increase in number ofmedications can, in fact, represent appropriate use of resources when it resulted from identifyingand resolving an inadequate drug regimen.Numerous studies provided information on the number of medications at followup inintervention and control arms or on the change in number of medications between baseline andfollowup. 37,49-51,55,57,61,62,64,73,74,82,84-86,93,94 , The use of generic medications, by contrast, can beinterpreted as cost-saving.Three cohort studies examined the use of generic medications (Table 45). One cohort study,designed to identify the impact of 2010 Part D MTM programs, compared cohorts (standalonePrescription Drug Plan or Medicare Advantage Prescription Drug Plan) receiving MTM with acomprehensive medication review with cohorts receiving usual care for congestive heart failure,chronic obstructive pulmonary disease, and diabetes, after limiting the sample to those newlyeligible or enrolled for MTM and controlling for characteristics such as demographics, medicalcomorbidities, condition severity, and intensity of provider care. 62 The study found very lowgeneric substitution ratios, likely because many patients were already on generic medications. Ina small number of instances, intervention arms were statistically significant from control arms,but the direction of change was inconsistent and the total magnitude of change was small. Twostudies evaluated telephone-based MTM 42,57 and one also compared community pharmacy-basedMTM with educational mailings. 57 We assessed both studies as high risk of bias owing to lack of77


adjustment for potential confounding from study design (intervention refusers versus acceptors) 42or lack of capacity of pharmacists or inability to reach patients. 57Table 45. Use of generic medications: Summary of resultsStudyDesign/Risk of Bias Study Arms N analyzed Outcome and TimePeriodPerlroth et al.,2013 62aCohort/MediumPindolia et al., 2009 42Cohort/HighCongestive heartfailureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD,receiving MTM withCMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD,receiving MTM withCMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD, receiving MTMwith CMRComparison—congestive heartfailureG13: enrolled in PDP,usual careG14: enrolled in MA-PD, usual careComparison—Chronicobstructive pulmonarydiseaseG15: enrolled in PDP,usual careG16: enrolled in MA-PD, usual careComparison—DiabetesG17: enrolled in PDP,usual careG18: enrolled in MA-PD, usual careG1: Telephone basedMTM program(acceptors)G2: Usual medicalcare (refusers)G1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912G1: 292G2: 1081Generic substitutionratio within 365 daysafter date of MTMenrollment (forinterventions) orrandomly-assigneddate in 2010 (forcomparators)Increase in the overalluse of generic drugsResultsOdds (95% CI)For CHF/COPD/diabetesdrugsCongestive heart failureG1 vs. G13: 0.001 (-0.000,0.002), p>0.05G3 vs. G14: 0.005 (0.003,0.006), p0.05G7 vs. G16: 0.000 (-0.002,0.002), p>0.05DiabetesG9 vs. G17: -0.000 (-0.000,0.000), p>0.05G11 vs. G18: 0.000 (-0.000,0.000), p>0.05For non-CHF/COPD/diabetesdrugsCongestive heart failureG1 vs., G13: 0.000 (-.002,0.002), p>0.05G3 vs., G14: -0.010 (-0.013,-0.008), p0.05G7 vs., G16: 0.006 (0.003,0.009), p0.05G11 vs., G18: -0.002 (-0.003,-0.001), p


Table 45. Use of generic medications: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outcome and TimePeriodWinston et al.,2009 57Cohort/HighG1: Community G1: 21,336pharmacy MTM G2: 3,436G2: Pharmacist-staffed G3: 49,021call center-based MTMG3: EducationalmailingsWeighted genericsubstitution ratio: 30-day equivalent claimsdivided by total numberof claimsResultsCalculated mean differencesfor G1 vs. G3: 1.2 (95% CI:0.724 to1.676; p


Table 47. Measures used in studies of costs of medicationsTotal ExpendituresPrescription CostsTotal Outlays onStudyon Medications byto PatientsMedicationHealth PlanChristensen etal., 2007 49Difference in patientcopayment forprescriptions over 6monthsFox et al., Mean Medicare Part2009 37 D copayment costsper patient per monthMean Medicare PartD and non- Part DcopaymentsPindolia et al., Out-of-pocket2009 42 prescription costs perhealth plan memberShimp et al., Annualized2012 60 prescription drugcosts for patient-paidamountDifference in insurerpayment forprescriptions over 6monthsNANAAnnualizedprescription drugcosts for University ofMichigan-paidamountMean Medicare Part D NAdrug costs (total MedicarePart D drug costs (patientcopays + insurance planmedication costs +dispensing fees)Total prescription drugcosts per health planmember (2006)NAMedication and OtherCosts CombinedNANAChrischilles etal., 2004 79 NA Mean amount billedper patient for activedrugs (based onMedicaid claims)NANAHirsch et al,2011 76 ; Hirschet al, 2009 77 NA Paid claims amountfor all prescriptionmedications, ARTmedications, andnon-ART medications(total cost-cost ofART medications)Jameson et al., NA Cost of prescription1995 50 drugs over 6 months,based on maximumallowable cost forMedicaidreimbursementMoore et al., NA Total plan-paid costs2013 45 for all dispensedmedications in preandpost-periodsMoczygemba etal., 2012 39Moczygemba etal., 2011 40Moczygemba etal., 2008 41 NA Total Part D drugcosts (based onprescription claimrecords, excludesnon-Part D drug costsNANANANAPaid claims amounts forinpatient, hospitaloutpatient (includesemergency department),outpatient, mental health,laboratory/X-ray, and AIDSWaiver ProgramNATotal plan-paid costs for alldispensed medications inpre- and post-periods+totalplan-paid costs for allcovered medical servicesin pre- and post-periodsNA80


Table 47. Measures used in studies of costs of medications (continued)StudyTotal ExpendituresPrescription CostsTotal Outlays onon Medications byto PatientsMedicationHealth PlanPerlroth et al., NA Total payments NA2013 62 recorded on Part Dclaims for allprescriptionmedications not usedfor treatment ofcondition specific toMTM eligibility (CHF,COPD, or diabetes)Sellors et al.,2003 55 NA Mean dailymedication costs tothe Ontario DrugBenefit ProgramSellors et al., NA Mean daily2001 61 medication costs tothe Ontario DrugBenefit ProgramWittayanukorn NA Mean pharmacyet al., 2013 80 expendituresMean daily medicationcostsMean daily medicationcostsHanlon et al.,1996 84Cowper et al.,1998 94 NA NA Price to the VA for theagent, plus the averagecost of filling prescriptionsJeong et al.,2009 38 NA NA Total prescription costs(full retail cost ofmedication had thepatient not had insurancecoverage)NA NA Average monthly costs ofNA NA Mean drug costsKrska et al.,NA2001 91 prescribed medication perpatient (excluding costs ofprescribed medicines nottaken)Malone et al.,NA2000 532000 51 ;Ellis et al.,2000 52 ;Malone et al.,2001 54 ;Ellis et al.,(calculated from DenverVAMC pharmacydepartment, individualsites, or the VA PharmacyBenefits Managementgroup)NAMedication and OtherCosts CombinedNAMean cost of health careresources per senior (totalcosts, including all hospitalstays)NAMean pharmacyexpenditures+meanmedical expendituresDrug outlays + average perdiem cost of inpatient carebased on annual outputand expenditure data forbed sections in the costdistribution report + costsof surgery based onrelative value weights, andVA costs per relative valueweight + health servicesvalued using 1991estimates of VAMC unitcosts; costs for non-VAhospital care were imputedusing logic underlying VAcost methodologyNA81


Table 47. Measures used in studies of costs of medications (continued)StudyTotal ExpendituresPrescription CostsTotal Outlays onon Medications byto PatientsMedicationHealth PlanPai et al., NA NA Mean drug costsPai, 2009 74 wholesale price)2009 73 ;(calculated from averageMedication and OtherCosts CombinedNA NA Total prescription cost perNA NA Mean medication coststotal cost per patientStaresinic et al.,NA2007 43 MTM program beneficiaryper month ([gross drugcost=ingredient cost paid+ dispensing fee + salestax]/member months inPart D contract)Welch et al.,NA2009 44 per day (from data onstudy beneficiaries’purchases of ambulatoryprescription medications)Williams, 2004 92 NA NA Average monthly NAwholesale price ofprescription and nonprescriptiondrugsWinston et al., NA NA Mean drug cost per NA2009 57 patient per month (baseddrug claims processingdata, total allowedcharges, includingingredient cost paid,dispensing fee, and salestax, before subtractingany patient cost-sharingamounts)Yamada, 2012 48 NA NA Change in annual NAprescription cost (detailsnot specified)Bernsten et al., NA NA NA Mean2001 64 ;Sturgess et al.,2003 65 including (1) costassociatedwith additional time spentby pharmacists; (2) costassociated with contactswith GPs, specialists andnurses; and (3) cost ofhospitalizations and drugsFischer et al.,2002 82 NA NA NA Change in total charges forinpatient care, outpatientcare, and pharmacychargesAbbreviations: ART = antiretroviral therapy; CHF = congestive heart failure; COPD = chronic obstructive pulmonary disease;GPs = general practitioners; MTM = medication therapy management; NA = not applicable; VA = Veterans’ Administration;VAMC = Veterans’ Administration Medical CenterMedication Costs: Patient CopaymentsFour studies (one medium risk-of-bias RCT, 60 one nonrandomized controlled trial [NRCT] ofmedium risk of bias 49 and two cohort studies of high risk of bias 37,42 ) compared the copaymentsfor patients who refused MTM with patients who accepted MTM enrollment. These studiesprovided inconsistent evidence that patient medication co-payments increased following MTM.Table 48 documents the main findings. We calculated mean differences between groups whenNA82


the original authors did not provide those data; all currencies are rounded to two decimals (i.e.,for U.S. currency, cents). The trial showed a decrease in costs of $234 in the intervention armand $170 in the control arm, but without information on numbers in each arm, we cannotcalculate variance.Table 48. Patient copayments: Summary of resultsStudyN Prescription Costs toDesign/Risk of Study ArmsAnalyzed PatientsBiasShimp et al.,2012 60RCT/MediumChristensen et al.,2007 49NRCT/MediumFox et al., 2009 37Cohort/HighPindolia et al.,2009 42Cohort/HighG1: MTM program for G1: NRUniversity of Michigan G2: NRbeneficiaries, entitled Focuson MedicinesG2: Usual care (notdescribed)G1: Patients receiving G1: 67pharmacist-provided MTM G2: 669servicesG3: 870G2: Patients from samecounties as G1 who did notreceive intervention (controlgroup 1)G3: Patients from adifferent county than G1who did not receiveintervention (control group2)G1: MTM program(acceptors)G2: Opt-out from MTMprogram (refusers)G1: Telephone-based MTMprogram (acceptors)G2: Usual medical care(refusers)G1: 247G2: 50G1: 292G2: 1,081Annualized prescription drugcosts for patient-paidamountMean difference in patientcopayment for prescriptionsover 6 months in USD (SD)Mean difference in MedicarePart D medicationcopayment costs per patientper monthMean difference in allmedication copayments(Medicare Part D and notPart D) per patient permonthResults12 months before first visitG1: 1,334 ± 593G2: 1,293 ± 68095% CI: NRp: NR at baseline12 months after secondvisitG1: 1,100 ± 645G2: 1,123 ± 64395% CI: NRp: NR at followupMean difference withvariance between armsnot calculable without N,reported P for G1 frombaseline to followup: 0.004p for G2 from baseline tofollowup: 0.062Calculated meandifference for G1 vs. G2:80.40 USD; 95% CI, 10.43to 150.37 p=0.024Calculated meandifference for G1 vs. G3=88.60 USD;95% CI, 24.61 to 152.59p=0.007Calculated meandifference:-3.92 USD,95% CI, -25.71 to 17.87p=0.724Calculated meandifference: -1.71 USD95% CI, -24.53 to 21.11p=0.883Mean out-of-pocket Calculated meanprescription costs per health difference: 77.00 USD;plan member in USD 95% CI, -71.82 to 225.82(assumed per year, as NR in p=0.311study) (SD)Abbreviations: CI = confidence interval; G = group; MTM = medication therapy management; NR = not reported; NRCT =nonrandomized controlled trial; RCT = randomized controlled trial; SD = standard deviation; USD = United States dollars.83


The NRCT compared patients in the MTM arm with controls within and outside theintervention county; the control arms had declines in copayments and the MTM had increases incopayments. The two cohort studies had inconsistent and imprecise estimates of effect; one studyshowed an increase in copayments for the MTM arm and a decline for the control arm, 42 and theother reported a smaller increase in the MTM arm than in the control arm. 37 None of thesestudies explained whether the increase in copayment was a result of an appropriate change inmedication therapy or the desired effect of the intervention. Although the results were precise inthe NRCT and suggested an increase in medication copayments following MTM, the lack ofdirectness in interpreting this outcome as a measure of appropriate resource utilization and theabsence of other low and medium risk-of-bias studies with sufficient information to assessconsistency of findings suggests insufficient evidence to judge the effect of MTM interventionson patient medication co-payment (Table 49).Table 49. Patient copayments: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects Limitations Consistency(Analyzed)RCT 1, NR (NR) Medium Consistencyunknown,single studyNRCT 1; 1,639(1,606)HighConsistencyunknown,single studyFindings and DirectionDirectness Precision[Magnitude] of EffectIndirectNotreportedCalculated meandifference= -64 USD,variance not calculableIndirect Precise Calculated meandifference for MTM vs.same country control:80.40 USD; 95% CI,10.43 to 150.37 p=0.024Strength ofEvidenceInsufficientInsufficientCalculated meandifference for MTM vs.different county control:88.60 USD; 95% CI,24.61 to 152.59 p=0.007Abbreviations: CI = confidence interval; MTM = medication therapy management; NR = not reported; NRCT = nonrandomizedcontrolled trial; USD = United States dollars.Medication Costs: Expenditures by InsurersThree RCTs (all medium risk of bias), 50,55,61 the NRCT reported on above, 49 and six cohortstudies (four medium 39-41,45,62,80 and two high risk of bias 76,77,79 ) measured the net effect of MTMon expenditures incurred by insurers on medications (Table 50). Changes in health plan drugexpenditures attributable to MTM depend on the net effect of MTM activities, which can entailadding clinically needed drugs, increasing doses or frequency, substituting therapeuticallyequivalent lower cost drugs, and simplifying regimens (singly or in combination). For individualpatients, a net increase in expenditures may be the outcome of a more appropriate drug regimen.Included studies provided only the net effect on expenditures at the study arm level. All trialsdemonstrated that MTM either reduced health plan expenditures or limited the increase inexpenditures over time for patients receiving the MTM intervention when compared withpatients in the control or comparison arm. These results were not precise, however; confidenceintervals included the null effect for all but one trial. Results from the nonrandomized studiescame from very disparate studies: the smallest included 120 patients and the largest as many as200,722 patients. The inherent heterogeneity within and across these studies likely explains thelack of consistency in direction, magnitude, and precision of effects.84


Table 50. Total expenditures on medications by insurers: Summary of resultsStudyDesign/Risk of BiasJameson et al., 1995 50RCT/MediumSellors et al., 2003 55RCT/MediumSellors et al., 2001 61RCT/MediumStudy ArmsG1:PharmacotherapyconsultationG2: Usual careN AnalyzedG1: 27G2: 29G1: Pharmacist G1: 379consultation program G2: 409G2: Usual careG1: Pharmacist G1: 61consultation program G2: 60G2: Usual carePrescription Coststo InsurersResultsChange in cost of Calculated mean difference:prescription drugs USD -293.00over 6 months, based 95% CI: -501.50 to -84.50on maximum p


Table 50. Total expenditures on medications by insurers: Summary of results (continued)StudyDesign/Risk of BiasPerlroth et al., 2013 62aCohort/MediumStudy ArmsCongestive heartfailureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD, receiving MTMwith CMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD, receiving MTMwith CMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD, receiving MTMwith CMRComparison—congestive heartfailureG13: enrolled inPDP, usual careG14: enrolled in MA-PD, usual careComparison—Chronic obstructivepulmonary diseaseG15: enrolled inPDP, usual careG16: enrolled in MA-PD, usual careComparison—DiabetesG17: enrolled inPDP, usual careG18: enrolled in MA-PD, usual careN AnalyzedG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13:156,441G14: 51,938G15:184,350G16: 73,623G17:133,925G18: 53,912Prescription Coststo InsurersTotal paymentsrecorded on Part Dclaims for allprescriptionmedications not usedfor treatment ofcondition specific toMTM eligibility (CHF,COPD, or diabetes)within 365 days afterdate of MTMenrollment (forinterventions) orrandomly-assigneddate in 2010 (forcomparators)ResultsAdjusted costs in USD (95%CI)Congestive heart failurecohortG1 vs. G13: 87.05 (7.33,166.78), p


Table 50. Total expenditures on medications by insurers: Summary of results (continued)StudyDesign/Risk of BiasWittayanukorn et al., 2013 80Cohort/MediumChrischilles et al., 2004 79Cohort/HighHirsch et al., 2011 76Hirsch et al., 2009 77Cohort/HighStudy ArmsN AnalyzedG1: Pharmacist G1: 63provided face-to-face G2: 62MTM services for30–60 minutes perencounter, notalways including afollowup visitG2: Patients who didnot receive MTMservicesG1: PCM-eligiblepatients whoreceived PCMservicesG2: PCM-eligiblepatients who did notreceive PCMservicesG1: Patients servedat pilot pharmaciesG2: Patients servedat nonpilotpharmaciesG1: 524G2: 1,6872005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Prescription Coststo InsurersResultsMean pharmacy Mean between-group costexpenditures, during difference in USD (SD):the 6 months prior to -31.9 (25.1)the initial MTM visit p


Table 51. Health plan expenditures: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects Limitations(Analyzed)Consistency Directness PrecisionFindings and Direction[Magnitude] of EffectRCT 3; 1,085 (965) Medium Consistent Indirect Imprecise Mean difference variesfrom -34 CAD to -293USD over 6 monthsNRCTandCohort5; 125–200,722(120–200,722)High Inconsistent Indirect Imprecise Mean difference variesfrom -800 USD over 1year to 425 USD over 2yearsStrength ofEvidenceLow forbenefitInsufficientAbbreviations: CAD = Canadian dollars; NRCT = nonrandomized controlled trial; RCT = randomized controlled trial; USD =United States dollars.Medication Costs: Total Outlays on MedicationsSeven RCTs (one low risk of bias, 84,94 five medium risk of bias 51-55,84,91,92,94 and one high riskof bias 73,74 ) and seven cohort studies, two medium 38,48 and five high risk of bias, 37,42-44,57measured the effect of MTM on total outlays on medications. As with other data on resourceutilization, we found it challenging to interpret inconsistent results when studies did not specifythe expected mechanism of action and predicted direction of effect. An additional challengerelates to the wide variation in data sources and degree of clarity on how investigators calculatedoutlays. In some studies, the specific measure used includes the combination of expendituresincurred by insurers and patients for prescription medications (Table 52). In other studies, themeasure is based on wholesale costs or full retail costs in the absence of insurance, but whetherand how the cost is split between the insurer and the patient is unclear, nor is it clear how thesewholesale or retail costs relate to actual incurred costs. Calculated differences in Table 52 arerounded to two decimals.Table 52. Total outlays on medications: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Outlays onprescriptionsHanlon et al.,1996 84,94RCT/LowKrska et al., 2001 91RCT/MediumG1: Usual care, plusclinical pharmacistcare.G2: Usual care in theGeneral MedicineClinicG1: 105G2: 103G1: Pharmacist-led G1: 168medication review G2: 164G2: Usual careincluding identificationof pharmaceutical careissues, but no planAnnual price to the VAfor the agent, plus theaverage cost of fillingprescriptions (timeperiod NR) (25th-75thpercentile)Average monthly costsof prescribedmedication per patientin British pounds (£)(SD) at 3 months(calculated usinginformation frompatients on actual use)ResultsMean cost in USDG1: 1,006 (574–1,285)G2: 1,096 (566–1,456)95% CI: NRp: NS at 0.05 level, specificsNRCalculated mean differenceper month: - 90/12= -7.5 USDCalculated mean difference:-£.0.19,95% CI: -6.69 to 6.49p=0.956.88


Table 52. Total outlays on medications: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outlays onprescriptionsMalone et al., 2000 51 ; G1: PharmaceuticalEllis et al., 2000 52 ; careMalone et al., 2001 54 ; G2: Usual careEllis et al., 2000 53RCT/MediumSellors et al., 2003 55RCT/MediumSellors et al., 2001 61RCT/MediumWilliams, 2004 92RCT/MediumJeong et al., 2009 38Cohort/MediumYamada et al.,2012 48Cohortstudy/MediumPai, 2009 73 ; Pai,2009 74RCT/HighFox et al., 2009 37Cohort/HighG1: Pharmacistconsultation programG2: Usual careG1: Pharmacistconsultation programG2: Usual careG1: 523G2: 531G1: 379G2: 409G1: 61G2: 60G1: Modification of G1: 57patient’s medication G2: 76regimen by aninterdisciplinarymedication adjustmentteamG2: Usual medicalcareG1: Participants in Part G1: 2,780D Medicare MTM G2: 2,251program (opted in toMTM program)G2: Control subjectswithout Part DMedicare as theirprimary drug benefitbut otherwise similar tointervention subjects.G1: MTM enrolledpatientsG2: Eligible MTMpatients not enrolledbut matched on age,gender, region andDCG riskG1: PharmaceuticalcareG2: Usual careG1: MTM program(acceptors)G2: Opt-out from MTMprogram (refusers)G1: 34,352G2: 138,182G1: NRG2: NRG1: 247G2: 50Mean change in annualdrug costs in(calculated from DenverVAMC pharmacydepartment, individualsites, or the VAPharmacyBenefits Managementgroup)ResultsCalculated mean difference:USD 63.0095% CI: -5.08 to 131.08;p=0.07Calculated mean differenceper month: USD 63/12=USD5.25Mean daily medication Calculated mean difference:costs per patient at 5 0.19 (assumed CAD)months (assumed CAD) 95% CI: -0.85 to 1.23p=0.72Mean daily medicationcosts (assumed CAD)at 6 monthsAverage monthlywholesale price ofprescription andnonprescription drugsFull retail cost ofmedication had thepatient not hadinsurance coverage.at6 months before and 6months after enrollmentin USDChange in annualprescription costsMean drug costs(calculated fromaverage wholesaleprice) over 2 yearsMean difference inannual Medicare Part Ddrug costs (patientcopayment + insuranceplan medication costs +dispensing fee)Calculated mean difference:-0.41 CAD (-1.40 to 0.58)p: 0.42Reported mean difference:-20.16 USD95% CI: -5.78 to -34.54p: 0.006Calculated mean difference inUSD (95% CI): -563 (-735.33to -390.67)p


Table 52. Total outlays on medications: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Outlays onprescriptionsPindolia et al., 2009 42Cohort/HighStaresinic et al.,2007 43Cohort/HighWelch et al., 2009 44Cohort/HighWinston et al.,2009 57Cohort/HighG1: Telephone basedMTM program(acceptors)G2: Usual medicalcare (refusers)G1: MTP program(acceptors)G2: Usual care(refusers)G1: MTM programprovided to homebasedbeneficiariesG2: Opt-out amonghome-based patientseligible for MTMG1: Communitypharmacy MTMG2: Pharmaciststaffed,call-centerbasedMTMG3: EducationalmailingsG1: 292G2: 1,081G1: 282G2: 1,544G1: 459G2: 336G1: 21,336G2: 3,436G3: 49,021ResultsTotal annualCalculated mean difference:prescription drug cost -62.22, 95% CI -112.469 toper health plan member -11.971; p=0.015in USDTotal prescription costper MTM programbeneficiary per month(gross drugcost=ingredient costpaid + dispensing fee +sales tax per membermonthsin Part Dcontract)Mean change inmedication costs perday at 6 months.(Estimates come fromdata on studybeneficiaries’purchasesof ambulatoryprescriptionmedications)Mean percentageincrease in medicationcosts per day at 6months (no SDreported)Mean (SD) drug costper patient per monthafter 8 months ofservices (based ondrug claims processingdata, total allowedcharges, includingingredient cost paid,dispensing fee, andsales tax, beforesubtractingany patient cost-sharingamounts)Participants spent less onprescription medications onaverage (described as permember per month drugspending) thannonparticipants. Figureprovided suggested adecrease in spending ofbetween USD 100 and USD150 in the intervention group,but exact numbers notreported.Difference in difference: USD3.62, SD NR, adjustedp=0.203NOTE: Age, sex, chronicdisease score, and preperioddrug cost included inmultivariate regressionmodeling for adjusted PAdjusted OR : 1.4 95% CI: 1.1to 1.9NOTE: Model adjusted forage, sex, chronic diseasescore, and baselinemedication costCalculated mean difference forG1 vs. G3: USD -35.00,95% CI -43.390 to -26.610;p


demonstrated inconsistent results; 38,48 in fact, the same study demonstrated a difference inoutcomes based on the specific criteria for MTM enrolment by year within the program. 48 Thehigh risk-of-bias studies similar reported inconsistent results: some reported reducedoutlays 42,43,57 and others showed increased outlays 37,44 or no effect 73,74 following MTM.Based on the lack of consistency, directness, and precision, we rated the evidence from fivemedium risk-of-bias trials and two cohort studies as insufficient to evaluate the effect of MTMon total outlays on medications (Table 53).Table 53. Total outlays on medications: Strength of evidenceNumber ofFindings andStudy Studies; StudyStrength ofConsistency Directness Precision Direction [Magnitude]Design Subjects LimitationsEvidenceof Effect(Analyzed)RCT 6; 2,804 (2,636) Medium Inconsistent Indirect Imprecise Mean difference varies Insufficientfrom -20.16 USD to+5.25 USD per monthCohort 2, 177,565(177,565)High Inconsistent Indirect Imprecise Mean difference variesfrom -563 USD to +310USD annuallyAbbreviations: RCT=randomized controlled trial; USD= United States dollars.InsufficientMedication Costs: Combined Medication and Other CostsThree trials (one low, 84,94 one medium, 55 and one high risk of bias 64,65 ), one NRCT (medium riskof bias 82 ), and three cohort studies (two medium 45,80 and one high risk of bias 76,77 ) providedinconsistent evidence of change in combined medication and other costs (variably defined ineach study) (Table 54). Studies did not report their results in sufficient detail to allow pooling.Based on available information, we judged the evidence to be insufficient to evaluate the effectof MTM on combined medication and other costs (Table 55).Table 54. Medication and other costs: Summary of resultsStudyDesign/Risk of Bias Study Arms N Analyzed Medication and OtherCostsHanlon et al.,1996 84,94RCT/LowG1: Usual care, plusclinical pharmacistcare.G2: Usual care in theGeneral MedicineClinicG1: 105G2: 103ResultsTotal cost = drug outlays Mean cost in USD+ average per diem cost G1: 7,873 (1,622–6,649)of inpatient care based on G2: 5,926 (1,608–7,036)annual output and 95% CI: NRexpenditure data for bed p: NS at 0.05 level, specificssections in the cost NRdistribution report + costsof surgery based onrelative value weights,and VA costs per relativevalue weight + healthservices valued using1991 estimates of VAMCunit costs; costs for non-VA hospital care wereimputed using logicunderlying VA costmethodology (time periodNR) (25th-75thpercentile)91


Table 54. Medication and other costs: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Medication and othercostsSellors et al., 2003 55RCT/MediumBernsten et al.,2001 64,65RCT/HighG1: Pharmacistconsultation programG2: Usual careG1: Structuredcommunity pharmacybasedpharmaceuticalcare programG2: Usual communitypharmacy servicesG1: 379G2: 409BaselineG1: 867G2: 7486 monthsG1: NRG2: NR12 monthsG1: NRG2: NRMean cost of health careresources per patient,including all hospitalstays at 5 months (CADassumed)Mean cost of health careresources per patient,including only drug (i.e.,medication)-relatedhospital stays at 5months (CAD assumed)ResultsCalculated mean difference:249.41 (assumed CAD),95% CI: -338.39 to 837.21;p=0.406Calculated mean difference:-8.10 (assumed CAD),95% CI: -386.72 to 4,350.52;p=0.923Mean total cost per Cost data not pooled andpatient including (1) cost analyzed for costs becauseassociated with additional health care systems differedtime spent bybetween 7 countriespharmacists; (2) cost included in the study.associated with contacts However, authors reportedwith GPs, specialists and no significant between-groupnurses; and (3) cost of differences in any countryhospitalizations and (p=NS)drugsFischer et al., 2002 82NRCT/MediumMoore et al.,2013 45Cohort/MediumWittayanukorn et al.,2013 80Cohort/MediumG1: PharmaceuticalcareG2: Usual careG1: MTM program (opt-in)G2: control group (refusers)G1: Pharmacistprovided face-to-faceMTM services for 30-60 minutes perencounter, not alwaysincluding a followupvisitG2: Patients who didnot receive MTMservices18 monthsG1: NRG2: NRG1: 231G2: 444G1: 2250G2: 2250G1: 63G2: 62Change in total chargesfor inpatient care,outpatient care, andpharmacy chargesG1: -900G2: -2,00095% CI: NRP: NS, no details reportedCalculated mean difference:USD 1,100.Total plan-paid costs for Calculated meanall dispenseddifference in USD (95%medications + total planpaidcosts for all covered 6.849)CI): -1,039 (-2,084.85 tomedical services 1 year p=0.052before invitation to MTMprogram and 1 year afterMean total expenditures(pharmacy+medical),during the 6 months priorto the initial MTM visitand costs during the 6months after the initialMTM visit, in USDMean between-group costdifference in USD (SD): -359.3 (219.2)p


Table 54. Medication and other costs: Summary of results (continued)StudyDesign/Risk of Bias Study Arms N Analyzed Medication and othercostsHirsch et al., 2011 76Hirsch et al., 2009 77Cohort/HighG1: Patients served atpilot pharmaciesG2: Patients served atnonpilot pharmacies2005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Paid claims amounts forinpatient, hospitaloutpatient (includesemergency department),outpatient, mental health,laboratory/X-ray, andAIDS Waiver ProgramcResultsMean USD (SE)2005G1: 35,546 (1,093)G2: 33,501 (505)p=0.0792006G1: 36,806 (980)G2: 35,230 (575)p= 0.1572007G1: 38,983 (1,023)G2: 38,856 (633)p= 0.915Abbreviations: CAD = Canadian dollars; CI = confidence interval; G = group; GP = general practitioner; MTM = medicationtherapy management; NR = not reported, NS = not significant; RCT = randomized controlled trial; SE = standard error; USD =United States dollars.Table 55. Medication and other costs: Strength of evidenceNumber ofFindings andStudy Studies; Studydesign Subjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCTNRCTandCohort2; 1097, N(996)3; 5,300(5,300)Medium Inconsistent Indirect Imprecise Differences inmean costsranging from-8.1 CAD to1,947 USDHigh Inconsistent Indirect Imprecise Differences inmean costsranging from to-1,039 to 1,100USDStrength ofEvidenceInsufficientInsufficientAbbreviations: CAD = Canadian dollars; NRCT = nonrandomized controlled trial; RCT = randomized controlled trial; USD =United States dollars.Number of Outpatient VisitsEleven studies examined the effect of MTM interventions, when compared with usual care,on outpatient visits. These studies varied in geographic setting (seven Western Europeancountries, 64,65 the United States, 45,51-54,56,69-71,79,82,84,94 the United Kingdom, 91 Canada 55 ), period ofevaluation (3 months to 36 months), specific outcome measure (ranging from a focus on visitswith physicians to total ambulatory care visits or contacts with physicians and nurses), and riskof bias. They are described in Table 56. No study indicated whether the intervention wasspecifically designed to increase or to decrease outpatient visits; as a result, the directionality ofthe results cannot be interpreted as a benefit or a harm.93


Table 56. Number of outpatient visits: Summary of resultsStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowKrska et al.,2001 91RCT/MediumMalone, 2000 51 ;Ellis, 2000 52 ;Malone, 2001 54 ;Ellis, 2000 53RCT/MediumSellors et al.,2003 55RCT/MediumSidel, 1990 56RCT/MediumTouchette et al.,2012 69RCT/MediumBernsten et al.,2001 64,65RCT/HighG1: Usual care, plusclinical pharmacist care.G2: Usual care in theGeneral Medicine ClinicG1: Pharmacist-ledmedication reviewG2: Usual care includingidentification ofpharmaceutical careissues, but no planG1: Pharmaceutical careG2: Usual careG1: Pharmacistconsultation programG2: Usual careG1: Patients received atleast 2 pharmacist visitsinvolving medicationreview, patient-specificeducation andcounseling; followuppatient telephone callsand contact ofphysicians as neededG2: Patients contactedonly to complete thesurvey.G1: 105G2: 103G1: NRG2: NRG1: 523G2: 531G1: 379G2: 409G1: 92G2: 104G1: MTM basic G1: 183(comprehensive G2: 190medication review and G3: 183DRP assessment)G2: MTM enhanced(MTM plus 2 pageclinical summaryabstracted from patient’smedical chart).G3: Usual careG1: Structuredcommunity pharmacybasedpharmaceuticalcare programG2: Usual communitypharmacy servicesG1: 1,024G2: 953Mean general medicineclinic visits (time periodNR) (25th–75thpercentile)Mean other clinic visits(time period NR) (25th–75th percentile)Hospital clinicattendance, use ofsocial services orcontacts with districtnurses and healthvisitors before and afterthe pharmacist reviewMean change in numberof clinic visits (includingvisits with thepharmacists in theintervention arm) over12 monthsNumber of clinic visitsover 5 mothsChange in number ofambulatory visits overpast 3 months,measured at baselineand again at 36 months3-6 monthsG1: 183G2: 190G3: 183ResultsG1: 5.5 (3–6)G2: 5.8 (3–7)95% CI: NRp: NS at 0.05 level, specificsNRG1: 7.7 (3–10)G2: 10.9 (2–15)95% CI: NRp: NS at 0.05 level, specificsNRNo differences; details NRCalculated mean difference:2.0,95% CI: -0.415 to 4.415,p= 0.104Calculated mean difference:-0.02,95% CI: -1.274 to 1.234,p=0.975Calculated mean difference:-1.41,95% CI: -2.98 to 0.160,p=0.078G1 vs. G3 Calculated meandifference: 0.50, 95% CI: -0.388 to 0.488, p=0.823G2 vs. G3 Calculated meandifference: -0.50, 95% CI: -0.383 to 0.483, p=0.821Mean number of Calculated mean difference:contacts with primary 0.120care providers, including 95% CI: -0.461 to 0.701,home visits and office p=0.686appointments at 6months94


Table 56. Number of outpatient visits: Summary of results (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasMoore et al.,2013 45Cohort/MediumFischer et al.,2002 82NRCT/HighCarter et al.,1997 70,71Cohort/HighChrischilles et al.,2004 79Cohort/HighG1: MTM program (optin)G2: control group(refusers)G1: Pharmaceutical careG2: Usual careG1: Pharmaceutical careG2: Usual careG1: PCM-eligiblepatients who receivedPCM servicesG2: PCM-eligiblepatients who did notreceive PCM servicesG1: 2,250G2: 2,250G1: 231G2: 444G1: 25G2: 26G1: 524G2: 1,687ResultsChange in number of Calculated mean differencephysician visits from 365 (95% CI): 2.48 (1.674 to 3.286,days preceding the p


Table 57. Number of outpatient visits: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects Limitations Consistency(Analyzed)Directness Precision Findings andDirection[Magnitude] of EffectRCT 3; 2,362 (2,208) Medium Inconsistent Indirect Precise Standardized meandifference: 0.049; 95%CI, -0.034 to 0.133,p=0.247; I 2 =0Cohort 1; 2250 (2250) High Consistencyunknown-singlestudyIndirectAbbreviations: CI = confidence interval; RCT= randomized controlled trial.Imprecise Calculated meandifference (95% CI):2.48 (1.674 to 3.286,p


Table 58. Costs of outpatient visits: Summary of results (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasSellors et al.,2003 55RCT/MediumG1: Pharmacistconsultation programG2: Usual careG1: 379G2: 409Mean cost of physicianvisits (assumed CAD) at5 monthsMean cost of clinic visits(assumed CAD) at 5monthsResultsCalculated mean difference:5.66, 95% CI: -24.22 to 35.54,p=0.71Monthly cost difference:5.66/5= 1.132 CADCalculated mean difference:-2.13, 95% CI: -20.46 to 16.20,p=0.82Carter et al.,1997 70,71Cohort/HighG1: PharmaceuticalcareG2: Usual careG1: 25G2: 26Monthly cost difference: -2.13/5= - 0.43 CADMean cost of other Calculated mean difference:health care services and -4.70visits to health care 95% CI: -140.22 to 130.82;professionals (assumed p=0.95CAD) at 5 monthsMonthly cost difference: -4.70/5= - 0.94 CADHypertension-relatedcharges (SD) at 6monthsCalculated mean difference:USD 70.00, 95% CI: 15.97 to124.03, p=0.011Mean visit charges at 6monthsMonthly cost difference: 70/6=11.67 USDCalculated mean difference:USD 487.00, 95% CI: 44.87 to929.14, p=0.031Chrischilles et al.,2004 79Cohort/HighHirsch et al.,2011 76Hirsch et al.,2009 77Cohort/HighG1: PCM-eligiblepatients who receivedPCM servicesG2: PCM-eligiblepatients who did notreceive PCM servicesG1: Patients served atpilot pharmaciesG2: Patients served atnonpilot pharmaciesG1: 524G2: 1,6872005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Outpatient facility claimsat 12 monthsOutpatient costs (notdefined)Monthly cost difference: 487/6=81.17 USDResults NR, p=0.107Mean USD (SE)2005G1: 112 (11)G2: 44 (2)p


One low risk-of-bias and two medium risk-of-bias trials offered inconsistent evidence on theeffect of MTM interventions on outpatient costs. Two U.S.-based VA studies 51-54,84,94 foundresults favoring the intervention group, but the results were not statistically significant in onestudy. 84,94 The Canadian study found no significant differences by study arm. 55 Two cohortstudies (high risk of bias) found significantly higher costs for the intervention arm than the usualcare arm. 70,71,76,77 Another U.S.-based cohort study of the Medicaid program in Iowa (high risk ofbias) found no statistically significant differences in cost of outpatient visits by intervention armbut did not report details to determine direction of effect. 79Based on the lack of consistency and precision, we graded the body of evidence from the twotrials as being insufficient to evaluate the effect of MTM interventions on the costs of outpatientvisits (Table 59).Table 59. Costs of outpatient resource utilization: Strength of evidenceNumber ofFindings andStudyStudyStudies; SubjectsConsistency Directness Precision DirectionDesignLimitations(Analyzed)[Magnitude] of EffectStrength ofEvidenceRCT 3; 2,151 (2,050) Medium Inconsistent Indirect Imprecise Variable estimates a Insufficienta Dissimilar time periods and measures, ranges are not meaningful.Abbreviation: CAD = , Canadian dollar; RCT= randomized controlled trial; USD = , U.S. dollars.Number of Laboratory and Diagnostic TestsUnderstanding whether a change in the number and costs of laboratory tests as a result of anMTM intervention measures appropriate resource use requires knowledge of the goals of drugtherapy. MTM could raise numbers and costs of laboratory and diagnostic tests by identifyingpatients who should be receiving more frequent laboratory monitoring or by starting patients onnew drugs that require laboratory monitoring based on their clinical situation. However, MTMcould also lower numbers and costs of laboratory and diagnostic tests if it produces bettercoordination of care and prevents duplicative testing. Included studies did not specify theexpected direction of effect from MTM on the number and costs of laboratory and diagnostictests.Two trials (both medium risk of bias; one set in the United States 51-54 the other in Canada 55 )reported on the number of laboratory tests following MTM interventions (Table 60). TheCanadian study included the number and costs of imaging procedures over a 5-month period; 55the U.S.-based study did not specify the inclusion of imaging procedures and evaluated tests andcosts over a 12-month period. The U.S.-based found statistically significant differences; theCanadian study failed to find any significant differences.98


Table 60. Number of laboratory and diagnostic tests: Summary of resultsStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedResultsPeriodBiasMalone, 2000 51 ;Ellis, 2000 52(interventions);Malone, 2001 54(detailed QOLoutcomes);Ellis, 2000 53RCT/MediumSellors et al.,2003 55RCT/MediumG1: PharmaceuticalcareG2: Usual careG1: Pharmacistconsultation programG2: Usual careG1: 523G2: 531G1: 379G2: 409Mean change in annualnumber of laboratorytestsCalculated mean difference:-1.6, 95% CI: -2.550 to -0.650,p=0.001Mean number of Calculated mean difference:laboratory tests and 0.15, 95% CI: -0.959 to 1.259,imaging procedures at 5 p=0.791monthsAbbreviations: CI = confidence interval; G =group; N = number; QOL = quality of life; RCT= randomized controlled trial.The small number of studies limits our ability to explore causes for the observedheterogeneity. Factors such as differences in health systems, period of evaluation, and definitionof the outcome could explain differences in results. Based on lack of consistency, we graded thebody of evidence from these two medium risk-of-bias trials as insufficient to evaluate either theeffect of MTM interventions on the number of laboratory and diagnostic and diagnostic tests(Table 61).Table 61. Number of laboratory and diagnostic tests: Strength of evidenceStudyDesignNumber ofStudyStudies; SubjectsLimitations(Analyzed)Consistency Directness PrecisionFindings andDirection[Magnitude] ofEffectRCT 2; 1,943 (1,842) Medium Inconsistent Indirect Imprecise Differences rangefrom +0.15 to -1.6testsAbbreviations: RCT = randomized controlled trial.Strength ofEvidenceInsufficientCosts of Laboratory and Diagnostic TestsThe two medium risk-of-bias studies reporting data on number of laboratory and diagnostictests also provided information on costs. 51-55 Two other studies, one low risk of bias 84,94 and onehigh risk of bias, 76,77 provided information on costs. The challenges associated with interpretingevidence on number of laboratory and diagnostic tests apply to costs as well (Table 62). Thethree U.S.-based studies showed a trend of reduced costs of laboratory and diagnostic tests in theintervention arm, but these results were not consistent in magnitude of effect or precision. 51-54,76,77,84,94 The Canadian study showed an increase in laboratory and imaging costs in theintervention arm. 55 99


Table 62. Costs of laboratory tests: Summary of resultsStudyOutcome and TimeDesign/Risk of Study Arms N AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowMalone, 2000 51 ;Ellis, 2000 52 ;Malone, 2001; 54Ellis, 2000 53RCT/MediumSellors et al.,2003 55RCT/MediumHirsch et al.,2011 76Hirsch et al.,2009 77Cohort/HighG1: Usual care, plusclinical pharmacistcare.G2: Usual care in theGeneral MedicineClinicG1: PharmaceuticalcareG2: Usual careG1: Pharmacistconsultation programG2: Usual careG1: Patients servedat pilot pharmaciesG2: Patients servedat nonpilotpharmaciesG1: 105G2: 103G1: 523G2: 531G1: 379G2: 4092005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606ResultsAnnual health care costs G1: 214 (52–318)for diagnostic tests at 1- G2: 354 (55–473)year closeout or adjusted 95% CI: NRto a 1-year followup and p: NS at 0.05 level, specifics NRweighted by actual timefor censored patients,(25th–75th percentile)Mean change in annualcosts for laboratory testsMean cost of all lab andimaging procedures at 5months (assumed CAD)Costs of laboratory/x-rayservicesCalculated mean difference:USD -33.0095% CI: -65.96 to -0.04, p=0.05Calculated mean difference:6.24 (assumed CAD) over 5months, 14.98 over 12 months95% CI: -46.34 to 58.88 p=0.816Mean USD (SE)2005G1: 389 (34)G2: 402 (17)p=0.7362006G1: 387 (28)G2: 407 (18)p=0.5302007G1: 401 (29)G2: 402 (18)p=0.974Abbreviations: CAD = Canadian dollars; CI = confidence interval; G = group; N = number; NR = not reported; NS = notsignificant; RCT = randomized controlled trial; SE = standard error; USD = United States dollars.Based on lack of consistency, we graded the body of evidence from these three medium riskof-biastrials as insufficient to evaluate either the effect of MTM interventions on the costs oflaboratory tests (Table 63).Table 63. Costs of laboratory tests: Strength of evidenceStudyDesignNumber ofStudies;Subjects(Analyzed)RCT 3; 2051(2,050)StudyLimitationsConsistency Directness PrecisionFindings andDirection[Magnitude] ofEffectMedium Inconsistent Indirect Imprecise Differencesrange from +15CAD to -140USDAbbreviations: CAD= Canadian dollars; RCT = randomized controlled trial; USD = United States dollars.StrengthofEvidenceInsufficientEmergency Department VisitsNine studies reported changes in emergency department (ED) visits following MTMinterventions: four trials (one low risk of bias, 84,94 two medium risk of bias, 55,69 and one high risk100


of bias 85 ) and five cohort studies (four medium risk of bias 44,45,62,80 and one high risk of bias 46,47 )(Table 64). We excluded data from one study in the analysis below because it reported totalnumber of events by each intervention arm rather than by patients within intervention arm. 80 Wecould not pool results from the eight studies with complete data because we did not havesufficient numbers of studies with similar designs. Across all low and medium risk-of-biasstudies, the confidence intervals for the effects from the medium risk-of-bias studies spanned thenull effect for the entire study 44,55,84,94 or a subset of analyses. 62,69 Studies with multiplecomparisons found some signals of benefit. For example, the low risk-of-bias trial found a lowermean number of ED visits in the intervention arm when compared with the usual care, but thiseffect was statistically significant only for the basic MTM arm. 69 Likewise, a medium risk-ofbiascohort study of 2010 Medicare Part D found consistently lower odds of all-cause ED visitsacross all clinical conditions but did not find the same consistency of direction, magnitude, orprecision for condition-specific ED visits. 62 One trial, rated high risk of bias for this outcome,reported a decline in ED visits in the intervention arm and no change in the control arm; 85 it didnot, however, provide patient-level means. As a result, we are unable to judge the variancewithin the sample. Another high risk-of-bias cohort study found lower odds of ED visits amongpatients accepting MTM when compared with patients refusing or disenrolling from MTM, butthe study did not account for confounders that may have influenced MTM participation andoutcomes. 46,47Table 64. Emergency department visits: Summary of resultsStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowSellors et al.,2003 55RCT/MediumTouchette et al.,2012 69RCT/MediumTaylor et al., 2003 85RCT/HighMoore et al.,2013 45Cohort/MediumG1: Usual care, plusclinical pharmacistcare.G1: 105G2: 103G2: Usual care in theGeneral Medicine ClinicG1: Pharmacist G1: 379consultation program G2: 409G2: Usual careG1: MTM basic(comprehensivemedication review andDRP assessment)G2: MTM enhanced(MTM plus 2-pageclinical summaryabstracted frompatient’s medical chart).G3: Usual careG1: Pharmaceuticalcare groupG2: Standard careG1: MTM program (optin)G2: control group(refusers)G1: 183G2: 190G3: 183G1: 33G2: 36G1: 2,250G2: 2,250Mean emergency roomvisits (time period NR)(25th-75th percentile)Mean number of ED orurgent care visits andambulance use at 5monthsMean number of EDvisits per participantbetween 3–6 monthsafter interventionChange in number ofED visits from 12months before baselinethrough 12 months afterChange in number ofED visits from 365 dayspreceding the patient'sprogram invitation dateto 365 days followingpatient's programinvitation dateResultsG1: 1.6 (0–2)G2: 2.3 (0–3)95% CI: NRp: NS at 0.05 level, specificsNRCalculated mean difference:-0.0395% CI: -0.113 to 0.053 p=0.48G1 vs. G3 calculated meandifference: -0.13895% CI: -0.258 to -0.018p=0.025G2 vs. G3 calculated meandifference: -0.118,95% CI: -0.242 to 0.006p=0.062G1: -12G2: 0p=0.044Calculated mean difference(95% CI): 0.04 (-0.043 to0.123, p=0.346101


Table 64. Emergency department visits: Summary of results (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedResultsPeriodBiasPerlroth et al.,2013 62aCohort/MediumCongestive heart failureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD,receiving MTM withCMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD,receiving MTM withCMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD, receiving MTM withCMRComparison—congestive heart failureG13: enrolled in PDP,usual careG14: enrolled in MA-PD, usual careComparison—Chronicobstructive pulmonarydiseaseG15: enrolled in PDP,usual careG16: enrolled in MA-PD, usual careG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Odds of any all-causeemergency room visitswithin 365 days afterdate of MTM enrollment(for interventions) orrandomly-assigned datein 2010 (forcomparators) (95% CI)Odds of conditionspecificemergencyroom visits within 365days after date of MTMenrollment (forinterventions) orrandomly-assigned datein 2010 (forcomparators) (95% CI)For PDPCongestive heart failureG1 vs. G13: 0.94 (0.90, 0.98),p0.05Chronic obstructive pulmonarydiseaseG5 vs. G15: 1.09 (1.04, 1.15),p0.05Welch et al., 2009 44Cohort/mediumComparison—DiabetesG17: enrolled in PDP,usual careG18: enrolled in MA-PD, usual careG1: MTM programprovided to homebasedbeneficiariesG2: No-MTM controlgroup (voluntary optout)G1: 459G2: 336Adjusted OR of ED visitfrom 6 month beforeMTM through 6 monthsafter MTM (adjusted forage, sex, chronicdisease score, specificbaseline utilization)Reported adjusted OR: 0.995% CI: 0.6 to 1.3, p NR102


Table 64. Emergency department visits: Summary of results (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedResultsPeriodBiasJeong, 46 ; Jeong,2012 47Cohort/HighG1: Kaiser-Permanente G1: 23,638MTM program G2: 14,232participants (2010) G3: 1,810G2: Kaiser-Permanentepatients eligible forMTM, but who declinedenrollment ordisenrolled with a PCPvisit during first half of2010G3: Kaiser-Permanentepatients eligible forMTM, but who declinedenrollment ordisenrolled without aPCP visit during firsthalf of 2010Percentage with EDvisits within 12 monthsof CMRCalculated OR G1 vs. G2:0.867 (0.832–0.904, p


offer consistency or precision of results, we graded it as insufficient to evaluate the effect ofMTM on ED costs (Table 67).Table 66. Costs of emergency department visits: Summary of resultsStudyOutcome andDesign/Risk Study ArmsN analyzedTime Periodof BiasHanlon et al.,G1: 1051996 84,94G2: 103RCT/LowSellors et al.,2003 55RCT/MediumChrischilleset al., 2004 79Cohort/HighPerlroth etal., 2013 62aCohort/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: PCM-eligible patients whoreceived PCM servicesG2: PCM-eligible patients who didnot receive PCM servicesCongestive heart failureG1: enrolled in PDP receiving MTMwith CMRG3: enrolled in MA-PD, receivingMTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receiving MTMwith CMRG7: enrolled in MA-PD, receivingMTM with CMRDiabetesG9: enrolled in PDP receiving MTMwith CMRG11: enrolled in MA-PD, receivingMTM with CMRComparison—congestive heartfailureG13: enrolled in PDP, usual careG14: enrolled in MA-PD, usual careComparison—Chronic obstructivepulmonary diseaseG15: enrolled in PDP, usual careG16: enrolled in MA-PD, usual careComparison—DiabetesG17: enrolled in PDP, usual careG18: enrolled in MA-PD, usual careG1: 379G2: 409G1: 524G2: 1,687G1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Annual health carecosts foremergency roomvisits (25th-75thpercentile) in USD(95% CI)Mean cost of ED orurgent care visitsand ambulance useat 5 months in $(assumed CAD)(SE)Charges for EDclaims at 12monthsChange in costs ofany all-causeemergency roomvisits within 365days after date ofMTM enrollment(for interventions)or randomlyassigned date in2010 (forcomparators) inUSD (95% CI)Change in costs ofcondition-specificemergency roomvisits within 365days after date ofMTM enrollment(for interventions)or randomlyassigned date in2010 (forcomparators) (95%CI)ResultsG1: 119 (0–146)G2: 171 (0–219)95% CI: NRp = NS at 0.05 level,specifics NRCalculated meandifference:-$5.60 (assumed CAD)95% CI:-$23.06 to$11.86p=0.53Results NRp=0.513For PDPCongestive heartfailureG1 vs. G13: -12.66 (-33.61, 8.30), p>0.05Chronic obstructivepulmonary diseaseG5 vs. G15: -16.21 (-35.37, 2.96), p>0.05DiabetesG9 vs. G17: -8.76 (-23.65, 6.12), p>0.05For PDPCongestive heartfailureG1 vs. G13: -3.17 (-14.59, 8.25), p>0.05Chronic obstructivepulmonary diseaseG5 vs. G15: 12.81USD (-.14, 25.76),p>0.05DiabetesG9 vs. G17: -3.27 (-15.37, 8.84), p>0.05a Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement forCMR. We have excluded groups 2, 4, 6, 8, 10, and 12 from our summary tables.Abbreviations: CAD = Canadian dollars, CI = confidence interval; CMR = comprehensive medication review; ED = Emergencydepartment; G = group; MA-PD = Medicare Advantage Part D Plan; MTM = medication therapy management; NR = notreported; PCM = pharmaceutical care management; PDP = Medicare Part D Plan; SE = standard error; USD = United Statesdollars.104


Table 67. Cost of emergency department visits: Strength of evidenceNumber ofFindings andStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 2; 1097(996)Cohort 1: 150,470–200,722(150,470–200,722)Strength ofEvidenceMedium Consistent Direct Imprecise Mean differenceranges from -52 USDto -5.6 CADInsufficientHigh Consistency Direct Imprecise Difference ranges Insufficientunknownsinglefrom -16 USD tostudy+12.8USDAbbreviations: CAD = Canadian dollars; RCT = randomized controlled trial; USD = United States dollars.HospitalizationsTwelve studies measured hospitalizations as an outcome following MTMinterventions. 44,47,48,51-55,62,64,65,69,73,74,78,80 ,84,91,94 Of these, we have excluded data from twostudies in the analysis below because they reported total number of events by each interventionarm rather than by patients within intervention arm. As a result, we are unable to assessvariance. 80,91 We report on the mean number of hospitalizations, the risk of hospitalization, andthe rates of hospitalization (Table 68).Table 68. Hospitalizations: Mean number, risk and ratesStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowMalone, 2000 51 ;Ellis, 2000 52(interventions);Malone, 2001 54(detailed QOLoutcomes);Ellis, 2000 53RCT/MediumSellors et al.,2003 55RCT/MediumG1: Usual care, plusclinical pharmacistcare.G2: Usual care in theGeneral MedicineClinicG1: PharmaceuticalcareG2: Usual careG1: Pharmacistconsultation programG2: Usual careG1: 105G2: 103G1: 523G2: 531G1: 379G2: 409Mean hospitaladmission (time periodNR) (25th–75thpercentile)Mean change in numberof hospitalizationsMean number of allcausehospitalizationsMean drug-(medications) relatedhospitalizationsResultsG1: 0.7 (0–1)G2: 0.8 (0–1)95% CI: NRp= NS at 0.05 level, specificsNRCalculated mean difference:0.06, 95% CI: -0.051 to 0.171,p=0.29Calculated mean difference:0.0395% CI: -0.085 to 0.025,p=0.289Calculated mean difference: 095% CI: -0.28 to 0.28p=1.0105


Table 68. Hospitalizations: Mean number, risk and rates (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasTouchette et al.,2012 69RCT/MediumG1: MTM basic(comprehensivemedication review andDRP assessment)G2: MTM enhanced(MTM plus 2-pageclinical summaryabstracted frompatient’s medical chart)G3: Usual careTime OneG1: 180G2: 190G3: 193Time TwoG1: 183G2: 190G3: 183Percentage ofparticipants with at leastone hospital visit at 3 to6 monthsMean number ofhospital visits perparticipantResultsG1 vs. G3: Calculated OR:2.069,95% CI: 1.104 to 3.878 p=0.02G2 vs. G3: Calculated OR:1.34595% CI: 0.693 to 2.609p=0.381G1 vs. G3: Calculated meandifference: 0.039,95% CI: -0.047 to 0.125,p=0.37Bernsten et al.,2001 64,65RCT/HighPai, 2009 73 ; Pai,2009 74RCT/HighMoore et al.,2013 45Cohort/MediumG1: Structuredcommunity pharmacybasedpharmaceuticalcare programG2: Usual communitypharmacy servicesG1: PharmaceuticalcareG2: Usual careG1: MTM program (optin)G2: control group(refusers)BaselineG1: 867G2: 7486 monthsG1: NRG2: NR12 monthsG1: NRG2: NR18 monthsG1: NRG2: NRG1: NRG2: NRG1: 2,250G2: 2,250Percentage with ≥1hospitalization in theprior 18 monthsMean number of allcausehospitalizationsover 2 yearsCumulative hospitalizedtime in days (SD)Change in number ofinpatient visits from 365days preceding thepatient's programinvitation date to 365days following patient'sprogram invitation dateG2 vs. G3: Calculated meandifference: 0.045, 95 % CI: -0.037 to 0.127, p=0.28Pooled sampleBaseline (during 18 monthsbefore study)Overall: NRG1: 41.7G2: 41.3p=NS18 monthsOverall: NRG1: 35.6G2: 40.4p=NS, cannot be calculatedwithout NG1: 1.8 (2.4)G2: 3.1 (3)95% CI: NR, cannot becalculated without Np: 0.02Cumulative hospital timeG1: 9.7 days (14.7)G2: 15.5 days (16.3)95% CI: NR, cannot becalculated without Np=0.06Calculated mean difference(95% CI): -0.21(-0.265 to -0.155, p


Table 68. Hospitalizations: Mean number, risk and rates (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasPerlroth et al.,2013 62aCohort/MediumCongestive heartfailureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD,receiving MTM withCMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD,receiving MTM withCMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD, receiving MTMwith CMRComparison—congestive heart failureG13: enrolled in PDP,usual careG14: enrolled in MA-PD, usual careComparison—Chronicobstructive pulmonarydiseaseG15: enrolled in PDP,usual careG16: enrolled in MA-PD, usual careG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Odds of any all-causehospitalization within365 days after date ofMTM enrollment (forinterventions) orrandomly assigned datein 2010 (forcomparators) (95% CI)Odds of anyCHF/COPD/diabetesrelatedhospitalizationwithin 365 days afterdate of MTM enrollment(for interventions) orrandomly assigned datein 2010 (forcomparators) (95% CI)ResultsCongestive heart failureG1 vs. G13: 0.90 (0.86, 0.94),p0.05Chronic obstructive pulmonarydiseaseG5 vs. G15: 0.90 (0.87, 0.94),p0.05DiabetesG9 vs. G17: 0.91 (0.87, 0.95),p


Table 68. Hospitalizations: Mean number, risk and rates (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasJeong, 46 ; Jeong,2012 47Cohort/HighRoughead, 2009 78Cohort/MediumWelch et al., 2009 44Cohortstudy/MediumG1: Kaiser-Permanente G1: 23,638MTM program G2: 14,232participants (2010) G3: 1,810G2: Kaiser-Permanentepatients eligible forMTM, but who declinedenrollment ordisenrolled with a PCPvisit during first half of2010G3: Kaiser-Permanentepatients eligible forMTM, but who declinedenrollment ordisenrolled without aPCP visit during firsthalf of 2010G1: Collaborativehome-basedmedication reviewG2: No medicationreview receivedG1: MTM programprovided to homebasedbeneficiariesG2: No-MTM controlgroup (voluntary optout)G1: 273G2: 5,444G1: 459G2: 336Percentage hospitalizedwithin 12 months ofCMRRate of hospitalizationfor heart failure at anytime during studyAdjusted OR ofhospitalization from 6month before MTMthrough 6 months after(adjusted for age, sex,chronic disease score,specific baselineutilization)ResultsCalculated OR G1 vs. G2:0.794 (0.760–0.830);calculated OR for G1 vs. G3:0.606 (0.550–0.668)Adjusted HR (95% CI): 0.55,:0.39 to 0.77p: NRNOTE: Model adjusted for age,sex, comorbidity, SES, season,region of residence, andnumbers of prescriptions,prescribers, pharmacies,changes in medications,hospitalizations, occupationaltherapy visits, and speechtherapy visitsReported adjusted OR: 1.495% CI: 1.1 to 2.0; p valuesNRNOTE: Model adjusted for age,sex, chronic disease score,specific baseline utilizationa Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement forCMR. We have excluded groups 2, 4, 6, 8, 10, and 12 from our summary tables.Abbreviations: CHF = congestive heart failure; CI = confidence interval; CMR = comprehensive medication review; DCG =diagnostic cost group (a measure of health care use and comorbidity); DRP = drug-related problems; ESRD = end-stage renaldisease; G = group; HR = hazard ratio; MA-PD = Medicare Advantage Part D Plan; MTM = medication therapy management, N= number; NR = not reported; NS = not significant; OR = odds ratio; PCP = primary care physician; PDP = Medicare Part DPlan; QOL = quality of life; RCT = randomized controlled trials; RR = relative risk; SES = socioeconomic status.Five trials (one low risk-of-bias study, 84,94 three medium risk-of-bias study, 51-55,69 and onehigh risk-of-bias study 74 ) and one medium risk-of-bias cohort study 45 reported on the change innumber of hospitalizations or mean number of hospitalizations following MTM interventions(Table 68). The low risk-of-bias study did not provide sufficient information on variance toallow pooling. 84,94 Using a random-effects model, we pooled results for three medium risk-ofbiastrials 51-55,69 for all-cause hospitalizations and obtained a mean difference of 0.037, 95% CI,-0.006 to 0.080; p=0.094; I 2 =0 (Appendix F-10). We obtained similarly small effect sizes and108


wide confidence intervals spanning the null when including the enhanced arm of the Touchette etal. study (0.038, 95% CI -0.004 to 0.081; p=0.076; I 2 =0) 69 or including the single high risk-ofbiastrial (0.033, 95% CI -0.046 to 0.112; p=0.412; I 2 =54.484). 74 One study also provided data tocalculate an effect size and confidence intervals for drug-related hospitalizations that alsooverlapped the null effect. 55 These results are consistent with the findings of the low risk-of-biasstudy, which reported no statistically significant differences in number of hospitalizations. Themedium risk-of-bias cohort study was not consistent with the trial evidence: it foundsignificantly lower inpatient visits among MTM acceptors compared with MTM refusers. 45Six studies (one medium-risk-of-bias RCT, 69 three medium risk-of-bias cohort studies, 44,48,62one high risk-of-bias RCT, 64,65 and one high risk-of-bias cohort study 46,47 ) reported on thepercentage hospitalized following MTM (Table 68) and odds or hazard ratios of hospitalization.Not all studies provided sufficient data to allow the generation of a summary estimate of effectwith confidence intervals, nor we did find sufficient numbers of studies of similar design topermit pooling. The results are inconsistent; two studies (one low risk-of-bias trial and onemedium risk-of-bias cohort) suggested higher hospitalizations with MTM rather than usualcare, 44,69 and three studies (one medium risk-of-bias cohort study, 48 one high risk-of-biastrial, 64,65 and one high risk-of-bias cohort study) 46,47 suggested lower hospitalizations for theMTM arm (but with confidence intervals overlapping the null in one instance). 64,65 The sixthstudy, a large medium risk-of-bias study of Medicare Part D in 2010, conducted separateanalyses for cohorts by plan type (standalone Prescription Drug Plan or Medicare AdvantagePrescription Drug Plan) and clinical condition (congestive heart failure, chronic obstructivepulmonary disease, and diabetes) for the odds of all-cause hospitalization and condition-specifichospitalization. 62 The diabetes cohort had lower risks of hospitalization (all-cause or conditionspecific)regardless of plan type. For the other cohorts in this study, the magnitude, direction, andprecision of the effect varied by specific analysis. The inconsistency in results may be aconsequence of the wide range of included populations and interventions.One cohort study (medium risk of bias) reported a decreased rate of hospitalization for heartfailure at any time during study. This study of home medications review was designedspecifically to delay the next hospitalization among patients with heart failure in Australia. 78Based on inconsistent results from trials with medium study limitations, we rated MTM ashaving insufficient on the mean number of hospitalizations. One cohort study offers low strengthof evidence of reduced number of inpatient visits (Table 69). The lack of consistency betweenthe trial and cohort results and the higher risk of bias from cohort studies of acceptors andrefusers suggest that the overall strength of evidence from all designs is insufficient for lack ofconsistency. We rated the evidence on the risk of hospitalization as insufficient based oninconsistent (or unknown consistency) and imprecise evidence (Table 70) for unspecified clinicalconditions, COPD, or CHF alone. We rated the evidence as low for benefit for diabetes cohort.We draw attention to the risk of selective analysis reporting in the evidence on diabetes. Onlyone study elected to provide condition-specific outcomes 62 ; further analysis on diabetes cohortsfrom existing studies or new studies with fewer limitations may well change the direction,magnitude, and precision of effect from available evidence. By contrast, we rated the evidenceon the rate of hospitalization as low based on a precise estimate from a large cohort study (Table71); we note that the findings from a single study of a very specific intervention (homemedicines review) of heart failure patients limits its applicability to patients with othermorbidities and settings. Together, the lack of consistency across these measures ofhospitalization likely reflects heterogeneity in numerous factors in this evidence base.109


Table 69. Mean number of hospitalizations: Strength of evidenceStudy DesignNumber ofStudies;Subjects(Analyzed)RCT 3; 2,580(2,208)Cohort 1, 4,500(4,500)StudyLimitationsConsistency Directness PrecisionFindingsand StrengthDirection of[Magnitude] Evidenceof EffectMedium Consistent Direct Imprecise Meandifference of0.037 (95%CI -0.006 to0.080)HighConsistencyunknownsinglestudyAbbreviations: CI = confidence interval; RCT = randomized controlled trial.Table 70. Risk of hospitalization: Strength of evidenceStudyDesignNumber ofStudies;Subjects(Analyzed)RCT 1; 637(556)CohortCohortCHF orCOPD orunspecified:3; 904-200,722(795 -200,722)Diabetes: 1;150,470(150,470)StudyLimitationsLowHighConsistencyConsistencyunknown,single studyCHF or COPDor unspecified:InconsistentDiabetes: Consistencyunknown,single studyDirect Precise Calculatedmeandifference(95% CI):-0.21(-0.265to -0.155,p


Table 71. Rate of hospitalization: Strength of evidenceNumber ofFindings andStudies; StudyStudy DesignSubjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectCohort 1; 5,717 High Consistency(5,717)unknown,single studyAbbreviations: CI = confidence interval; HR = hazard ratio.Direct Precise Adjusted HR(95% CI): 0.55(0.39 to 0.77)StrengthofEvidenceLow forbenefitHospitalization CostsTwo trials (medium risk of bias) 51-55 and two cohort studies (one medium risk of bias 62 andone high risk of bias 79 ) reported changes in costs of hospitalization following MTM interventions(Table 72). Although two studies were set in the United States, one evaluated outcomes fromVeteran Affairs Medical Centers 51-54 and the other evaluated claims from the Iowa Medicaidprogram. 79 The third study was set in Canada. 55 The period of evaluation of outcomes rangedfrom 5 months 55 to 12 months. 51-54,79 The other cohort study, designed to identify the impact of2010 Part D MTM programs, compared cohorts (standalone Prescription Drug Plan or MedicareAdvantage Prescription Drug Plan) receiving MTM with a comprehensive medication review withcohorts receiving usual care for congestive heart failure, chronic obstructive pulmonary disease,and diabetes, after limiting the sample to those newly eligible or enrolled for MTM andcontrolling for characteristics such as demographics, medical comorbidities, condition severity,and intensity of provider care. 62 Three were consistent in demonstrating no effect of MTMinterventions on the costs of hospitalization; the large Part D evaluation demonstratedinconsistent results by clinical condition. MTM appeared to consistently reduce costs of all-causeand condition-specific hospitalization costs for the diabetes cohort only. Based on lack ofconsistency in direction of effect and lack of precision, we graded the body of evidence as beinginsufficient to evaluate the effect of MTM interventions on the cost of hospitalization overall andlow for diabetes (Table 73).Table 72. Costs of hospitalization: Summary of resultsStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowSellors et al.,2003 55RCT/MediumMalone, 2000 51 ;Ellis, 2000 52Malone, 2001 54(Ellis, 2000 53RCT/MediumG1: Usual care, plusclinical pharmacist care.G2: Usual care in theGeneral Medicine ClinicG1: Pharmacistconsultation programG2: Usual careG1: Pharmaceutical careG2: Usual careG1: 105G2: 103G1: 379G2: 409G1: 523G2: 531Annual health carecosts for inpatients 1-year closeout oradjusted to a 1-yearfollowup and weightedby actual time forcensored patientsMean cost of alladmissions to hospital(assumed CAD) overwhat time periodMean change in annualhospitalization costsResultsMean USD (25th–75thpercentile)G1: 5751 (0–3780)G2: 3349 (0–4824)95% CI: NRp: NS at 0.05 level, specificsNRCalculated mean difference:$159.74 (assumed CAD)95% CI: -$281.99 to $601.47p=0.478Calculated mean difference:-$221.0095% CI: -$566.33 to $124.33p=-0.21111


Table 72. Costs of hospitalization: Summary of results (continued)StudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasPerlroth et al.,2013 62 aCohort/MediumChrischilles et al.,2004 79Cohort/HighCongestive heart failureG1: enrolled in PDPreceiving MTM withCMRG3: enrolled in MA-PD,receiving MTM withCMRChronic obstructivepulmonary diseaseG5: enrolled in PDPreceiving MTM withCMRG7: enrolled in MA-PD,receiving MTM withCMRDiabetesG9: enrolled in PDPreceiving MTM withCMRG11: enrolled in MA-PD,receiving MTM withCMRComparison—Congestive heart failureG13: enrolled in PDP,usual careG14: enrolled in MA-PD,usual careComparison—Chronicobstructive pulmonarydiseaseG15: enrolled in PDP,usual careG16: enrolled in MA-PD,usual careComparison—DiabetesG17: enrolled in PDP,usual careG18: enrolled in MA-PD,usual careG1: PCM-eligiblepatients who receivedPCM servicesG2: PCM-eligiblepatients who did notreceive PCM servicesG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912G1: 524G2: 1,687ResultsAll hospitalization costs: Risk adjusted costs in USD forGeneric substitution PDP (95% CI)ratio within 365 days G1 vs. G13: -526.19 (-.919.71,after date of MTM -132.66), p0.05randomly assigned date G9 vs. G17: -398.98 (-651.21, -in 2010 (for146.75), p0.05G5 vs. G15: 200.21 (-55.81,456.23), p>0.05G9 vs. G17: -363.45 (-562.00,-164.91), p


Table 73. Cost of hospitalization: Strength of evidenceNumber ofFindings andStudy Studies; StudyDesign Subjects Limitations Consistency Directness Precision Direction[Magnitude] of(Analyzed)EffectRCT 3; 2,151(2,050)Cohort CHF or COPD:1; 169,099-200,722(169,099 -200,722)Diabetes: 1;150,470(150,470)Medium Inconsistent Direct Imprecise Inconsistentdirection of effectbut consistent inlack of significanteffectHighConsistencyunknown(single study)DirectImprecise forCHF orCOPD,precise fordiabetesDifferences rangefrom -526 USD to200 USD for CHFand COPDDifferences rangefrom -363 USD to-399 USD fordiabetesStrength ofEvidenceInsufficientInsufficientfor CHF orCOPDLow forbenefit fordiabetesAbbreviations: CAD = Canadian dollars; CHF = congestive heart failure; CI = confidence interval; COPD = chronic obstructivepulmonary disease; RCT= randomized controlled trial; USD = United States dollars.Hospital Length of StayTwo trials (one low 84,94 and one high risk of bias 73,74 ) reported inconsistent results on theeffects of MTM interventions on length of hospital stay. Neither study reported statisticallysignificant results, but the low risk-of-bias study found longer stays in the intervention arm andthe high risk-of-bias study found shorter stays in the intervention arm (Table 74). Based on lackof precision of the results, we graded this outcome as having insufficient evidence to evaluate theeffect of MTM interventions on the length of hospital visits (Table 75).Table 74. Length of hospitalization: Summary of resultsStudyOutcome and TimeDesign/Risk of Study ArmsN AnalyzedPeriodBiasHanlon et al.,1996 84,94RCT/LowPai et al., 2009 73 ;Pai et al., 2009 74RCT/HighG1: Usual care, plusclinical pharmacist care.G2: Usual care in theGeneral Medicine ClinicG1: Pharmaceuticalcare, consisting of oneon-onecare, with indepthdrug therapyreviews conducted by aclinical pharmacistG2: Standard of care,consisting of brieftherapy reviewsconducted by a nurseG1: 105G2: 103BaselineG1: 61G2: 44Year 1:G1: 44G2: 36Year 2:G1: 24G2: 32Hospitalized days(time period NR)Cumulative hospitaltime (days) over 2yearsResultsAbbreviations: G = group; NR = not reported; NS = not significant; RCT= randomized controlled trial.Mean (25th–75th percentile)G1: 6.7 (0–5)G2: 4.9 (0–6)95% CI: NRp: NS at 0.05 level, specifics NRCumulative hospital timeG1: 9.7 days (14.7)G2: 15.5 days (16.3)p: 0.06Pharmaceutical care reducedlength of stay by 21% comparedwith the standard of care group.p=NS113


Table 75. Length of hospital stay: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency(Analyzed)RCT 1; 208 (208) Low Consistency Directunknown—singlestudyDirectness PrecisionAbbreviations: MTM = medication therapy management; RCT= randomized controlled trial.Findings andDirection[Magnitude] ofEffectImprecise MTM reducedlength of stay 1.8daysStrength ofEvidenceInsufficientKey Question 3: Outcomes of Medication TherapyManagement by Intervention FeaturesKey Points• Studies do not routinely report outcomes of MTM by intervention features.• We found evidence on five intervention features informed by a single study for eachfeature: access to patient data, 69 intensity of care coordination and followup aftercomprehensive medication review, 88 pharmacy intensity of adoption of the intervention, 79community pharmacy versus call-center pharmacy, 57 and private versus Medicaidcoverage of pharmaceutical care. 58 With the exception of the investigation reporting onaccess to patient data, these studies had a high risk of bias.• Evidence was insufficient on access to patient data, intensity of care coordination andfollowup after comprehensive medication review, pharmacy intensity of adoption of theintervention, community pharmacy versus call-center pharmacy, and private versusMedicaid coverage of pharmaceutical care for most outcomes.• MTM programs with pharmacist access to patient records reduces the number of adversedrug events (low strength of evidence) when compared basic MTM programs.• Community pharmacists increase the generic dispensing ratio more than call-centerbasedpharmacists (low strength of evidence).Detailed Synthesis: Intervention FeaturesAccess to Patient RecordsA single trial (medium risk of bias) of 556 patients overall (373 in the two MTM arms)evaluated differences between two MTM intervention arms; one without access to patientrecords (denoted “basic” MTM) and one specifically with such access in the form of a two-pageclinical synopsis containing basic data on a patient’s medical history, laboratory values, andcurrent medications, including over-the-counter and herbal medications (denoted “enhancedMTM”). 69 Table 76 provides the effect size and strength of evidence for the seven outcomesassessed in this trial. In all instances, we rated the trial as medium for study limitations andunknown for consistency; we do not repeat these ratings in the table. With the exception of meannumber of adverse drug events, which suggested benefit for enhanced MTM when comparedwith basic MTM (low strength of evidence), we found insufficient evidence to evaluate thecomparative effectiveness of the two arms.114


Table 76. Access to patient records (basic MTM versus enhanced MTM): Strength of evidenceOutcome Directness Precision Findings and Direction [Magnitude] of EffectStrength ofEvidencePercentage with ≥ 1 Direct Imprecise Calculated OR: 1.294, 95% CI: 0.768 to 2.180, InsufficientADEp=0.333Percentage with ≥ 1 Direct Imprecise Calculated OR: 1.222, 95% CI: 0.795 to 1.878, Insufficientemergency departmentvisitp=0.360Percentage with ≥ 1 Direct Imprecise Calculated OR: 1.539, 95% CI: 0.862 to 2.746, Insufficienthospitalizationp=0.145Mean number of ADEs Direct Imprecise Calculated mean difference: 0.346, 95% CI: 0.112 Low for benefitto 0.580, p=0.004Mean number ofemergency departmentvisitsMean number ofhospitalizationsMean number ofphysician visitsDirect Imprecise Calculated mean difference: -0.001 , 95% CI:-0.119 to 0.117, p=0.987Direct Imprecise Calculated mean difference: 0.055 , 95% CI:-0.038 to 0.148, p=0.244Indirect Imprecise Calculated mean difference: 0.100 , 95% CI:-0.322 to 0.522 , p=0.643Abbreviations: ADE = adverse drug event; CI = confidence interval; OR = odds ratio.InsufficientInsufficientInsufficientIntensity of Care Coordination and Followup FollowingComprehensive Medication ReviewOne RCT (high risk of bias) of 131 patients aged 65 or older compared comprehensive drugtherapy review and subsequent coordination and followup with the patient and physician withcomprehensive review and subsequent referral to the usual pharmacist only (Table 77). 88 For alloutcomes, we rated this study as high for study limitations and unknown for consistency (notrepeated in table). We found insufficient evidence to judge the effectiveness of MTM byintensity of adoption on all reported outcomes.Table 77. Intensity of care coordination and followup following comprehensive medication review:Strength of evidenceOutcomeFindings and Direction [Magnitude] Strength ofDirectness Precisionof EffectEvidenceHoarding drugs Indirect Imprecise Calculated OR: 1.08, 95% CI (0.52– Insufficient2.27); p=0.830Non-prescribed drugs discontinued Indirect Imprecise Calculated OR: 1.00, 95% CI (0.07– Insufficient13.77); p=1.000Taking home remedies Indirect Imprecise Calculated OR: 1.48, 95% CI (0.65– Insufficient3.34); p=0.348Mean symptoms reported Direct Imprecise Calculated mean: 0.70, 95% CI Insufficient(-0.73–2.13); p=0.34Mean medication adherence Indirect Imprecise Calculated mean: 1.60, 95% CI Insufficient(-14.40–-17.60); p=0.84Estimated annual prescription costs in Direct Imprecise Calculated mean: -65.00, 95% CI InsufficientUSD per client(-305.67–175.67); p=0.60Abbreviations: CI = confidence interval; OR = odds ratio; USD = United States dollars.Pharmacy Intensity of AdoptionOne cohort study (high risk of bias) of 2,211 patients evaluated eight outcomes based onpharmacy intensity of adoption of the MTM intervention (Table 78). 79 Specifically, the authorscategorized pharmacies that completed recommendations in at least one quarter into four groups:(1) for at least 50 percent of patients, high-intensity pharmacy (2) 25 to 49 percent as moderate115


intensity; (3) 1 to 24 percent as low intensity; and (4) no recommendations study as no intensity.For all outcomes, we rated this study as high for study limitations and unknown for consistency(not repeated in table). Outcomes for which we can infer a benefit or a harm from the effect arerated as direct outcomes. We found insufficient evidence to judge the effectiveness of MTM byintensity of adoption on all reported outcomes.Table 78. Pharmacy intensity of adoption: Strength of evidenceOutcomeFindings and Direction [Magnitude] Strength ofDirectness Precisionof EffectEvidenceNumber of emergency department Direct Imprecise Findings NR, p=0.330 InsufficientclaimsNumber of inpatient institutional claims Direct Imprecise Findings NR, p=0.839 InsufficientNumber of outpatient facility claims Indirect Imprecise Findings NR, p=0.112 InsufficientNumber of pharmacy, institutional, and Indirect Imprecise Findings NR, p=0.616 Insufficientmedical servicesEmergency department claims Direct Imprecise Findings NR, p=0.652 InsufficientInpatient institutional claims Direct Imprecise Findings NR, p=0.862 InsufficientOutpatient facility claims Indirect Imprecise Findings NR, p=0.212 InsufficientPharmacy, institutional, and medical Indirect Imprecise Findings NR, p=0.166 InsufficientservicesAbbreviation: NR = not reported.Community Pharmacy Versus Call CenterOne large cohort study (high risk of bias) of the MirixiaPro platform (95,736 patientsenrolled, 73,793 analyzed) compared patients using a community pharmacy, which included bothface-to-face and telephone interactions, with patients using a call center pharmacy (Table 79). 57The investigators measured three diverse outcomes. In all instances, we rated the study as highfor study limitations and unknown for consistency (not repeated in table).Outcomes for which wecan infer a benefit or a harm from the effect are rated as direct outcomes. We found insufficientevidence for drug cost and drug use outcomes, which we rated as indirect evidence with highstudy limitations. MTM delivered by community pharmacists increases the weighted genericdispensing ratio (GDR) when compared with MTM delivered by call-center pharmacists (lowstrength of evidence). The study defines the weighted GDR as the number of generic 30-dayequivalent claims divided by the total number of claims, and then weighted for each patient by afactor equal to the individual’s total prescription volume multiplied by a constant to hold samplesize unchanged.Table 79. Community pharmacy versus call center: Strength of evidenceOutcomeFindings and Direction [Magnitude] of Strength ofDirectness PrecisionEffectEvidenceDrug cost per patient per month Indirect Precise Calculated mean difference: -20.0, 95% CI: Insufficient(USD)-32.826 to -7.174, p=0.002Drug use per patient per month Indirect Precise Calculated mean difference: -0.370, 95% CI: Insufficient-0.477 to -0.263, p


outcomes. In all instances, we rated the study as high for study limitations and unknown forconsistency (not repeated for each outcome in the table). We found insufficient evidence to judgethe effectiveness of MTM by type of payer on all reported outcomes.Table 80. Type of payer: Strength of evidenceOutcomeFindings and Direction [Magnitude] Strength ofDirectness Precisionof EffectEvidencePer-patient MedicationDirect Precise Calculated mean difference: 0.81, InsufficientAppropriateness Index at followup95% CI: -1.303 to 2.923, p=0.452Proportion of patients for whom cost Direct Precise Calculated OR: 1.498, 95% CI: 0.807 Insufficientwas a problem at followupto 2.778, p=0.20Drug therapy problems identified Direct Precise 2.6 in both arms, p=1.0 InsufficientAbbreviations: CI = confidence interval; OR = odds ratio.Key Question 4. Outcomes of MTM by Patient CharacteristicsWe did not identify any studies that analyzed outcomes of MTM by patient characteristics.Key Question 5. Harms of Medication Therapy ManagementInterventionsKey Points• Studies do not routinely report harms that result from MTM interventions. One studyreported on inconvenience from information received through an MTM intervention.Study limitations and lack of precision of these results suggested that evidence wasinsufficient to evaluate the effect of MTM interventions on harms.Detailed Synthesis: InconvenienceA single cohort study (high risk of bias) compared pharmaceutical care with usual care(Table 81). 70,71 The investigators reported that patients in the intervention arm were less likely toagree or strongly agree with the statement that they were inconvenienced by monthlyappointments with the pharmacists than patients in the control arm (40 percent versus 69 percent;calculated OR, 0.278; 95% CI, 0.088 to 0.875; p=0.029). The sample size does not meet optimalinformation size criteria, suggesting lack of precision of the results.Table 81. Inconvenience: Strength of evidenceNumber ofStudy Studies; StudyDesign Subjects LimitationsConsistency(Analyzed)Cohort 1; 55 (51) High Consistencyunknown-singlestudyAbbreviations: CI = confidence interval; OR = odds ratio.Findings andDirectionDirectness Precision[Magnitude] ofEffectDirect Imprecise Calculated OR:0.278, 95% CI:0.088 to 0.875;p=0.029Strength ofEvidenceInsufficient117


DiscussionWe conducted a systematic review of benefits and harms of medication therapy management(MTM) programs. Because of the wide variation in types of interventions classified as MTM, wefirst catalogued intervention components and implementation features of MTM interventions(Key Question [KQ] 1). We then evaluated the effect of MTM on intermediate, patient-centered,and resource utilization outcomes (KQ 2). We also reviewed the evidence to identify how theseeffects might vary by specific intervention components and features (KQ 3) and patientcharacteristics (KQ 4). Finally, we reviewed the evidence on harms associated with MTM (KQ5).Below, we summarize the main findings and strength of evidence, where applicable. We thendiscuss the findings in relationship to what is already known, applicability of the findings,implications for decisionmaking, limitations, research gaps, and conclusions.This evidence base consisted of 44 studies (21 randomized controlled trials [RCTs], 4controlled clinical trials, and 19 cohort studies) reported in 61 articles. Most RCTs compared anMTM intervention with usual care rather than with a different active intervention; mostobservational studies were cohort studies. Numerous studies had methods problems that led us torate them as having a medium or high risk of bias; only a few studies were of low risk of bias.When possible (enough studies similar in intervention, populations, and outcomes measured), weconducted meta-analyses of data from RCTs; in some cases, wwe did two sets, one with and onewithout the high risk-of-bias trials.Key Findings and Strength of EvidenceKQ 1: Intervention Components and Implementation FeaturesOf the 44 included studies, over three-quarters were broadly focused MTM interventionswith patients that had a wide-ranging collection of conditions; the remaining studies werenarrowly focused MTM interventions with patients that had a specific condition. All studies useda pharmacist as the interventionist. Services were provided face-to-face in half of includedstudies. Included studies provided interventions in a variety of clinical settings, includingcommunity pharmacies, centralized pharmacies or pharmacy call centers, and outpatient medicalclinics, and some used home visits; half of the narrowly focused interventions were deliveredexclusively in an outpatient medical clinic.Whether termed “pharmaceutical care” or “MTM,” studies did not describe interventioncomponents and features in a consistent manner or in sufficient detail. These drawbacks wereespecially prevalent for intervention intensity and frequency of followup, method of patientenrollment for services, level of integration with usual care, and reimbursement characteristicsfor rendered MTM services. KQ 1 was descriptive in nature, so we did not grade strength ofevidence.KQ 2: Overall EffectivenessOf the 44 studies included in this review, we rated 16 as high risk of bias overall; that is,concerns about randomization failure, confounding, or overall attrition increased the risk of biasfor all outcomes. In addition, we rated some studies that were otherwise of low or medium riskof bias as high for individual outcomes, chiefly because of measurement or detection bias relatedto the specific outcome. These instances are specified in the relevant section the Results chapter118


We rated the strength of evidence for each outcome from low- or medium risk-of-bias studieswhen available. MTM significantly improved some measures of medication adherence,medication appropriateness assessed in general and medication dosing (Table 82). However, wedid not find evidence of benefit for any other intermediate outcomes on which we had data. Nostudies addressed either goals of therapy or patient engagement.Table 82. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventionsStudy Design:Intermediate No. Studies Strength ofSupporting Judgment Findings and Direction of EffectOutcome (N Patients EvidenceAnalyzed)Anticoagulation RCT: 1 (10) Insufficient Medium studylimitations, consistencyunknown-single study,direct, impreciseTherapeutic INR achieved, 100%vs. 16.7%; p = 0.048.HbA1c RCT: 2 (102) Insufficient Medium study One trial with significantly greaterlimitations, inconsistent, percentage of patients with HbA1cdirect, imprecise


Table 82. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventions (continued)Study Design:Intermediate No. Studies (N Strength ofSupporting Judgment Findings and Direction of EffectOutcome Patients EvidenceAnalyzed)Medicationadherencemeasured asproportion adherentto a thresholdMedicationadherencemeasured aspercentage ofprescribed dosestakenMedicationadherence usingself-reportmeasuresMedicationadherence,miscellaneousmeasuresMedicationAppropriatenessGeneral IndexScoresMedication-specificappropriatenessRCT: 1 (69) Insufficient Medium studylimitations, consistencyunknown, direct,preciseCohort: 2 (224to 200,722) )Cohort: 2 (120 -4,500)Low for benefitLow for benefitfor adherence totreatment forhypertesion anddyslipidemiaHigh study limitations,inconsistent, direct,preciseHigh study limitations,inconsistent, direct,impreciseInsufficient fortreatment ofpatients withdiabetes,depression andasthmaRCT: 1 (292) Insufficient Medium studylimitations, consistencyunknown, direct,impreciseRCT: 2 (365) Insufficient Medium studylimitations,inconsistent,direct,impreciseRCT: 1 (208) Low for benefit Low study limitations,consistency unknown,direct, precise100% of intervention patients and88.9% of controls were adherent;p = 0.115.Two studies with findings inopposite direction; larger studyshowing range of ORs formedication-specific adherencedepending on medication.For comparison of PDP vs. controls,ORs ranged from 0.99 to 1.43; 95%CIs ranged from (0.90, 1.08) to(1.26, 1.62).For comparison of MA-PD vs.controls ORs ranged from 1.10 to1.40; 95% CIs ranged from (0.83,1.24) to (1.29, 1.52).For clinic-based MTM vs. usual carefor adherence to aspirin, odds ofadherence ranged from 5.981 (95%CI, 0.284 to 126.030; p = 0.250)during the intervention to 1.17 1year after the intervention (95% CI,0.072 to 18.903; p = 0.912).Calculated mean difference fromsmall study, -0.040; 95% CI, -0.101to 0.021; p = 0.201.Larger study found a small(difference in adherence ~4.6%) butstatistically significant effect of MTMon adherence to medications forsome (2 of 5) conditions but nosignificant effect for the otherconditions.Calculated mean difference, 0.090;95% CI, -0.076 to 0.256; p = 0.289.Two studies with opposite directionof effect, both with non-significantdifferences between groupsImprovement in MTM group fromscore of 17.7 to 13.4 at 3 monthsand 12.8 at 12 months.RCT: 2 (261) Insufficient Medium study Significant improvement inlimitations, inconsistent, appropriateness in the MTM groupdirect, imprecise for some medications but not others120


Table 82. Summary of findings and strength of evidence for intermediate outcomes of MTMinterventions (continued)Study Design:Intermediate No. Studies (N Strength ofSupporting Judgment Findings and Direction of EffectOutcome Patients EvidenceAnalyzed)Medication dosing RCT: 1 (56) Low for benefit Medium studylimitations, consistencyunknown, indirect,preciseMean difference, -2.2 doses;calculated 95% CI, -3.738 to-0.662Goals of therapy 0 NA NA NAPatient0 NA NA NAengagementAbbreviations: BP = blood pressure; CI = confidence interval, HbA1c = Hemoglobin A1c; INR = International NormalizedRatio; LDL = low density lipoprotein; MA-PD = Medicare Advantage Part D Plan; MTM = medication therapy management;NA = not applicable; OR = odds ratio; PDP = Medicare Part D Plan; RCT = randomized controlled trial.Similarly, we did not have evidence of benefit for most patient-centered outcomes, includingadverse drug events or mortality (Table 83). MTM did not improve most measures of healthrelatedquality of life (low strength of evidence for no benefit). We graded the “vitality” and“emotional role functioning” domains of the Medical Outcomes Study Short-Form (SF36)questionnaire as insufficient for this domain. For the SF-36, neither the other six domains nor thetwo component scores (physical health, mental health) showed significant benefit from MTMinterventions. The various patient satisfaction items also showed no impact from MTM programs(low strength of evidence for no benefit). We found no evidence on activities of daily living,work or school absenteeism, and patient and caregiver participation in medical care anddecisionmaking.Table 83. Summary of findings and strength of evidence for patient-centered outcomes of MTMinterventionsStudy Design:Patient-CenteredStrength ofFindings and DirectionNo. Studies (NSupporting JudgmentOutcomeEvidenceof EffectPatients Analyzed)Adverse drug events RCT: 2 (806) Insufficient Medium study limitations,inconsistent, direct, impreciseCognitive and physicalfunctionRCT: 1 (133) Insufficient ImpreciseMedium study limitations,consistency unknown, direct,impreciseAffective function RCT: 2 (181) Insufficient Medium study limitations,inconsistent, direct, impreciseDirection and magnitudeof effect differs betweenthe 2 trialsNo significant differencesbetween armsOne study with nosignificant calculatedmean differences indepression or anxietyscores; the other studywith significantdifferences in meandepression and anxietyscores, but no significantdifference in percentageachieving a depressionremission.121


Table 83. Summary of findings and strength of evidence for patient-centered outcomes of MTMinterventions (continued)Study Design:Patient-CenteredOutcomeNo. Studies (NPatientsAnalyzed)Strength ofEvidenceSupporting JudgmentMortality RCT: 1 (181) Insufficient Medium study limitationsconsistency unknown, direct,impreciseCohort: 2 (173,329) Insufficient High study limitations,inconsistent (magnitude),direct, impreciseGastrointestinalbleeding eventsGeneral health-relatedquality of life domainsother than vitality andemotional rolefunctioningGeneral health-relatedquality of life for vitalityand emotional rolefunctioning domainCondition-specifichealth-related qualityof life (diabetes)Cohort: 1 (unclear) InsufficientRCT: 3 (1,169)Low for nobenefitHigh study limitations,consistency unknown, direct,impreciseMedium study limitations,consistent for physical rolefunctioning, general healthperceptions, and socialfunctioning domains,inconsistent for physicalfunctioning, bodily pain, andmental health domains, direct,preciseRCT: 3 (1,169) Insufficient Medium study limitations,consistent, direct, imprecise(not corrected for multiplecomparisons or wide CIs)RCT: 1 (73) Insufficient Medium study limitations,consistency unknown, direct,imprecisePatient satisfaction RCT: 3. (1,463) Low for nobenefitNo differenceMedium study limitations,consistent, direct, preciseFindings and Directionof EffectOR, 0.59; 95% CI, 0.12to 2.49; p = 0.48.One study: OR, 0.5;95% CI, 0.3 to 0.9.Second study: adjustedHR, 0.92;95% CI, 0.87 to 0.96;p < 0.001.RRR, 60%; p = 0.001.Variable mean differencewith CIs consistentlyspanning the null effect.Vitality: Mean differenceof 2.797; 95% CI, 0.655to 4.939; p = 0.010.Emotional rolefunctioning: Meandifference of 5.386, 95%CI, -7.244 to 18.013)Nonsignificantimprovement of 0.1 pointon a 5-point scale in theintervention groupcompared with nochange in the controlgroup.No differences on 17 of21 items of patientsatisfaction; 4 statisticallysignificant differencesranged in magnitude from-0.15 to -0.36, favoringMTM.Activities of daily living 0 NA NA NAWork or school 0 NA NA NAabsenteeismPatient and caregiver 0 NA NA NAparticipation in medicalcare anddecisionmakingAbbreviations: CI = confidence interval, HR = hazard ratio; MTM = medication therapy management; NA = not applicable; OR= odds ratio; RCT = randomized controlled trial.122


Outcomes related to using health resources were similarly not much influenced by MTMinterventions (Table 84). Two exceptions may merit attention: (1) health plan expenditures onmedication costs and (2) the proportion and costs of hospitalization for patient with diabetes. Inboth instances, MTM interventions improved outcomes. MTM trials implemented in settingswith a broad range of patients did not show a consistent signal of reduction in the number ofhospitalizations but a single cohort study that partially addressed confounding inherent in studiesof refusers and acceptors found a lower mean number of inpatient visits for patients acceptingMTM when compared with patients refusing MTM. Overall, we judge the strength of evidencefor this outcome to be insufficient owing to lack of consistency.Table 84. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventionsStudy Design:Resource-UtilizationStrength ofStudy Findings andNo. Studies (NSupporting JudgmentOutcomesEvidenceDirection of EffectPatients Analyzed)Use of generics Cohort: 1 (63,198to Insufficient High study limitations, consistency Odds range from200,722)unknown, direct, imprecise -0.01 to 0.006.Medication costs:patient copaymentsRCT: 1 (NR) Insufficient Medium study limitations,consistency unknown, indirect,precision cannot be determinedCohort: 1 (1,606) Insufficient High study limitations, consistencyunknown, indirect, preciseCalculated meandifference, -64 USD;variance notcalculable.Calculated meandifference for MTMvs. same-countrycontrol, 80.40 USD;95% CI, 10.43 to150.37; p = 0.024.Medication costs:health planexpendituresMedication costs:total outlaysMedication costs:medication costs plusother expendituresRCT: 3; (965)NRCT and cohort:5; (120 to 200,722)Low forbenefitInsufficientMedium study limitations,consistent, indirect, impreciseHigh study limitations,inconsistent, indirect, impreciseRCT 6 (2,636) Insufficient Medium study limitations,inconsistent, indirect, impreciseCohort: 2 (177,565) InsufficientHigh study limitations,inconsistent, indirect, impreciseRCT: 2 (996) Insufficient Medium study limitations,inconsistent, indirect, impreciseNRCT and cohort: 3(5,300)InsufficientHigh study limitations,inconsistent, indirect, impreciseCalculated meandifference for MTMvs. different countrycontrol, 88.60 USD;95% CI, 24.61 to152.59; p = 0.007.Mean differencevaries from -34 CADto -293 USD over 6months.Mean differencevaries from -800 USDover 1 year to 425USD over 2 years.Mean differencevaries from -20.16USD to +5.25 USDper month.Mean differencevaries from -563 USDto +310 USDannually.Differences in meancosts range from -8.1CAD to 1,947 USD.Differences in meancosts range from-1,039 to 1,100 USD.123


Table 84. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventions (continued)Study Design:Resource-UtilizationStrength ofStudy Findings andNo. Studies (NSupporting JudgmentOutcomeEvidenceDirection of EffectPatients Analyzed)Number of outpatientvisitsRCT: 3; (2,208) Insufficient Medium study limitations,inconsistent, indirect, preciseCohort: 1 (4,500) Insufficient High study limitations, consistencyunknown, indirect, impreciseOutpatient costs RCT: 3 (2,050) Insufficient Medium study limitations,inconsistent, indirect, impreciseNumber of laboratorytestsCosts of laboratorytestsNumber ofemergencydepartment visitsCosts of emergencydepartment visitsHospitalization:numberRCT: 2 (1,842) Insufficient Medium study limitations,inconsistent, indirect, impreciseRCT: 3 (2,050) Insufficient Medium study limitations,inconsistent, indirect, impreciseRCT: 3 (1,552) Insufficient Medium study limitations,inconsistent, direct, impreciseObservational: 3(795 to 200,722)InsufficientHigh study limitations,inconsistent, direct, impreciseRCT: 2 (996) Insufficient Medium study limitations,consistency unknown, direct,impreciseCohort: 1 (150,470to 200,722)RCT: 3 (2, 208)Cohort: 1 (4,500)InsufficientLow for nobenefitLow forbenefitHigh study limitations, consistencyunknown, direct, impreciseMedium study limitations,consistent, direct, preciseHigh study limitations, consistencyunknown, direct, preciseStandardized meandifference, 0.049;95% CI, -0.034 to0.133; p = 0.247; I 2 =0.Calculated meandifference, 2.48; 95%CI, 1.674 to 3.286; p


Table 84. Summary of findings and strength of evidence for resource-utilization outcomes of MTMinterventions (continued)Study Design:Use of ResourcesStrength ofStudy Findings andNo. Studies (N ofSupporting JudgmentOutcomesEvidenceDirection of EffectPatients Analyzed)Hospitalization: risk RCT: 1 (556) Insufficient Low study limitations, consistencyunknown, direct, impreciseOR for basic MTM vs.usual care, 2.069;95% CI, 1.104 to3.878; p = 0.02.Cohort—CHF, COPD, orunspecified: 3 (795to 200,722)CHF, COPD,orunspecified:insufficientHigh study limitations,inconsistent, direct, impreciseOR for enhancedMTM vs. usual care,1.345; 95% CI, -0.693 to 2.609; p =0.381.Adjusted OR rangesfrom 0.90 to 1.4.Hospitalization: rate(patients with heartfailure and homemedicine review)Costs ofhospitalizationDiabetes: 1(150,470)Cohort: 1 (5,717)Diabetes: lowfor benefitLow forbenefitHigh study limitations, consistency OR ranges from 0.91unknown, direct, precise to 0.93.High study limitations, consistency Adjusted HR, 0.55;unknown, direct, precise 95% CI, 0.39 to 0.77.3; 2,151 (2,050) Insufficient Medium study limitations,inconsistent, direct, impreciseCHF or COPD: 1(169,099 to200,722)Diabetes: 1(150,470)Insufficient forCHF or COPDLow forbenefit fordiabetesHigh study limitations, consistencyunknown, direct, impreciseHigh study limitations, consistencyunknown, direct, preciseInconsistent directionof effect butconsistent in lack ofsignificant effectDifferences rangefrom -526 USD to 200USD for CHF andCOPDDifferences rangefrom -363 USD to-399 USD for diabetesLength of hospital RCT: 1 (208) Insufficient Low study limitations, consistency MTM reduced lengthstayunknown, direct, imprecise of stay 1.8 daysAbbreviations: CAD = Canadian dollar; CHF = congestive heart failure; CI = confidence interval; COPD = chronic obstructivepulmonary disease; HR = hazard ratio; MTM = medication therapy management; NR = not reported; NRCT = nonrandomizedcontrolled trial; OR = odds ratio; RCT = randomized controlled trial; USD = U.S. dollar.Over all three categories of outcomes, each of which had a substantial number of individualmeasures, MTM improved outcomes in only a few instances. Study limitations, lack ofconsistency, and lack of precision of the estimates of effects limited the strength of evidenceconsiderably. As discussed later, even the minimal findings of effectiveness are at best onlynarrowly applicable.KQ 3: Effectiveness of MTM by Intervention FeaturesWe found evidence from one study each on five intervention features: (1) access ofpharmacists to patient records, 69 (2) intensity of care coordination and followup aftercomprehensive medication review, 88 (3) community pharmacy versus call center, 79 (4) level ofintensity of intervention, 57 and (5) type of payer (private vs. Medicaid). 58 With the exception ofthe study on pharmacists’ access to patient records, we rated these studies as high risk of bias.125


Evidence was insufficient for most outcomes for the first two intervention features, with twoexceptions. First, MTM delivered by community pharmacists increased the weighted genericdispensing ratio when compared with call-center pharmacists (low strength of evidence). Second,enhanced MTM with pharmacists’ access to patient records reduced the mean number of adversedrug events; this finding suggested benefit when compared with basic MTM (low strength ofevidence). We found insufficient evidence for all outcomes for intensity of intervention and typeof payer.KQ 4: Effectiveness of MTM by Patient CharacteristicsWe did not identify any studies that analyzed outcomes of MTM by patient characteristics.KQ 5: Harms of MTM InterventionsLack of precision and the limitations of a single high risk-of-bias study meant that evidencewas insufficient to judge whether MTM resulted in greater or inconvenience 70,71 than usual care.We found no evidence on other prespecified harms, specifically including care fragmentation,patient decisional conflict, patient anxiety, increased (actual) adverse drug events, prescriberconfusion, and prescriber dissatisfaction.Findings in Relation to What Is Already KnownOur findings contrast with conclusions that Chisholm-Burns and colleagues reached in arecent systematic review. 95 In that review, the authors concluded that “Pharmacist-provideddirect patient care has favorable effects across various patient outcomes, health care settings, anddisease states.” 95, p. 923 Several differences between the Chisholm-Burns review and the currentreview may account for the discrepant conclusions. First, the Chisholm-Burns review includedall studies that cited evidence of pharmacist involvement in direct patient care. The interventionsexamined included chronic disease management and prospective and retrospective drugutilization review; we excluded these types of efforts because of the clinical heterogeneity thoseinterventions would have introduced into the review. Notably, the Chisholm-Burns review didnot use the term “medication therapy management” to categorize the interventions in the articlesthey reviewed. Second, approximately 30 percent of the papers in the Chisholm-Burns reviewwere conducted entirely in institutional settings. In contrast, we did not identify any studieswithin institutional settings that met our MTM intervention definition criteria. Third, theChisholm-Burns review included a total of 298 articles and did not omit from their analysesstudies with a high risk of bias; by contrast, we based our strength-of-evidence grades in thisreview on only those studies with no more than medium risk of bias. Thus, a direct comparisonof findings between these two reviews would be ill advised.The striking differences between the conclusions reached in these two reviews emphasizetwo important needs for efforts to systematically review MTM programs. The first is forresearchers to specify the MTM intervention based on existing definitions, taxonomies, orservice models. The second is to develop consensus guidelines for describing interventionfeatures and fidelity of intervention delivery in publications reporting findings from evaluationstudies. Progress on these two steps would enable systematic reviews to differentiate betterbetween different types of services and avoid the problem of overgeneralizing review results.126


Applicability of the FindingsThis body of evidence has significant clinical and methodological heterogeneity, which limitsthe ability to make any universal statements about effectiveness. 36 However, the range of studydesigns, which includes RCTs, nonrandomized trials, and cohort studies, enhances theapplicability of findings for real-world settings. Included studies ranged from relatively smallinterventions in single clinics provided by a single interventionist to evaluations of MTMservices delivered on a large scale through integrated health systems or health plans as aMedicare Part D or other drug plan benefit. This diversity of studies enhanced the applicabilityof findings to a wide variety of settings, including outpatient clinics, community pharmacies, andcentralized pharmacy call centers. A few studies conducted outside the United States includedMTM as part of a home visits program; findings from this model may not be directly applicablewithin the United States.The studies in this review are broadly applicable to a range of chronically ill, adult patientpopulations. The majority of interventions were directed at populations with multiple andcommon chronic conditions, such as diabetes, chronic heart failure, and hypertension. Severalspecifically targeted adults aged 65 years or older. Few studies reported sociodemographiccharacteristics beyond age and sex; thus, the applicability of findings to specific populations(e.g., rural, low socioeconomic status, cognitively impaired, uninsured) is unknown. The natureof the MTM intervention, which includes involving patients as active participants in the process,limits the extent to which findings can be generalized beyond patients who agreed to participatein such interventions. Patients who agree to participate may be systematically different fromthose who decline to be in such a program. For that reason, the impact of such interventions at apopulation or health-plan level may be limited by the degree of uptake among interested patients.The intervention used across most studies can be characterized as complex and moderatelyresource intensive. Components involve identifying applicable patients; initially assessingpatients; providing counseling, education, and care coordination; and following patients overtime. These services were provided per protocol in some studies and as needed or ad hoc inothers. Most studies described intervention components in terms of “pharmaceutical care model”components or Medicare Part D MTM program criteria, but few provided detailed descriptionsor measurement of implementation fidelity.All studies included comparator arms with usual medical care or pharmacy care. Usual caredoes not typically include distinct MTM services by health care providers other than prescribingproviders (not common for the time period covered by most of the studies). Models ofcollaborative health care delivery are evolving, and the changing roles and training ofpharmacists increase the potential applicability of MTM interventions in future models of healthcare.The broad sets of outcomes evaluated across this body of evidence spanned a substantialrange of both intermediate and health outcomes as well as outcomes related to resource use.Proximal and intermediate outcomes included number of drugs, identification of drug therapyproblems, appropriateness of medication prescribing, and laboratory or biometric markers ofdisease control (e.g., hypertension, hemoglobin A1c, low-density lipoprotein cholesterol).Patient-centered outcomes focused on numerous measures of quality of life as well as adversedrug events. Many studies also reported outcomes involving health care resource use andexpenditures (e.g., number and costs of hospitalizations, emergency department visits, outpatientvisits).127


Most studies did not, however, clearly indicate the expected, desired, or intended direction ofeffect on most resource use outcomes, making the applicability of using these interventions toreduce drug-related or overall health care costs or expenditures difficult to assess. For example,whether one should expect the number of medications prescribed or drug costs for heart failureto increase or decrease under the careful scrutiny of an MTM intervention is not clear becausethe desired impact will be based on the goal of therapy for each individual.The focus of outcome measurement in many studies was the short-term identification andcharacterization of drug therapy problems and their resolution; these endpoints are thought to bethe outcomes most sensitive to change as a result of receiving MTM services. However, bydesign, because identification of drug therapy problems is a part of the MTM intervention itself,differences between the nature of the intervention and that of the control group mean thatmeasuring these outcomes cannot be as rigorous in a usual care comparison group as it is in theintervention group. In fact, many studies were able to measure only changes in this outcome inthe intervention group. Hence, many studies failed to demonstrate a direct analytic link betweenthe resolution of drug therapy problems as a result of MTM and impact on intermediateoutcomes, patient-centered outcomes, and resource utilization. Thus, the applicability of studiesthat demonstrate an impact on the resolution of drug therapy problems is limited.Implications for Clinical Practice and PolicymakersAlthough we found the evidence insufficient in general to draw definitive conclusions aboutthe comparative effectiveness of MTM for most outcomes that we evaluated, our findings dosuggest some implications for practice and policy. MTM is already in widespread practice and isnow shaped in the United States largely by Medicare Part D policy: this presents both challengesand opportunities. MTM programs sponsored and administered by Part D drug benefit plans areoften centrally administered and delivered primarily by phone and may be less integrated intoroutine health care than some of the interventions included in our review. MTM programs of thefuture have the potential to be more integrated into routine health care through participation inaccountable care organizations or patient-centered medical home models. We were unable toanswer definitively whether level of integration matters for effectiveness, but policymakers mayneed to consider expectations about the impact that MTM might have on patient-centeredoutcomes and resource use in the context of other health care delivery transformation activitiesor quality improvement initiatives that are also occurring. More integration of MTM serviceswith other activities may be effective; however, the more integrated MTM becomes withinroutine medical care, the more difficult it becomes to isolate it as a discrete intervention forevaluation.Policymakers could thus consider whether MTM services should be positioned as acontributor to overall improvement in processes of care, health status, and costs or positioned asan intervention to which effects can be discretely attributed. As noted earlier, improvements inmedication appropriateness or drug therapy regimens may not always translate intoimprovements in health or costs, and even if they do, secular trends in related qualityimprovement (e.g., medication adherence interventions, regulatory requirements for medicationreconciliation, electronic health record meaningful use incentives) may make measuringoutcomes attributable to MTM very challenging.Future training of MTM providers would benefit from a better understanding of which MTMcomponents really matter. At the moment, such information is lacking. Policymakers and funderswho wish to understand the comparative effectiveness of different MTM components could128


encourage rigorous program evaluation designs that fit within the context of the real-worldimplementation of these programs. For example, positive deviance analyses 96 with rigorousmeasurement of implementation features or stepped wedge trial designs 97 may be usefulapproaches.A typical approach for evaluating complex interventions is to identify the “core” componentsfor standardization, while allowing for flexibility for peripheral components or variations inimplementation. In complex practice-based innovations, such flexibility may reflect desirable (orunavoidable) adaptations to local circumstances. Policy governing MTM programs may warrantmodifications to permit investigators to conduct rigorous and innovative evaluative designs toidentify core components or effectiveness-enhancing modifications. As future research andevaluation elucidates these components or enhancements, policy will need to evolve to keep pacewith best practices.Finally, considering both patients’ and prescribers’ perspectives in future design and deliveryof MTM services may be needed. In our current analytic framework, MTM interventions requirea significant element of engagement by both patients and prescribers if the interventions are tohave a reasonable likelihood of improving outcomes. Although “opt in” strategies may increasethe reach of such interventions, keeping patients (and their prescribing providers) engaged in theintervention over a reasonable amount of time may be the key to translating the potential ofMTM interventions into actual improvements. Further refinement of eligibility criteria based onevidence to provide interventions to those most at risk from drug-related problems and thereforemost likely to benefit may also be warranted.Limitations of the Comparative Effectiveness ReviewProcessThe constraints for populations, interventions, and settings that we imposed on thissystematic review may limit its applicability as discussed above. During topic refinement andbased on technical expert panel inputs and public comment, we expanded the scope by removingan exclusion criterion that would have required MTM interventions to have been directed at apatient population with two or more chronic conditions. As a result, we did include studies thatfocused on one chronic condition. Because of the prevalence of certain chronic conditions in theadult population, and particularly among Medicare beneficiaries, we think this decision wassensible and permitted us to examine a broader evidence base than would otherwise have beenthe case.Although we tried to distinguish MTM from disease or case management interventions,making this distinction was challenging. We created a threshold for what interventioncomponents were required to be present for this distinction. Specifically, we elected toemphasize whether the intervention entailed a comprehensive review of all medications; for thatreason, we did not constrain studies of interest to those that targeted a single medication or drugregimen or that focused on a single condition such as diabetes or hypertension.As described in the Methods chapter, when we were unable to determine which medicationsthe interventionist had reviewed, we wrote to the authors for additional information. We chose topursue authors in an effort to permit us to use studies that had been designed as MTM but did notdescribe the comprehensive medication review component in detail.Our approach may have been overly inclusive because it led us to include studies thataddressed a single disease, as long as the pharmacist reviewed all medications. For example, 10of the 44 studies were relatively narrowly focused; two of these addressed patients with chronic129


heart failure and two addressed patients with either hypertension or hypertension and diabetes.The remaining six studies focused on, patients with diabetes, HIV, glucocorticoid-inducedosteoporosis, or hemodialysis. The fact that did not require patients to have more than oneclinical condition resulted in an approach that was inclusive of these more narrowly focused(albeit often termed “MTM”) studies and may render our results less applicable to MTMinterventions targeted to patients with a wide range of chronic conditions.Also based on feedback during the process of setting out the scope of this review, we choseto include interventions that were broader than the Medicare Part D MTM-defined interventions.Put another way, we broadened our view of patient populations and intervention criteria, and weallowed studies not conducted in the United States into the evidence base. This decision led us toinclude interventions described as “pharmaceutical care,” which were generally based on thepharmaceutical care model principles; 10 it also permitted us to examine investigations withelements of pharmaceutical care or MTM that did not specifically label the intervention as eitherMTM or pharmaceutical care. These studies were often described as “clinical pharmacistinterventions.”Furthermore, all the non-U.S. studies involved interventions within single-payer healthsystems. Hence, the interventions in this review constitute a more heterogeneous group than ifwe had allowed only those labeled as Medicare Part D MTM programs. This is both a limitationand a strength. Although our approach makes results more challenging to interpret, it enhancesour ability not to miss interventions that include MTM components but lack the descriptor termMTM.Studies did not often explicitly describe certain MTM components. In cases when we couldnot determine whether investigators had provided certain MTM components (such as patienteducation and counseling, or coordination with other health care providers), we contacted theauthors to gain additional information that would allow us to make an informed decision. Wewere fairly permissive in interpreting the presence of the MTM intervention components otherthan comprehensive medication review. The main reason is that we recognized that termsdescribing some components have evolved over time and may have been absent from the lexiconin earlier years or implicitly conveyed by authors by simply using the terms “MTM” orpharmaceutical care to describe their intervention.Our approach to categorizing interventions for KQ 1 relied primarily on the shortdescriptions in published manuscripts and those we were able to obtain via email inquiries. Theirsimilarities or differences substituted for any overarching taxonomy, because none that weconsidered seemed to fit our purpose. Thus, we have introduced intervention labels that,admittedly, do not fully describe or account for clinical heterogeneity among interventions. Thisapproach limits our ability to make definitive statements about the effectiveness of variousintervention components. We believe that the clusters and categorizations we used are usefulheuristics, but some may regard them more as hypothesis generating than as reflecting settledprinciples of classification.Finally, our search process was complicated by having to ensure coverage of all terms thatcould be used to describe MTM interventions over time. Adding to this challenge was our effortto examine the gray literature, where we thought we might find studies tilted towardeffectiveness and real-world program evaluation. As it turned out, studies of these types ofinterventions were not indexed similarly; for that reason, we needed to rely heavily on handsearches of citation lists from key background articles to identify possibly relevant studies forinclusion. Thus, we may have missed some studies that might have qualified for inclusion. Given130


the considerable diversity in the evidence base we did have, however, we do not think that anypotentially missed studies would have changed our conclusions in any material way. No metaanalysesincluded more than five studies; as a result, we did not examine included studies forpublication bias quantitatively.Limitations of the EvidenceAs a body of evidence, the MTM literature evaluated in this review has measured numerousoutcomes. As indicated in previous sections, very few outcomes, with the exception of harms,remain completely unexamined. Of the 44 studies in this review, we rated 28 as having medium,low, or mixed risk of bias. The 44 studies included 21 trials and 4 nonrandomized controlledstudies. In other words, the literature on this topic is not marked by failure to consider importantoutcomes, universally high risk of bias, or pervasively weak designs.Despite these advantages, we were unable to identify sufficient evidence on the majority ofhypothesized outcomes of MTM. In several instances, our inability to rate evidence as higherthan insufficient came from indirect, inconsistent, and imprecise evidence. The choice ofoutcome measures in this body of evidence limited our ability to come to conclusions in someinstances. For example, some studies did not focus on changes that proponents might expectMTM services to produce. Because effective MTM can either increase or decrease expendituresor use of services based on the needs of the patient, studies that did not prespecify the expecteddirection of change had no way to interpret their results as an appropriate change. Studies thatdemonstrated inconsistent results in direction of change (i.e., some showing an increase inresource use and others showing a decrease) may well have been consistent in terms ofappropriate change, but because they generally failed to establish a priori the direction in whichthey expected to find an effect, we rated such evidence as indirect and inconsistent.Similarly, studies often used nonstandardized or idiosyncratic measures for outcomes such asadverse events, adherence, and expenditures or costs; this tendency limited our ability to metaanalyzeresults. When studies focused on specific outcomes, they were often significantlyunderpowered to detect differences between groups (that is, they did not meet optimalinformation size criteria). As a result, we rated several studies as imprecise.MTM intervention studies are largely practice based and incorporate substantialheterogeneity in specific intervention elements and in patient populations targeted. Yet theevidence is sharply constrained in its ability to inform questions of the effectiveness of specificMTM components or intervention features (KQ 3 in our review) because study designs did notoften capitalize on variants in MTM programs for a prospective evaluation of outcomes by thosevariants. Neither did they measure fidelity to intended MTM elements for post-hoc evaluation.Similarly, the relatively untargeted nature of the MTM interventions meant that, in many studies,only small numbers of patients had any one specific condition, and most studies did not measurepatient characteristics beyond age and sex, thus limiting our ability to address KQ 4 in ourreview. For this reason, the evidence we identified for this review was most relevant for KQ 2.Research GapsIn many bodies of research, questions regarding the comparative effectiveness of specificintervention components or implementation features are best answered after clear evidence of theeffectiveness of the intervention relative to usual care has been established. Our review largelyindicates insufficient evidence on the primary question of effectiveness relative to usual care. Bydefinition, this limited what we could say about comparative effectiveness.131


Nonetheless, the widespread implementation of MTM coexists with the urgent need foractionable information for policy, program policies, and training. This clinical and policyenvironment means that new research cannot afford to address causal claims relative to usualcare first, followed by comparative effectiveness of the intervention elements in a relativelycontrolled environment, and finally, program evaluation of real-world implementation, all insequential order.In prioritizing among various research goals, therefore, funders may wish to consider therelative value of new evidence on overall effectiveness, effectiveness of implementation features,and program implementation and accountability. Trial research in narrow clinical settings canaddress questions of effectiveness but may lack applicability to real-world implementation.Likewise, evaluations of real-world programs with variable fidelity to interventions can answerquestions about process and implementation, but they offer limited information on effectiveness.Research prioritization exercises will also need to account for already commissioned MTMintervention studies.For new studies focusing on causal claims, a critical gap relates to the failure to specify theexpected direction of effect. New research requires a strong theoretical foundation to helpspecify causal mechanisms and hypothesized effects. Without such an edifice, future researchwill continue to produce inconsistent and uninterpretable results.Heightened attention to causal mechanisms will also help researchers convey theirunderstanding of what outcomes these types of interventions are likely to influence. For instance,how should researchers wishing to establish direct causal links between MTM programs andoutcomes evaluate distal outcomes such as patient-centered outcomes and resource utilization?This effort requires a better understanding of the relationship between proximal outcomes like“drug therapy problems identified and resolved” and distal outcomes. For instance, MTM mayreduce outpatient visits to address side effects. MTM may also result in the need for furthertesting and evaluation for some patients, which could, in turn, result in more rather than feweroutpatient visits. Unless the nature of change resulting from MTM is specified in relation togoals of drug therapy, studies cannot assert benefit or harm. Further, drug therapy problems arediverse and may not all have the same causal relationship to health, quality of life, patientsatisfaction, or resource use outcomes. Furthermore, a causal model of these distal outcomes mayneed to take into account the competing or complementary contributions of MTM, new modelsof health care delivery (e.g., patient-centered medical homes), and other quality improvementinterventions.Investigators embarking on new studies focusing on causal links between MTM andoutcomes may wish to consider the limitations of studies based on secondary data from existingMTM programs that use opt-in/opt-out patient enrollment mechanisms. Although these studiesmay provide invaluable information on process measures such as patient engagement, underlyingissues of confounding severely limit the validity of causal claims from such studies.Regardless of the goal of their future research, investigators should consider issues of samplesize to ensure precision of their results. This issue is particularly relevant when evaluatingoutcomes likely to occur in smaller subgroups defined by patient risk, context, or highly adaptedintervention features. Innovative designs (e.g., stepped wedge trials, statistical process control,time series analysis, simulations, and factorial experiments) may permit both rigor and adequatesample size within the context of real-world implementation. With careful attention to fidelity,new studies may also inform questions of the effectiveness of intervention components andimplementation features. Mixed-methods approaches may allow more information on variations132


in context and implementation. Such designs may also help inform our understanding of criticaltraining elements for MTM service providers.Regarding research gaps for specific outcomes such as patient satisfaction, measures specificto the types of services provided through MTM (e.g., patient education about medications) or tothe proximal outcomes that MTM is intended to achieve (e.g., reduced medication side effects,improved disease control) may offer better insights into the effects of MTM. Similarly, amedication-related instrument may better measure patients’ concerns that are directly related tomedication use (e.g., experience of side-effects, intrusiveness of the medication regimen) thangeneric tools.ConclusionsWe included 44 eligible studies (21 randomized controlled trials, 4 controlled clinical trials,and 19 cohort studies) reported in 61 articles, described in detail in the report (KQ 1). Evidencewas insufficient on the effect of MTM on most outcomes (KQ 2). In a few instances, describedbelow, the evidence led us to conclude with a low strength of evidence either a benefit or lack ofbenefit. Specifically, we found evidence that MTM results in improvement when compared withusual care for some measures of medication adherence and appropriateness, medication dosing,health plan expenditures on medication costs, and the proportion and costs of hospitalization forpatient with diabetes. Similarly, we conclude based on a low strength of evidence that MTMconferred no benefit for patient satisfaction and most measures of health-related quality of life.We found evidence on five intervention components and intervention features (KQ 3): one studyprovided information on each feature and yielded insufficient evidence for most outcomes withtwo exceptions. MTM programs with pharmacist access to brief clinical summaries from themedical record reduced the mean number of adverse drug events when compared with basicMTM programs without such access (low strength of evidence). Community pharmacistsincrease the generic dispensing ratio more than call-center–based pharmacists (low strength ofevidence). We found no relevant studies on patient characteristics moderating the effect of MTMinterventions (KQ 4). Similarly, the evidence on harms associated with MTM was limited to onestudy on inconvenience and was rated as insufficient (KQ 5).The evidence base offers low evidence of benefit for a limited number of intermediate andhealth utilization outcomes. We graded the evidence for most outcomes as insufficient becauseof inconsistency in direction, magnitude, and precision, rather than lack of evidence. Widevariations in populations and interventions, both within and across studies, likely explain theseinconsistencies. Given the widespread implementation of MTM and urgent need for actionableinformation, optimal investments in new research require a process of research prioritization inwhich the value of information from each proposed study is carefully considered. Studiesdesigned to identify causal relationships between MTM interventions and their outcomes requireadequate controls for confounding but may offer limited information on what explains programsuccess or failure. Studies designed to explore the reasons for program success or failure usingqualitative or single-arm designs may offer hypotheses-generating rather than hypothesesconfirminginsights on MTM effectiveness. New research, regardless of specific focus, willlikely continue to find inconsistent results until underlying sources of heterogeneity areaccounted for.133


1. Adams K, Corrigan E. Priority Areas for QualityImprovement. Washington, DC: Institute ofMedicine of the National Academies, TheNational Academies Press; 2003.References11. Lewin Group. Medication therapy managementservices: a critical review. J Am Pharm Assoc(2003). 2005 Sep-Oct;45(5):580-7. Epub:2005/11/22. PMID: 16295643.2. Kohn LT, Corrigan J, Donaldson MS. To Err isHuman: Building A Safer Health System.Washington, DC: National Academy Press;2000.3. McGlynn EA, Asch SM, Adams J, et al. Thequality of health care delivered to adults in theUnited States. N Engl J Med. 2003 Jun26;348(26):2635-45. Epub: 2003/06/27. PMID:12826639.4. Zhan C, Arispe I, Kelley E, et al. Ambulatorycare visits for treating adverse drug effects in theUnited States, 1995-2001. Jt Comm J QualPatient Saf. 2005 Jul;31(7):372-8. Epub:2005/09/01. PMID: 16130980.5. Gurwitz JH, Rochon P, Food and DrugAdministration. Improving the quality ofmedication use in elderly patients: a not-sosimpleprescription. Arch Intern Med. 2002 Aug12-26;162(15):1670-2. Epub: 2002/08/03.PMID: 12153368.6. Lazarou J, Pomeranz BH, Corey PN. Incidenceof adverse drug reactions in hospitalizedpatients: a meta-analysis of prospective studies.JAMA. 1998 Apr 15;279(15):1200-5. Epub:1998/04/29. PMID: 9555760.7. Warholak TL, Nau DP. Quality and Safety inPharmacy Practice. New York, NY: McGraw-Hill Medical; 2010.8. Mahan D. Out-of-Bounds: Rising PrescriptionDrug Prices for Seniors. Families USAFoundation Publication No. 03-106. Washington,DC: July 2003.9. Kaufman DW, Kelly JP, Rosenberg L, et al.Recent patterns of medication use in theambulatory adult population of the United States:the Slone survey. JAMA. 2002 Jan16;287(3):337-44. Epub: 2002/01/16. PMID:11790213.10. Pellegrino AN, Martin MT, Tilton JJ, et al.Medication therapy management services:definitions and outcomes. Drugs.2009;69(4):393-406. Epub: 2009/03/28. PMID:19323584.12. 108th U.S. Congress. Medicare PrescriptionDrug, Improvement, and Modernization Act of2003. 2003.13. Bluml BM. Definition of medication therapymanagement: development of professionwideconsensus. J Am Pharm Assoc (2003). 2005 Sep-Oct;45(5):566-72. Epub: 2005/11/22. PMID:16295641.14. American Pharmacists Association. APhA MTMCentral. What is medication therapymanagement? Washington, DC: AmericanPharmacists Association; 2014www.pharmacist.com/mtm. Accessed March 6,2014.15. Association of Chain Drug Stores Foundation.Medication therapy management in pharmacypractice: core elements of an MTM servicemodel (version 2.0). J Am Pharm Assoc (2003).2008 May-Jun;48(3):341-53. Epub: 2008/07/04.PMID: 18595820.16. Pharmacist Services Technical AdvisoryCoalition. Medication Therapy ManagementService Codes. PSTAC; 2010www.pstac.org/services/mtms-codes.html.Accessed March 5, 2013.17. McInnis T, Strand LM, Webb CE, et al. ThePatient-Centered Medical Home: IntegratingComprehensive Medication Management toOptimize Patient Outcomes. Resource Guide.Patient-Centered Primary Care CollaborativeWashington, DC: June 2012.www.pcpcc.org/sites/default/files/media/medmanagement.pdf.18. Department of Veterans Affairs, Veterans HealthAdministration. Pharmacy GeneralRequirements. Washington, DC: Veterans HealthAdministration; 2008 April 17.www.va.gov/vhapublications/ViewPublication.asp?pub_ID=1678. Accessed February 20, 2013.134


19. Tudor CG. Contract Year 2014 MedicationTherapy Management Program Guidance andSubmission Instructions. Baltimore, MD:Centers for Medicare & Medicaid Services; 2013www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/Memo-Contract-Year-2014-Medication-Therapy-Management-MTM-Program-Submission-v040513.pdf. Accessed May 2,2014.20. Isetts BJ, Buffington DE. CPT code-changeproposal: national data on pharmacists'medication therapy management services.Consult Pharm. 2007 Aug;22(8):684-9. Epub:2008/01/22. PMID: 18203409.21. American Medical Association. CPT changes2006: an insider's view. Chicago, IL: AmericanMedical Association; 2005. p. 309-12.22. CPT Assistant. Transitional Care ManagementServices. CPT Assistant; 2013 www.sccmamcms.org/Portals/19/assets/docs/TCM%20CPT%20Assist%20Article.pdf.23. Bodenheimer T, Berry-Millett R. Caremanagement of patients with complex healthcare needs. Robert Wood Johnson ResearchSynthesis Report No. 19. San Francisco, CA:December 2009.www.rwjf.org/content/dam/farm/reports/issue_briefs/2009/rwjf49853/subassets/rwjf49853_1.24. Hassol A, Shoemaker SJ. Exploratory Researchon Medication Therapy Management. (Preparedby Abt Associates under Contract #HHSM-500-2005-00181/TO#3.) Baltimore, MD: Centers forMedicare & Medicaid Services; July 8, 2008.www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Reports/downloads/blackwell.pdf.25. Agency for Healthcare Research and Quality.Evidence-based Practice Center SystematicReview Protocol. Project Title: MedicationTherapy Management. Rockville, MD: AHRQ;2013 July 24.http://effectivehealthcare.ahrq.gov/ehc/products/516/1601/medication-therapy-managementprotocol-130724.pdf.Accessed September 7,2013.26. Moher D, Liberati A, Tetzlaff J, et al. Preferredreporting items for systematic reviews and metaanalyses:the PRISMA statement. J ClinEpidemiol. 2009 Oct;62(10):1006-12. Epub:2009/07/28. PMID: 19631508.27. Shea BJ, Grimshaw JM, Wells GA, et al.Development of AMSTAR: a measurement toolto assess the methodological quality ofsystematic reviews. BMC Med Res Methodol.2007;7:10. Epub: 2007/02/17. PMID: 17302989.28. Viswanathan M, Ansari MT, Berkman ND, et al.Assessing the Risk of Bias of Individual Studiesin Systematic Reviews of Health CareInterventions. Agency for Healthcare Researchand Quality Methods Guide for ComparativeEffectiveness Reviews AHRQ Publication No.12-EHC047-EF. Rockville, MD: March 2012.www.effectivehealthcare.ahrq.gov/.29. Higgins JP, Altman DG, Gotzsche PC, et al. TheCochrane Collaboration's tool for assessing riskof bias in randomised trials. BMJ.2011;343:d5928. Epub: 2011/10/20. PMID:22008217.30. Viswanathan M, Berkman ND. Development ofthe RTI item bank on risk of bias and precisionof observational studies. J Clin Epidemiol. 2012Feb;65(2):163-78. Epub: 2011/10/01. PMID:21959223.31. West SL, Gartlehner G, Mansfield AJ, et al.Comparative Effectiveness Review Methods:Clinical Heterogeneity. Agency for HealthcareResearch and Quality Methods Research Paper.AHRQ Publication No. 10-EHC070-EF.Rockville, MD: September 2010.http://effectivehealthcare.ahrq.gov/.32. Berkman ND, Lohr KN, Ansari M, et al. Gradingthe Strength of a Body of Evidence WhenAssessing Health Care Interventions for theEffective Health Care Program of the Agency forHealthcare Research and Quality: An Update.Methods Guide for Comparative EffectivenessReviews (Prepared by the RTI-UNC EvidencebasedPractice Center under Contract No. 290-2007-10056-I) AHRQ Publication No. 13(14)-EHC130-EF. Rockville, MD: Agency forHealthcare Research and Quality; November2013.www.effectivehealthcare.ahrq.gov/reports/final.cfm.33. Guyatt GH, Oxman AD, Kunz R, et al. GRADEguidelines 6. Rating the quality of evidence--imprecision. J Clin Epidemiol. 2011Dec;64(12):1283-93. Epub: 2011/08/16. PMID:21839614.135


34. Guyatt GH, Oxman AD, Vist G, et al. GRADEguidelines: 4. Rating the quality of evidence--study limitations (risk of bias). J Clin Epidemiol.2011 Apr;64(4):407-15. Epub: 2011/01/21.PMID: 21247734.35. Owens DK, Lohr KN, Atkins D, et al. AHRQseries paper 5: grading the strength of a body ofevidence when comparing medical interventions--Agency for Healthcare Research and Qualityand the Effective Health-Care Program. J ClinEpidemiol. 2010 May;63(5):513-23. Epub:2009/07/15. PMID: 19595577.36. Atkins D, Chang SM, Gartlehner G, et al.Assessing applicability when comparing medicalinterventions: AHRQ and the Effective HealthCare Program. J Clin Epidemiol. 2011Nov;64(11):1198-207. Epub: 2011/04/06. PMID:21463926.37. Fox D, Ried LD, Klein GE, et al. A medicationtherapy management program's impact on lowdensitylipoprotein cholesterol goal attainment inMedicare Part D patients with diabetes. J AmPharm Assoc (2003). 2009 Mar-Apr;49(2):192-9. Epub: 2009/03/18. PMID: 19289345.38. Jeong E, Hui R, Gallagher M, et al. Clinical,Economical and Humanistic Outcomes from aPharmacist-managed Medication TherapyManagement Program (MTMP) in MedicarePatients from an Integrated Healthcare DeliverySystem. Kaiser Permanante; 2009. p. 1-31.39. Moczygemba LR, Barner JC, Gabrillo ER.Outcomes of a Medicare Part D telephonemedication therapy management program.2012;52:E144.40. Moczygemba LR, Barner JC, Lawson KA, et al.Impact of telephone medication therapymanagement on medication and health-relatedproblems, medication adherence, and MedicarePart D drug costs: a 6-month follow up. Am JGeriatr Pharmacother. 2011 Oct;9(5):328-38.Epub: 2011/08/26. PMID: 21865093.41. Moczygemba LR, Barner JC, Gabrillo ER, et al.Development and implementation of a telephonemedication therapy management program forMedicare beneficiaries. Am J Health Syst Pharm.2008 Sep 1;65(17):1655-60. Epub: 2008/08/21.PMID: 18714113.42. Pindolia VK, Stebelsky L, Romain TM, et al.Mitigation of medication mishaps via medicationtherapy management. Ann Pharmacother. 2009Apr;43(4):611-20. Epub: 2009/04/02. PMID:19336646.43. Staresinic AG. Early experience with medicationmanagement - Description and preliminaryoutcomes of a Medicare Part D medicationtherapy management program. J Pharm Soc Wis.2007(SEP-OCT):8-0. PMID: 45-01183.44. Welch EK, Delate T, Chester EA, et al.Assessment of the impact of medication therapymanagement delivered to home-based Medicarebeneficiaries. Ann Pharmacother. 2009Apr;43(4):603-10. Epub: 2009/03/26. PMID:19318600.45. Moore JM, Shartle D, Faudskar L, et al. Impactof a patient-centered pharmacy program andintervention in a high-risk group. J Manag CarePharm. 2013 Apr;19(3):228-36. Epub:2013/03/30. PMID: 23537457.46. Jeong E, Pitcher K, Fredriks D, et al. Clinical &Safety Outcomes from a Pharmacist-managedMedicare Medication Therapy ManagementProgram. Kaiser Permanente.47. Jeong E, Deguzman L, Fredriks D, et al.Comprehensive Medication Reviews ImproveHospitalization and Emergency DepartmentOutcomes. Kaiser Permanente; 2012.48. Yamada B. Comparison of Outcomes in theMedicare Medication Therapy ManagementProgram within Kaiser Permanente California.2012.49. Christensen DB, Roth M, Trygstad T, et al.Evaluation of a pilot medication therapymanagement project within the North CarolinaState Health Plan. J Am Pharm Assoc (2003).2007 Jul-Aug;47(4):471-83. Epub: 2007/07/10.PMID: 17616493.50. Jameson J, VanNoord G, Vanderwoud K. Theimpact of a pharmacotherapy consultation on thecost and outcome of medical therapy. J FamPract. 1995 Nov;41(5):469-72. Epub:1995/11/01. PMID: 7595265.51. Malone DC, Carter BL, Billups SJ, et al. Aneconomic analysis of a randomized, controlled,multicenter study of clinical pharmacistinterventions for high-risk veterans: theIMPROVE study. Impact of ManagedPharmaceutical Care Resource Utilization andOutcomes in Veterans Affairs Medical Centers.Pharmacotherapy. 2000 Oct;20(10):1149-58.Epub: 2000/10/18. PMID: 11034037.136


52. Ellis SL, Billups SJ, Malone DC, et al. Types ofinterventions made by clinical pharmacists in theIMPROVE study. Impact of ManagedPharmaceutical Care on Resource Utilization andOutcomes in Veterans Affairs Medical Centers.Pharmacotherapy. 2000 Apr;20(4):429-35. Epub:2000/04/20. PMID: 10772374.53. Ellis SL, Carter BL, Malone DC, et al. Clinicaland economic impact of ambulatory care clinicalpharmacists in management of dyslipidemia inolder adults: the IMPROVE study. Impact ofManaged Pharmaceutical Care on ResourceUtilization and Outcomes in Veterans AffairsMedical Centers. Pharmacotherapy. 2000Dec;20(12):1508-16. Epub: 2000/12/29. PMID:11130223.54. Malone DC, Carter BL, Billups SJ, et al. Canclinical pharmacists affect SF-36 scores inveterans at high risk for medication-relatedproblems? Med Care. 2001 Feb;39(2):113-22.Epub: 2001/02/15. PMID: 11176549.55. Sellors J, Kaczorowski J, Sellors C, et al. Arandomized controlled trial of a pharmacistconsultation program for family physicians andtheir elderly patients. CMAJ. 2003 Jul8;169(1):17-22. Epub: 2003/07/09. PMID:12847034.56. Sidel VW, Beizer JL, Lisi-Fazio D, et al.Controlled study of the impact of educationalhome visits by pharmacists to high-risk olderpatients. J Community Health. 1990Jun;15(3):163-74. Epub: 1990/06/01. PMID:2195066.57. Winston S, Lin YS. Impact on drug cost and useof Medicare part D of medication therapymanagement services delivered in 2007. J AmPharm Assoc (2003). 2009 Nov-Dec;49(6):813-20. Epub: 2009/10/28. PMID: 19858047.58. Witry MJ, Doucette WR, Gainer KL. Evaluationof the pharmaceutical case management programimplemented in a private sector health plan. JAm Pharm Assoc (2003). 2011 Sep-Oct;51(5):631-5. Epub: 2011/09/08. PMID:21896463.59. Gattis WA, Hasselblad V, Whellan DJ, et al.Reduction in heart failure events by the additionof a clinical pharmacist to the heart failuremanagement team: results of the Pharmacist inHeart Failure Assessment Recommendation andMonitoring (PHARM) Study. Arch Intern Med.1999 Sep 13;159(16):1939-45. Epub:1999/09/24. PMID: 10493325.60. Shimp LA, Kucukarslan SN, Elder J, et al.Employer-based patient-centered medicationtherapy management program: evidence andrecommendations for future programs. J AmPharm Assoc (2003). 2012;52(6):768-76. Epub:2012/12/12. PMID: 23229963.61. Sellors C, Dalby DM, Howard M, et al. Apharmacist consultation service in communitybasedfamily practices: a randomized, controlledtrial in seniors. J Pharm Technol. 2001(6):264-9.PMID: CN-00425489.62. Perlroth D, Marrufo G, Montesinos A, et al.Medication Therapy Management in ChronicallyIll Populations: Final Report. Contract #HHSM-500-2011-00012I/TOT0001. Baltimore,MD: Centers for Medicare & Medicaid Services(CMS) Center for Medicare & MedicaidInnovation; 2013.http://innovation.cms.gov/Files/reports/MTM_Final_Report.pdf.63. McDonough RP, Doucette WR, Kumbera P, etal. An evaluation of managing and educatingpatients on the risk of glucocorticoid-inducedosteoporosis. Value Health. 2005 Jan-Feb;8(1):24-31. Epub: 2005/04/22. PMID:15841891.64. Bernsten C, Bjorkman I, Caramona M, et al.Improving the well-being of elderly patients viacommunity pharmacy-based provision ofpharmaceutical care: a multicentre study in sevenEuropean countries. Drugs Aging.2001;18(1):63-77. Epub: 2001/03/10. PMID:11232739.65. Sturgess IK, McElnay JC, Hughes CM, et al.Community pharmacy based provision ofpharmaceutical care to older patients. PharmWorld Sci. 2003 Oct;25(5):218-26. Epub:2003/10/31. PMID: 14584229.66. Brummel AR, Soliman AM, Carlson AM, et al.Optimal diabetes care outcomes following faceto-facemedication therapy management services.Popul Health Manag. 2013 Feb;16(1):28-34.Epub: 2012/11/02. PMID: 23113628.67. Soliman AM, Carlson AM, MacLehose RF, et al.Patient characteristics predicting the frequencyof medication therapy management visits forpatients with diabetes. Clin Ther. 2013Apr;35(4):534-40. Epub: 2013/04/02. PMID:23541131.137


68. Ramalho de Oliveira D, Brummel AR, MillerDB. Medication therapy management: 10 yearsof experience in a large integrated health caresystem. J Manag Care Pharm. 2010Apr;16(3):185-95. Epub: 2010/03/25. PMID:20331323.69. Touchette DR, Masica AL, Dolor RJ, et al.Safety-focused medication therapy management:a randomized controlled trial. J Am Pharm Assoc(2003). 2012 Sep-Oct;52(5):603-12. Epub:2012/10/02. PMID: 23023840.70. Carter BL, Barnette DJ, Chrischilles E, et al.Evaluation of hypertensive patients after careprovided by community pharmacists in a ruralsetting. Pharmacotherapy. 1997 Nov-Dec;17(6):1274-85. Epub: 1997/12/17. PMID:9399611.71. Barnette DJ, Murphy CM, Carter BL. Clinicalskill development for community pharmacists. JAm Pharm Assoc (Wash). 1996Sep;NS36(9):573-80. Epub: 1996/09/01. PMID:8824077.72. Clifford RM, Batty KT, Davis TME, et al. Arandomised controlled trial of a pharmaceuticalcare programme in high-risk diabetic patients inan outpatient clinic. Int J Pharm Pract.2002;10(2):85-9.73. Pai AB, Boyd A, Depczynski J, et al. Reduceddrug use and hospitalization rates in patientsundergoing hemodialysis who receivedpharmaceutical care: a 2-year, randomized,controlled study. Pharmacotherapy. 2009Dec;29(12):1433-40. Epub: 2009/12/02. PMID:19947803.74. Pai AB, Boyd A, Chavez A, et al. Health-relatedquality of life is maintained in hemodialysispatients receiving pharmaceutical care: a 2-yearrandomized, controlled study. Hemodial Int.2009 Jan;13(1):72-9. Epub: 2009/02/13. PMID:19210281.75. Marques LA, Galduroz JC, Fernandes MR, et al.Assessment of the effectiveness ofpharmacotherapy follow-up in patients treatedfor depression. J Manag Care Pharm. 2013Apr;19(3):218-27. Epub: 2013/03/30. PMID:23537456.76. Hirsch JD, Gonzales M, Rosenquist A, et al.Antiretroviral therapy adherence, medicationuse, and health care costs during 3 years of acommunity pharmacy medication therapymanagement program for Medi-Cal beneficiarieswith HIV/AIDS. J Manag Care Pharm. 2011Apr;17(3):213-23. Epub: 2011/03/26. PMID:21434698.77. Hirsch JD, Rosenquist A, Best BM, et al.Evaluation of the first year of a pilot program incommunity pharmacy: HIV/AIDS medicationtherapy management for Medi-Cal beneficiaries.J Manag Care Pharm. 2009 Jan-Feb;15(1):32-41.Epub: 2009/01/08. PMID: 19125548.78. Roughead EE, Barratt JD, Ramsay E, et al. Theeffectiveness of collaborative medicine reviewsin delaying time to next hospitalization forpatients with heart failure in the practice setting:results of a cohort study. Circ Heart Fail. 2009Sep;2(5):424-8. Epub: 2009/10/08. PMID:19808372.79. Chrischilles EA, Carter BL, Lund BC, et al.Evaluation of the Iowa Medicaid pharmaceuticalcase management program. J Am Pharm Assoc(2003). 2004 May-Jun;44(3):337-49. Epub:2004/06/12. PMID: 15191244.80. Wittayanukorn S, Westrick SC, Hansen RA, etal. Evaluation of medication therapymanagement services for patients withcardiovascular disease in a self-insured employerhealth plan. J Manag Care Pharm. 2013Jun;19(5):385-95. Epub: 2013/05/24. PMID:23697476.81. Isetts BJ, Schondelmeyer SW, Artz MB, et al.Clinical and economic outcomes of medicationtherapy management services: the Minnesotaexperience. J Am Pharm Assoc. 2008 Mar-Apr;48(2):203-11; 3 p following 11. Epub:2008/03/25. PMID: 18359733.82. Fischer LR, Defor TA, Cooper S, et al.Pharmaceutical care and health care utilization inan HMO. Eff Clin Pract. 2002 Mar-Apr;5(2):49-57. Epub: 2002/05/07. PMID: 11990212.83. Fischer LR, Scott LM, Boonstra DM, et al.Pharmaceutical care for patients with chronicconditions. J Am Pharm Assoc (Wash). 2000Mar-Apr;40(2):174-80. Epub: 2000/03/24.PMID: 10730021.138


84. Hanlon JT, Weinberger M, Samsa GP, et al. Arandomized, controlled trial of a clinicalpharmacist intervention to improve inappropriateprescribing in elderly outpatients withpolypharmacy. Am J Med. 1996Apr;100(4):428-37. Epub: 1996/04/01. PMID:8610730.85. Taylor CT, Byrd DC, Krueger K. Improvingprimary care in rural Alabama with a pharmacyinitiative. Am J Health Syst Pharm. 2003 Jun1;60(11):1123-9. Epub: 2003/06/21. PMID:12816022.86. Volume CI, Farris KB, Kassam R, et al.Pharmaceutical care research and educationproject: patient outcomes. J Am Pharm Assoc(Wash). 2001 May-Jun;41(3):411-20. Epub:2001/05/25. PMID: 11372906.87. Kassam R, Farris KB, Burback L, et al.Pharmaceutical care research and educationproject: pharmacists' interventions. J Am PharmAssoc (Wash). 2001 May-Jun;41(3):401-10.Epub: 2001/05/25. PMID: 11372905.88. Grymonpre RE, Williamson DA, MontgomeryPR. Impact of a pharmaceutical care model fornon-institutionalised elderly: results of arandomised, controlled trial. Int J Pharm Pract.2001(4):235-41. PMID: CN-00425461.89. Park JJ, Kelly P, Carter BL, et al.Comprehensive pharmaceutical care in the chainsetting. J Am Pharm Assoc (Wash). 1996Jul;NS36(7):443-51. Epub: 1996/07/01. PMID:8840744.90. Planas LG, Crosby KM, Mitchell KD, et al.Evaluation of a hypertension medication therapymanagement program in patients with diabetes. JAm Pharm Assoc (2003). 2009 Mar-Apr;49(2):164-70. Epub: 2009/03/18. PMID:19289342.91. Krska J, Cromarty JA, Arris F, et al. Pharmacistledmedication review in patients over 65: arandomized, controlled trial in primary care. AgeAgeing. 2001 May;30(3):205-11. Epub:2001/07/10. PMID: 11443021.92. Williams ME, Pulliam CC, Hunter R, et al. Theshort-term effect of interdisciplinary medicationreview on function and cost in ambulatoryelderly people. J Am Geriatr Soc. 2004Jan;52(1):93-8. Epub: 2003/12/23. PMID:14687321.93. Blakey SA, Hixson-Wallace JA. Clinical andeconomic effects of pharmacy services in ageriatric ambulatory clinic (Structured abstract).Pharmacotherapy. 2000(10):1198-203. PMID:NHSEED-22000001609.94. Cowper PA, Weinberger M, Hanlon JT, et al.The cost-effectiveness of a clinical pharmacistintervention among elderly outpatients.Pharmacotherapy. 1998 Mar-Apr;18(2):327-32.Epub: 1998/04/17. PMID: 9545151.95. Chisholm-Burns MA, Kim Lee J, Spivey CA, etal. US pharmacists' effect as team members onpatient care: systematic review and metaanalyses.Med Care. 2010 Oct;48(10):923-33.Epub: 2010/08/20. PMID: 20720510.96. Bradley EH, Curry LA, Ramanadhan S, et al.Research in action: using positive deviance toimprove quality of health care. Implement Sci.2009;4:25. Epub: 2009/05/12. PMID: 19426507.97. Brown CA, Lilford RJ. The stepped wedge trialdesign: A systematic review. BMC Med ResMethodol. 2006;6:54. Epub: 2006/11/10. PMID:17092344.139


Appendix A. Literature Search StrategiesPublished LiteraturePubMed. Total of 1961 records retrieved; 1709 records imported after removingduplicates.PubMed Search Update 1-9-14211 additional results; 49 importedSearchNumber ofSearch TermsStringResults#1 Search "Medication Therapy Management"[Mesh] 610#2 Search "medication therapy management" 795#3 Search "comprehensive medication review" 22#4 Search "personal medication record" 14#5 Search ("medication" AND "action plan") 152#6 Search "medication therapy review" 11#7 Search "Medication Reconciliation"[Mesh] 252#8 Search (med* AND reconciliation) 335#9 Search "medication-related problems" 218#10 Search MTMP 34#11 Search prescriber intervention* 243#12 Search "drug utilization management" 6#13 Search "chronic care improvement " 13#14 Search "drug therapy services" 4#15 Search ("utilization management strategies" OR "utilization management strategy") 22#16 Search "optimized treatment outcomes" 6#17 Search ((patient OR patients) AND "medication understanding") 12#18 Search ("drug therapy outcome" OR "drug therapy outcomes") 34#19 Search "medication counseling" 133#20 Search "pharmaceutical case management" 12#21 Search "drug therapy management" 105#22 Search ("drug therapy problem" OR "drug therapy problems") 90#23 Search ("medicine management”[tiab] OR “medicines management”[tiab]) 223#25 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 2113or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Humans#26 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 1961or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Humans; English#27 Search ((#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #140or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) AND (“retraction”[All Fields]OR “Retracted Publication”[pt])) Filters: Humans; English#28 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Publication date from2012/11/04 to 2014/12/31; Humans; English211A-1


PubMed Search Update 11-4-13390 additional results; 236 importedSearchNumber ofSearch TermsStringResults#1 Search "Medication Therapy Management"[Mesh] 579#2 Search "medication therapy management" 765#3 Search "comprehensive medication review" 22#4 Search "personal medication record" 13#5 Search ("medication" AND "action plan") 149#6 Search "medication therapy review" 11#7 Search "Medication Reconciliation"[Mesh] 234#8 Search (med* AND reconciliation) 326#9 Search "medication-related problems" 215#10 Search MTMP 34#11 Search prescriber intervention* 242#12 Search "drug utilization management" 6#13 Search "chronic care improvement " 13#14 Search "drug therapy services" 4#15 Search ("utilization management strategies" OR "utilization management strategy") 22#16 Search "optimized treatment outcomes" 6#17 Search (patient OR patients) AND "medication understanding") 12#18 Search ("drug therapy outcome" OR "drug therapy outcomes") 33#19 Search "medication counseling" 130#20 Search "pharmaceutical case management" 12#21 Search "drug therapy management" 105#22 Search ("drug therapy problem" OR "drug therapy problems") 88#23 Search (“medicine management”[tiab] OR “medicines management”[tiab]) 218#24 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 2414or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23)#25 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 2042or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Humans#26 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 1894or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Humans; English#27 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23) Filters: Publication date from2012/02/27 to 2013/12/31; Humans; English390PubMed search revision 6-27-13: added British terms for MTM to account for the MEDMAN study.149 additional results; all importedSearchSearch TermsString#1 Search “medicine management”[tiab] OR “medicines management”[tiab] Filters:Humans; EnglishNumber ofResults149A-2


PubMed search revision 2-27-13: search re-run while keeping “wildcard” search terms.26 additional results; all importedSearchNumber ofSearch TermsStringResults#1 Search "Medication Therapy Management"[Mesh] 475#2 Search "medication therapy management" 622#3 Search "comprehensive medication review" 18#4 Search "personal medication record" 13#5 Search ("medication" AND "action plan") 139#6 Search "medication therapy review" 10#7 Search "Medication Reconciliation"[Mesh] 168#8 Search (med* AND reconciliation) 27#9 Search "medication-related problems" 197#10 Search MTMP 31#11 Search prescriber intervention* 223#12 Search "drug utilization management" 5#13 Search "chronic care improvement " 13#14 Search "drug therapy services" 4#15 Search ("utilization management strategies" OR "utilization management strategy") 17#16 Search "optimized treatment outcomes" 6#17 Search ((patient OR patients) AND "medication understanding") 12#18 Search ("drug therapy outcome" OR "drug therapy outcomes") 33#19 Search "medication counseling" 122#20 Search "pharmaceutical case management" 11#21 Search “drug therapy management” 97#22 Search ("drug therapy problem" OR “drug therapy problems") 82#23 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or 1694#14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22)#24 Search #23 Filters: Humans 1491#25 Search #23 Filters: Humans; English 1387#26 Search (#25 AND (2012/10:2013/12[edat])) 26A-3


PubMed search revision 2-18-13: updated final PubMed/Medline “specific” MTM-and-MTMcomponentssearch conducted on 11/26/12 by using Entrez date limit of October 2012 to February2013, which is the date each record was entered into PubMed, as opposed to limiting bypublication date.17 additional results; all importedSearchNumber ofSearch TermsStringResults#1 Search "Medication Therapy Management"[Mesh] 472#2 Search "medication therapy management" 621#3 Search "comprehensive medication review" 18#4 Search "personal medication record" 13#5 Search ("medication" AND "action plan") 139#6 Search "medication therapy review" 10#7 Search "Medication Reconciliation"[Mesh] 162#8 Search (med* AND reconciliation) 27#9 Search "medication-related problems" 197#10 Search MTMP 31#11 Search prescriber intervention* 223#12 Search "drug utilization management" 5#13 Search "chronic care improvement " 13#14 Search "drug therapy services" 4#15 Search ("utilization management strategies" OR "utilization management strategy") 17#16 Search "optimized treatment outcomes" 6#17 Search ((patient OR patients) AND "medication understanding") 12#18 Search ("drug therapy outcome" OR "drug therapy outcomes") 33#19 Search "medication counseling" 122#20 Search "pharmaceutical case management" 11#21 Search “drug therapy management” 97#22 Search (“drug therapy problem" OR “drug therapy problems") 82#23 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or 1687#14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22)#24 Search #23 Filters: Humans 1476#25 Search #23 Filters: Humans; English 1372#26 Search (#25 AND (2012/10:2013/02[edat])) 17A-4


PubMed primary search 11-26-12 – 1190 results, all importedSearchNumber ofSearch TermsStringResults#1 Search "Medication Therapy Management"[Mesh] 433#2 Search "medication therapy management" 582#3 Search "comprehensive medication review" 17#4 Search "personal medication record" 13#5 Search ("medication" AND "action plan") 134#6 Search "medication therapy review" 10#7 Search "Medication Reconciliation"[Mesh] 135#8 Search (med* AND reconciliation) 27#9 Search "medication-related problems" 193#10 Search MTMP 31#11 Search prescriber intervention* 217#12 Search "drug utilization management" 5#13 Search "chronic care improvement " 13#14 Search "drug therapy services" 4#15 Search ("utilization management strategies" OR "utilization management strategy") 17#16 Search "optimized treatment outcomes" 6#17 Search ((patient OR patients) AND "medication understanding") 9#18 Search ("drug therapy outcome" OR "drug therapy outcomes") 33#19 Search "medication counseling" 120#20 Search "pharmaceutical case management" 11#21 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or 1473#14 or #15 or #16 or #17 or #18 or #19 or #20)#22 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or 1280#14 or #15 or #16 or #17 or #18 or #19 or #20) Filters: Humans#23 Search (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or#14 or #15 or #16 or #17 or #18 or #19 or #20) Filters: Humans; EnglishMTM terms and specific component terms.1190A-5


Cochrane Library. Total of 408 records retrieved; 299 imported after removing duplicates.Cochrane Library search update 1-10-2014 – 97 results, 9 importedSearchNumber ofSearch TermsStringResults#1 MeSH descriptor: [Medication Therapy Management] explode all trees 31#2 "medication therapy management" 43#3 "comprehensive medication review" 4#4 "personal medication record" 1#5 "medication" and "action plan" 94#6 "medication therapy review" 0#7 MeSH descriptor: [Medication Reconciliation] explode all trees 18#8 "medication reconciliation" 45#9 "medication-related problems" 35#10 MTMP 0#11 "prescriber intervention" or "prescriber interventions" 0#12 "drug utilization management" 0#13 "chronic care improvement" 0#14 "drug therapy services" 0#15 "utilization management strategies" or "utilization management strategy" 0#16 "optimized treatment outcomes" 0#17 (patient or patients) and "medication understanding" 3#18 "drug therapy outcome" or "drug therapy outcomes" 156#19 "medication counseling" 21#20 "pharmaceutical case management" 1#21 "drug therapy problem" or "drug therapy problems" 16#22 "drug therapy management" 10#23 ("medicine management":ti or "medicine management":ab or "medicines management":ti 23or "medicines management":ab)#24 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or 421#15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23#25 MeSH descriptor: [Congresses] explode all trees 4#26 MeSH descriptor: [Congresses as Topic] explode all trees 40#27 congresses:pt 45#28 #25 or #26 or #27 85#29 #24 not #28 from 2012 to 2014 97A-6


Cochrane search update 11-4-2013 – 84 results, 40 importedSearchNumber ofSearch TermsStringResults#1 MeSH descriptor: [Medication Therapy Management] explode all trees 29#2 "medication therapy management" 41#3 "comprehensive medication review" 4#4 "personal medication record" 1#5 "medication" and "action plan" 92#6 "medication therapy review" 0#7 MeSH descriptor: [Medication Reconciliation] explode all trees 16#8 "medication reconciliation" 42#9 "medication-related problems" 35#10 MTMP 0#11 "prescriber intervention" or "prescriber interventions" 0#12 "drug utilization management" 0#13 "chronic care improvement" 0#14 "drug therapy services" 0#15 "utilization management strategies" or "utilization management strategy" 0#16 "optimized treatment outcomes" 0#17 (patient or patients) and "medication understanding" 3#18 "drug therapy outcome" or "drug therapy outcomes" 152#19 "medication counseling" 21#20 "pharmaceutical case management" 1#21 "drug therapy problem" or "drug therapy problems" 16#22 "drug therapy management" 9#23 ("medicine management":ti or "medicine management":ab or "medicines management":ti 23or "medicines management":ab)#24 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or 410#15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23#25 MeSH descriptor: [Congresses] explode all trees 4#26 MeSH descriptor: [Congresses as Topic] explode all trees 40#27 congresses:pt 45#28 #25 or #26 or #27 85#29 #24 not #28 410#30 #24 not #28 from 2012 to 2014 84Cochrane Library search revision 6-27-13: added British terms for MTM to account for theMEDMAN study.21 additional results; all importedSearchSearch TermsString#1 "medicine management":ti or "medicine management":ab or "medicinesmanagement":ti or "medicines management":abNumber ofResults21A-7


Cochrane Library search revision 2-27-13: search re-run while removing “wildcard” search termsand conference papers and abstracts.338 additional results; 337 importedSearchNumber ofSearch TermsStringResults#1 MeSH descriptor: [Medication Therapy Management] explode all trees 19#2 "medication therapy management" 30#3 "comprehensive medication review" 3#4 "personal medication record" 1#5 "medication" and "action plan" 81#6 "medication therapy review" 0#7 MeSH descriptor: [Medication Reconciliation] explode all trees 5#8 "medication reconciliation" 21#9 "medication-related problems" 32#10 MTMP 0#11 "prescriber intervention" or "prescriber interventions" 0#12 "drug utilization management" 0#13 "chronic care improvement" 0#14 "drug therapy services" 0#15 "utilization management strategies" or "utilization management strategy" 0#16 "optimized treatment outcomes" 0#17 (patient or patients) and "medication understanding" 3#18 "drug therapy outcome" or "drug therapy outcomes" 142#19 "medication counseling" 19#20 "pharmaceutical case management" 1#21 "drug therapy problem" or "drug therapy problems" 16#22 "drug therapy management" 8#23 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or 338#15 or #16 or #17 or #18 or #19 or #20 or #21 or #22#24 MeSH descriptor: [Congresses] explode all trees 4#25 MeSH descriptor: [Congresses as Topic] explode all trees 38#26 congresses:pt 45#27 #24 or #25 or #26 83#28 #23 not #27 338A-8


Cochrane Library primary search 2-18-13: run concurrently with revised PubMed search, buteventually replaced with 2-27-13 search described above.534 additional results; 532 importedSearchNumber ofSearch TermsStringResults#1 MeSH descriptor: [Medication Therapy Management] explode all trees 19#2 "medication therapy management" 30#3 "comprehensive medication review" 3#4 "personal medication record" 1#5 "medication" and "action plan" 81#6 "medication therapy review" 0#7 MeSH descriptor: [Medication Reconciliation] explode all trees 5#8 med* and reconciliation 47#9 "medication-related problems" 32#10 MTMP 0#11 prescriber intervention* 180#12 "drug utilization management" 0#13 "chronic care improvement" 0#14 "drug therapy services" 0#15 "utilization management strategies" or "utilization management strategy" 0#16 "optimized treatment outcomes" 0#17 (patient or patients) and "medication understanding" 3#18 "drug therapy outcome" or "drug therapy outcomes" 142#19 "medication counseling" 19#20 "pharmaceutical case management" 1#21 "drug therapy problem" or "drug therapy problems" 16#22 "drug therapy management" 8#23 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or#15 or #16 or #17 or #18 or #19 or #20 or #21 or #22534A-9


International Pharmaceutical Abstracts (IPA): total of 756 records retrieved; 508 imported afterremoving duplicates.IPA search update 1-10-2014 – 63 results, 16 importedSearchNumber ofSearch Terms Limiters/Expanders Last Run ViaStringResultsS24 S23 Limiters - Published Date:20121101-20141231Search modes -Boolean/PhraseS23 S22 Limiters - Language: English;Articles about Human StudiesSearch modes -Boolean/PhraseS22S21S1 OR S2 OR S3 OR S4 OR S5 OR Search modes -S6 OR S7 OR S8 OR S9 OR S10 OR Boolean/PhraseS11 OR S12 OR S13 OR S14 ORS15 OR S16 OR S17 OR S18 ORS19 OR S20 OR S21TI ( "medicine management" OR"medicines management" ) AND AB ("medicine management" OR"medicines management" )Search modes -Boolean/PhraseS20 "drug therapy management" Search modes -Boolean/PhraseS19"drug therapy problem" OR "drugtherapy problems"Search modes -Boolean/PhraseInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstracts637561,66540247149A-10


SearchStringSearch Terms Limiters/Expanders Last Run ViaS18 "pharmaceutical case management" Search modes -Interface -Boolean/PhraseEBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS17 "medication counseling" Search modes -Boolean/PhraseS16S15"drug therapy outcome" OR "drugtherapy outcomes"(patient OR patients) AND"medication understanding"Search modes -Boolean/PhraseSearch modes -Boolean/PhraseS14 "optimized treatment outcomes" Search modes -Boolean/PhraseS13"utilization management strategies"OR "utilization management strategy"Search modes -Boolean/PhraseInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults1423673406A-11


SearchStringSearch Terms Limiters/Expanders Last Run ViaS12 "drug therapy services" Search modes -Interface -Boolean/PhraseEBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS11 "chronic care improvement" Search modes -Boolean/PhraseS10 "drug utilization management" Search modes -Boolean/PhraseS9"prescriber intervention" OR"prescriber interventions"Search modes -Boolean/PhraseS8 MTMP Search modes -Boolean/PhraseS7 "medication-related problems" Search modes -Boolean/PhraseInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults2317410204A-12


SearchStringSearch Terms Limiters/Expanders Last Run ViaS6 "medication reconciliation" Search modes -Interface -Boolean/PhraseEBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS5 "medication therapy review" Search modes -Boolean/PhraseS4 "medication" AND "action plan" Search modes -Boolean/PhraseS3 "personal medication record" Search modes -Boolean/PhraseS2 "comprehensive medication review" Search modes -Boolean/PhraseS1 "medication therapy management" Search modes -Boolean/PhraseInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearch DatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults36710821215315A-13


IPA search update 11-4-13 – 105 results, 40 importedSearchStringSearch Terms Limiters/Expanders Last Run ViaS24 S23 Limiters - Published Interface -Date: 20120201- EBSCOhost20131231ResearchSearch modes - DatabasesBoolean/Phrase Search Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS23 S22 Limiters - Language:English; Articles aboutHuman StudiesSearch modes -Boolean/PhraseS22S21S1 OR S2 OR S3 OR S4 OR S5 OR S6 Search modes -OR S7 OR S8 OR S9 OR S10 OR S11 OR Boolean/PhraseS12 OR S13 OR S14 OR S15 OR S16 ORS17 OR S18 OR S19 OR S20 OR S21TI ( "medicine management" OR"medicines management" ) AND AB ("medicine management" OR "medicinesmanagement" )Search modes -Boolean/PhraseS20 "drug therapy management" Search modes -Boolean/PhraseInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults1057391,64840246A-14


SearchStringS19Search Terms Limiters/Expanders Last Run Via"drug therapy problem" OR "drug therapyproblems"Search modes -Boolean/PhraseS18 "pharmaceutical case management" Search modes -Boolean/PhraseS17 "medication counseling" Search modes -Boolean/PhraseS16S15"drug therapy outcome" OR "drug therapyoutcomes"(patient OR patients) AND "medicationunderstanding"Search modes -Boolean/PhraseSearch modes -Boolean/PhraseInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults14814235734A-15


SearchStringSearch Terms Limiters/Expanders Last Run ViaS14 "optimized treatment outcomes" Search modes - Interface -Boolean/Phrase EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS13 "utilization management strategies" OR Search modes - Interface -"utilization management strategy" Boolean/Phrase EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS12 "drug therapy services" Search modes -Boolean/PhraseS11 "chronic care improvement" Search modes -Boolean/PhraseS10 "drug utilization management" Search modes -Boolean/PhraseInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults052317A-16


SearchStringS9Search Terms Limiters/Expanders Last Run Via"prescriber intervention" OR "prescriberinterventions"Search modes -Boolean/PhraseS8 MTMP Search modes -Boolean/PhraseS7 "medication-related problems" Search modes -Boolean/PhraseS6 "medication reconciliation" Search modes -Boolean/PhraseS5 "medication therapy review" Search modes -Boolean/PhraseInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsNumber ofResults41020236410A-17


SearchStringSearch Terms Limiters/Expanders Last Run ViaS4 "medication" AND "action plan" Search modes - Interface -Boolean/Phrase EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS3 "personal medication record" Search modes - Interface -Boolean/Phrase EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsS2 "comprehensive medication review" Search modes -Boolean/PhraseS1 "medication therapy management" Search modes -Boolean/PhraseInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsInterface -EBSCOhostResearchDatabasesSearch Screen -Advanced SearchDatabase -InternationalPharmaceuticalAbstractsIPA search revision 6-27-13: added British terms for MTM to account for the MEDMAN study.19 additional results; 18 importedSearchSearch Terms Limiters/Expanders Last Run ViaStringS1TI ( "medicine management" OR"medicines management" ) ANDAB ( "medicine management" OR"medicines management" )Limiters - Language:English; Articles aboutHuman StudiesSearch modes -Boolean/PhraseInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsNumber ofResults811214308Number ofResults19A-18


IPA search revision 2-27-13: search re-run while removing “wildcard” search terms.673 additional results; 666 importedSearchSearch Terms Limiters/Expanders Last Run ViaStringS22 S21 Limiters - Language:English; Articles aboutHuman StudiesSearch modes -Boolean/PhraseS21 S1 OR S2 OR S3 OR S4 OR S5 Search modes -OR S6 OR S7 OR S8 OR S9 OR Boolean/PhraseS10 OR S11 OR S12 OR S13 ORS14 OR S15 OR S16 OR S17 ORS18 OR S19 OR S20S20 "drug therapy management" Search modes -Boolean/PhraseS19S18"drug therapy problem" OR "drugtherapy problems""pharmaceutical casemanagement"Search modes -Boolean/PhraseSearch modes -Boolean/PhraseS17 "medication counseling" Search modes -Boolean/PhraseS16S15"drug therapy outcome" OR "drugtherapy outcomes"(patient OR patients) AND"medication understanding"Search modes -Boolean/PhraseSearch modes -Boolean/PhraseS14 "optimized treatment outcomes" Search modes -Boolean/PhraseS13"utilization managementstrategies" OR "utilizationmanagement strategy"Search modes -Boolean/PhraseS12 "drug therapy services" Search modes -Boolean/PhraseInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsNumber ofResults6731,55824314514232724042A-19


SearchStringSearch Terms Limiters/Expanders Last Run ViaS11 "chronic care improvement" Search modes - Interface - EBSCOhostBoolean/Phrase Search Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsS10 "drug utilization management" Search modes -Boolean/PhraseS9"prescriber intervention" OR"prescriber interventions"Search modes -Boolean/PhraseS8 MTMP Search modes -Boolean/PhraseS7 "medication-related problems" Search modes -Boolean/PhraseS6 "medication reconciliation" Search modes -Boolean/PhraseS5 "medication therapy review" Search modes -Boolean/PhraseS4 "medication" AND "action plan" Search modes -Boolean/PhraseS3 "personal medication record" Search modes -Boolean/PhraseS2S1"comprehensive medicationreview""medication therapymanagement"Search modes -Boolean/PhraseSearch modes -Boolean/PhraseInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsInterface - EBSCOhostSearch Screen - AdvancedSearchDatabase - InternationalPharmaceutical AbstractsNumber ofResults31641019934110801212289A-20


IPA primary search 2-18-13: run concurrently with revised PubMed search, but eventually replacedwith 2-27-13 search described above.739 additional results; 679 importedSearchNumber ofSearch TermsLimiters/ExpandersStringResultsS23S22S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 ORS17 OR S18 OR S19 OR S20 OR S21S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 ORS17 OR S18 OR S19 OR S20 OR S21Limiters - Language:English; Articles aboutHuman StudiesSearch modes -Boolean/PhraseSearch modes -Boolean/PhraseS21 "drug therapy management" Search modes -Boolean/PhraseS20 "drug therapy problem" OR "drug therapy problems" Search modes -Boolean/PhraseS19 "pharmaceutical case management" Search modes -Boolean/PhraseS18 "medication counseling" Search modes -Boolean/PhraseS17 "drug therapy outcome" OR "drug therapy outcomes" Search modes -Boolean/PhraseS16 (patient OR patients) AND "medication understanding" Search modes -Boolean/PhraseS15 "optimized treatment outcomes" Search modes -Boolean/PhraseS14 "utilization management strategies" OR "utilization management Search modes -strategy"Boolean/PhraseS13 "drug therapy services" Search modes -Boolean/PhraseS12 "chronic care improvement" Search modes -Boolean/PhraseS11 "drug utilization management" Search modes -Boolean/PhraseS10 prescriber intervention* Search modes -Boolean/PhraseS9 MTMP Search modes -Boolean/PhraseS8 "medication-related problems" Search modes -Boolean/PhraseS7 med* AND reconciliation Search modes -Boolean/PhraseS6 "medication reconciliation" Search modes -Boolean/PhraseS5 "medication therapy review" Search modes -Boolean/PhraseS4 "medication" AND "action plan" Search modes -Boolean/PhraseS3 "personal medication record" Search modes -Boolean/PhraseS2 "comprehensive medication review" Search modes -Boolean/PhraseS1 "medication therapy management" Search modes -Boolean/Phrase7391,80324314514232724042316951019950834110801212289A-21


Gray LiteratureSearch update 11-4-13Source Search Terms Limits or AdjustmentsClinicalTrials.govExpert SearchStrategyWHO ICTRPHSRProjAdvanced searchDOPHER(Database ofPromoting HealthEffectivenessReviews)( “medication therapy management” OR“comprehensive medication review” OR“Medication Reconciliation” OR “pharmaceuticalcase management” OR “drug therapymanagement” OR “drug therapy problem” OR“drug therapy problems” OR “medicinemanagement” OR “medicines management” )Title search: “medication therapy management” OR“comprehensive medication review” OR“Medication Reconciliation” OR “pharmaceuticalcase management” OR “drug therapymanagement” OR “drug therapy problem” OR“drug therapy problems” OR “medicinemanagement” OR “medicines management”[ALL-FIELDS] AND ( NOTNOTEXT ) [FIRST-RECEIVED-RESULTS-DATE] AND ("03/04/2013" :"11/04/2013" ) [FIRST-RECEIVED-DATE]Registered from 3/4/2013to 11/4/13Intervention search: “medication therapymanagement”“medication therapy management” ORInitial Year Range = 2011 18 (7)“comprehensive medication review” OR “personal – 2013medication record” OR (medication AND “actionplan”) OR “medication therapy review” OR“Medication Reconciliation” OR “medication-relatedproblems” OR “prescriber intervention” OR “drugutilization management” OR “chronic careimprovement” OR “drug therapy services” OR“utilization management strategies” OR “utilizationmanagement strategy” OR “optimized treatmentoutcomes” OR ((patients OR patient) AND“medication understanding”) OR “drug therapyoutcome” OR “drug therapy outcomes” OR“medication counseling” OR “pharmaceutical casemanagement” OR “drug therapy management” OR“drug therapy problem” OR “drug therapyproblems” OR “medicine management” OR“medicines management” [Limited toOngoing/Completed status]1) medication therapy management ORNone 0comprehensive medication review OR personalmedication record OR (medication AND actionplan) OR medication therapy review ORMedication Reconciliation OR medication-relatedproblems OR medication relation problems ORprescriber intervention OR drug utilizationmanagement OR chronic care improvement ORdrug therapy services OR utilization managementstrategies OR utilization management strategy ORoptimized treatment outcomes OR (patients ORpatient) AND medication understanding) OR drugtherapy outcome OR drug therapy outcomes ORmedication counseling OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problems2) “MTM” or “Medication Therapy Management”Number of ResultsRetrieved (Imported)5 (5)5 (5) (Title search);2 (1) (Interventionsearch)A-22


Source Search Terms Limits or AdjustmentsNew YorkAcademy ofMedicine GrayLiterature Report(greylit.org)1) medication therapy management ORcomprehensive medication review OR personalmedication record OR (medication AND actionplan) OR medication therapy review ORMedication Reconciliation OR medication-relatedproblems OR medication relation problems ORprescriber intervention OR drug utilizationmanagement OR chronic care improvement ORdrug therapy services OR utilization managementstrategies OR utilization management strategy ORoptimized treatment outcomes OR (patients ORpatient) AND medication understanding) OR drugtherapy outcome OR drug therapy outcomes ORmedication counseling OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problemsPublished from 2012-2013Number of ResultsRetrieved (Imported)0 for search string #1;1 (0) for search string#2CMS.gov2) “MTM” or “Medication Therapy Management”1) allintitle: "medication therapy management"site:cms.gov“allintitle”, which limitedresults to those in which“medication therapymanagement” appearedin title of retrievedwebsites82 (82) total:82 through CMS.govdirectly;6 indirectly throughGoogleSearch revision 6-28-13: added British terms (“medicine management” OR “medicinesmanagement”) for MTM to account for the MEDMAN study.Total of 14 records retrieved, 13 imported after initial screening.Source Search Terms Limits or AdjustmentsClinicalTrials.govExpert SearchStrategyWHO ICTRPHSRProj AdvancedsearchNIH RePORTERAdvanced searchDOPHER (Databaseof Promoting HealthEffectivenessReviews)New York Academy ofMedicine GrayLiterature Report(greylit.org)CMS.gov“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”“medicine management” OR “medicinesmanagement”[ALL-FIELDS] AND ( NOTNOTEXT ) [FIRST-RECEIVED-RESULTS-DATE]Number ofResultsRetrieved(Imported)2 (2)None 10 (10)None 0None 0None 0None 0“allintitle”, which limited results tothose in which “medicationtherapy management” appearedin title of retrieved websites2 (1)A-23


Primary searches 3-4-13: 750 records retrieved, 596 imported after removing duplicates.SourceSearch TermsLimits or Number of ResultsAdjustments Retrieved (Imported)ClinicalTrials.govExpert SearchStrategy119 (119)WHO ICTRP( “medication therapy management” OR“comprehensive medication review” OR“Medication Reconciliation” OR “pharmaceuticalcase management” OR “drug therapymanagement” OR “drug therapy problem” OR“drug therapy problems” )Title search: medication therapy management ORcomprehensive medication review OR MedicationReconciliation OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problems[ALL-FIELDS] AND( NOT NOTEXT )[FIRST-RECEIVED-RESULTS-DATE]None5 (5) (Title search);0 (Intervention search)HSRProj AdvancedsearchNIH RePORTERAdvanced searchIntervention search: was either 41,000+, with theshorter search (see Search Strings #1c and #1d),or no results for “medication therapy management”by itself.“medication therapy management” ORNone 87 (82)“comprehensive medication review” OR “personalmedication record” OR (medication AND “actionplan”) OR “medication therapy review” OR“Medication Reconciliation” OR “medication-relatedproblems” OR “prescriber intervention” OR “drugutilization management” OR “chronic careimprovement” OR “drug therapy services” OR“utilization management strategies” OR “utilizationmanagement strategy” OR “optimized treatmentoutcomes” OR ((patients OR patient) AND“medication understanding”) OR “drug therapyoutcome” OR “drug therapy outcomes” OR“medication counseling” OR “pharmaceutical casemanagement” OR “drug therapy management” OR“drug therapy problem” OR “drug therapyproblems” [Limited to Ongoing/Completed status]medication therapy management ORNone 234 (85)comprehensive medication review OR personalmedication record OR (medication AND actionplan) OR medication therapy review ORMedication Reconciliation OR medication-relatedproblems OR medication relation problems ORprescriber intervention OR drug utilizationmanagement OR chronic care improvement ORdrug therapy services OR utilization managementstrategies OR utilization management strategy ORoptimized treatment outcomes OR (patients ORpatient) AND medication understanding) OR drugtherapy outcome OR drug therapy outcomes ORmedication counseling OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problemsA-24


SourceDOPHER (Databaseof Promoting HealthEffectivenessReviews)Search Terms1) medication therapy management ORcomprehensive medication review OR personalmedication record OR (medication AND actionplan) OR medication therapy review ORMedication Reconciliation OR medication-relatedproblems OR medication relation problems ORprescriber intervention OR drug utilizationmanagement OR chronic care improvement ORdrug therapy services OR utilization managementstrategies OR utilization management strategy ORoptimized treatment outcomes OR (patients ORpatient) AND medication understanding) OR drugtherapy outcome OR drug therapy outcomes ORmedication counseling OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problemsLimits orAdjustmentsNoneNumber of ResultsRetrieved (Imported)0 for all search stringsNew York Academyof Medicine GrayLiterature Report(greylit.org)2) “MTM” or “Medication Therapy Management”1) medication therapy management ORcomprehensive medication review OR personalmedication record OR (medication AND actionplan) OR medication therapy review ORMedication Reconciliation OR medication-relatedproblems OR medication relation problems ORprescriber intervention OR drug utilizationmanagement OR chronic care improvement ORdrug therapy services OR utilization managementstrategies OR utilization management strategy ORoptimized treatment outcomes OR (patients ORpatient) AND medication understanding) OR drugtherapy outcome OR drug therapy outcomes ORmedication counseling OR pharmaceutical casemanagement OR drug therapy management ORdrug therapy problem OR drug therapy problemsNone 0 for search string #1;1 (1) for search string#2CMS.gov2) “MTM” or “Medication Therapy Management”1) “medication therapy management” OR “allintitle”, which“comprehensive medication review” OR “personal limited results tomedication record” OR (medication AND “action those in whichplan”) OR “medication therapy review” OR “medication“Medication Reconciliation” OR “medication-related therapyproblems” OR “prescriber intervention” OR “drug management”utilization management” OR “chronic care appeared in title ofimprovement” OR “drug therapy services” OR retrieved websites“utilization management strategies” OR “utilizationmanagement strategy” OR “optimized treatmentoutcomes” OR ((patients OR patient) AND“medication understanding”) OR “drug therapyoutcome” OR “drug therapy outcomes” OR“medication counseling” OR “pharmaceutical casemanagement” OR “drug therapy management” OR“drug therapy problem” OR “drug therapyproblems”304 (304) total:295 through CMS.govdirectly;9 indirectly throughGoogle2) allintitle: "medication therapy management"site:cms.govA-25


Appendix B. Abstract and Full-Text Review Form TemplatesAbstract Review FormRef ID Author Year Include orExclude?(separateexclusioncodes forpublicationtype, PICOTS,and studydesign)If ineligible, ismanualreview orhand searchof full-textneeded?If ineligible,potentialbackgroundreference?NOTE: Thefollowingcolumnsapply only tostudiesmeeting ourinclusioncriteriaStudy Design(RCT, NRCT,Other StudyDesign)If "Other StudyDesign", whichspecific designdoes it use?(Cohort, Case-Control,NonconcurrentTime Series, Other –describe inComments column)Comments (e.g., ifreviewer includedan abstract due toa lack of claritywithin the abstract)Full-Text Review FormRefYeaIDrFirstauthor's lastnameStudyname (ifapplicable)Include orExclude?(separateexclusioncodes forpublicationtype, PICOTS,and studydesign)Handsearchreferences? (If so,markedwith an"X")BKG?(If so,marked withan "X")CommentsforINELIGIBLE studies(e.g.,additionaldetailaboutexclusionreasons)NOTE: Thefollowingcolumnsapply onlyto studiesmeetingourinclusioncriteriaStudyDesign(Dropdownlistoptions:RCT,NRCT,Cohort,Case-Control)KQ(s)(separatesub-columnsfor KQs 1, 2a,2b, 2c, 3, 4,and 5, andrelevantquestionsmarked withan "X")Commentsfor ELIGIBLEstudies (e.g.,for reviewersto describe"Other"studydesigns)B-1


Appendix C. Studies Excluded After Full-Text LevelReviewX1 = Ineligible PublicationX2 = Ineligible or No InterventionX3 = Ineligible PopulationX4 = Ineligible Study DesignX5 = Ineligible ComparatorX6 = Ineligible OutcomesX7 = Ineligible SettingX8 = Insufficient Information to Determine Eligibility1. Patients confirm that medication counselinghelps. Am J Hosp Pharm. 1994 Jul1;51(13):1606, 8. PMID: 7942886. ExclusionCode: X2.2. The MEDMAN study: a randomized controlledtrial of community pharmacy-led medicinesmanagement for patients with coronary heartdisease. Fam Pract. 2007 Apr;24(2):189-200.PMID: 17272285. Exclusion Code: X2.3. MTM program increased statin use. Dis ManagAdvis. 2008 Oct;14(10):suppl 1-3, 1. PMID:19031586. Exclusion Code: X1.4. What's expected for med reconciliation? ORManager. 2008 Mar;24(3):21, 3. PMID:18438074. Exclusion Code: X1.5. Hospitals collaborate to reduce ED overuse.Hosp Case Manag. 2012 Oct;20(10):151-3.PMID: 23091842. Exclusion Code: X1.6. First, do no harm: avoiding medication mishaps.Johns Hopkins Med Lett Health After 50. 2013Summer;24(6):1-4. PMID: 24000429. ExclusionCode: X1.7. Cut readmissions through med adherence. HospPeer Rev. 2013 Feb;38(2):19-20. PMID:23513301. Exclusion Code: X1.8. Optimising medicines management? Drug TherBull. 2013 Apr;51(4):37. PMID: 23575602.Exclusion Code: X1.9. Aguiar PM, Balisa-Rocha BJ, Brito GC, et al.Pharmaceutical care in hypertensive patients: asystematic literature review (Provisionalabstract). DARE. 2012(4):383-96. PMID:DARE-12013008032. Exclusion Code: X2.10. Al-Ghamdi SA, Mahmoud MA, Alammari MA,et al. The outcome of pharmacist counseling atthe time of hospital discharge: an observationalnonrandomized study. Ann Saudi Med. 2012Sep-Oct;32(5):492-7. PMID: 22871618.Exclusion Code: X2.11. Alsuwaidan S, Malone DC, Billups SJ, et al.Characteristics of ambulatory care clinics andpharmacists in Veterans Affairs medical centers.IMPROVE investigators. Impact of ManagedPharmaceutical Care on Resource Utilization andOutcomes in Veterans Affairs Medical Centers.Am J Health Syst Pharm. 1998 Jan 1;55(1):68-72. PMID: 9437478. Exclusion Code: X2.12. Altman JS. Medication therapy management andthe new practitioner. Am J Health Syst Pharm.2007 Mar 15;64(6):590-2. PMID: 17353567.Exclusion Code: X1.13. Anonymous. Prescribing and research inmedicines management (UK & Ireland)Conference 2013 Imperial Hotel London January24th 2013 "Intelligent polypharmacy ... It's notall about the number" abstract.Pharmacoepidemiol Drug Saf.22:670. ExclusionCode: X1.14. Armour CL, Reddel HK, LeMay KS, et al.Feasibility and effectiveness of an evidencebasedasthma service in Australian communitypharmacies: a pragmatic cluster randomized trial.J Asthma. 2013 Apr;50(3):302-9. PMID:23270495. Exclusion Code: X2.15. Asis ML, Greene R. A cost-effectivenessanalysis of a peak flow-based asthma educationand self-management plan in a high-costpopulation (Structured abstract). J Asthma.2004(5):559-65. PMID: NHSEED-22004001183. Exclusion Code: X2.16. Atkinson WL, Frey D. Integration of amedication management model into outcomebasedquality improvement: a pilot program in arural propriety home healthcare agency. HomeC-1


Health Care Serv Q. 2005;24(1-2):29-45. PMID:16236657. Exclusion Code: X4.17. Avery AJ, Rodgers S, Cantrill JA, et al. Protocolfor the PINCER trial: a cluster randomised trialcomparing the effectiveness of a pharmacist-ledIT-based intervention with simple feedback inreducing rates of clinically important errors inmedicines management in general practices.Trials. 2009;10:28. PMID: 19409095. ExclusionCode: X2.18. Avery AJ, Rodgers S, Cantrill JA, et al.Correction: Protocol for the PINCER trial: acluster randomised trial comparing theeffectiveness of a pharmacist-led IT-basedintervention with simple feedback in reducingrates of clinically important errors in medicinesmanagement in general practices. Trials.2010:23. PMID: CN-00789806. Exclusion Code:X1.19. Bandres MA, Mendoza MA, Nicolas FG, et al.Pharmacist-led medication reconciliation toreduce discrepancies in transitions of care inSpain. p. 1083. Exclusion Code: X2.20. Bates DW. Role of pharmacists in the medicalhome. Am J Health Syst Pharm. 2009 Jun15;66(12):1116-8. PMID: 19498128. ExclusionCode: X1.21. Bayoumi I, Howard M, Holbrook AM, et al.Interventions to improve medicationreconciliation in primary care (Structuredabstract). Ann Pharmacother. 2009(10):1667-75.PMID: DARE-12010000178. Exclusion Code:X2.22. Bell JS, Vaananen M, Ovaskainen H, et al.Providing patient care in community pharmacies:practice and research in Finland. AnnPharmacother. 2007 Jun;41(6):1039-46. PMID:17504836. Exclusion Code: X1.23. Bellone JM, Barner JC, Lopez DA.Postdischarge interventions by pharmacists andimpact on hospital readmission rates. J AmPharm Assoc (2003). 2012 May-Jun;52(3):358-62. PMID: 22618976. Exclusion Code: X2.24. Bennett MI, Bagnall AM, Raine G, et al.Educational interventions by pharmacists topatients with chronic pain: systematic review andmeta-analysis. Clin J Pain. 2011 Sep;27(7):623-30. PMID: 21610491. Exclusion Code: X3.25. Benrimoj SI, Langford JH, Berry G, et al.Economic impact of increased clinicalintervention rates in community pharmacy. Arandomised trial of the effect of education and aprofessional allowance. Pharmacoeconomics.2000 Nov;18(5):459-68. PMID: 11151399.Exclusion Code: X2.26. Bergkvist A, Midlov P, Hoglund P, et al.Improved quality in the hospital dischargesummary reduces medication errors--LIMM:Landskrona Integrated Medicines Management.Eur J Clin Pharmacol. 2009 Oct;65(10):1037-46.PMID: 19557400. Exclusion Code: X7.27. Bergkvist A, Midlov P, Hoglund P, et al. Amulti-intervention approach on drug therapy canlead to a more appropriate drug use in theelderly. LIMM-Landskrona Integrated MedicinesManagement. J Eval Clin Pract. 2009Aug;15(4):660-7. PMID: 19674217. ExclusionCode: X7.28. Berjis M, Ho MI, Gray DR. Evaluation of patientoutcome and pharmaceutical care in an HIVclinic. ASHP Midyear Clinical Meeting.1996;31(Dec):P-E. PMID: 33-13178. ExclusionCode: X6.29. Bertino J, Villa T, Corbett A, et al. Effects ofoptimal medication therapy management ofdepression in human immunodeficiency virus(HIV) infected patients on associated virologicand immunologic outcomes. J Pharm Pract.2008;21(Jan):78. PMID: 45-14933. ExclusionCode: X2.30. Billups SJ, Okano G, Malone D, et al. Assessingthe structure and process for providingpharmaceutical care in Veterans Affairs medicalcenters. Am J Health Syst Pharm. 2000 Jan1;57(1):29-39. PMID: 10630554. ExclusionCode: X6.31. Bilyeu KM, Gumm CJ, Fitzgerald JM, et al.Cultivating quality: Reducing the use ofpotentially inappropriate medications in olderadults. Am J Nurs. 2011 Jan;111(1):47-52.PMID: 21191234. Exclusion Code: X2.32. Bland CM, Tritsch AM, Bookstaver DA, et al.Hypertension and diabetes mellitus medicationmanagement in sleeve gastrectomy patients. AmJ Health Syst Pharm. 2013 Jun 15;70(12):1018-20. PMID: 23719877. Exclusion Code: X2.33. Blennerhassett JD, Cusack BM, Smith CD, et al.A novel medicines management pathway. JPharm Pract Res. 2006;36(Mar):175-9. PMID:44-00783. Exclusion Code: X7.34. Blozik E, Born AM, Stuck AE, et al. Reductionof inappropriate medications among olderC-2


nursing-home residents: a nurse-led, pre/postdesign,intervention study. Drugs Aging. 2010Dec 1;27(12):1009-17. PMID: 21087070.Exclusion Code: X7.35. Bonnet-Zamponi D, d'Arailh L, Konrat C, et al.Drug-related readmissions to medical units ofolder adults discharged from acute geriatricunits: results of the Optimization of Medicationin AGEd multicenter randomized controlled trial.J Am Geriatr Soc. 2013 Jan;61(1):113-21.PMID: 23252914. Exclusion Code: X7.36. Borges AP, Guidoni CM, Ferreira LD, et al. Thepharmaceutical care of patients with type 2diabetes mellitus. Pharm World Sci. 2010Dec;32(6):730-6. PMID: 20734138. ExclusionCode: X8.37. Borges AP, Guidoni CM, Freitas O, et al.Economic evaluation of outpatients with type 2diabetes mellitus assisted by a pharmaceuticalcare service. Arq Bras Endocrinol Metabol. 2011Dec;55(9):686-91. PMID: 22231970. ExclusionCode: X8.38. Borgsdorf LR, Miano JS, Knapp KK.Pharmacist-managed medication review in amanaged care system. Am J Hosp Pharm. 1994Mar 15;51(6):772-7. PMID: 8010315. ExclusionCode: X4.39. Branham AR, Katz AJ, Moose JS, et al.Retrospective Analysis of Estimated CostAvoidance Following Pharmacist-ProvidedMedication Therapy Management Services. p.420. Exclusion Code: X4.40. Bratcher CR, Bello E. Traditional or centralizedmodels of diabetes care: the multidisciplinarydiabetes team approach. J Fam Pract. 2011Nov;60(11 Suppl):S6-11. PMID: 22336928.Exclusion Code: X1.41. Brooks JM, Unni EJ, Klepser DG, et al. Factorsaffecting demand among older adults formedication therapy management services. ResSocial Adm Pharm. 2008 Dec;4(4):309-19.PMID: 19064238. Exclusion Code: X2.42. Brown DL, Wolff DJ, Smith LD, et al.Computer-supported medication counseling fordischarged patients. Am J Hosp Pharm. 1986Sep;43(9):2232-5. PMID: 3766581. ExclusionCode: X2.43. Brown EV. Reconcilable differences. HealthManag Technol. 2008 Jan;29(1):12-4, 6. PMID:18286931. Exclusion Code: X1.44. Brusig E, Davies W. PSTAC survey in 2006 wasinstrumental in obtaining permanent (category I)CPT codes for MTM services performed bypharmacists. J Manag Care Pharm. 2008 Jan-Feb;14(1):86-7. PMID: 18240887. ExclusionCode: X1.45. Bucci C, Jackevicius C, McFarlane K, et al.Pharmacist's contribution in a heart functionclinic: patient perception and medicationappropriateness. Can J Cardiol. 2003 Mar31;19(4):391-6. PMID: 12704485. ExclusionCode: X8.46. Bunting BA, Cranor CW. The Asheville Project:long-term clinical, humanistic, and economicoutcomes of a community-based medicationtherapy management program for asthma. J AmPharm Assoc (2003). 2006 Mar-Apr;46(2):133-47. PMID: 16602223. Exclusion Code: X2.47. Bunting BA, Smith BH, Sutherland SE. TheAsheville Project: clinical and economicoutcomes of a community-based long-termmedication therapy management program forhypertension and dyslipidemia. J Am PharmAssoc (2003). 2008 Jan-Feb;48(1):23-31. PMID:18192127. Exclusion Code: X4.48. Burnett KM, Scott MG, Fleming GF, et al.Effects of an integrated medicines managementprogram on medication appropriateness inhospitalized patients. Am J Health Syst Pharm.2009 May 1;66(9):854-9. PMID: 19386949.Exclusion Code: X2.49. Butler CD. Medication therapy management foradverse drug event preventive care. J Am PharmAssoc (2003). 2010 Jul-Aug;50(4):449. PMID:20621858. Exclusion Code: X1.50. Cadth. Medication reconciliation at discharge: areview of the clinical evidence and guidelines(Structured abstract). Health TechnologyAssessment Database. 2012(1)PMID: HTA-32012000668. Exclusion Code: X2.51. Cahill JA, Manasse Jr HR. Medication therapymanagement programs: to optimize pharmacyoutcomes. J Manag Care Pharm. 2005Mar;11(2):179. PMID: 15766326. ExclusionCode: X1.52. Cantwell KM. Collaborating for successfulmedication therapy management programs. Am JHealth Syst Pharm. 2005 Mar 15;62(6):583.PMID: 15757878. Exclusion Code: X1.53. Carey N, Courtenay M, James J, et al. Anevaluation of a Diabetes Specialist NurseC-3


prescriber on the system of delivering medicinesto patients with diabetes. J Clin Nurs. 2008Jun;17(12):1635-44. PMID: 18482124.Exclusion Code: X7.54. Carter BL, Ardery G, Dawson JD, et al.Physician and pharmacist collaboration toimprove blood pressure control. Arch InternMed. 2009 Nov 23;169(21):1996-2002. PMID:19933962. Exclusion Code: X2.55. Carter BL, Lund BC, Hayase N, et al. Alongitudinal analysis of antihypertensive druginteractions in a Medicaid population. Am JHypertens. 2004 May;17(5 Pt 1):421-7. PMID:15110901. Exclusion Code: X5.56. Carter BL, Malone DC, Billups SJ, et al.Interpreting the findings of the IMPROVE study.Am J Health Syst Pharm. 2001 Jul15;58(14):1330-7. PMID: 11471481. ExclusionCode: X1.57. Carter BL, Malone DC, Ellis SL, et al.Antihypertensive Drug Utilization inHypertensive Veterans With ComplexMedication Profiles. J Clin Hypertens(Greenwich). 2000 May;2(3):172-80. PMID:11416643. Exclusion Code: X5.58. Carter BL, Malone DC, Valuck RJ, et al. TheIMPROVE study: background and study design.Impact of Managed Pharmaceutical Care onResource Utilization and Outcomes in VeteransAffairs Medical Centers. Am J Health SystPharm. 1998 Jan 1;55(1):62-7. PMID: 9437477.Exclusion Code: X6.59. Carter BL, Zillich AJ, Elliott WJ. Howpharmacists can assist physicians withcontrolling blood pressure. J Clin Hypertens(Greenwich). 2003 Jan-Feb;5(1):31-7. PMID:12556651. Exclusion Code: X1.60. Casteel C, Blalock SJ, Ferreri S, et al.Implementation of a community pharmacy-basedfalls prevention program. Am J GeriatrPharmacother. 2011 Oct;9(5):310-9 e2. PMID:21925959. Exclusion Code: X2.61. Castro R, Leung J, Song J, et al. Outcomes ofimplementing a medication therapy managementservice for dialysis patients. p. 056. ExclusionCode: X2.62. Cauchi R. Medication therapy management:catching errors, saving lives and money. NCSLLegisbrief. 2010 Jan;18(4):1-2. PMID:20196247. Exclusion Code: X1.63. Centers for Medicare & Medicaid Services.Medicare Part D Medication TherapyManagement (MTM) Programs: 2009 FactSheet. 2009.Exclusion Code: X4.64. Centers for Medicare & Medicaid Services. 2010Medicare Part D Medication TherapyManagement (MTM) Programs Fact Sheet.2010.Exclusion Code: X4.65. Centers for Medicare & Medicaid Services. 2012Medicare Part D Medication TherapyManagement (MTM) Programs Fact Sheet.2012.Exclusion Code: X4.66. Cestac P, Tavassoli N, Vellas B, et al. Improvingmedication use in the nursing homes: a Europeanperspective. J Am Med Dir Assoc. 2013Jan;14(1):6-9. PMID: 23158846. ExclusionCode: X1.67. Chapman NR, Fotis MA, Yarnold PR, et al.Pharmacist interventions to improve themanagement of coronary artery disease. Am JHealth Syst Pharm. 2004 Dec 15;61(24):2672-8.PMID: 15646702. Exclusion Code: X2.68. Charrois TL, Zolezzi M, Koshman SL, et al. Asystematic review of the evidence for pharmacistcare of patients with dyslipidemia.Pharmacotherapy. 2012 Mar;32(3):222-33.PMID: 22392455. Exclusion Code: X2.69. Cheng Y, Raisch DW, Borrego ME, et al.Economic, clinical, and humanistic outcomes(ECHOs) of pharmaceutical care services forminority patients: A literature review. p. 311.Exclusion Code: X2.70. Ching MM, Chen T, Bounthavong M.Medication therapy management programs at aVA medical facility: effectson blood cholesterolin high risk patients. CA J Health-SysPharm.23:15. Exclusion Code: X2.71. Chisholm MA, Mulloy LL, Jagadeesan M, et al.Effect of clinical pharmacy services on the bloodpressure of African-American renal transplantpatients. Ethn Dis. 2002 Summer;12(3):392-7.PMID: 12148711. Exclusion Code: X7.72. Chisholm MA, Reinhardt BO, Vollenweider LJ,et al. Effect of pharmaceutical care services onrenal transplant patients' blood glucose levels.ASHP Midyear Clinical Meeting.1999;34(Dec):P-E. PMID: 36-12509. ExclusionCode: X4.73. Chisholm MA, Spivey CA, Mulloy LL. Effectsof a medication assistance program withmedication therapy management on the health ofC-4


enal transplant recipients. Am J Health SystPharm. 2007 Jul 15;64(14):1506-12. PMID:17617501. Exclusion Code: X5.74. Choe HM, Mitrovich S, Dubay D, et al.Proactive case management of high-risk patientswith type 2 diabetes mellitus by a clinicalpharmacist: a randomized controlled trial. Am JManag Care. 2005 Apr;11(4):253-60. PMID:15839185. Exclusion Code: X2.75. Christensen DB, Campbell WH, Madsen S, et al.Documenting outpatient problem interventionactivities of pharmacists in an HMO. Med Care.1981 Jan;19(1):104-17. PMID: 7464303.Exclusion Code: X4.76. Christensen DB, Holmes G, Fassett WE, et al.Principal findings from the Washington Statecognitive services demonstration project. ManagCare Interface. 1998 Jul;11(7):60-2, 4. PMID:10181572. Exclusion Code: X2.77. Cioffi ST, Caron MF, Kalus JS, et al.Glycosylated hemoglobin, cardiovascular, andrenal outcomes in a pharmacist-managed clinic.Ann Pharmacother. 2004 May;38(5):771-5.PMID: 15031417. Exclusion Code: X4.78. Coast-Senior EA, Kroner BA, Kelley CL, et al.Management of patients with type 2 diabetes bypharmacists in primary care clinics. AnnPharmacother. 1998 Jun;32(6):636-41. PMID:9640480. Exclusion Code: X4.79. Collins C, Kramer A, O'Day ME, et al.Evaluation of patient and provider satisfactionwith a pharmacist-managed lipid clinic in aVeterans Affairs medical center. Am J HealthSyst Pharm. 2006 Sep 15;63(18):1723-7. PMID:16960256. Exclusion Code: X4.80. Cordina M, McElnay JC, Hughes CM.Assessment of a community pharmacy-basedprogram for patients with asthma.Pharmacotherapy. 2001 Oct;21(10):1196-203.PMID: 11601666. Exclusion Code: X2.81. Cote I, Gregoire JP, Moisan J, et al. Apharmacy-based health promotion programme inhypertension: cost-benefit analysis.Pharmacoeconomics. 2003;21(6):415-28. PMID:12678568. Exclusion Code: X2.82. Cotterell CC, Dombroske L, Fischermann EA.Comprehensive drug-use evaluation program in ahealth maintenance organization. Am J HospPharm. 1991 Aug;48(8):1712-7. PMID:1897551. Exclusion Code: X1.83. Couturaud F, Proust A, Frachon I, et al.Education and self-management: a one-yearrandomized trial in stable adult asthmaticpatients. J Asthma. 2002 Sep;39(6):493-500.PMID: 12375708. Exclusion Code: X2.84. Cranor CW, Bunting BA, Christensen DB. TheAsheville Project: long-term clinical andeconomic outcomes of a community pharmacydiabetes care program. J Am Pharm Assoc(Wash). 2003 Mar-Apr;43(2):173-84. PMID:12688435. Exclusion Code: X4.85. Curwain B. Hampshire medicines managementproject saves NHS over British pound sterling1million. Pharm J (England).2007;278(Feb):PM1-PM2. PMID: 44-11008.Exclusion Code: X2.86. Degnin FD. Difficult patients, overmedication,and groupthink. J Clin Ethics. 2009Spring;20(1):64-74. PMID: 19385324. ExclusionCode: X1.87. Dersch D. Patient assessment tool forpharmacists. Am J Health Syst Pharm. 2008 Apr15;65(8):702-3. PMID: 18387897. ExclusionCode: X1.88. Donohue JM, Zhang Y, Aiju M, et al. Impact ofMedicare Part D on antidepressant treatment,medication choice, and adherence among olderadults with depression. Am J Geriatr Psychiatry.2011 Dec;19(12):989-97. PMID: 22123272.Exclusion Code: X2.89. Doucette WR, McDonough RP, Klepser D, et al.Comprehensive medication therapymanagement: identifying and resolving drugrelatedissues in a community pharmacy. ClinTher. 2005 Jul;27(7):1104-11. PMID: 16154490.Exclusion Code: X4.90. Doucette WR, Witry MJ, Alkhateeb F, et al.Attitudes of Medicare beneficiaries towardpharmacist-provided medication therapymanagement activities as part of the MedicarePart D benefit. J Am Pharm Assoc (2003). 2007Nov-Dec;47(6):758-62. PMID: 18032140.Exclusion Code: X4.91. Dougherty L. Medicine management andinfection prevention. Br J Nurs. 2011 Jul 28-Aug4;20(14):S3. PMID: 21841665. Exclusion Code:X1.92. Downie G, Cromarty E, Ellis G, et al.Assessment of medicine management in peopleages 75 years or over living in the community inGrampian, Scotland. ASHP Midyear ClinicalC-5


Meeting. 2001;36(Dec):P-E. PMID: 38-12661.Exclusion Code: X4.93. Dudas V, Bookwalter T, Kerr KM, et al. Theimpact of follow-up telephone calls to patientsafter hospitalization. Dis Mon. 2002Apr;48(4):239-48. PMID: 12021756. ExclusionCode: X2.94. Eden M, Avery A, Rodgers S, et al. Ensuringaccuracy in routinely collected data in a clusterrandomised controlled trial (RCT) comparing theeffectiveness of a pharmacist-led IT-basedintervention with simple feedback in reducingrates of clinically important errors in medicinesmanagement: The PINCER Trial. Int J PharmPract. 2010:60-1. PMID: CN-00790010.Exclusion Code: X1.95. Elkjaer M. E-health: Web-guided therapy anddisease self-management in ulcerative colitis.Impact on disease outcome, quality of life andcompliance. Dan Med J. 2012 Jul;59(7):B4478.PMID: 22759851. Exclusion Code: X2.96. Elliott RA, Martinac G, Campbell S, et al.Pharmacist-led medication review to identifymedication-related problems in older peoplereferred to an Aged Care Assessment Team: arandomized comparative study. Drugs Aging.2012 Jul 1;29(7):593-605. PMID: 22715865.Exclusion Code: X2.97. Emmerton L, Shaw J, Kheir N. Asthmamanagement by New Zealand pharmacists: apharmaceutical care demonstration project. JClin Pharm Ther. 2003 Oct;28(5):395-402.PMID: 14632964. Exclusion Code: X2.98. Epplen K, Dusing-Wiest M, Freedlund J, et al.Stepwise approach to implementing ambulatoryclinical pharmacy services. Am J Health SystPharm. 2007 May 1;64(9):945-51. PMID:17468149. Exclusion Code: X2.99. Epplen K, Marsden LM, Murphy C.Reimbursement for medication therapymanagement services. Am J Health Syst Pharm.2008 May 15;65(10):906-8. PMID: 18463337.Exclusion Code: X1.100. Ermer M, Harder S. Medication review ofcommunity-dwelling seniors using intensifiedhome-care service. Int J Clin Pharmacol Ther.2011 Mar;49(3):179-84. PMID: 21329619.Exclusion Code: X2.101. Farris KB, Ganther-Urmie JM, Fang G, et al.Population-based medication reviews: adescriptive analysis of the medication issuesidentified in a medicare not-for-profitprescription discount program. AnnPharmacother. 2004 Nov;38(11):1823-9. PMID:15479770. Exclusion Code: X2.102. Farris KB, Kumbera P, Halterman T, et al.Outcomes-based pharmacist reimbursement:Reimbursing pharmacists for cognitive services -(Part 1 of a 2-part series). J Manag Care Pharm.2002;8(May):383-93. PMID: 40-02496.Exclusion Code: X4.103. Feldman PH, Totten AM, Foust J, et al.Medication management: evidence brief. Centerfor Home Care Policy & Research. HomeHealthc Nurse. 2009 Jun;27(6):379-86. PMID:19509525. Exclusion Code: X1.104. Feldman SM. Pharmaceutical care and theconsultant pharmacist. A model for drug therapymanagement. Am Pharm. 1995;Suppl:19-21.PMID: 7677061. Exclusion Code: X1.105. Fera T, Bluml BM, Ellis WM, et al. TheDiabetes Ten City Challenge: interim clinicaland humanistic outcomes of a multisitecommunity pharmacy diabetes care program. JAm Pharm Assoc (2003). 2008 Mar-Apr;48(2):181-90. PMID: 18359731. ExclusionCode: X2.106. Figge HL. Interoperable health informationexchange between medication therapymanagement services and the medical home. AmJ Health Syst Pharm. 2010 Feb 1;67(3):190-1.PMID: 20101060. Exclusion Code: X1.107. Finley PR, Bluml BM, Bunting BA, et al.Clinical and economic outcomes of a pilotproject examining pharmacist-focusedcollaborative care treatment for depression. J AmPharm Assoc (2003). 2011 Jan-Feb;51(1):40-9.PMID: 21247825. Exclusion Code: X2.108. Finley PR, Rens HR, Pont JT, et al. Impact of acollaborative care model on depression in aprimary care setting: a randomized controlledtrial. Pharmacotherapy. 2003 Sep;23(9):1175-85.PMID: 14524649. Exclusion Code: X2.109. Fisher JE, Zhang Y, Sketris I, et al. The effect ofan educational intervention on meperidine use inNova Scotia, Canada: a time series analysis.Pharmacoepidemiol Drug Saf. 2012Feb;21(2):177-83. PMID: 22081471. ExclusionCode: X2.110. Fletcher J, Hogg W, Farrell B, et al. Effect ofnurse practitioner and pharmacist counseling oninappropriate medication use in family practice.C-6


Can Fam Physician. 2012 Aug;58(8):862-8.PMID: 22893340. Exclusion Code: X2.111. Forster AJ, Auger C. Using informationtechnology to improve the monitoring ofoutpatient prescribing. JAMA Intern Med. 2013Mar 11;173(5):382-4. PMID: 23381469.Exclusion Code: X2.112. Frew EJ, Kluettgens B, Gilday N, et al. Theimpact of a medicines management service onpatients with cystic fibrosis. p. 686. ExclusionCode: X4.113. Gadzhanova SV, Roughead EE, Bartlett MJ.Improving cardiovascular disease managementin Australia: NPS MedicineWise. Med J Aust.2013 Aug 5;199(3):192-5. PMID: 23909542.Exclusion Code: X5.114. Garcao JA, Cabrita J. Evaluation of apharmaceutical care program for hypertensivepatients in rural Portugal. J Am Pharm Assoc(Wash). 2002 Nov-Dec;42(6):858-64. PMID:12482009. Exclusion Code: X2.115. Gardella JE, Cardwell TB, Nnadi M. Improvingmedication safety with accurate preadmissionmedication lists and postdischarge education. JtComm J Qual Patient Saf. 2012 Oct;38(10):452-8. PMID: 23130391. Exclusion Code: X2.116. Garrett DG, Bluml BM. Patient self-managementprogram for diabetes: first-year clinical,humanistic, and economic outcomes. J AmPharm Assoc (2003). 2005 Mar-Apr;45(2):130-7. PMID: 15868754. Exclusion Code: X2.117. Garrett DG, Bluml BM. Patient self-managementprogram for diabetes. J Pharm Soc Wis.2005(C):17-27. PMID: 43-16326. ExclusionCode: X4.118. Geletko SM, Poulakos MN. Pharmaceuticalservices in an HIV clinic. Am J Health SystPharm. 2002 Apr 15;59(8):709-13. PMID:11977854. Exclusion Code: X4.119. Gerber RA, Liu G, McCombs JS. Impact ofpharmacist consultations provided to patientswith diabetes on healthcare costs in a healthmaintenance organization. Am J Manag Care.1998 Jul;4(7):991-1000. PMID: 10181997.Exclusion Code: X2.120. Gilbert AL, Roughead EE, Beilby J, et al.Collaborative medication management services:improving patient care. Med J Aust. 2002 Aug19;177(4):189-92. PMID: 12175322. ExclusionCode: X4.121. Gillespie U, Alassaad A, Hammarlund-UdenaesM, et al. Effects of pharmacists' interventions onappropriateness of prescribing and evaluation ofthe instruments' (MAI, STOPP and STARTs')ability to predict hospitalization--analyses from arandomized controlled trial. PLoS One.2013;8(5):e62401. PMID: 23690938. ExclusionCode: X7.122. Gizzi L, Slain D, Hare J, et al. An assessment ofa pharmacy-based enhancement to the hospitalmedication reconciliation process. ExclusionCode: X2.123. Goode JV, Swiger K, Bluml BM. Regionalosteoporosis screening, referral, and monitoringprogram in community pharmacies: findingsfrom Project ImPACT: Osteoporosis. J AmPharm Assoc (2003). 2004 Mar-Apr;44(2):152-60. PMID: 15098849. Exclusion Code: X2.124. Gourley GA, Portner TS, Gourley DR, et al.Humanistic outcomes in the hypertension andCOPD arms of a multicenter outcomes study. JAm Pharm Assoc (Wash). 1998(5):586-97.PMID: CN-00155944. Exclusion Code: X2.125. Grantham D. Idaho behavioral health optimizesmed management visits using telehealth. BehavHealthc. 2011 Oct;31(7):46-7. PMID: 22117294.Exclusion Code: X1.126. Griffiths M. Medicines management. NursStand. 2008 Oct 1-7;23(4):28. PMID: 19051531.Exclusion Code: X1.127. Griffiths M. Medicines management. NursStand. 2009 Sep 23-29;24(3):28. PMID:19856642. Exclusion Code: X1.128. Griffiths M. Medicines management. NursStand. 2010 Nov 17-23;25(11):28. PMID:21189818. Exclusion Code: X1.129. Griffiths M. Medicines management. Clearcriteria are needed for any dose adjustment. NursStand. 2010 Jan 13-19;24(19):28. PMID:20175358. Exclusion Code: X1.130. Gruffydd-Jones K, Hollinghurst S, Ward S, et al.Targeted routine asthma care in general practiceusing telephone triage (Structured abstract). Br JGen Pract. 2005(1):918-23. PMID: NHSEED-22006008031. Exclusion Code: X2.131. Guirguis AB, Zingone MM. Medication therapymanagement and cost-related medicationnonadherence. JAMA. 2008 Sep10;300(10):1130; author reply -1. PMID:18780839. Exclusion Code: X1.C-7


132. Haines LA, Putney KS, Varkey DA, et al. Pilotof a patient-centered pharmacy practice model.Am J Health Syst Pharm. 2012 Nov1;69(21):1860-1. PMID: 23111668. ExclusionCode: X5.133. Hall D, Buchanan J, Helms B, et al. Health careexpenditures and therapeutic outcomes of apharmacist-managed anticoagulation serviceversus usual medical care. Pharmacotherapy.2011 Jul;31(7):686-94. PMID: 21923456.Exclusion Code: X2.134. Hall DL, Pater KS. Implementation of aMedication Therapy Management Program in aHospital-Based Outpatient Pharmacy. p. 512.Exclusion Code: X4.135. Hallam A. Does peer review improve patientcare and documentation in medication therapymanagement? : Kaiser Permanente. p. 1-33.Exclusion Code: X2.136. Hardin HC, Hall AM, Roane TE, et al. AnAdvanced Pharmacy Practice Experience in aStudent-Staffed Medication TherapyManagement Call Center. p. NIL. ExclusionCode: X4.137. Harris A, Gospodarevskaya E, Callaghan J, et al.The cost effectiveness of a pharmacist reviewingmedication among the elderly in the community.Australas J Ageing. 2001;20(4):179-86.Exclusion Code: X2.138. Harris IM, Westberg SM, Frakes MJ, et al.Outcomes of medication therapy review in afamily medicine clinic. J Am Pharm Assoc(2003). 2009 Sep-Oct;49(5):623-7. PMID:19748869. Exclusion Code: X4.139. Harrison JJ, Wang J, Cervenko J, et al. Pilotstudy of a pharmaceutical care intervention in anoutpatient lung transplant clinic. Clin Transplant.2012 Mar-Apr;26(2):E149-57. PMID: 22507355.Exclusion Code: X7.140. Hassol A, Shoemaker SJ. Exploratory Researchon Medication Therapy Management. (Preparedby Abt Associates under Contract #HHSM-500-2005-00181/TO#3.) Baltimore, MD: Centers forMedicare & Medicaid Services; July 8, 2008.http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Reports/downloads/blackwell.pdf.Exclusion Code: X4.141. Hata M, Klotz R, Sylvies R, et al. Medicationtherapy management services provided bystudent pharmacists. Am J Pharm Educ. 2012Apr 10;76(3):51. PMID: 22544968. ExclusionCode: X4.142. Hellstrom LM, Bondesson A, Hoglund P, et al.Impact of the Lund Integrated MedicinesManagement (LIMM) model on medicationappropriateness and drug-related hospitalrevisits. Eur J Clin Pharmacol. 2011Jul;67(7):741-52. PMID: 21318595. ExclusionCode: X2.143. Hemens BJ, Holbrook A, Tonkin M, et al.Computerized clinical decision support systemsfor drug prescribing and management: adecision-maker-researcher partnershipsystematic review. Implement Sci. 2011;6:89.PMID: 21824383. Exclusion Code: X2.144. Hilario Deguzman LF. Intervening On HospitalDischarge Patients: Results from a Medicare PartD MTM program. Kaiser Permanente; 2013. p.1-32. Exclusion Code: X2.145. Hogan EG, Leal S, Slack M, et al. Comparisonof pharmacist led collaborative drug therapymanagement to standard physician providedtherapy for type 2 diabetes mellitus. ASHPMidyear Clinical Meeting. 2005;40:P518E.PMID: 43-01390. Exclusion Code: X2.146. Holloway KA, Ivanovska V, Wagner AK, et al.Have we improved use of medicines indeveloping and transitional countries and do weknow how to? Two decades of evidence. TropMed Int Health. 2013 Jun;18(6):656-64. PMID:23648177. Exclusion Code: X2.147. Holtorf AP, McAdam-Marx C, Schaaf D, et al.Systematic review on quality control for drugmanagement programs: is quality reported in theliterature? BMC Health Serv Res. 2009;9:38.PMID: 19243591. Exclusion Code: X2.148. Hong N, Jackson L, Pisupati R. Evaluation ofclinical pharmacy services in an ambulatory careHIV/AIDS Clinic. p. 056. Exclusion Code: X4.149. Howard-Thompson A, Farland MZ, Byrd DC, etal. Pharmacist-Physician Collaboration forDiabetes Care: Cardiovascular Outcomes. AnnPharmacother.47:1471. Exclusion Code: X2.150. Humphries TL, Carroll N, Chester EA, et al.Evaluation of an electronic critical druginteraction program coupled with activepharmacist intervention. Ann Pharmacother.2007 Dec;41(12):1979-85. PMID: 17986517.Exclusion Code: X2.151. Hunteman LM. Potential role of medicationtherapy management for bariatric surgeryC-8


patients. J Am Pharm Assoc (2003). 2008 Jul-Aug;48(4):440-2. PMID: 18653412. ExclusionCode: X1.152. Hussein M, Benner JS, Lee D, et al. Propensityscore matching in the evaluation of drug therapymanagement programs: an illustrative analysis ofa program for patients with hepatitis C virus.Qual Manag Health Care. 2010 Jan-Mar;19(1):25-33. PMID: 20042931. ExclusionCode: X2.153. Isetts BJ. Evaluating Effectiveness of theMinnesota Medication Therapy ManagementCare Program: Final Report. University ofMinnesota College of Pharmacy. Minneapolis,MN: 2007. Exclusion Code: X4.154. Isetts BJ, Brown LM, Schondelmeyer SW, et al.Quality assessment of a collaborative approachfor decreasing drug-related morbidity andachieving therapeutic goals. Arch Intern Med.2003 Aug 11-25;163(15):1813-20. PMID:12912717. Exclusion Code: X4.155. Isetts BJ, Brummel AR, de Oliveira DR, et al.Managing drug-related morbidity and mortalityin the patient-centered medical home. Med Care.2012 Nov;50(11):997-1001. PMID: 23047790.Exclusion Code: X2.156. Isetts BJ, Schondelmeyer SW, Heaton AH, et al.Effects of collaborative drug therapymanagement on patients' perceptions of care andhealth-related quality of life. Res Social AdmPharm. 2006 Mar;2(1):129-42. PMID:17138505. Exclusion Code: X2.157. Jaber LA, Halapy H, Fernet M, et al. Evaluationof a pharmaceutical care model on diabetesmanagement. Ann Pharmacother. 1996Mar;30(3):238-43. PMID: 8833557. ExclusionCode: X2.158. Jack BW, Chetty VK, Anthony D, et al. Areengineered hospital discharge program todecrease rehospitalization: a randomized trial.Ann Intern Med. 2009(3):178-87. PMID: CN-00666837. Exclusion Code: X2.159. Jackevicius CA, Li P, Tu JV. Prevalence,predictors, and outcomes of primarynonadherence after acute myocardial infarction.Circulation. 2008 Feb 26;117(8):1028-36.PMID: 18299512. Exclusion Code: X2.160. Jackson GL, Oddone EZ, Olsen MK, et al.Racial differences in the effect of a telephonedeliveredhypertension disease managementprogram. J Gen Intern Med. 2012Dec;27(12):1682-9. PMID: 22865016. ExclusionCode: X2.161. Jameson JP, VanNoord GR. Pharmacotherapyconsultation on polypharmacy patients inambulatory care. Ann Pharmacother. 2001 Jul-Aug;35(7-8):835-40. PMID: 11485129.Exclusion Code: X2.162. Jayaprakash A. ASCP reports. Consult Pharm.2010 Feb;25(2):128. PMID: 20211826.Exclusion Code: X1.163. Johnson CL, Nicholas A, Divine H, et al.Outcomes from DiabetesCARE: a pharmacistprovideddiabetes management service. J AmPharm Assoc (2003). 2008 Nov-Dec;48(6):722-30. PMID: 19019800. Exclusion Code: X4.164. Johnson JA, Bootman JL. Drug-relatedmorbidity and mortality and the economicimpact of pharmaceutical care. Am J Health SystPharm. 1997 Mar 1;54(5):554-8. PMID:9066865. Exclusion Code: X2.165. Johnston AM, Doane K, Phipps S, et al.Outcomes of pharmacists' cognitive services inthe long-term care setting. Consult Pharm.1996;11:41-50. Exclusion Code: X4.166. Kaboli PJ, Fernandes O. Medicationreconciliation: moving forward. Arch InternMed. 2012 Jul 23;172(14):1069-70. PMID:22733283. Exclusion Code: X1.167. Karapinar-Carkit F, Borgsteede SD, Zoer J, et al.The effect of the COACH program (ContinuityOf Appropriate pharmacotherapy, patientCounselling and information transfer inHealthcare) on readmission rates in amulticultural population of internal medicinepatients. BMC Health Serv Res. 2010;10:39.PMID: 20156368. Exclusion Code: X7.168. Kimberlin CL, Berardo DH, Pendergast JF, et al.Effects of an education program for communitypharmacists on detecting drug-related problemsin elderly patients. Med Care. 1993May;31(5):451-68. PMID: 8501992. ExclusionCode: X2.169. Kliethermes MA. Adherence in an MTM clinic.2009. p. 204. Exclusion Code: X8.170. Klimek JS. MTM services standards improvepatient safety. Medication therapy managementsafeguards, systems and services can reduce thetotal number of medication-relatedcomplications. Health Manag Technol. 2009Jul;30(7):20-1. PMID: 19739561. ExclusionCode: X1.C-9


171. Kocarnik BM, Liu CF, Wong ES, et al. Does thepresence of a pharmacist in primary care clinicsimprove diabetes medication adherence? BMCHealth Serv Res. 2012;12:391. PMID:23148570. Exclusion Code: X2.172. Krass I, Taylor SJ, Smith C, et al. Impact onmedication use and adherence of Australianpharmacists' diabetes care services. J Am PharmAssoc (2003). 2005 Jan-Feb;45(1):33-40. PMID:15730115. Exclusion Code: X2.173. Kripalani S, Roumie CL, Dalal AK, et al. Effectof a pharmacist intervention on clinicallyimportant medication errors after hospitaldischarge: a randomized trial. Ann Intern Med.2012 Jul 3;157(1):1-10. PMID: 22751755.Exclusion Code: X7.174. Kucukarslan SN, Hagan AM, Shimp LA, et al.Integrating medication therapy management inthe primary care medical home: A review ofrandomized controlled trials. Am J Health SystPharm. 2011 Feb 15;68(4):335-45. PMID:21289329. Exclusion Code: X6.175. Kumbera P. A new addition to pharmacy'salphabet soup: MTMS. Pharmacy Times (USA).2005;71(Mar):90. PMID: 43-01425. ExclusionCode: X1.176. Kuzuya M, Hirakawa Y, Suzuki Y, et al.Association between unmet needs for medicationsupport and all-cause hospitalization incommunity-dwelling disabled elderly people. JAm Geriatr Soc. 2008 May;56(5):881-6. PMID:18384585. Exclusion Code: X2.177. Kwint HF, Bermingham L, Faber A, et al. Therelationship between the extent of collaborationof general practitioners and pharmacists and theimplementation of recommendations arisingfrom medication review: a systematic review.Drugs Aging. 2013 Feb;30(2):91-102. PMID:23322285. Exclusion Code: X2.178. Lai LL, Sorkin AL. Cost benefit analysis ofpharmaceutical care in a Medicaidpopulation\M/from a budgetary perspective. JManag Care Pharm. 1998;4(May-Jun):303-8.PMID: 36-06642. Exclusion Code: X6.179. Lai PS, Chua SS, Chan SP. Pharmaceutical careissues encountered by post-menopausalosteoporotic women prescribed bisphosphonates.J Clin Pharm Ther. 2012;37(5):536-43. PMID:CN-00881017. Exclusion Code: X2.180. Lam A. Practice innovations: Deliveringmedication therapy management services viavideoconference interviews. Consultantpharmacist. 2011(10):764-73. PMID: CN-00833961. Exclusion Code: X4.181. Lam A, Odegard PS, Gardner J. School ofpharmacy-based medication therapymanagement program: development and initialexperience. J Am Pharm Assoc (2003). 2012May-Jun;52(3):398-404. PMID: 22618981.Exclusion Code: X4.182. Lam S, Ruby CM. Impact of an interdisciplinaryteam on drug therapy outcomes in a geriatricclinic. Am J Health Syst Pharm. 2005 Mar15;62(6):626-9. PMID: 15757885. ExclusionCode: X2.183. LaMarr B, Valdez C, Driscoll K, et al. Influenceof pharmacist intervention on prescribing ofangiotensin-converting-enzyme inhibitors,angiotensin II-receptor blockers, and aspirin fordiabetic patients. Am J Health Syst Pharm. 2010Feb 15;67(4):290-4. PMID: 20133534.Exclusion Code: X2.184. Lambert-Kerzner A, Del Giacco EJ, Fahdi IE, etal. Patient-centered adherence intervention afteracute coronary syndrome hospitalization. CircCardiovasc Qual Outcomes. 2012 Jul 1;5(4):571-6. PMID: 22811499. Exclusion Code: X6.185. Lapane KL, Hughes CM. Pharmacotherapyinterventions undertaken by pharmacists in theFleetwood phase III study: the role of processcontrol. Ann Pharmacother. 2006Sep;40(9):1522-6. PMID: 16882872. ExclusionCode: X5.186. Lapane KL, Hughes CM, Christian JB, et al.Evaluation of the fleetwood model of long-termcare pharmacy. J Am Med Dir Assoc. 2011Jun;12(5):355-63. PMID: 21450170. ExclusionCode: X2.187. Lau DT. Consumer medication management anderror. Clin Ther. 2008 Nov;30(11):2156-8.PMID: 19108803. Exclusion Code: X1.188. Ledwidge M, Barry M, Cahill J, et al. Ismultidisciplinary care of heart failure costbeneficialwhen combined with optimal medicalcare? (Structured abstract). Eur J Heart Fail.2003(3):381-9. PMID: NHSEED-22003000934.Exclusion Code: X2.189. Lee E, Braund R, Tordoff J. Examining the firstyear of Medicines Use Review services providedby pharmacists in New Zealand: 2008. N Z MedJ. 2009 Apr 24;122(1293):3566. PMID:19448788. Exclusion Code: X6.C-10


190. Lee JK, Grace KA, Taylor AJ. Effect of apharmacy care program on medication adherenceand persistence, blood pressure, and low-densitylipoprotein cholesterol: a randomized controlledtrial. JAMA. 2006 Dec 6;296(21):2563-71.PMID: 17101639. Exclusion Code: X2.191. Lee KL, Peng YL, Chou JL, et al. Economicevaluation of therapeutic drug monitoringservices in renal transplant recipients treatedwith cyclosporine (Structured abstract).Transplant Proc. 2000(7):1801-6. PMID:NHSEED-22001000188. Exclusion Code: X3.192. Lee M, Kemp JA, Canning A, et al. Arandomized controlled trial of an enhancedpatient compliance program for Helicobacterpylori therapy. Arch Intern Med. 1999 Oct25;159(19):2312-6. PMID: 10547171. ExclusionCode: X3.193. Leikola SN, Virolainen J, Tuomainen L, et al.Comprehensive medication reviews for elderlypatients: findings and recommendations tophysicians. J Am Pharm Assoc (2003). 2012Sep-Oct;52(5):630-3. PMID: 23023843.Exclusion Code: X4.194. Lemay G. Medication therapy management incommunity pharmacy practice. Med Health R I.2012 Sep;95(9):281-2. PMID: 23094424.Exclusion Code: X1.195. Li X, Mao M, Ping Q. Effect of pharmaceuticalcare programs on glycemic control in patientswith diabetes mellitus: a meta-analysis ofrandomized controlled trials (Structuredabstract). J Pharm Technol. 2010(5):255-63.PMID: DARE-12011001815. Exclusion Code:X2.196. Liang C. The Impact of Post-DischargeMedication Therapy Management (MTM)Encounters on Readmission Rate. KaiserPermanente. Exclusion Code: X2.197. Libby EA, Laub JJ. Economic and clinicalimpact of a pharmacy-based antihypertensivereplacement program in primary care. Am JHealth Syst Pharm. 1997 Sep 15;54(18):2079-83. PMID: 9377207. Exclusion Code: X2.198. Lin KE, Lei EX, Chretien SD. Pharmacotherapyassessment and education for newly enrolledpatients in a primary care clinic. ASHP MidyearClinical Meeting. 2001;36(Dec):P-E. PMID: 38-12877. Exclusion Code: X3.199. Lingaratnam S, Aranda S, Pearce T, et al. Acontrolled before and after study to evaluate apatient and health professional partnership modeltowards effective medication reconciliation. JOncol Pharm Pract. 2013 Mar;19(1):48-56.PMID: 22849967. Exclusion Code: X4.200. Linton A, Bacon TA, Trice S, et al. Results froma mailed promotion of medication reviewsamong Department of Defense beneficiariesreceiving 10 or more chronic medications. JManag Care Pharm. 2010 Oct;16(8):578-92.PMID: 20866163. Exclusion Code: X2.201. Lipton HL, Bero LA, Bird JA, et al. The impactof clinical pharmacists' consultations onphysicians' geriatric drug prescribing. Arandomized controlled trial. Med Care. 1992Jul;30(7):646-58. PMID: 1614233. ExclusionCode: X2.202. Lobeck F, Traxler WT, Bobinet DD. The costeffectivenessof a clinical pharmacy service in anoutpatient mental health clinic. Hosp CommunityPsychiatry. 1989 Jun;40(6):643-5. PMID:2500390. Exclusion Code: X2.203. Lu Z, Cao S, Chai Y, et al. Effectiveness ofinterventions for hypertension care in thecommunity: a meta-analysis of controlled studiesin China (Provisional abstract). DARE.2012(4)PMID: DARE-12013002399. ExclusionCode: X2.204. Ma CB, Nakasato SK. Effectiveness of aneducational program to improve patientunderstanding of and compliance with drugtherapies used to treat HIV disease. ASHPMidyear Clinical Meeting. 1991;26(Dec):P-R.PMID: 29-00124. Exclusion Code: X2.205. Maack B, Miller DR, Johnson T, et al. Economicimpact of a pharmacy resident in an assistedliving facility-based medication therapymanagement program. Ann Pharmacother. 2008Nov;42(11):1613-20. PMID: 18940921.Exclusion Code: X4.206. Malaeb DN, Aoun J. Physician communitypharmacist collaborative care in diabetesmanagement.2011:P. Exclusion Code: X2.207. Martinez AS, Saef J, Paszczuk A, et al.Implementation of a pharmacist-managed heartfailure medication titration clinic. Am J HealthSyst Pharm. 2013 Jun 15;70(12):1070-6. PMID:23719886. Exclusion Code: X4.208. McAdam-Marx C, Schaaf DT, Holtorf AP, et al.Systematic analysis of outcomes evaluationsapplied to drug management programs. Am JC-11


Manag Care. 2008 Nov;14(11 Suppl):SP36-45.PMID: 18991480. Exclusion Code: X2.209. McAllister D, Schommer JC, McAuley JW, et al.Comparison of skilled nursing and assisted livingresidents to determine potential benefits ofpharmacist intervention. Consultant Pharmacist(USA). 2000;15(Nov):1110-6. PMID: 38-05993.Exclusion Code: X2.210. McCall N, Cromwell J, Urato C. Evaluation ofMedicare Care Management for High CostBeneficiaries (CMHCB) Demonstration:VillageHealth's Key to Better Health (KTBH).(Prepared by RTI International under CMSContract No. #500-00-0024/TO#25). Baltimore,MD: Centers for Medicare & Medicaid Services;2010. Exclusion Code: X2.211. McCombs JS, Liu G, Shi J, et al. The KaiserPermanente/USC Patient Consultation Study:change in use and cost of health care services.Am J Health Syst Pharm. 1998 Dec1;55(23):2485-99. PMID: 9853633. ExclusionCode: X2.212. McCord AD. Clinical impact of a pharmacistmanageddiabetes mellitus drug therapymanagement service. Pharmacotherapy. 2006Feb;26(2):248-53. PMID: 16466329. ExclusionCode: X2.213. McDonough RP. Medication therapymanagement and collaborative practice: How tomake it work. Drug Topics.155:52. ExclusionCode: X1.214. McDonough RP, Doucette WR. Drug therapymanagement: an empirical report of drug therapyproblems, pharmacists' interventions, and resultsof pharmacists' actions. J Am Pharm Assoc(2003). 2003 Jul-Aug;43(4):511-8. PMID:12952316. Exclusion Code: X2.215. McFarland M, Davis K, Wallace J, et al. Use ofhome telehealth monitoring with activemedication therapy management by clinicalpharmacists in veterans with poorly controlledtype 2 diabetes mellitus. Pharmacotherapy. 2012May;32(5):420-6. PMID: 22488512. ExclusionCode: X2.216. McIntosh K. Medicines management. NursTimes. 2008 Nov 4-10;104(44):18-20. PMID:19014127. Exclusion Code: X1.217. McKibbon KA, Lokker C, Handler SM, et al.The effectiveness of integrated healthinformation technologies across the phases ofmedication management: a systematic review ofrandomized controlled trials. J Am Med InformAssoc. 2012 Jan-Feb;19(1):22-30. PMID:21852412. Exclusion Code: X2.218. McLean W, Gillis J, Waller R. The BCCommunity Pharmacy Asthma Study: A study ofclinical, economic and holistic outcomesinfluenced by an asthma care protocol providedby specially trained community pharmacists inBritish Columbia. Can Respir J. 2003 May-Jun;10(4):195-202. PMID: 12851665. ExclusionCode: X2.219. McMahan R. Medication therapy managementservices: assessing year 1, gauging year 2. J AmPharm Assoc (2003). 2006 Nov-Dec;46(6):680-1. PMID: 17176682. Exclusion Code: X1.220. Meindl CL, Duncan AL, Achusim LE, et al.Impact of a clinical pharmacist in an internalmedicine resident clinic. ASHP Midyear ClinicalMeeting. 1998;33(Dec):P-R. PMID: 35-12686.Exclusion Code: X4.221. Michaels NM, Jenkins GF, Pruss DL, et al.Retrospective analysis of communitypharmacists' recommendations in the NorthCarolina Medicaid medication therapymanagement program. J Am Pharm Assoc(2003). 2010 May-Jun;50(3):347-53. PMID:20452907. Exclusion Code: X4.222. Michel B, Quelennec B, Andres E. Medicationreconciliation practices and potential clinicalimpact of unintentional discrepancies. JAMAIntern Med. 2013 Feb 11;173(3):246-7. PMID:23400661. Exclusion Code: X2.223. Midlov P, Bahrani L, Seyfali M, et al. The effectof medication reconciliation in elderly patients athospital discharge. Int J Clin Pharm. 2012Feb;34(1):113-9. PMID: 22207271. ExclusionCode: X2.224. Monte SV, Slazak EM, Albanese NP, et al.Clinical and economic impact of a diabetesclinical pharmacy service program in auniversity and primary care-based collaborationmodel. J Am Pharm Assoc (2003). 2009 Mar-Apr;49(2):200-8. PMID: 19289346. ExclusionCode: X4.225. Mort J. Medication therapy management: aMedicare Part D benefit available to qualifyingbeneficiaries. S D Med. 2007 Mar;60(3):121.PMID: 17455583. Exclusion Code: X1.226. Mortimer C, Emmerton L, Lum E. The impact ofan aged care pharmacist in a department ofemergency medicine. J Eval Clin Pract. 2011C-12


Jun;17(3):478-85. PMID: 21040247. ExclusionCode: X3.227. Motz J. The implementation of MTM incommunity pharmacies - Communication andplanning are key factors for success. J Pharm SocWis. 2007(SEP-OCT):11-2. PMID: 45-01241.Exclusion Code: X1.228. Mueller SK, Kripalani S, Stein J, et al. A toolkitto disseminate best practices in inpatientmedication reconciliation: multi-centermedication reconciliation quality improvementstudy (MARQUIS). Jt Comm J Qual Patient Saf.2013 Aug;39(8):371-82. PMID: 23991510.Exclusion Code: X2.229. Mueller SK, Sponsler KC, Kripalani S, et al.Hospital-based medication reconciliationpractices: a systematic review. Arch Intern Med.2012 Jul 23;172(14):1057-69. PMID: 22733210.Exclusion Code: X2.230. Murphy P, Cocohoba J, Tang A, et al. Impact ofHIV-specialized pharmacies on adherence andpersistence with antiretroviral therapy. AidsPatient Care STDS. 2012 Sep;26(9):526-31.PMID: 22860900. Exclusion Code: X2.231. Murray MD, Young J, Hoke S, et al. Pharmacistintervention to improve medication adherence inheart failure: a randomized trial. Ann InternMed. 2007 May 15;146(10):714-25. PMID:17502632. Exclusion Code: X2.232. Nau DP, Pacholski AM. Impact of pharmacycare services on patients' perceptions of healthcare quality for diabetes. J Am Pharm Assoc(2003). 2007 May-Jun;47(3):358-65. PMID:17510030. Exclusion Code: X4.233. Nguyen J. Evaluating the Medication TherapyManagement Program Impact on Hospital andED Visits. Kaiser Permanente. p. 1-34.Exclusion Code: X7.234. Nguyen J, Matsuoka B, Morodomi L, et al.Evaluating the Medication Therapy ManagementProgram impact. CA J Health-Sys Pharm.2011;23:11. Exclusion Code: X8.235. Nkansah N, Mostovetsky O, Yu C, et al. Effectof outpatient pharmacists' non-dispensing roleson patient outcomes and prescribing patterns.Cochrane Database Syst Rev.2010(7):CD000336. PMID: 20614422.Exclusion Code: X6.236. Nola KM, Gourley DR, Portner TS, et al.Clinical and humanistic outcomes of a lipidmanagement program in the communitypharmacy setting. J Am Pharm Assoc (Wash).2000(2):166-73. PMID: CN-00276497.Exclusion Code: X2.237. Osterberg L, Blaschke T. Adherence tomedication. N Engl J Med. 2005 Aug4;353(5):487-97. PMID: 16079372. ExclusionCode: X1.238. Oyetayo OO, James C, Martinez A, et al. TheHispanic Diabetes Management Program: Impactof community pharmacists on clinical outcomes.J Am Pharm Assoc (2003). 2011 Sep-Oct;51(5):623-6. PMID: 21896461. ExclusionCode: X4.239. Packard K, Herink M, Kuhlman P. Pharmacist'srole in an interdisciplinary cardiac rehabilitationteam. J Allied Health. 2012 Fall;41(3):113-7, 7a,7b. PMID: 22968772. Exclusion Code: X4.240. Padiyara RS, Rabi SM. Physician perceptions ofpharmacist provision of outpatient medicationtherapy management services. J Am PharmAssoc (2003). 2006 Nov-Dec;46(6):660, 3.PMID: 17176679. Exclusion Code: X4.241. Pakes GE. Group medication counselingconducted by a pharmacist for severely disturbedclients. Hosp Community Psychiatry. 1979Apr;30(4):237-8. PMID: 422130. ExclusionCode: X2.242. Parks PM. A patient medication counselinggroup in an adult psychiatric facility. HospPharm. 1977 Feb;12(2):63-4. PMID: 10305432.Exclusion Code: X4.243. Patel P, Roberts S, Cooper S, et al. Evaluation ofMedication Therapy Management Program in aHealth Maintenance Organization. KaiserPermanente. p. 1-13. Exclusion Code: X4.244. Patterson SM, Hughes C, Kerse N, et al.Interventions to improve the appropriate use ofpolypharmacy for older people. CochraneDatabase Syst Rev. 2012;5:CD008165. PMID:22592727. Exclusion Code: X2.245. Pecora PG, Flint NB, Aleman MA. Improvingdisease control in patients with hypertensionand/or diabetes mellitus with the implementationof ambulatory clinical pharmacy services. ASHPMidyear Clinical Meeting. 1998;33(Dec):P-E.PMID: 35-13312. Exclusion Code: X2.246. Perera PN, Guy MC, Sweaney AM, et al.Evaluation of prescriber responses to pharmacistrecommendations communicated by fax in amedication therapy management program(MTMP). J Manag Care Pharm. 2011C-13


Jun;17(5):345-54. PMID: 21657804. ExclusionCode: X4.247. Persell SD, Eder M, Friesema E, et al. EHRbasedmedication support and nurse-ledmedication therapy management: rationale anddesign for a three-arm clinic randomized trial. JAm Heart Assoc. 2013 Oct;2(5):e000311.PMID: 24157649. Exclusion Code: X6.248. Petty DR, Zermansky AG, Raynor DK, et al.Clinical medication review by a pharmacist ofelderly patients on repeat medications in generalpractice - Pharmacist interventions and reviewoutcomes. Int J Pharm Pract. 2002(1):39-45.PMID: CN-00443126. Exclusion Code: X5.249. Pickard AS, Johnson JA, Farris KB. The impactof pharmacist interventions on health-relatedquality of life. Ann Pharmacother. 1999Nov;33(11):1167-72. PMID: 10573313.Exclusion Code: X2.250. Pinto SL, Bechtol RA, Partha G. Evaluation ofoutcomes of a medication therapy managementprogram for patients with diabetes. p. 519.Exclusion Code: X4.251. Pinto SL, Kumar J, Partha G, et al. Pharmacist-Provided Medication Therapy Management(MTM) Program Impacts Outcomes forEmployees With Diabetes. Popul Health Manag.2013 Jul 12PMID: 23848476. Exclusion Code:X4.252. Pinto SL, Kumar J, Partha G, et al. Improvingthe economic and humanistic outcomes fordiabetic patients: making a case for employersponsoredmedication therapy management.Clinicoecon Outcomes Res. 2013;5:153-9.PMID: 23610526. Exclusion Code: X4.253. Ponnusankar S, Surulivelrajan M,Anandamoorthy N, et al. Assessment of impactof medication counseling on patients' medicationknowledge and compliance in an outpatientclinic in South India. Patient Educ Couns. 2004Jul;54(1):55-60. PMID: 15210260. ExclusionCode: X2.254. Powers MF. Recently approved medicareprovisions for prescription drugs and medicationmanagement services. J Pharm Tech.2005;21(Feb):63-8. PMID: 42-10642. ExclusionCode: X2.255. Raimondi GA, Menga G, Botas C, et al. Declineof outpatient asthma management in Argentina.Respirology. 2008 Jan;13(1):134-7. PMID:18197924. Exclusion Code: X2.256. Reinke T. Medication therapy managementprogram in N.C. saves $13 million. Manag Care.2011 Oct;20(10):17-8. PMID: 22111478.Exclusion Code: X1.257. Rennke S, Nguyen OK, Shoeb MH, et al.Hospital-initiated transitional care interventionsas a patient safety strategy: a systematic review(Structured abstract). DARE. 2013(4):433-40.PMID: DARE-12013013336. Exclusion Code:X2.258. Riley D, Dodson P. Medicine management forthe elderly. Tex Hosp. 1983 Apr;38(11):20.PMID: 10317312. Exclusion Code: X1.259. Roberts S, Gainsbrugh R. Medication therapymanagement and collaborative drug therapymanagement. J Manag Care Pharm. 2010 Jan-Feb;16(1):67-8. PMID: 20044847. ExclusionCode: X6.260. Rodgers S, Avery AJ, Meechan D, et al.Controlled trial of pharmacist intervention ingeneral practice: the effect on prescribing costs.Br J Gen Pract. 1999 Sep;49(446):717-20.PMID: 10756613. Exclusion Code: X2.261. Ross LA, Bloodworth LS. Patient-centeredhealth care using pharmacist-deliveredmedication therapy management in ruralMississippi. JAPhA.52:802. Exclusion Code:X4.262. Roughead EE, Barratt JD, Ramsay E, et al.Collaborative home medicines review delaystime to next hospitalization for warfarinassociated bleeding in Australian war veterans. JClin Pharm Ther. 2011 Feb;36(1):27-32. PMID:21108651. Exclusion Code: X2.263. Roughead EE, Semple SJ, Vitry AI.Pharmaceutical care services: a systematicreview of published studies, 1990 to 2003,examining effectiveness in improving patientoutcomes (Structured abstract). Int J PharmPract. 2005(1):53-70. PMID: DARE-12005009527. Exclusion Code: X3.264. Rozenfeld V, Pflomm J, Singh KK, et al.Assessing the impact of medication consultationswith a Medication Event Monitoring System.Hosp Pharm. 1999(5):539-49. PMID: CN-00415071. Exclusion Code: X2.265. Rucker NL. Medicare Part D's MedicationTherapy Management: Shifting from Neutral toDrive. AARP Public Policy Institute.Washington, DC: 2012. Exclusion Code: X1.C-14


266. Sadler MA, Force RW. Pharmacy services in anambulatory FFS setting can increase physicianrevenue. Formulary (USA). 2002;37(Feb):97-01.PMID: 39-13925. Exclusion Code: X2.267. Saez-Benito L, Fernandez-Llimos F, Feletto E, etal. Evidence of the clinical effectiveness ofcognitive pharmaceutical services for agedpatients. Age Ageing. 2013 Jul;42(4):442-9.PMID: 23676212. Exclusion Code: X2.268. Saini B, Filipovska J, Bosnic-Anticevich S, et al.An evaluation of a community pharmacy-basedrural asthma management service. Aust J RuralHealth. 2008 Apr;16(2):100-8. PMID:18318852. Exclusion Code: X2.269. Saini B, Krass I, Armour C. Development,implementation, and evaluation of a communitypharmacy-based asthma care model (Structuredabstract). Ann Pharmacother. 2004(11):1954-60.PMID: NHSEED-22004001446. ExclusionCode: X2.270. Salvador CH, Ruiz-Sanchez A, Gonzalez deMingo MA, et al. Evaluation of a telemedicinebasedservice for the follow-up and monitoringof patients treated with oral anticoagulanttherapy. IEEE Trans Inf Technol Biomed. 2008Nov;12(6):696-706. PMID: 19000948.Exclusion Code: X2.271. Sando KR, Elliott J, Stanton ML, et al. Aneducational tool for teaching medication historytaking to pharmacy students. Am J Pharm Educ.2013 Jun 12;77(5):105. PMID: 23788816.Exclusion Code: X2.272. Schnipper JL, Roumie CL, Cawthon C, et al.Rationale and design of the PharmacistIntervention for Low Literacy in CardiovascularDisease (PILL-CVD) study. Circulation.2010(2):212-9. PMID: CN-00750924. ExclusionCode: X7.273. Schroeder SD. Quality focus: medication therapymanagement services. A multi-statecollaborative. S D Med. 2007 Dec;60(12):499.PMID: 18232257. Exclusion Code: X4.274. Schulz M. Pharmacist care for reactive airwaydisease. JAMA. 2003 Mar 5;289(9):1104; authorreply 5-6. PMID: 12622576. Exclusion Code:X1.275. Schulz M, Verheyen F, Muhlig S, et al.Pharmaceutical care services for asthma patients:a controlled intervention study. J ClinPharmacol. 2001 Jun;41(6):668-76. PMID:11402636. Exclusion Code: X2.276. Schulz RM, Porter C, Lane M, et al. Impact of amedication management system on nursinghome admission rate in a community-dwellingnursing home-eligible Medicaid population. AmJ Geriatr Pharmacother. 2011 Feb;9(1):69-79.PMID: 21459310. Exclusion Code: X2.277. Scott A, Tinelli M, Bond C. Costs of acommunity pharmacist-led medicinesmanagement service for patients with coronaryheart disease in England: healthcare system andpatient perspectives. Pharmacoeconomics.2007;25(5):397-411. PMID: 17488138.Exclusion Code: X2.278. Shah C, Coyne T. Medicines managementprogramme for non-medical prescribers. NursManag (Harrow). 2012 Dec;19(8):34-7. PMID:23285791. Exclusion Code: X2.279. Shillam CR, Orton VJ, Waring D, et al. Faithcommunity nurses & brown bag events helpolder adults manage meds. J Christ Nurs. 2013Apr-Jun;30(2):90-6. PMID: 23607155.Exclusion Code: X4.280. Simon GE, Ludman EJ, Operskalski BH.Randomized trial of a telephone caremanagement program for outpatients startingantidepressant treatment. Psychiatr Serv. 2006Oct;57(10):1441-5. PMID: 17035563. ExclusionCode: X2.281. Simpson SH, Johnson JA, Tsuyuki RT.Economic impact of community pharmacistintervention in cholesterol risk management: anevaluation of the study of cardiovascular riskintervention by pharmacists. Pharmacotherapy.2001 May;21(5):627-35. PMID: 11349751.Exclusion Code: X2.282. Singhal PK, Raisch DW, Gupchup GV. Theimpact of pharmaceutical services in communityand ambulatory care settings: evidence andrecommendations for future research. AnnPharmacother. 1999 Dec;33(12):1336-55. PMID:10630834. Exclusion Code: X4.283. Sklenicka JM, Proffitt LS, Dare D. The role of ageriatric clinical pharmacist in improvingmedication safety in outpatient clinics. ASHPMidyear Clinical Meeting. 2005;40:P-D. PMID:43-01370. Exclusion Code: X4.284. Smith CP, Christensen DB. Identification andclarification of drug therapy problems by Indianhealth service pharmacists. Ann Pharmacother.1996 Feb;30(2):119-24. PMID: 8835041.Exclusion Code: X4.C-15


285. Smith DH, Fassett WE, Christensen DB.Washington State CARE Project: downstreamcost changes associated with the provision ofcognitive services by pharmacists. J Am PharmAssoc (Wash). 1999 Sep-Oct;39(5):650-7.PMID: 10533346. Exclusion Code: X2.286. Smith SR, Catellier DJ, Conlisk EA, et al. Effecton health outcomes of a community-basedmedication therapy management program forseniors with limited incomes. Am J Health SystPharm. 2006 Feb 15;63(4):372-9. PMID:16452523. Exclusion Code: X2.287. Smith SR, Clancy CM. Medication therapymanagement programs: forming a newcornerstone for quality and safety in medicare.Am J Med Qual. 2006 Jul-Aug;21(4):276-9.PMID: 16849785. Exclusion Code: X1.288. Soller RW, Vogt E. Positioning patientcounselling as a measurable cornerstone ofmedication therapy management. Int J PharmPract. 2010 Oct;18(5):249-51. PMID: 20840679.Exclusion Code: X1.289. Solomon DK, Portner TS, Bass GE, et al.Clinical and economic outcomes in thehypertension and COPD arms of a multicenteroutcomes study. J Am Pharm Assoc (Wash).1998(5):574-85. PMID: CN-00155943.Exclusion Code: X2.290. Solow B. Medication therapy management--notjust for seniors anymore. Manag Care. 2009Oct;18(10):7. PMID: 19902682. ExclusionCode: X1.291. Sperling S, Neal K, Hales K, et al. A qualityimprovement project to reduce falls and improvemedication management. Home Health CareServ Q. 2005;24(1-2):13-28. PMID: 16236656.Exclusion Code: X3.292. Spooner JJ. Medication therapy managementprograms: when will the outcomes come out? JManag Care Pharm. 2007 Apr;13(3):276-7.PMID: 17407394. Exclusion Code: X1.293. Stang P, Morris L, Kempf J, et al. Thecoprescription of contraindicated drugs withstatins: continuing potential for increased risk ofadverse events. Am J Ther. 2007 Jan-Feb;14(1):30-40. PMID: 17303973. ExclusionCode: X2.294. Stebbins MR, Kaufman DJ, Lipton HL. ThePRICE clinic for low-income elderly: a managedcare model for implementing pharmacistdirectedservices. J Manag Care Pharm. 2005May;11(4):333-41. PMID: 15871644. ExclusionCode: X4.295. Stefanacci RG, Haimowitz D. Medicationmanagement in assisted living. Geriatr Nurs.2012 Jul-Aug;33(4):304 e1-12. PMID:22841458. Exclusion Code: X1.296. Stell R, Bonollo M, Fiddes K, et al. Successfulintegration of a clinical pharmacist into a diseasemanagement unit. J Pharm Pract Res.2008;38(Feb):132-6. PMID: 45-17110.Exclusion Code: X2.297. Stemer G, Lemmens-Gruber R. Clinicalpharmacy services and solid organtransplantation: a literature review (Structuredabstract). Pharm World Sci. 2010(1):7-18.PMID: DARE-12010001751. Exclusion Code:X2.298. Stergachis A, Gardner JS, Anderson MT, et al.Improving pediatric asthma outcomes in thecommunity setting: does pharmaceutical caremake a difference? J Am Pharm Assoc (Wash).2002 Sep-Oct;42(5):743-52. PMID: 12269709.Exclusion Code: X3.299. Stewart S, Pearson S, Luke CG, et al. Effects ofhome-based intervention on unplannedreadmissions and out-of-hospital deaths(Structured abstract). J Am Geriatr Soc.1998(2):174-80. PMID: NHSEED-21998000321. Exclusion Code: X2.300. Stockl KM, Shin JS, Gong S, et al. Improvingpatient self-management of multiple sclerosisthrough a disease therapy management program.Am J Manag Care. 2010 Feb;16(2):139-44.PMID: 20148619. Exclusion Code: X2.301. Stockl KM, Tjioe D, Gong S, et al. Effect of anintervention to increase statin use in medicaremembers who qualified for a medication therapymanagement program. J Manag Care Pharm.2008 Jul-Aug;14(6):532-40. PMID: 18693777.Exclusion Code: X5.302. Strand LM, Cipolle RJ, Morley PC, et al. Theimpact of pharmaceutical care practice on thepractitioner and the patient in the ambulatorypractice setting: twenty-five years of experience.Curr Pharm Des. 2004;10(31):3987-4001.PMID: 15579084. Exclusion Code: X4.303. Stratton TP, Cernohous T, Hager K, et al.College of pharmacy-based medication therapymanagement program for a university system. JAm Pharm Assoc (2003). 2012 Sep-C-16


Oct;52(5):653-60. PMID: 23023847. ExclusionCode: X4.304. Strom BL, Hennessy S. Pharmacist care andclinical outcomes for patients with reactiveairways disease. JAMA. 2002 Oct2;288(13):1642-3. PMID: 12350196. ExclusionCode: X1.305. Summers KH. Medication therapy managementversus drug regimen review. J Manag CarePharm. 2007 Jan-Feb;13(1):68-9. PMID:17269840. Exclusion Code: X1.306. Swims MP, Reed P. A pharmacotherapy clinic ina neurologic practice. J Pharm Tech.2006;22(May):254-60. PMID: 44-00819.Exclusion Code: X4.307. Tarn YH, Chou SF, Wang TY. Drug regimenreview at nursing homes in Taiwan (CPS-O-013). Int Pharm Fed. 2002;62:78. PMID: 40-01282. Exclusion Code: X2.308. Taylor J, Davies P. The medicines managementneeds of carers during an episode of mentalhealth crisis. J Psychiatr Ment Health Nurs. 2012Mar;19(2):190-2. PMID: 22070543. ExclusionCode: X1.309. Tett SE, Higgins GM, Armour CL. Impact ofpharmacist interventions on medicationmanagement by the elderly: a review of theliterature. Ann Pharmacother. 1993 Jan;27(1):80-6. PMID: 8431627. Exclusion Code: X2.310. Thomas MC. A Practical Approach toMedication Reconciliation by Pharmacists in aCommunity Hospital Emergency Department. p.41. Exclusion Code: X1.311. Thomas R, Kilbey C. Complex medicinesmanagement. Nurs Times. 2011 Mar 1-7;107(8):20-2. PMID: 21667662. ExclusionCode: X4.312. Thompson CA. State-paid medication therapymanagement services succeed. Am J Health SystPharm. 2008 Mar 15;65(6):490, 3, 8. PMID:18319489. Exclusion Code: X1.313. Tinelli M, Bond C, Blenkinsopp A, et al. Patientevaluation of a community pharmacymedications management service. AnnPharmacother. 2007 Dec;41(12):1962-70. PMID:17971403. Exclusion Code: X2.314. Tinelli M, Ryan M, Bond C, et al. Valuingbenefits to inform a clinical trial in pharmacy: dodifferences in utility measures at baseline affectthe effectiveness of the intervention?Pharmacoeconomics. 2013 Feb;31(2):163-71.PMID: 23329427. Exclusion Code: X2.315. Touchette DR. Improving adherence in thecommunity and clinic pharmacy settings: anemerging opportunity. Pharmacotherapy. 2010May;30(5):425-7. PMID: 20411993. ExclusionCode: X1.316. Trask SJ, Szabo SA. Polypharmacy interventionprogram in a tertiary care VA medical center.ASHP Midyear Clinical Meeting.1991;26(Dec):P-R. PMID: 29-00493. ExclusionCode: X2.317. Traynor K. Maryland School of Pharmacy,XLHealth team up on medication managementproject. Am J Health Syst Pharm. 2010 Jun1;67(11):868, 70. PMID: 20484207. ExclusionCode: X1.318. Triller DM, Hamilton RA. Effect ofpharmaceutical care services on outcomes forhome care patients with heart failure. Am JHealth Syst Pharm. 2007 Nov 1;64(21):2244-9.PMID: 17959576. Exclusion Code: X7.319. Truong AC. Evaluating the impact of amedication teaching clinic in a hospital basedoutpatient pharmacy. ASHP Midyear ClinicalMeeting. 2006;41(Dec)PMID: 44-06528.Exclusion Code: X4.320. Trygstad TK, Christensen D, Garmise J, et al.Pharmacist response to alerts generated fromMedicaid pharmacy claims in a long-term caresetting: results from the North Carolinapolypharmacy initiative. J Manag Care Pharm.2005 Sep;11(7):575-83. PMID: 16137215.Exclusion Code: X2.321. Trygstad TK, Christensen DB, Wegner SE, et al.Analysis of the North Carolina long-term carepolypharmacy initiative: a multiple-cohortapproach using propensity-score matching forboth evaluation and targeting. Clin Ther. 2009Sep;31(9):2018-37. PMID: 19843492. ExclusionCode: X2.322. Tsuyuki RT, Johnson JA, Teo KK, et al. Arandomized trial of the effect of communitypharmacist intervention on cholesterol riskmanagement: the Study of Cardiovascular RiskIntervention by Pharmacists (SCRIP). ArchIntern Med. 2002 May 27;162(10):1149-55.PMID: 12020186. Exclusion Code: X2.323. VanHaaren A. Pharmaceutical care in a multisitemedical group practice: reaching goals in thetreatment of hyperlipidemia and diabetes. ASHPC-17


Midyear Clinical Meeting. 2006;41(Dec)PMID:44-06529. Exclusion Code: X4.324. Vinks TH, Egberts TC, de Lange TM, et al.Pharmacist-based medication review reducespotential drug-related problems in the elderly:the SMOG controlled trial. Drugs Aging.2009;26(2):123-33. PMID: 19220069. ExclusionCode: X2.325. Viswanathan M, Golin CE, Jones CD, et al.Medication Adherence Interventions:Comparative Effectiveness. Closing the QualityGap: Revisiting the State of the Science.Evidence Report No. 208. Rockville, MD:Agency for Healthcare Research and Quality;September 2012.www.effectivehealthcare.ahrq.gov/reports/final.cfm. Exclusion Code: X1.326. Vuong T, Marriott JL, Kong DCM, et al.Implementation of a community liaisonpharmacy service: a randomised controlled trial.Int J Pharm Pract. 2008(3):127-35. PMID: CN-00708167. Exclusion Code: X2.327. Walberg SJ, Haight R. Discharge follow-up witha pharmacist study: Incorporating pharmacistprovideddischarge education and medicationreconciliation with an outpatient follow-upservice for patients with psychiatric illnessdischarged to an outpatient day treatmentprogram. J Pharm Pract. 2008;21(Jan):75-6.PMID: 45-15441. Exclusion Code: X4.328. Walker S, Willey CW. Impact on drug costs andutilization of a clinical pharmacist in a multisiteprimary care medical group. J Manag CarePharm. 2004 Jul-Aug;10(4):345-54. PMID:15298532. Exclusion Code: X2.329. WALTON T, HOLLOWAY KP, KNAUSS MD.Pharmacist-managed anemia program in anoutpatient hemodialysis population. Hosp Pharm.2005;40(12):1051-6. Exclusion Code: X2.330. Ward MA, Xu Y. Pharmacist-providedtelephonic medication therapy management in anMAPD plan. Am J Manag Care. 2011Oct;17(10):e399-409. PMID: 21999720.Exclusion Code: X2.331. Watkins JL, Landgraf A, Barnett CM, et al.Evaluation of pharmacist-provided medicationtherapy management services in an oncologyambulatory setting. J Am Pharm Assoc (2003).2012 Mar-Apr;52(2):170-4. PMID: 22370379.Exclusion Code: X4.332. Weinberger M, Murray MD, Marrero DG, et al.Effectiveness of pharmacist care for patientswith reactive airways disease: a randomizedcontrolled trial. JAMA. 2002 Oct2;288(13):1594-602. PMID: 12350190.Exclusion Code: X2.333. Weinhandl ED, Arneson TJ, St Peter WL.Clinical outcomes associated with receipt ofintegrated pharmacy services by hemodialysispatients: a quality improvement report. Am JKidney Dis. 2013 Sep;62(3):557-67. PMID:23597860. Exclusion Code: X2.334. Wessell AM, Ornstein SM, Jenkins RG, et al.Medication Safety in Primary Care Practice:results from a PPRNet quality improvementintervention. Am J Med Qual. 2013 Jan-Feb;28(1):16-24. PMID: 22679129. ExclusionCode: X2.335. Westfall GR, Narducci WA. A communitypharmacy-basedcallback program for antibiotictherapy. J Am Pharm Assoc (Wash). 1997 May-Jun;NS37(3):330-4. PMID: 9170810. ExclusionCode: X2.336. While A. Making the most of good medicinesmanagement. Br J Community Nurs. 2007Sep;12(9):434. PMID: 18026009. ExclusionCode: X1.337. White B, Truax D. The long and winding road.Arriving at safe medication management in LTCsetting. Adv Nurse Pract. 2007 Sep;15(9):28-35;quiz -6. PMID: 19998897. Exclusion Code: X1.338. Wilk JE, West JC, Rae DS, et al. Medicare PartD prescription drug benefits and administrativeburden in the care of dually eligible psychiatricpatients. Psychiatr Serv. 2008 Jan;59(1):34-9.PMID: 18182537. Exclusion Code: X2.339. Williams A. Medicine management. NursMirror. 1984 Oct 3;159(12):i-iii, vi. PMID:6567175. Exclusion Code: X1.340. Williams BR, Lopez S. Reaching the homeboundelderly: the Prescription Intervention andLifelong Learning (PILL) program. HomeHealth Care Serv Q. 2005;24(1-2):61-72. PMID:16236659. Exclusion Code: X4.341. Williams T, Acton C, Hicks RW. Deploying medreconciliation. Nurs Manage. 2008 Apr;39(4):54-7. PMID: 18391828. Exclusion Code: X1.342. Williford SL, Johnson DF. Impact of pharmacistcounseling on medication knowledge andcompliance. Mil Med. 1995 Nov;160(11):561-4.PMID: 8538891. Exclusion Code: X2.C-18


343. Wilson J. Breaking down barriers. Am Pharm.2005;127(Sep):24-6. PMID: 43-01410.Exclusion Code: X1.344. Wolfe SC, Schirm V. Medication counseling forthe elderly: effects on knowledge andcompliance after hospital discharge. GeriatrNurs. 1992 May-Jun;13(3):134-8. PMID:1319935. Exclusion Code: X2.345. Wong E. CDTM and smoking cessation. HospPharm. 1999;34(Nov):1295-303. PMID: 37-05144. Exclusion Code: X1.346. Wong MC, Kong AP, So WY, et al.Pharmacoepidemiological profiles of oralhypoglycemic agents among 28,773 Chinesepatients with diabetes. Diabetes Res Clin Pract.2012 Jun;96(3):319-25. PMID: 22305941.Exclusion Code: X2.347. Wu JC, Chin M, Pitcher K, et al. Identifyinghospitalization risk factors to prioritizemedication therapy management enrollment. CAJ Health-Sys Pharm.22:19. Exclusion Code: X2.348. Yamamura K, Osada T, Yano K, et al.Evaluation of pharmacist-managedanticoagulation classroom. ASHP MidyearClinical Meeting. 2006;41(Dec)PMID: 44-06505. Exclusion Code: X2.349. Yeoh TT, Si P, Chew L. The impact ofmedication therapy management in olderoncology patients. Support Care Cancer. 2013May;21(5):1287-93. PMID: 23179492.Exclusion Code: X4.350. Young D. Pharmaceutical case managementhelps Iowa Medicaid patients. Am J Health SystPharm. 2003 Mar 1;60(5):414, 6. PMID:12635443. Exclusion Code: X1.351. Zagaria MA. Medication therapy managementservices. US Pharmacist (USA).2004;30(Apr):35-42. PMID: 42-10763.Exclusion Code: X1.352. Zarowitz BJ, Stebelsky LA, Muma BK, et al.Reduction of high-risk polypharmacy drugcombinations in patients in a managed caresetting. Pharmacotherapy. 2005Nov;25(11):1636-45. PMID: 16232025.Exclusion Code: X4.353. Zellmer WA. Collaborative drug therapymanagement. Am J Health Syst Pharm. 1995Aug 1;52(15):1732. PMID: 7583839. ExclusionCode: X1.354. Zermansky AG, Petty DR, Raynor DK, et al.Randomised controlled trial of clinicalmedication review by a pharmacist of elderlypatients receiving repeat prescriptions in generalpractice. BMJ. 2001 Dec 8;323(7325):1340-3.PMID: 11739221. Exclusion Code: X2.355. Zermansky AG, Petty DR, Raynor DK, et al.Clinical medication review by a pharmacist ofpatients on repeat prescriptions in generalpractice: a randomised controlled trial. HealthTechnol Assess. 2002;6(20):1-86. PMID:12234455. Exclusion Code: X2.356. Zierler-Brown S, Coll RE, Llewellyn K.Implementation and evaluation of a pharmacistmanagedambulatory, indigent care practice,focused on cardiovascular disease (CVD)prevention and management. ASHP MidyearClinical Meeting. 2005;40:P-R. PMID: 43-00015. Exclusion Code: X4.357. Zillich AJ, Jaynes HA, Snyder ME, et al.Evaluation of specialized medication packagingcombined with medication therapy management:adherence, outcomes, and costs among Medicaidpatients. Med Care. 2012 Jun;50(6):485-93.PMID: 22498687. Exclusion Code: X2.358. Zillich AJ, Sutherland JM, Kumbera PA, et al.Hypertension outcomes through blood pressuremonitoring and evaluation by pharmacists(HOME study). J Gen Intern Med. 2005Dec;20(12):1091-6. PMID: 16423096. ExclusionCode: X2.C-19


Appendix D. Evidence TablesTable D1. Study and patient-level characteristicsAuthor, InterventionsYear andInterventionInclusion Criteria ExclusionTrial Comparator GoalCriteriaName DescriptionsBernsten G1: Structuredet al., community2001 1 ; pharmacybasedSturgesset al., pharmaceutical2003 2care programG2: NormalpharmaceuticalUsualcommunitypharmacyservicesBlakey, G1: Pharmacist2000 3 evaluation plususual medicalcareG2: Usualmedical careTo identifyactual andpotentialDRPs using astructuredapproach,and toresolve thoseproblems incollaborationwith PCPsusingpharmacybasedinterventionsTo recognizeand resolvedrug therapyproblems,decreasedrug therapycosts, andmaintainpositiveclinicaloutcomes.1) Aged ≥652) Taking ≥4prescribedmedications3) Oriented withrespect to self,time, and place4) Communitydwelling5) Regular visitorsto recruitedcommunitypharmacyPatients age 65and receiving carein VA geriatricclinic due toassistancerequired inactivities of dailyliving, memoryimpairment, poorjudgment,diagnosis ofdementia, historyof falls or difficultywalking,incontinence ofbowel or bladder,or polypharmacy.Houseboundor residentin nursing/residentialhomeStudyDesignRCT: 18clusterrandomimonthszedStudy FundingDuration Source(s)Multiple(Government,foundation,professionalorganizations,pharmaceuticalcompanies)Baseline% RuralBaseline Age –Mean (SD) orMedian (Range)Baseline%FemaleNR NR PooledsampleOverall:NRG1: 57.9G2: 57.3NorthernIrelandOverall:NRG1: 63.6G2: 61.0NR NRCT 8 months Unspecified NR NR Overall:0%G1: 0%G2: 0%Race/Ethnicity %NRNRD-1


Table D1. Study and patient-level characteristics (continued)Author, InterventionsYear andInterventionInclusion Criteria ExclusionTrial Comparator GoalCriteriaName DescriptionsG1: FairviewBrummel,Achievement2010 62013 4: ;Soliman,PharmacyServices' MTMof OptimalDiabetes2013 5 ; program (opt-in) CareRamalhodeOliveira,G2: controlgroup (did notopt-in)benchmarksStudyDesignPatients withdiabetes whoparticipated in anMTMdemonstrationproject and hadMTM visits to anyFairview clinicoffering MTMservices betweenJanuary 1, 2007,and December31, 2007. Arandom selectionof 121 patientswith diabetes whowere eligible forthe demonstrationproject but whodid not activelyparticipate inMTM servicesserved as thecontrol group. Thefinal analysisincluded data onpatients for whomall information onmedications wasavailable atbaseline and allD5 qualitymeasurecomponents wereavailable for 2006,2007, and 2008.NR Cohort 36monthsStudy FundingDuration Source(s)Pharmaceutical NRBaseline% RuralBaseline Age –Mean (SD) orMedian (Range)Overall: NRG1: 58.44G2: 58.19Baseline%FemaleOverall:NRG1:52.07G2:51.46Race/Ethnicity %NRD-2


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameCarter et G1:al., 1997 7 ;Barnette,Murphy,andCarter,1996 8InterventionsandComparatorDescriptionsPharmaceuticalcare providedby pharmacistswithin aninterdisciplinarypractice model.Patienteducation(lifestyle, riskfactormodifications,and drugtherapy) wasstandardized.G2: Usual careInterventionGoalTo traincommunitypharmaciststo provideHTNmonitoringand directconsultationto physiciansand nursesInclusionCriteria(1) Greater than (1)18 years of age,with essentialHTN (one of thefollowing:average diastolicblood pressure90 mm Hg orabove, averagesystolic bloodpressure 140mm Hg or above,or currenttherapy withantihypertensivedrugs [controlledor uncontrolledblood pressure]);(2) Receivingcare from aphysician in themedical center orannex andprescriptionsfrom the clinicpharmacyExclusionCriteriaSecondarycauses ofHTN; (2)Unwilling orunable toreturn toclinicpharmacy forscheduledappointment;(3) Spouse orsiblingenrolled instudy; (4) BP>210 mm Hgsystolic or>115 mm Hgdiastolic; (5)Seriouscomplicatingdisease sodisabling thatBP controlwassecondary orminorconcern (e.g.,terminalcancer, NewYork HeartAssociationclass III or IVCHF)StudyDesignStudy FundingDuration Source(s)Cohort 6 months UnspecifiedBaseline% RuralOverall:NR, butlikely100%ruralG1: NRG2: NRBaselineAge – Mean(SD) orMedian(Range)Mean (range)Overall: NRG1: 67.3 (47-80)G2: 68.5 (40-92)Baseline %FemaleNRRace/Ethnicity %NRD-3


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameChrischillesetal., 2004 9InterventionsandComparatorDescriptionsG1: PCMprovided bypharmacistsG2: Did notreceive PCMservicesInterventionGoalAvoidadverse drugevents andthe healthsystem costsassociatedwith theseadverseevents in aMedicaidpopulation athigh risk foradverseeffectsInclusion Criteria ExclusionCriteriaNoninstitutionalizedIowa Medicaidpatients taking fouror more long-termmedications,including at leastone medicationrepresenting 1 ofspecified 12diseases, wereeligible (the 12diseases werecongestive heartdisease, ischemicheart disease,diabetes mellitus,HTN,hyperlipidemia,asthma,depression, atrialfibrillation,osteoarthritis,gastroesophagealreflux, peptic ulcerdisease, andchronic obstructivepulmonarydisease) withpharmacy claims atone (or more) ofthe 117participatingpharmacies.StudyDesignAll patients Cohort 21who weremonthsnotcontinuously eligibleforMedicaidfrom 6monthsbeforethrough 12monthsafter thedate onwhich theybecameeligible forPCM.Study FundingDuration Source(s)Multiple(Governmentand foundationfunding)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)NR Overall: 52.5(20.2)G1: 54.1 (0.8)G2: 48.4 (0.5)Baseline %FemaleOverall: 71.4G1: 80.0G2: 69.3Race/Ethnicity %Overall: NRWhiteG1: 89.1G2: 90.0BlackG1: 5.9G2: 5.5OtherG1: 1.0G2: 2.1UnknownG1: 4.0G2: 2.4D-4


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsChristensenet al., servicesG1: MTM2007 10designed by ahealth plan forits beneficiariesand provided byeithercommunitypharmacists ormedical clinicbasedpharmacists.G2: Patientsfrom samecounties as G1who did notreceiveintervention(control group1)G3: Patientsfrom a differentcounty than G1who did notreceiveintervention(control group2)InterventionGoalTo assessthe feasibilityof apharmacistbasedmedicationtherapymanagementservice forNorthCarolinaState HealthPlanenrollees.Inclusion Criteria ExclusionCriteria(1) Residence inOrange or DurhamCounty, NC;(2) Among the1,000 highestnumber ofprescriptions usedduring the first 6months of 2004.StudyDesignStudy FundingDuration Source(s)NR NRCT 6 months Multiple (Thirdpartypayor andfoundation)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)NR G1: 67.7(11.4)G2: 67.6(12.2)G3:66.0(12.1)Baseline %FemaleG1: 62.3G2: 68.9G3: 71.3Race/Ethnicity %NRD-5


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsClifford et G1:al., 2002 11 Pharmaceuticalcare providedby a clinicalpharmacist ,which includedacomprehensivereview relatingto pharmacotherapyanddiabetes, use ofproprietary andnon-proprietarymedications,such ascomplementarymedicines, andidentification ofdrug therapyproblems.G2: Standardoutpatient carefor diabetesInterventionGoalTo improveglycemiccontrol indiabeticpatientswithoutadverselyaffecting QOLor satisfactionwith healthcare providedInclusion Criteria ExclusionCriteriaAdult patients ≥18 Seeyears with type 1 or inclusiontype 2 diabetes criteriaand at least one ofthe followingfeatures indicatinghigh risk fordevelopment ofdiabetescomplications:1) Random bloodglucose levels >11mmol/L on ≥2occasions intertiary care settingwithin previous 12months;2) HbA1C >8% on≥2 occasions inprevious 12months;3) HTN (SBP >160mm Hg and/or DBP>90 mm Hg) and/ortaking drugtherapy;4) Dyslipidemia(total serumcholesterol >5.5mmol/L and/orserum triglycerides>4.0 mmol/L);5) Polypharmacy(>3 drugs)StudyDesignStudy FundingDuration Source(s)RCT: 6 months Multipleparallel, (Pharmaceutical,notclustere professionaldorganization)Baseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 60 (12)G2: 61 (12)p=NSBaseline %FemaleOverall: NRG1: 42G2: 52p=NSRace/Ethnicity %NRD-6


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventions andComparatorDescriptionsFischer et G1:al., 2000 12 Pharmaceuticalcare based onEncara PracticeSystem providedby onsite healthmaintenanceorganization staffpharmacistsG2: Standardcommunitypharmacy practiceG3: Patients ateligible clinics whodeclined to receiveintervention butwere included insome analyses.Fischer etal., 2002 13Pharmaceuticalcare based onEncara PracticeSystem providedby pharmacists.Pharmacistphysiciancommunicationabout pharmacistidentifiedDTPs.G2: Usual carewith no additionalinterventionsIntervention Goal(1) Toimprove theamount ofinformationpatientsreceived; (2)To improvethe waypatients selfadministermedication;(3) Toenhanceawarenessof sideeffects.To assesswhetherpharmaceutical careprogramdecreaseshealth careutilization,medicationuse, orchargesInclusion Criteria ExclusionCriteria(1) HMO enrolleesenrolled in aparticipating clinic;(2) Had asthma,COPD or heartdisease identifiedvia pharmacy orhospital data basemedicationrecords.(1) Age ≥18; (2)Enrolled inparticipating HMOfor ≥2 years withactive prescriptionstreating heart orlung disease; (3)Obtainedprescriptions fromparticipatingpharmacy; (4) Musthave filledprescriptions forone of several prespecifiedmedication typesfor heart or lungdisease in 6months beforestudyStudyDesignStudy FundingDuration Source(s)NR NRCT 6 months Foundation ornon-profitDied, NRCTdisenrolled,ordiscontinuedpharmacybenefitbefore endof studyperiod2 years Multiple(1997- (Pharmaceutical98) [oneyear companies,before third-partyintervene payors)-tioninitiationand oneyearafter]Baseline% RuralNRNRBaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 67.2G2: 68.3G3: 58.9Overall: NRG1: 57G2: 58Baseline %FemaleOverall: NRG1: 54G2: 52G3: 50Overall: NRG1: 50G2: 51Race/Ethnicity %% WhiteOverall: NRG1: 98G2: 96G3: 92NRD-7


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsFox et al., G1: Florida2009 14 Health CarePlans MTMprogram,consisting of amedicationtherapy reviewand evaluationby a clinicalpharmacist thatwasdocumentedand sent to thepatient'sphysicianthrough healthplan reviewG2: Opt-outfrom MTMprogramInterventionGoalTo reduceLDL-C andimproveHEDIS goalattainmentamongpatients withdiabetes onlipid-loweringmedicationsInclusion Criteria ExclusionCriteriaFHCP enrolleeswho:1) Were MedicarePart D members;2) Were diagnosedwith ≥3 chronicdiseases;3) Used ≥4maintenancemedications;4) Were likely tohave Part Dmedication costs≥$4000 per year;5) Were eligible forinclusion in 2008HEDIS CDCadministrativedatasetNonespecifiedStudyDesignCohort 21monthsStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleUnspecified NR Overall: NR Overall: NRG1: 67.6 (7.2) G1: 45.5G2: 68.3 (6.1) G2: 57.9Race/Ethnicity %NRD-8


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameGattis etal., 1999 15Interventionsand ComparatorDescriptionsG1: Clinicalpharmacyservices, includingan assessment ofprescribedregimen,compliance, andadverse effects,and symptomsand response totherapy. Providingpatient educationabout the purposeof each drug andreinforcingadherence.Detailed writteninformation wasalso provided topatients.G2: Usual medicalcareInterventionGoalTo improveoutcomes inoutpatientswith heartfailure.InclusionCriteriaPatients with a Lifediagnosis ofheart failurewith LVEF


Table D1. Study and patient-level characteristics (continued)Author,YearTrial NameInterventionsandComparatorDescriptionsGrymonpre, G1:2001 16 Comprehensivedrug therapyreview, thenissuesaddressed withthe client and/orthe client'sphysician, withfollow-up asrequired.G2:Comprehensivedrug therapyreview only withreferral to usualpharmacistInterventionGoalInclusionCriteriaIdentify and Age 65 or older, NRaddress drugrelatedissues institutionalized,non-with an noninstitutionalizmoretaking 2 ored elderly prescribed orpopulation non-prescribedmedicationsExclusionCriteriaStudyDesignRCT:parallel,notclusteredStudy FundingDuration Source(s)Baseline% Rural6 months Unspecified Overall:NRG1: NRG2: NRBaselineAge – Mean(SD) orMedian(Range)Overall: N-RG1: 76.9(8.4)G2: 77.2(8.8)Baseline %FemaleOverall: NRG1: 75G2: 83Race/Ethnicity %Overall: NRG1: 100%CaucasianG2: 100%CaucasianD-10


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsHanlon et G1:al., 1996 17 PharmaceuticalCowper, care provided1998 18 by a clinicalpharmacistG2: Usual carein the GeneralMedicine ClinicInterventionGoalTo evaluate theeffect ofsustainedclinicalpharmacistinterventionsinvolving elderlyoutpatients withpolypharmacyand theirprimary carephysicians on:prescribingappropriateness,healthrelatedqualityof life, adversedrug events,medicationcompliance andknowledge,number ofmedicationsused, patientsatisfaction,and physicianreceptivityInclusionCriteria(1) Age ≥65; (2)Had evidenceofpolypharmacyoperationallydefined asprescribed 5+regularlyscheduledmedications bya VA physician;(3) Receivingprimary care inthe GeneralMedicine Clinic.ExclusionCriteria1) Nursinghomeresidence;2) Patientswithcognitiveimpairment, asdeterminedby theMentalStatusQuestionnaire;3)Nocaregiveravailable tobe involvedin theinterventionStudyDesignRCT:parallel,notclusteredStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)One year Government NR Overall: NRG1: 69.7 (3.5)G2: 69.9 (4.1)Baseline %FemaleOverall: NRG1: 1.9G2: 0Race/Ethnicity %WhiteOverall: NRG1: 79G2: 74.8D-11


Table D1. Study and patient-level characteristics (continued)Author,YearTrial NameInterventionsandComparatorDescriptionsHirsch, G1: Patients2011 19 served atHirsch, nonpilot2009 20 pharmaciesRosenquist, G2: Patients2010 21 served at pilot(methods) pharmaciesIntervention GoalExamining Medi-Calan HIV/AIDSpharmacyMTMcompensation pilotprogram in asample ofMedi-Calbeneficiariesbydescribingtheassociationsbetween useof pilotpharmaciesand (a)adherenceto ARTregimens;(b)medicationutilization;(c)occurrenceofopportunisticinfections;and (d) allcausepharmacyand medicalcostsInclusion Criteria ExclusionCriteriabeneficiaries aged≥18, who werecontinuouslyenrolled fromJanuary 1, 2004through December31, 2007 anddiagnosed withHIV/AIDS. In eachstudy year, patientswere identified aspilot pharmacypatients if theyfilled 50% or moreof their ARTprescriptions at 1of the 10 pilotpharmacies.Comparison grouppatients met thesame inclusion andexclusion criteriaas study patientsexcept they filledless than 50% oftheir ARTprescriptions at 1of the 10 pilotpharmaciesIndividualsenrolled inmanagedcare plans,dualeligibleMedicarepatients,patientswho died atany timefrom 2004through2007StudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Cohort 3 years Government NR Overall: NRG1: 44.7 (8.1)G2: 45.4 (7.8)Baseline %FemaleOverall: NRG1: 32.8G2: 28.0Race/Ethnicity %Non-LatinoWhiteOverall: NRG1: 40.3G2: 42.7AfricanAmericanOverall: NRG1: 33.0G2: 31.4LatinoOverall: NRG1: 16.2G2: 16.3OtherOverall: NRG1: 10.5G2: 9.6D-12


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameIsetts etal., 2008 22InterventionsandComparatorDescriptionsG1: MTMservicesprovided bystaffpharmacists,including theestablishmentof goals oftherapy, incollaborationwith primarycare providers.G2: Usualmedical carewithout MTMInterventionGoal(1) Toprovide MTMservices topatients; (2)to measureclinicaleffectsassociatedwith MTM, (3)to measurepercent ofpatientsachievinggoals for HTNandhyperlipidemiain MTMvs.comparison;and 4) tocomparepatients' totalhealthexpendituresfor the yearbefore andafter MTMInclusion Criteria ExclusionCriteriaPatients ininterventiongroup: 1) Enrolledin Blue Plusinsurance productof Blue CrossBlueShield ofMinnesota; (2)Age ≥18 years;(3) Receivingmedical care atone of 6 clinics inFairview, MNwhere MTMservices provided;(4) Diagnosedwith ≥1 of 12study medicalconditions, (5) ≥2health care claimsrelated to 12study conditionsin 6-month periodbefore the start ofthe study.StudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)NR Cohort 12 Academic NR Overall: NRG1: 14%were age 65or olderG2: NRBaseline %FemaleOverall: NRG1: 66G2: NRRace/Ethnicity %NRD-13


Table D1. Study and patient-level characteristics (continued)Interventions andAuthor, YearIntervention Inclusion ExclusionComparatorTrial NameGoal Criteria CriteriaDescriptionsJameson, PharmacothVanNoord, erapyand consultationVanderwoud, and followup1995 23provided byclinicalambulatorycarepharmacist.G2:Standardoffice-basedprimarycare.To simplify thepharmacologicregimen,improveeffectivenessof the regimen,and decreaseside effects.Secondarygoal todecrease costwithoutadverselyaffecting thefirst threegoals.Improvedclinicaloutcomes(decreasednumber ofmedications,decreasednumber ofdoses per day,monthly costofmedications,patient selfreportedcompliance,drug regimenconvenience,fewer sideeffects andproblems).All patients at aFamily HealthCenter seenduring a 1 yearperiod with 2 ormore riskfactors: 5 ormoremedications, 12or more dailydoses, 4 ormoremedicationchanges in last12 months,more than 3disease states,documentedmedicationnoncompliance,medicationsthat requiretherapeuticmonitoring.Activealcohol orillicit druguse,unwilling orunable toreturn for apharmacotherapyconsultation,medicationregimenprimarilymanaged byan outsideprovider,terminally ill,less than 18years ofage.StudyDesignStudy FundingDuration Source(s)RCT: 6 months Multipleparallel,notclustered(Foundation ornon-profit,academic, andpharmaceutical)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)NR Overall: 60.5G1: NRG2: NRBaseline %FemaleOverall: 80G1: NRG2: NRRace/Ethnicity %AfricanAmericanOverall: 28G1: NRG2: NRD-14


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsJeong et G1: Kaiseral., 2007 24 PermanenteJeong, 20062009 25 pharmacistmanagedMTMPG2: PatientswithoutMedicare PartD as theirprimary drugbenefit andlikely to incurdrug costsgreater than orequal to $4000per year with asimilar diseaseburden.InterventionGoalProvideMedicarePart D MTMbenefit toeligiblebeneficiariesto improvedmedicationuse anddecreaseadverseevents.Inclusion Criteria ExclusionCriteriaPatients who: See1) Were likely to inclusionincur >$4,000 in criteriadrug costs per year2) Received ≥2Part D medications3) Had ≥2 chronicconditions4) Had a diagnosisof hyperlipidemia,diabetes, or CADfor LDL-C analysis5) Had a diagnosisof diabetes forHbA1c analysis6) Had a lab (LDL-C or HbA1c) within6 months beforeand 6 months afterindex date7) were at least 65years old as of Jan1, 2006 and had adx of DM, HTN orCAD in 2005.8) continuousmembership withdrug benefitsduring 12 monthsstudy periodStudyDesignCohort 12monthsStudy FundingDuration Source(s)Integratedhealth caresystem (KaiserPermanente)Baseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleOverall: NR Overall: NRG1: 75.1 (6.5) G1: 54G2: 73.8 (7.0) G2:51p< 0.0001 p= 0.04Race/Ethnicity %NRD-15


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsJeong 26 G1: Kaiser-Jeong, Permanente2012 27MTM programparticipants(2010)G2: Kaiser-Permanentepatients eligiblefor MTM, butwho declinedenrollment ordisenrolled witha PCP visitduring first halfof 2010G3: Kaiser-Permanentepatients eligiblefor MTM, butwho declinedenrollment ordisenrolledwithout a PCPvisit during firsthalf of 2010InterventionGoalTo optimizetherapeuticoutcomesthroughimprovedmedicationuse and toreduce therisk ofadverseevents.Inclusion Criteria ExclusionCriteria1) at least 3chronic conditions2) on at least 5Medicare Part Ddrugs3) Incurred costsfor Part D drugs>=$3,000.StudyDesignNR Cohort 12monthsStudy FundingDuration Source(s)Integratedhealth caresystem (KaiserPermanente)Baseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 74.98(8.67)G2: 74.67(9.47)G3: 78.34(10.78)Baseline %FemaleNRRace/Ethnicity %NRD-16


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameKrska etal., 2001 28InterventionsandComparatorDescriptionsG1: Medicationreviews led byclinically-trainedpharmacists.G2: Usual careinvolvinginterviews andidentification ofpharmaceuticalcare issues butwith nopharmaceuticalcare planimplemented.InterventionGoalTo study theeffect ofmedicationreview led bya pharmaciston resolutionof pharmaceuticalcareissuesmedicinecosts, use ofhealth andsocialservices andhealth-relatedquality of lifeInclusion Criteria ExclusionCriteriaPatients aged at Dementia,least 65 years with beingat least two chronic considereddisease states, by the GP totaking at least 4 be unable toprescribed cope with themedicines regularly studyStudyDesignRCT:parallel,notclusteredStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)3 months Government NR Overall: NRG1: 74.8 (6.2)G2: 75.2 (6.6)p=0.972Baseline %FemaleOverall: NRG1: 56.5G2: 64.6p=0.132Race/Ethnicity %NRD-17


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameMalone,2000 29Ellis,2000 30Malone,2001 31Ellis,2000 32IMPROVEInterventionsandComparatorDescriptionsG1:Pharmaceuticalcare providedby clinicalpharmacistspracticingaccording toscope ofpractice withintheir respectivehealth carefacilities.G2: Usual carewithoutpharmaceuticalcareInterventionGoalDetermine ifclinicalpharmacistscould changeresource useandhumanisticoutcomesamongVeteransidentified athigh risk formedicationrelatedproblems.InclusionCriteria1) High risk for 1)drug-relatedproblems (weretaking 5 ormore drugs, 12or moredoses/day, had3 or morechronic medicalconditions, 4 ormorechanges in theirdrug regimenover the pastyear, history ofnonadherenceor taking anagent thatrequiredtherapeuticdrugmonitoring); 2)Received careat the VA withinthe past 12months andanticipatedcontinued VAcare for theduration of thestudy; 3) Livedclose/hadtransportationto VA.ExclusionCriteriaParticipationin apharmacistmanagedclinic withinprevious 12months; 2)Terminalcondition/poor lifeexpectancy;3) Requiredmental healthservices; 4)Poor spokenor writtenEnglish; 5)Visuallyimpaired.StudyDesignRCT: 12parallel, monthsnotclusteredStudy FundingDuration Source(s)Baseline% RuralPharmaceutical Overall:67 (10.1)G1: 66.8(10.2)G2: 66.6(10.0)BaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 3.6G2: 3.8Baseline %FemaleNRRace/Ethnicity %NRD-18


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsInterventionGoalMarques, G1: Intervention Clinical2013 33 group: Dadermethodpharmacotherapy follow-upinterventionmonthly over 3-month follow-upperiodG2: Controlgroup: monthlypharmacistvisits withoutpharmacotherapy follow-upinterventionimprovementof depressionand anxioussymptomsInclusionCriteriaFemale gender,age 18-65years,diagnosis ofdepression atthe initial stageof treatment(first treatment,no previousantidepressant),prescribed newantidepressantExclusionCriteriaInsurmountable difficultiesschedulingvisits, BeckDepressionInventory


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameMarrufo, CHF2013 34 ,Perlroth,2013 35InterventionsandComparatorDescriptionsG1: enrolled inMedicare PDPreceiving MTMwith a CMRG2: enrolled inPDP receivingMTM, no CMRG3: enrolled inMA-PDreceiving MTMwith CMRG4: enrolled inMA-PD,receiving MTM,no CMRCOPDG5: enrolled inMedicare PDPreceiving MTMwith a CMRG6: enrolled inPDP receivingMTM, no CMRG7: enrolled inMA-PDreceiving MTMwith CMRG8: enrolled inMA-PD,receiving MTM,no CMRInterventionGoalInvestigatehowenrollment ina standalonePDP or MA-PD MTMprogram withor withoutreceipt of aCMRinfluencedadherence,quality ofprescribing,resourceutilization,and costs forMedicarebeneficiarieswith CHF,COPD, anddiabetes.Inclusion Criteria ExclusionCriteriaG1-G12: Part Dbeneficiaries with2009 risk data;have CHF, COPD,or diabetes; haveat least one PDEclaim in 2010;enrolled in contractthat passed datavalidation for MTMsection; enrolled inone MTM programin 2010; enrolled ina MTM program atleast one day in2010; new to MTMin 2010; samecontract reported inMTM Beneficiary-Level file and PartD enrollment file;continuouslyenrolled in Part Dduring studyperiod; enrolled inthe same contractduring outcomeperiod.G13-G18:Constructed fromthe pool ofbeneficiaries in theDid nothaveESRD in2009; Non-LTI in2010; Notnew in riskfile;StudyDesignCohort 12monthsStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Government NR G1:≤65: 25.266-75: 33.476-85: 30.7>85: 10.7G2:≤65: 13.866-75: 31.876-85: 36.8>85: 17.6G3:≤65: 8.066-75: 35.876-85: 41.9>85: 14.3G4:≤65: 11.966-75: 35.176-85: 38.3>85: 14.6G5:≤65: 34.266-75: 34.776-85: 24.6>85: 6.5G6:≤65: 23.166-75: 35.776-85: 30.7>85: 10.4Baseline %FemaleG1: 68.2G2: 59.7G3: 52.3G4: 54.8G5: 69.2G6: 61.8G7: 54.6G8: 56.7G9: 68.8G10: 59.4G11: 53.3G12: 56.4G13: 63.3G14: 59.1G15: 64.7G16: 62.4G17: 60.9G18: 57.8Race/Ethnicity %WhiteG1: 74.8G2: 82.5G3: 79.8G4: 79.1G5: 81.4G6: 85.6G7: 84.0G8: 84.4G9: 76.3G10: 80.7G11: 75.1G12: 76.5G13: 81.8G14: 79.2G15: 86.1G16: 83.7G17: 77.3G18: 72.2BlackG1: 19.0G2: 11.3G3: 11.6G4: 13.3G5: 13.7G6: 9.1G7: 9.3G8: 9.8G9: 16.2D-20


Table D1. Study and patient-level characteristics (continued)Author,YearTrial NameMarrufo,2013 34 ,Perlroth,2013 35(continued)InterventionsandComparatorDescriptionsDiabetesG9: enrolled inMedicare PDPreceiving MTMwith a CMRG10: enrolledin PDPreceivingMTM, no CMRG11: enrolledin PA-PDreceiving MTMwith CMRG12: enrolledin MA-PD,receivingMTM, no CMRComparison -CHFG13: enrolledin PDP, usualcareG14: enrolledin MA-PD,usual careComparison -COPDG15: enrolledin PDP, usualcareG16: enrolledin MA-PD,usual careInterventionGoalInclusion Criteria ExclusionCriteriasame diseasecohort. To narrowthe set ofbeneficiaries in thecomparison groupto include onlybeneficiaries withchronic conditionsand drug utilizationlevels similar tothose experiencedby MTM enrollees,used variations inMTM eligibilityrules andimplementationmethods set byPart D sponsors.StudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)G7:≤65: 10.166-75: 38.876-85: 40.1>85: 10.9G8:≤65: 15.666-75: 40.976-85: 34.2>85: 9.2G9:≤65: 28.666-75: 36.176-85: 27.8>85: 7.6G10:≤65: 16.566-75: 40.176-85: 33.4>85: 10.0G11:≤65: 8.966-75: 45.576-85: 37.6>85: 8.0G12:≤65: 13.266-75: 45.876-85: 33.4>85: 7.6Baseline %FemaleRace/Ethnicity %G10: 10.6G11: 12.4G12: 12.9G13: 12.6G14: 13.9G15: 9.1G16: 10.1G17: 13.7G18: 16.5HispanicG1: 3.5G2: 2.6G3: 3.5G4: 4.0G5: 2.6G6: 2.1G7: 2.9G8: 3.0G9: 3.9G10: 3.0G11: 4.2G12: 5.0G13: 2.5G14: 3.4G15: 2.1G16: 2.9G17: 3.7G18: 4.6D-21


Table D1. Study and patient-level characteristics (continued)Author,YearTrial NameMarrufo,2013 34 ,Perlroth,2013 35(continued)InterventionsandComparatorDescriptionsComparison -DiabetesG17: enrolledin PDP, usualcareG18: enrolledin MA-PD,usual careInterventionGoalInclusion Criteria ExclusionCriteriaStudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)G13:≤65: 16.766-75: 28.776-85: 34.8>85: 19.8G14:≤65: 12.566-75: 32.676-85: 38.0>85: 16.9G15:≤65: 24.966-75: 33.076-85: 29.9>85: 12.2G16:≤65: 17.966-75: 37.376-85: 34.0>85: 10.7G17:≤65: 22.266-75: 37.676-85: 30.5>85: 9.7G18:≤65: 14.766-75: 43.976-85: 33.4>85: 8.0Baseline %FemaleRace/Ethnicity %Other/UnknownG1: 2.7G2: 3.6G3: 5.2G4: 3.7G5: 2.3G6: 3.2G7: 3.8G8: 2.8G9: 3.6G10: 5.7G11: 8.3G12: 5.6G13: 3.1G14: 3.4G15: 2.7G16: 3.2G17: 5.3G18: 6.6D-22


Table D1. Study and patient-level characteristics (continued)Author,YearTrial NameInterventionsandComparatorDescriptionsInterventionGoalInclusionCriteriaMcDonoughet al., 2005 36 G1:To reducePharmaceutical risk ofPatients 18years of age orcare providedby communitypharmacists.Drug therapymonitoringfocused on 5glucocorticoid older who hadosteoporosis. been on theequivalent of atleast 7.5 mg ofprednisone for atleast 6 monthsdrug therapyproblems:appropriatenessof dose, properregimen,potentialinteractions,nonadherence,and adverseeffects. Patienteducation alsoprovided.G2: Usual careExclusionCriteriaNRStudyDesignRCT:clusterrandomizedStudy FundingDuration Source(s)9 months Multiple(Pharmaceuticalcompany andacademic)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleNR NR Overall: NRG1: 57.7G2: 74.3Race/Ethnicity %Caucasianor AsianOverall: NRG1: 92.3G2: 84.3D-23


Table D1. Study and patient-level characteristics (continued)Author, YearTrial NameMoczygemba, G1: Opt-in2011 37 telephonebasedMTMMoczygemba,2008 38 program, inMoczygemba,2012 39Interventions and InterventionComparator GoalDescriptionswhich MTMservicesprovided byclinicalpharmacistsor a managedcarepharmacyresidentbased on theAmericanPharmacistsAssociationand NationalAssociationof Chain DrugStoresFoundationMTMframework.G2: No-MTMcontrol groupTo identifyand resolvemedicationand healthrelatedproblemsInclusion Criteria ExclusionCriteriaMedicare Part D Patientsbeneficiaries of ≥90 yearsthe Scott & White of age dueHealth Plan with: to patient1) ≥2 chronic privacydiseases concerns2) ≥2 Part D drugs3) ≥$4000 in PartD drug costs4) Received ≥1MTM consultationStudyDesignStudy FundingDuration Source(s)Cohort 9 months Foundation ornon-profitBaseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleMean (SD) Overall: NROverall: NR G1: 48.3G1: 71.2 (7.5) G2: 71.7(range: 53- p: 0.00986)G2: 73.9 (8.0)(range: 46-88)p: 0.06Race/Ethnicity %WhiteOverall: NRG1: 78.3G2: 91.7p: 0.29D-24


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsInterventionGoalMoore, G1: MTM To improve2013 40 program (opt-in) drug therapyG2: control adherencegroup (refusers) and clinicaloutcomesInclusion Criteria ExclusionCriteriaStudyDesignPatients over theage of 18 who had14 or more claimswithin a 120-dayperiod and/or hadclaims showing theabsence of arecommendedtherapy or thepresence of aconflicting therapyin the treatment ofconditions such as,but not limited to,asthma, diabetes,heart failure, orheart diseaseNR Cohort 24monthsStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Other NR Mean (SE)G1: 74.1(0.226)G2: 73.7(0.259)p: 0.277Baseline %FemaleOverall: 60G1: 60G2: 60Race/Ethnicity %NRD-25


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNamePai, G1:2009 41 ;Pai,2009 42InterventionsandComparatorDescriptionsPharmaceuticalcare includingdrug therapyreviewsconducted by anephrologytrainedclinicalpharmacist withthe patient. Alsoincluded patientand health careprovidereducation.G2: Standard ofcare, consistingof brief therapyreviewsconducted by anursePark et al., G1:1996 43 Comprehensivepharmaceuticalservicesincluding drugtherapymonitoring andpatienteducationprovided by acommunitypharmacyresident.G2: Usual careInterventionGoalTo investigatethe impact of apharmaceutical care programmanaged byclinicalpharmacistson drug use,drug costs,hospitalizationrates, anddrug-relatedproblems(DRPs) inambulatorypatientsundergoinghemodialysis.Improve bloodpressure andquality of lifefor patientswith HTN.Inclusion Criteria ExclusionCriteriaTo participate inthe study, patientshad to speakEnglish, and beolder than 18years andundergoing astablehemodialysisregimen for atleast 3 months.Informed consentwas obtained fromeach patientbefore starting thestudy, with aconsent rate of70%.If patientselected notto consentor wereunable toprovideinformedconsent,theycontinuedto receivethe carethat theirshift wasassigned;however,no datafrom thesepatientswerecollected oranalyzed.Patients with HTN Bedridden;either currently nontakingantihypertensivespeaking;Englishmedication or with hada BP > 140/90. anotherfamilymemberenrolled inthe study.StudyDesignRCT:clusterrandomizedRCT:parallel,notclusteredStudy FundingDuration Source(s)2 years Foundation ornon-profitBaseline% RuralBaselineAge – Mean(SD) orMedian(Range)NR Overall: 59.0(15.0)G1: 56.3 (15)G2: 60.5(14.7)4 months Unspecified NR Overall: NRG1: 57.3(range 29-82)G2: 63.0(Range 23-88)Baseline %FemaleOverall: 48.1G1: 38.6G2: 59.6Overall: NRG1: 44G2: 41Race/Ethnicity %CaucasianOverall: 27.9G1: 22.8G2: 34.0HispanicOverall: 30.8G1: 29.8G2: 31.9NativeAmericanOverall: 17.3G1: 22.8G2: 10.6OtherOverall: 24.0G1: 24.6G2: 23.4% whiteG1: 81G2: 69D-26


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNamePindolia etal., 2009 44InterventionsandComparatorDescriptionsG1: TelephonebasedMTMservicesprovided as partof a MedicarePart D MTMprogram bypharmacy caremanagementclinicalpharmacists(acceptors).G2: Usualmedical care(refusers)InterventionGoalTo 1) ensurethat safest,mostefficacious,and costeffectivedrugtherapy isprovided bycollaboratingwithphysiciansandpatients/caregivers in thedevelopmentof an optimaldrug regimenthat meetsboth medicaland patientneeds; (2)educatepatients on allaspects oftheir drugtherapy; and(3) improveadherence todrug therapyregimensInclusion Criteria ExclusionCriteriaIn 2006:1) Diagnosed with2 of 26 selectedchronic diseases;2) Filled ≥2prescriptions asidentified bypharmacy claimsdata;3) Likely to incurannual costs of≥$4000 for allMedicare Part D-coveredmedications basedon quarterlyprescription drugexpenditures of$1000In 2007:1) Diagnosed with3 of 21 selectedchronic diseases;2) Filled ≥4prescriptions asidentified bypharmacy claimsdata;3) Likely to incurannual costs of≥$4000 for allMedicare Part D-coveredmedications basedon monthlyprescription drugexpenditures of$334StudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleNR Cohort 2 years Unspecified NR 2006Mean (SD)[range]Overall: NR2006Overall: NRG1: 64G2: 60G1: 73.5 (9.7) p: 0.175[42-92] 2007G2: 74.2 (9.8) Overall: NR[32-96]p: 0.2292007G1: 54G2: 63p: 0.01Mean (SD)[range]Overall: NRG1: 73.0 (9.1)[39-93]G2: 73.9 (9.8)[33-98]p: 0.168D-27Race/Ethnicity %NR


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNamePlanas etal., 2009 45InterventionsandComparatorDescriptionsG1: MTMservicesprovided bycommunitypharmacists.Also includedpatienteducation ondiet and lifestylemodifications tolower bloodpressure.G2: No MTMreceived, butonly informed ofblood pressuregoals forpatients withdiabetesInterventionGoalTo improveBP andantihypertensivemedicationadherence inpatients withboth diabetesand HTNInclusion Criteria ExclusionCriteriaMembers of MCOalready enrolled inconcurrent study ofcommunitypharmacy-baseddiabetesmanagementprogram.Criteria fordiabetes study:(1) Lack ofdiabetes control(i.e., most recentA1C within last 6months >7.0%)(2) ≥18 years old(3) Currentlyinsured by MCO(4) Able and willingto come to periodicvisits during a 9-month period(1)Pregnant;(2)Currentlyenrolled inanotherdiabetesprogramStudyDesignRCT:parallel,notclusteredStudy FundingDuration Source(s)9 months Multiple(Foundationand pharmacychain)Baseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 64.2(10.5)G2: 65.2(14.1)Baseline %FemaleOverall: NRG1: 65.6G2: 60.0Race/Ethnicity %WhiteOverall: NRG1: 75.0G2: 90.0BlackOverall: NRG1: 21.9G2: 10.0HispanicOverall: NRG1: 3.1G2: 0Criteria for HTNstudy:(1) Present atbaseline diabetesstudy visit(2) BP ≥130/80 mmHg or currentlytakingantihypertensivetherapyD-28


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsRoughead G1: HMR, aet al., collaborative2009 46Sellors etal., 2003 47model ofpharmaceuticalcare, conductedby accreditedpharmacists.G2: Nomedicationreview receivedG1: Clinicalpharmacistconsultationsprovided tofamilyphysicians andtheir patients bycommunitypharmacists.G2: Usual carefor familyphysicians andtheir patientsfrom matchedpostal codesInterventionGoalTo reducetime to nexthospitalizationfor HFamongAustralianwar veteransand warwidows withHFReducingregimencomplexityandimprovingpatientoutcomesInclusion Criteria ExclusionCriteriaCommunitydwellingelderlywho:1) Had all healthservices fullysubsidized byAustralianGovernment'sDVA;2) Were dispensedbeta-blockersubsidized for HFin 6 months beforethe HMR;3) Were aged ≥65years at the time ofhome review1) Communitydwelling;2) 65 years orolder; 3) taking 5 ormore medications;4) had been seenby their physicianwithin the past 12months; 5) had noevidence ofcognitiveimpairment; and 6)could understandEnglish1)Residentsin agedcarefacilities1) Hadplannedsurgery, 2)were on anursinghomewaitinglist or 3)werereceivingpalliativecareStudyDesignCohort 40monthsStudy FundingDuration Source(s)GovernmentRCT: 3 months Multipleclusterrandomiand hospital)(GovernmentzedBaseline% RuralRegion ofresidenceRemoteOverall: NRG1: 0G2: 1OuterregionalOverall: NRG1: 12G2: 9InnerregionalOverall: NRG1: 29G2: 31NRBaselineAge – Mean(SD) orMedian(Range)Median (SD)Overall:G1: 81.6 (4.8)G2: 81.6 (4.8)Baseline %FemaleOverall: NRG1: 30G2: 26Overall: NR Overall: NRG1: 74.0 (6.1) G1: 64.3G2: 74.0 (6.0) G2: 61.4Race/Ethnicity %NRNRD-29


Table D1. Study and patient-level characteristics (continued)Author,YearTrialNameInterventionsandComparatorDescriptionsSellors, G1:2003 48 PharmaceuticalconsultationG2: Usual careShimp, G1: MTM2012 49 program forUniversity ofMichiganbeneficiaries,entitled FOMG2: Usual care(not described)InterventionGoalPilot study fora largerrandomizedtrialinvestigatingthe efficacyand costeffectivenessof apharmacistconsultationprogram forelderlypatients infamilypracticeTo leveragetheUniversity'sinvestment inemployeehealth byintroducing apatientcenteredMTMprogramoffered toemployees,retirees, andtheirdependentsto optimizedrug therapyInclusionCriteriaEligibleparticipants wereat least 65 yearsof age and weretaking at leastfour medicationsregularly, whichin additionprescribed oralmedicationsincluded insulin,inhalers, creams,drops andointments, OTCmedications, andherbal remedies.University ofMichiganbeneficiaries(employees,retirees, andtheirdependents),taking seven ormoreprescriptionmedications.Patients with aUniversity ofMichigan PCPwerepreferentiallyinvited.ExclusionCriteriaNRNRStudyDesignRCT:parallel,notclusteredRCT:parallel,notclusteredStudy FundingDuration Source(s)Baseline% Rural6 months Unspecified Overall:NRG1: NRG2: NR12monthsAcademicOverall:NRG1: NRG2: NRBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleOverall: NR Overall: NRG1: 76.4 (6.5) G1: 60.6G2: 75.5 (6.4) G2: 69.7Overall: NRG1: 70G2: NROverall: NRG1: 55G2: NRRace/Ethnicity %Overall: NRG1: NRG2: NROverall: NRG1: NRG2: NRD-30


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaSidel etal., 1990 50G1: Home visitsby pharmacistsand, whenneeded,consultationswith physiciansto identify andcorrectproblemsassociated withmedication use.G2: Standardcare withoutany visits orinformationprovided to G1.To determinetheprevalence ofuse ofprescriptionand OTCmedicationsand homeremedies, tocharacterizemedicationtakingbehaviorsandpractices,and to assessthe impact ofin-homepharmacistinterventionin identifyingandcorrectingproblems withmedicationuse(1) All Medicarerecipients 65 yearsor older living in theNorwood areaeligible; (2)Patients who wereconsidered "highrisk" by baselineRAP questionnaire(1) Patientsconsideredreluctant ordifficult; (2)Those whodied ormovedduringidentification andassignmentStudyDesignRCT: 6-11parallel, monthsnotclusteredStudy FundingDuration Source(s)GovernmentBaseline% RuralBaselineAge – Mean(SD) orMedian(Range)Overall: 0 65-74 yearsG1: 48.4%G2: 48.1%75-84 yearsG1: 38.5%G2: 41.4%85 years andolderG1: 13.2%G2: 10.6%Baseline %FemaleOverall: NRG1: 76.9G2: 77.9Race/Ethnicity %Non-WhiteG1: 7.7G2: 6.7HispanicG1: 4.4G2: 7.7D-31


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaStaresinic G1: MTMet al., services2007 51provided as partof a MedicarePart D MTMprogram by anMTMCoordinator(non-clinicalstaff) and apharmacistG2: Usual careprovided toMTM-eligibleenrollees whochose not toparticipateTo ensure (1) PDPthat drugsprescribed tobeneficiariesareappropriatelyused tooptimizetherapeuticoutcomesand lower therisks ofadverse drugevents anddruginteractions.beneficiaries whofill at least two PartD covered drugsfor ≥2 chronicdiseases ofinterest, including,but not limited to,asthma, CD, CHF,diabetes,dyslipidemia, andHTN (specificdisease statesvaried by PDP); (2)Additionalindependenteligibility criteriaincluded anextrapolatedannual drug cost(set by theSecretary ofDHHS) of $4000 ormore by the end ofthe plan year.Long-term careresidents eligible.CMSmandatedexclusionsincludingany one offollowing:use of OTCdrugs,vitamins,drugs forcosmeticuse,medications to treatcold orcoughsymptoms,fertilityagents,DESIdrugs, anddrugs notcoveredunder PartD.StudyDesignStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Cohort NR Unspecified NR


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion ExclusionYearTrialNameandComparatorDescriptionsGoal Criteria CriteriaTaylor, G1:Byrd, andKrueger,2003 52Pharmaceuticalcare providedby pharmacistsG2: Standardcare withoutadvice orrecommenddationsgiven topatients orphysiciansPrevention,detection andresolution ofmedicationrelatedproblems inhigh-riskpatients in aruralcommunity.(1) Age ≥18years; (2)Receiving careat participatingclinic; (3)Identified as highrisk formedicationrelatedadverseevent (defined as≥3 of followingrisk factors: 5+medications, 12+doses per day,4+ medicationschanges in theprevious year,3+ concurrentdiseases, historyof medicationnoncompliance,presence ofdrugs thatrequiretherapeuticmonitoring)(1)Significantcognitiveimpairment;(2) Historyof missedoffice visits;(3)Schedulingconflicts; (4)Lifeexpectancyof


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaTouchette G1: MTM basicet al., (comprehensive2012 53medicationreview andDRPassessment)G2: MTMenhanced(MTM plus 2-page clinicalsummaryabstracted frompatient'smedical chart)G3: Usual care,consisting ofmedicationcounseling perclinic’s normalroutine but noformal MTMfrom a studypharmacistTo improvethe safety ofmedication byreducingADEs andDRPs; Alsoto reducehealth care,especially EDvisits.1) Age ≥65 years; 1)(2) Primary use ofEnglish for writtenand oralcommunication; (3)Telephone accessfor the study’sduration; (4)Presence of ≥3comorbid chronicconditionsassociated withincreased healthcare use; (5) >2visits to clinicprovider duringprevious year; (6)>6 chronicprescriptionmedications during6 months beforeenrollment; (7) >1recent situationsplacing patient athigher risk of DRP(i.e., ≥3 differenthealth careproviders visited inthe last 12 months;any medicationchange, newphysician visit, EDvisit,hospitalization; orinvasive procedure[requiring stoppingmedications] inprevious 30 days).Presenceof aterminalconditionwith lifeexpectancyof 6months orless;2) Previousenrollmentin MTMprograminvolvingcomprehensivemedicationreview inprevious 12months.StudyDesignRCT:parallel,notclusteredD-34Study FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)6 months Government NR Overall: 74.6(6.7)G1: 74.5 (6.6)G2: 74.8 (6.8)G3: 74.6 (6.8)Baseline %FemaleOverall: 66.2G1: 63.0G2: 67.0G3: 68.3Race/Ethnicity %BlackOverall: 51.2G1: 48.3G2: 49.1G3: 56.3HispanicOverall: 4.4G1: 6.2G2: 2.3G3: 4.8AsianOverall: 0.8G1: 0.5G2: 0.9G3: 1.0AmericanIndianOverall: 0.3G1: 0G2: 0G3: 1.0


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaVolume etal., 2001 54 ;Kassam etal., 2001 55PREPG1:Comprehensivepharmaceutical care servicesusing a ninestepprocessas defined byHepler andStrandprovided bycommunitypharmacists.G2: Traditionalpharmacy careAuthorsdescribe thegoal ofpharmaceuticalcareas the"improvementofpatientoutcomesand quality oflife." Theyadd studyobjective tobe todescribechanges inintermediateand primaryoutcomesafter theprovision ofpharmaceuticalcare.Pharmacies: 1)Participation ofpharmacists working >8hours a weekdispensing medications;2) Agreement toparticipate in practiceenhancement program;3) Agreement toconform withprofessional standardsdeveloped by AlbertaPharmaceuticalAssociation; 4) AlbertaBlue Cross Billingsrepresented at leastone-third of pharmacybillings; 5) Located≤200 miles ofEdmonton.Patients: (1) >65 years;(2) Prescriptionmedication coverageunder Alberta Healthand Wellness' SeniorHealth Plan; (3) Use ≥3medicationsconcurrently; (4)Residing in Alberta for12 of 15 study months;(5) Agree to receiveprescriptionmedications only fromstudy pharmacy duringstudy periodStudyDesignPatients: 1) RCT:Individuals cluster-with rando-mizedterminaldisease; 2)Could notcommunicate inEnglish; 3)Could notcompletetelephoneinterviews.Study FundingDuration Source(s)12 to 13monthsMultiple NR(Government,foundation,andpharmaceutical)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleMean (SD) Mean (SD)Overall: 74 Overall: NR(NR) G1: 63.5G1: 73.9 (6.1) G2: 69.6G2: 73.2 (6.1)Race/Ethnicity%NRD-35


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaWelch etal., 2009 56G1: MTMprogramprovided tohome-basedbeneficiaries aspart of aMedicare PartD MTMprogramG2: No-MTMcontrol group(voluntary optout)To reducemortality,inpatienthospitalizations, ED visits,and Part D-coveredmedicationcosts1) MTM-eligibleKPCObeneficiaries;1) Had ≥2 chronicconditions, one ofwhich wasconsidered highrisk;2) Receiving 5 ormore Part D–coveredmedications;3) Likely to incur atleast $4000 in totalcosts for Part D–coveredmedications.KPCObeneficiaries with endstagerenaldisease(ESRD)StudyDesignStudy FundingDuration Source(s)Cohort 180 days Integratedhealth caresystem (KaiserPermanenteColorado)Baseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Mean (SD)Overall: NRG1: 68.8(10.7)G2: 68.9(11.3)p=0.949Baseline %FemaleOverall: NRG1: 56.6G2: 54.5p=0.541Race/Ethnicity %NRD-36


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaWilliams etal., 2004 57G1: Medicationreview andoptimization ofpatient'smedicationregimenconducted byaninterdisciplinarymedicationadjustmentteam in additionto usualmedical careand "Bound forHealth" booklet.G2: Usualmedical careplus provisionof "Bound forHealth" bookletTo promoteregimenchanges tosimplifyregimens inelders takingmultiplemedicationsand to seewhetherthesechangesimprovedfunctioning.1) Age ≥65 years;2) Cognitivelyintact (no evidenceof dementia orcognitivedysfunction in themedical record); 3)Minimum of 5prescriptionmedications, ofwhich 2 had to bepotentiallyproblematic forgeriatric patients.NRStudyDesignRCT:parallel,notclusteredStudy FundingDuration Source(s)Baseline% RuralBaselineAge – Mean(SD) orMedian(Range)6 weeks Unspecified NR G1: 73.5 (5.9)G2: 73.9 (5.6)Baseline %FemaleG1: 65.1G2: 50.6Race/Ethnicity %WhiteG1: 79.4G2: 76.6Non-WhiteG1: 20.6G2: 23.4D-37


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaWinston G1: MTMand Lin, provided in a2009 58communitypharmacy (i.e.,care in face-tofacemeetingsor bytelephone) aspart of aMedicare PartD MTMprogramG2: MTMprovided bypharmaciststaffedcallcenters as partof a MedicarePart D MTMprogramG3: Educationalmailings (i.e.,mailed lettercontainingpatient-specificmedicationrelatedinformation,personalmedicationrecord, and tipsto save moneyonprescriptions)Describeexperienceswith MTMservicesdelivered tobeneficiariesof Mirixa'shealth planclientsPatients whoqualified for MTMservices betweenApril 1, 2007 andJune 30, 2007.MTM qualificationdetermined byeach participatinghealth plan;generally patientswho had increasedcardiovascular riskdue to diabetesand HTN and/ordyslipidemiaStudyDesign(1) Patients Cohort Unclearwhosecoveragewasdiscontinued for anyreason; (2)Patientswhoreceivedadditionalpharmacist-providedservices(i.e.,formularyreview orother MTM)duringinterventionperiod.Study FundingDuration Source(s)Private MTM NRand pharmacydeliveredservice providerBaseline% RuralBaselineAge – Mean(SD) orMedian(Range)Overall: NRG1: 67.4(13.1)G2: 67.8(12.8)G3: 66.5(13.4)Baseline %FemaleOverall: NRG1: 70.4G2: 70.5G3: 69.5Race/Ethnicity %NRD-38


Table D1. Study and patient-level characteristics (continued)Author, Interventions Intervention Inclusion Criteria ExclusionYearTrialNameandComparatorDescriptionsGoalCriteriaWitry, G1: PCMDoucette, provided byand communityGainer, pharmacists to2011 59Iowa MedicaidenrolleesG2: PCMprovided bycommunitypharmacists topatients withprivateindividual-groupinsuranceTo decreasethe risk ofDRPs(1) Patients ≥1 Seechronic condition inclusion(i.e., who filled a criteriamedicationcommonly used totreat 1 of 12chronic conditions,as defined byMedicaid PCMprogram, at leasttwice during 3months prior toscreening date);(2) Must have filled≥4 unique,nontopicalmedications during3 months prior toscreening date;(3) Patrons ofstudy pharmacies,meaning that ≥50%of patients’prescription claimswere paid to thosepharmaciesStudyDesignCohort 21monthsStudy FundingDuration Source(s)Foundation ornon-profitBaseline% RuralNRBaselineAge – Mean(SD) orMedian(Range)Baseline %FemaleMean (SD) Mean (SD)Overall: NR Overall: NRG1: 54.1 (0.8) G1: 80G2: 58.9 G2: 68.1(7.51)Race/Ethnicity %NRD-39


Table D1. Study and patient-level characteristics (continued)Author, YearIntervention Inclusion CriteriaTrial NameGoalInterventions andComparatorDescriptionsWittayanukorn, G1:2013 60 Interventiongroup:Pharmacistprovidedface-to-faceMTMservices for30-60minutes perencounter,not alwaysincluding afollow-up visitG2: Controlgroup:Patients whodid notreceive MTMservices(economicanalysesonly)To improvecardiovascularoutcomesand improvedirect andindirect costsamongemployeesworking forlarge, selfinsuredemployers.ExclusionCriteriaIG: Medical claims Age


Table D1. Study and patient-level characteristics (continued)Author, YearTrial NameInterventions andComparatorDescriptionsInterventionGoalInclusionCriteriaExclusionCriteriaYamada 2012 61G1: Kaiser-PermanenteMTM enrolledpatientsG2: Kaiserpatientsenrolled inMedicare partD, but not inMTMprogrammatched tocontrol onage, gender,region andDCG riskNot stated,explicitlystated, but thiswas anevaluation ofan existingMTM program.InterventionGroup: Enrolledin the MTMprogrambetween Jan2006 and Dec2010. Receiveda CMR. Met allmatchingcriteria.For bothgroups:Cancerdiagnosiswithin oneyear of studyentryResiding in anursing homefor more than20 days, orless than 30days at indexdate and arepeat staywithin oneyear.Gap of morethan 2 monthsof membershipcoveragewithin 12months priorto index date.StudyDesignCohort Up to 4yearsStudy FundingDuration Source(s)OtherBaseline% RuralOverall:NRG1: NRG2: NRBaselineAge – Mean(SD) orMedian(Range)Overall: 75(8)G1: NRG2: NRBaseline %FemaleOverall: 58G1: NRG2: NRRace/Ethnicity %Overall: NRG1: NRG2: NRAbbreviations: ADE = adverse drug event; ART = antiretroviral therapy; BP = blood pressure; CAD = coronary artery disease; CDC = comprehensive diabetes care; CHF =chronic heart failure; CMR = comprehensive medication review; CMS = Centers for Medicare and Medicaid Services; COPD = chronic obstructive pulmonary disease; CVD =cardiovascular disease; DBP = diastolic blood pressure; DHHS = Department of Health and Human Services; DM = diabetes mellitus; DRP = drug-related problem; DTP = drugtherapy problem; DVA = Department of Veterans’ Affairs; ED = emergency department; ESRD = end-stage renal disease; FHCP = Federal Hazard Communication Program;FOM = Focus on Medicines; G = group; GP = general practitioner; HbA1c = hemoglobin A1c; HF = heart failure; HMO = health maintenance organization; HMR = homemedication review; HTN = hypertension; ICD-9-CM = International Classification of Diseases, Ninth Revision, Clinical Modification; IG = intervention group; IQR = interquartilerange; KPCO = Kaiser Permanente Colorado; LDL-C = low- density lipoprotein- cholesterol; MA-PD = Medicare Advantage Part D; MCO = managed care organization; Mg =milligram; mm Hg = milligrams mercury; mmoL = millgrams per liter; MTM = medication therapy management; MTMP = Part-D medication therapy management; NR = notreported; NRCT = non-randomized controlled trial; OTC = over-the-counter; PCM = pharmaceutical case management; PCP = primary care provider; PDP = Medicare Part DPlan; PREP = Pharmaceutical Care Research and Education Project; QOL = quality of life; RCT = randomized controlled trial; SBP = systolic blood pressure; SD = standarddeviation; VA = Veterans’ Administration.D-41


Table D2. Other patient-level and clinical characteristicsAuthor, YearTrial NameInterventionsand ComparatorDescriptionsStudy Design Other BaselineCharacteristicsBernsten et al., G1: Structured2001 1 ; communitySturgess et al., pharmacy-based2003 2 pharmaceuticalcare programG2: NormalpharmaceuticalUsual communitypharmacyservicesBlakey, 2000 3G1: Pharmacistevaluation plususual medicalcareG2: Usual medicalcareRCT: clusterrandomizedMeasure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsPooled samplePatients living alone(%)Overall: NRG1: 37.2G2: 37.7Patients requiring helpwith daily activities (%)Overall: NRG1: 50.9G2: 47.4Northern IrelandPatients living alone(%)Overall: NRG1: 30.9G2: 26.9Patients requiring helpwith daily activities (%)Overall: NRG1: 43.1G2: 55.7NR NR Pooled sampleOverall: NRG1: 7.1 (2.5)G2: 7.0 (2.5)p=NSNorthern IrelandOverall: NRG1: 5.9 (1.9)G2: 6.7 (1.9)p


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudy Design Other BaselineCharacteristicsInterventionsand ComparatorDescriptionsBrummel, G1: Fairview2013 4: ; Soliman, Pharmacy2013 5 ; Ramalho Services' MTMde Oliveira, program (opt-in)2010 6 G2: control group(did not opt-in)Carter et al., G1:1997 7 ;Barnette et al.,1996 8Pharmaceuticalcare provided bypharmacistswithin aninterdisciplinarypractice model.Patient education(lifestyle, riskfactormodifications, anddrug therapy) wasstandardized.G2: Usual careCohort% MedicareG1: 12.73G2: 20.59% MedicaidG1: 5.45G2: 1.96Measure of Co-MorbidityCharlson indexscoreG1: 3.7G2: 2.72p < 0.001% with diabeticcomplicationsG1: 95.04G2: 15.53p< 0.001% with insulintherapyG1: 53.33G2: 34.95P=0.005Cohort NR N of comorbidconditionsOverall: NRG1: 3.5 (2.4)G2: 3.2 (2.0)p=0.47DiagnosedConditions orDiseases (%)DiabetesG1: 100G2: 100No. (%) withcontrolled bloodpressure at baselineBaseline Numberof PrescribedMedications% on statins +othersG1: 36.36G2: 23.3p=0.03% on ACE/ARB+othersG1: 56.20G2: 58.25Overall: NRG1: 13 (52)G2: 14 (54)Other PatientClinicalCharacteristicsNRNRD-43


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityChrischilles etal., 2004 9Christensen etal., 2007 10G1: PCMprovided bypharmacistsG2: Did notreceive PCMservicesG1: MTM servicesdesigned by ahealth plan for itsbeneficiaries andprovided by eithercommunitypharmacists ormedical clinicbasedpharmacists.G2: Patients fromsame counties asG1 who did notreceiveintervention(control group 1)G3: Patients froma different countythan G1 who didnot receiveintervention(control group 2)DiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsCohort NR NR NR Overall: NRG1: 7.5 (0.2)G2: 6.9 (0.1)NRCT NR NR Patients youngerthan 65:HypertensionG1: 48.1G2: 47.9G3: 46.4>1 ConditionG1: 42.8G2: 34.0G3: 38.3DiabetesG1: 37G2: 31.7G3: 37.7Patients older than65:HypertensionG1: 62.5G2: 41.5G3: 48.5CardiovascularDiseaseG1: 55.0G2: 48.4G3: 50.2>1 ConditionG1: 46.3G2: 39.7G3: 39.9DiabetesG1: 45.0G2: 36.8G3: 33.7Patients youngerthan 65:G1: 40.3 (15.3)G2: 37.2 (17.5)G3: 36.9 (17.3)Patients older than65:G1: 41.7 (16.3)G2: 38.4 (16.3)G3: 41.7 (16.2)Other PatientClinicalCharacteristicsNRDifferences in %with selectedconditions and innumber of baselinemedications werenot significantamong the threegroups.D-44


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsClifford et al., G1:2002 11 Pharmaceuticalcare provided bya clinicalpharmacist ,which included acomprehensivereview relating topharmacotherapyand diabetes, useof proprietary andnon-proprietarymedications, suchas complementarymedicines, andidentification ofdrug therapyproblems.G2: Standardoutpatient care fordiabetesRCT:parallel,notclusteredMeasure of Co-MorbidityDiagnosedConditions orDiseases (%)NR NR Type 1 or 2DiabetesOverall: 100G1: 100G2: 100Type 1 DiabetesOverall: NRG1: 29.2G2: 20.0Type 2 DiabetesOverall: NRG1: 70.8G2: 80.0Hypertension: NRDyslipidemia: NRBaseline Numberof PrescribedMedicationsNROther PatientClinicalCharacteristicsNRD-45


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsFischer et al., G1:NRCT2000 12 Pharmaceuticalcare based on theEncara PracticeSystem providedby onsite healthmaintenanceorganization staffpharmacists.G2: StandardCommunityPharmacyPracticeG3: A set ofrefusers surveyedand included insome analysesamong those whowere at eligibleclinics but initiallydeclined toparticipate.Fischer et al., Pharmaceutical NRCT2002 13 care based on theEncara PracticeSystem providedby pharmacists.Communication ofpharmacist withthe patient'sphysician aboutdrug therapyproblemsidentified by thepharmacist.G2: Usual carewith no additionalinterventions.% MarriedOverall: NRG1: 68G2: 71G3: 72% Education < HSOverall: NRG1: 9G2: 18G3: 20% Income < 10KOverall: NRG1: 3G2: 9G3: 9Annual health care chargesOverall: NRG1: $9,600G2: $11,000Measure of Co-Morbidity% in Fair or PoorHealthOverall: NRG1: 28G2: 26G3: 35Charlson IndexG1: 1.2G2: 1.3DiagnosedConditions orDiseases (%)% Heart/HTNproblemsOverall: NRG1: 68G2: 61G3: 65% Asthma/LungProblemsOverall: NRG1: 49G2: 52G3: 42Heart disease (%)Overall: NRG1: 43G2: 40Baseline Numberof PrescribedMedicationsOverall: NRG1: 5.2G2: 4.6G3: 4.3Overall: NRG1: 9.1G2: 9.4Other PatientClinicalCharacteristicsMean N nonprescriptionmeds:Overall: NRG1: 2.2G2: 1.8G3: 1.7NRD-46


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsFox et al., G1: Florida Health2009 14 Care Plans MTMprogram,consisting of amedicationtherapy reviewand evaluation bya clinicalpharmacist thatwas documentedand sent to thepatient's physicianthrough healthplan reviewG2: Opt-out fromMTM programGattis et al., G1: Clinical1999 15 pharmacyservices, includingan assessment ofprescribedregimen,compliance, andadverse effects,and symptomsand response totherapy. Providingpatient educationabout the purposeof each drug andreinforcingadherence.Detailed writteninformation wasalso provided topatients.G2: Usual medicalcareMeasure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsCohort NR NR Diabetes: 100 Number of PMPM in2007Overall: NRG1: 9.4G2: 8.8RCT:parallel,notclusteredNR NR Heart FailureOverall: 100G1: 100G2: 100Overall: NRG1: 6.5 (25%: 5,75%: 8)G2: 6 (25%: 4.5,75%: 8)Other PatientClinicalCharacteristicsNRNRD-47


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsGrymonpre, G1:2001 16 Comprehensivedrug therapyreview, thenissues addressedwith the clientand/or the client'sphysician, withfollow-up asrequired.G2:Comprehensivedrug therapyreview only withreferral to usualpharmacist.RCT:parallel,notclusteredEducationOverall: NRG1:


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsHanlon, 1996 17Cowper, 1998 18 G1:Pharmaceuticalcare provided bya clinicalpharmacistG2: Usual care inthe GeneralMedicine ClinicHirsch, 2011 19Hirsch, 2009 20Rosenquist,2010 21(methods)G1: Patientsserved at nonpilotpharmaciesG2: Patientsserved at pilotpharmaciesRCT: Married (%)parallel, Overall: NRnot G1: 65.7clustered G2: 85.4Mean years of education(SD)Overall: NRG1: 10.2 (3.8)G2: 9.9 (4.2)Measure of Co-MorbidityN of chronicconditionsOverall: NRG1: 9.2 (3.7)G2: 9.0 (3.0)Cohort NR Overall: NRG1: NRG2: NRDiagnosedConditions orDiseases (%)NROverall: NRG1: NRG2: NRBaseline Numberof PrescribedMedicationsOverall: NRG1: 7.6 (2.8)G2: 8.2 (2.7)These were limitedto medicationsprescribed by a VAphysician.Overall: NRG1: 14.46 (5.16)G2: 14.79 (5.34)G3: 13.97 (5.74)Other PatientClinicalCharacteristics% of medications forwhich compliantOverall: NRG1: 73%G2: 74%NRD-49


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsIsetts et al., G1: MTM services2008 22 provided by staffpharmacists,including theestablishment ofgoals of therapy,in collaborationwith primary careproviders.G2: Usual medicalcare without MTMJameson, PharmacotherapyVanNoord, and consultation andVanderwoud, followup provided1995 23 by clinicalambulatory carepharmacist.G2: Standardoffice-basedJeong, 2007 24Jeong, 2009 25primary care.G1: KaiserPermanente 2006pharmacistmanagedMTMPG2: Patientswithout MedicarePart D as theirprimary drugbenefit and likelyto incur drug costsgreater than orequal to $4000per year with asimilar diseaseburdenMeasure of Co-MorbidityCohort NR Mean Number ofConditionsOverall: NRG1: 6.4 (NR)G2: NRRCT: NRparallel,notclusteredMore than 3 chronicdiseases Overall:NRG1: 70%G2: 76%DiagnosedConditions orDiseases (%)NRCohort NR NR DiabetesG1: 48G2:51NRHTNG1:86G2:83CAD OR DMG1: 54G2: 59Baseline Numberof PrescribedMedicationsOverall: NRG1: 14% were age65 or olderG2: NR5 or more long-termmedications (%)Overall: NRG1: 89G2: 90NROther PatientClinicalCharacteristicsThese variableswere not reportedfor the HEDIScomparison groupother than astatement that says"…were similar tointervention grouppatients in terms ofage, gender, andpresence of studymedical conditions."(bottom of page205)NRNRD-50


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator Design CharacteristicsMorbidityJeong 26Jeong, 2012 27DescriptionsG1: Kaiser-Permanente MTMprogramparticipants(2010)G2: Kaiser-Permanentepatients eligiblefor MTM, but whodeclinedenrollment ordisenrolled with aPCP visit duringfirst half of 2010G3: Kaiser-Permanentepatients eligiblefor MTM, but whodeclinedenrollment ordisenrolledwithout a PCPvisit during firsthalf of 2010Krska et al., G1: Medication2001 28 reviews led byclinically-trainedpharmacists.G2: Usual careinvolvinginterviews andidentification ofpharmaceuticalcare issues butwith nopharmaceuticalcare planimplemented.DiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsCohort NR NR NR Overall: NRG1: 14.46 (5.16)G2: 14.79 (5.34)G3: 13.97 (5.74)RCT: NRparallel,notclusteredOverall: NRG1: 3.9 (1.4)G2: 3.8 (1.4)p=0.968NRRepeat medicineson computer recordsOverall: NRG1: 7.4 (2.7)G2: 7.7 (2.8)p: 0.951Other PatientClinicalCharacteristicsNRNRD-51


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator Design CharacteristicsMorbidityMalone, 2000 29Ellis, 2000 30Malone, 2001 31Ellis, 2000 32IMPROVEDescriptionsG1:Pharmaceuticalcare provided byclinicalpharmacistspracticingaccording toscope of practicewithin theirrespective healthcare facilities.G2: Usual carewithoutpharmaceuticalcareRCT:parallel,notclustered% MarriedOverall: NRG1: 68.5G2: 67.8Mean number ofchronic conditionsOverall: NRG1: 4.0 (2.0)G2: 3.8 (1.9)DiagnosedConditions orDiseases (%)HypertensionOverall: NRG1: 68.5G2: 66.5AnginaOverall: NRG1: 46.1G2: 46.7HyperlipidemiaOverall: NRG1: 39.8G2: 43.1Baseline Numberof PrescribedMedicationsOverall: NRG1: 8.4 (4.4)G2: 8.0 (4.0)Other PatientClinicalCharacteristicsNRD-52


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityMarques, 2013 33 G1: Interventiongroup: Dadermethodpharmacotherapyfollow-upinterventionmonthly over 3-month follow-upperiodG2: Controlgroup: monthlypharmacist visitswithoutpharmacotherapyfollow-upinterventionRCT:parallel,notclusteredMarital statusOverall: NRG1:Married: 72.7%Not married: 27.3%G2:Married: 65.4%Not married: 34.6%SchoolingOverall: NRG1:Until 9 years: 59.1%>10 years: 40.9%G2:Until 9 years: 46.2%>10 years: 53.9%OccupationOverall: NRG1:Homemaker: 50.0%Other: 50.0%G2:Homemaker: 30.8%Other: 69.2%ReligionOverall: NRG1:Catholic: 40.9%Other: 59.1%G2:Catholic: 80.8%Other: 19.2%Overall: NRG1: NRG2: NRDiagnosedConditions orDiseases (%)Depression (mild,moderate, severe)Overall: NRG1:Mild:18.2%Moderate: 59.1%Severe: 22.7%G2:Mild: 30.8%Moderate: 50.0%Severe: 19.2%p=0.59Depression (firstepisode, relapse)Overall: NRG1:First episode: 31.8%Relapse: 68.2%G2:First episode: 23.1%Relapse: 76.9%p=0.53Baseline Numberof PrescribedMedicationsOverall: NRG1: 1.5 (0.6)G2: 1.7 (0.8)Other PatientClinicalCharacteristicsMean treatment time(on drugs) beforestarting study:G1: 60 daysG2: 30 daysp=0.53D-53


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityMarrufo, 2013 34 ,Perlroth, 2013 35CHFG1: enrolled inMedicare PDPreceiving MTMwith a CMRG2: enrolled inPDP receivingMTM, no CMRG3: enrolled inMA-PD receivingMTM with CMRG4: enrolled inMA-PD, receivingMTM, no CMRCOPDG5: enrolled inMedicare PDPreceiving MTMwith a CMRG6: enrolled inPDP receivingMTM, no CMRG7: enrolled inMA-PD receivingMTM with CMRG8: enrolled inMA-PD, receivingMTM, no CMRDiabetesG9: enrolled inMedicare PDPreceiving MTMwith a CMRG10: enrolled inPDP receivingMTM, no CMRG11: enrolled inPA-PD receivingCohortLIS eligibleG1: 71.1%G2: 44.1%G3: 22.4%G4: 36.7%G5: 77.0%G6: 54.1%G7: 25.0%G8: 38.9%G9: 68.0%G10: 42.8%G11: 22.5%G12: 36.1%G13: 52.1%G14: 37.0%G15: 56.1%G16: 41.7%G17: 56.4%G18: 36.5%DisabledG1: 27.9%G2: 15.6%G3: 9.8%G4: 14.0%G5: 37.1%G6: 25.4%G7: 12.2%G8: 18.0%G9: 31.0%G10: 18.3%G11: 10.7%G12: 15.3%G13: 18.5%G14: 14.6%G15: 27.1%G16: 20.4%G17: 24.3%G18: 16.7%NRDiagnosedConditions orDiseases (%)CHFG1-G4, G13, G14:100.0%COPDG5-G8, G15, G16:100.0%DiabetesG9-G12, G17, G18:100.0%Baseline Numberof PrescribedMedications≤8 MaintenanceDrugsG1: 8.2%G2: 17.4%G3: 15.5%G4: 20.6%G5: 11.5%G6: 23.5%G7: 22.5%G8: 30.3%G9: 16.8%G10: 29.7%G11: 27.6%G12: 34.7%G13: 20.8%G14: 26.3%G15: 28.7%G16: 34.7%G17: 29.0%G18: 33.0%9-10 MaintenanceDrugsG1: 17.4%G2: 22.4%G3: 21.9%G4: 23.4%G5: 19.1%G6: 21.9%G7: 22.3%G8: 22.8%G9: 26.6%G10: 27.9%G11: 28.1%G12: 27.8%G13: 27.0%G14: 29.3%G15: 25.9%G16: 27.9%Other PatientClinicalCharacteristicsNRD-54


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityMarrufo, 2013 34 ,Perlroth, 2013 35(continued)MTM with CMRG12: enrolled inMA-PD, receivingMTM, no CMRComparison -CHFG13: enrolled inPDP, usual careG14: enrolled inMA-PD, usualcareComparison -COPDG15: enrolled inPDP, usual careG16: enrolled inMA-PD, usualcareComparison -DiabetesG17: enrolled inPDP, usual careG18: enrolled inMA-PD, usualcareDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsG17: 34.0%G18: 35.7%11-12 MaintenanceDrugsG1: 22.0%G2: 22.1%G3: 24.4%G4: 23.0%G5: 21.9%G6: 19.8%G7: 22.5%G8: 19.3%G9: 24.0%G10: 20.5%G11: 22.4%G12: 19.7%G13: 22.4%G14: 21.4%G15: 19.7%G16: 18.8%G17: 19.4%G18: 18.2%>12 MaintenanceDrugsG1: 52.4%G2: 38.0%G3: 38.2%G4: 33.0%G5: 47.4%G6: 34.9%G7: 32.7%G8: 27.6%G9: 32.6%G10: 21.9%G11: 22.0%G12: 17.8%G13: 29.8%G14: 23.0%Other PatientClinicalCharacteristicsD-55


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsMarrufo, 2013 34 ,Perlroth, 2013 35(continued)Interventionsand ComparatorDescriptionsMcDonough et G1:al., 2005 36 Pharmaceuticalcare provided bycommunitypharmacists. Drugtherapymonitoringfocused on 5 drugtherapy problems:appropriatenessof does, properregimen, potentialinteractions,nonadherence,and adverseeffects. Patienteducation alsoprovided.G2: Usual careRCT:clusterrandomizedMeasure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsG15: 25.7%G16: 18.6%G17: 17.6%G18: 13.1%NR NR NR Overall:G1: 5.6 (3.1)G2: 7.0 (3.2)Other PatientClinicalCharacteristicsAt baseline, thetreatment group wassignificantly lesslikely to reportalcohol use andmore likely to bepost-menopausal.D-56


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityMoczygemba etal., 2011 37 ;Moczygemba etal., 2008 38Moore, 2013 40G1: Opt-intelephone-basedMTM program, inwhich MTMservices providedby clinicalpharmacists or amanaged carepharmacyresident based onthe AmericanPharmacistsAssociation andNationalAssociation ofChain DrugStores FoundationMTM framework.G2: No-MTMcontrol groupG1: MTM program(opt-in)G2: control group(refusers)Cohort NR Number of chronicdxMean (SD)Overall: NRG1: 6.5 (2.3)G2: 7.0 (2.1)p: 0.18DiagnosedConditions orDiseases (%)HypertensionOverall: NRG1: 95G2: 95DyslipidemiaOverall: NRG1: 77G2: 87DiabetesOverall: NRG1: 55G2: 60Cohort NR NR HypertensionG1: 87.6G2: 86.4DyslipidemiaG1: 67.3G2: 58.8DiabetesG1: 27.0G2: 30.4Baseline Numberof PrescribedMedicationsMean (SD)Overall: NRG1: 13.0 (3.2)G2: 13.2 (3.4)Mean (SE)G1: 55.3 (0.485)G2: 69.2 (0.656)p


Table D2. Other patient-level and clinical characteristics (continued)Author, YearStudy Design Other BaselineTrial NameCharacteristicsPai et al., G1:2009 41 ; Pai etal., 2009 42Interventionsand ComparatorDescriptionsPharmaceuticalcare includingdrug therapyreviewsconducted by anephrologytrainedclinicalpharmacist withthe patient. Alsoincluded patientand health careprovidereducation.G2: Standard ofcare, consisting ofbrief therapyreviewsconducted by anursePark et al., G1:1996 43 Comprehensivepharmaceuticalservices includingdrug therapymonitoring andpatient educationprovided by acommunitypharmacyresident.G2: Usual careRCT: clusterrandomizedRCT: parallel,not clusteredMean Time onHemodialysis in years(SD)Overall: 2.6 (2.0)G1: 2.8 (1.8)G2: 2.4 (2.2)Measure of Co-MorbidityDiagnosedConditions orDiseases (%)NR ESRD etiology -Diabetes mellitusOverall: 43.3G1: 38.6G2: 48.9ESRD etiology -HypertensionOverall: 28.9G1: 31.6G2: 25.5ESRD etiology -OtherOverall: 27.9G1: 29.8G2: 25.5Baseline Numberof PrescribedMedicationsOverall: 10 (4)G1: 10 (4)G2: 10 (4)Other PatientClinicalCharacteristicsNRNR NR NR NR Mean number ofantihypertensivesOverall: NRG1: 1.4G2: 1.3D-58


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsPindolia et al., G1: TelephonebasedMTM2009 44 services providedas part of aMedicare Part DMTM program bypharmacy caremanagementclinicalpharmacists(acceptors).G2: Usual medicalcare (opt-out)Cohort 2006 Part D type (%)Donut holeG1: 68G2: 60Nondonut hole or coveragegapG1: 6G2: 8Low income subsidyG1: 18G2: 24InstitutionalizedG1: 8G2: 7Overall p: 0.0542007 Part D type (%)Donut holeG1: 93G2: 63Nondonut hole or coveragegapG1: 1G2: 9Low income subsidyG1: 6G2: 26InstitutionalizedG1: 0G2: 2Overall p: 0.001Measure of Co-Morbidity2006N of qualifyingdiseases (mean,SD)Overall: NRG1: 5.9 (2.2)G2: 5.6 (2.1)p: 0.0472007N of qualifyingdiseases (mean,SD)Overall: NRG1: 5.8 (2.0)G2: 5.9 (2.0)Overall p: 0.701DiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsNR 2006Unique prescriptionsfilled (mean, SD)Overall: NRG1: 16.7 (7.2)G2: 14.8 (6.1)p: 0.0012007Unique prescriptionsfilled (mean, SD)Overall: NRG1: 14.4 (6.2)G2: 14.9 (6.2)p: 0.223Other PatientClinicalCharacteristicsNRD-59


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsPlanas et al., G1: MTM services2009 45 provided bycommunitypharmacists. Alsoincluded patienteducation on dietand lifestylemodifications tolower bloodpressure.G2: No MTMreceived, but onlyinformed of bloodpressure goals forpatients withdiabetesRoughead et al., G1: Home2009 46 MedicationReviews (HMR), acollaborativemodel ofpharmaceuticalcare, conductedby accreditedpharmacists.G2: Nomedication reviewreceivedRCT:parallel,notclusteredCohortOverweight (25-29.9 kg/m 2 ),%:Overall: NRG1: 15.6G2: 42.1p: NRObese (≥30 kg/m 2 ), %:Overall: NRG1: 68.8G2: 47.4p: NRSocioeconomic index ofdisadvantage (%)Lowest disadvantageOverall: NRG1: 31G2: 25Medium/low disadvantageOverall: NRG1: 25G2: 25Medium/high disadvantageOverall: NRG1: 24G2: 25Highest disadvantageOverall: NRG1: 20G2: 25Overall p: 0.01Measure of Co-MorbidityDiagnosedConditions orDiseases (%)NR Hypertension: 100Diabetes: 100N of co-morbidities(median, SD)Overall: NRG1: 8 (2)G2: 7 (2)p:


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudy Design Other BaselineCharacteristicsInterventionsand ComparatorDescriptionsSellors et al.,2003 47 G1: Clinicalpharmacistconsultationsprovided to familyphysicians andtheir patients bycommunitypharmacists.G2: Usual care forfamily physiciansand their patientsfrom matchedpostal codes.RCT: clusterrandomizedSellors, 2003 47 G1:PharmaceuticalconsultationG2: Usual careShimp, 2012 49G1: MTM programfor University ofMichiganbeneficiaries,entitled Focus onMedicinesG2: Usual care(not described)RCT: parallel,not clusteredRCT: parallel,not clusteredEducation: highestlevel attained (%)Elementary SchoolOverall: NRG1: 26.9G2: 24.1High school graduateOverall: NRG1: 50.8G2: 51.0Some collegeOverall: NRG1: 22.2G2: 24.9% married FPL/common-law spouseOverall: NRG1: 58.2G2: 63.1Measure of Co-MorbidityNRDiagnosedConditions orDiseases (%)HypertensionG1: 54.3G2: 55.7OsteoarthritisG1: 46.4G2: 48.3IHDG1: 36.0G2: 38.0Baseline Numberof PrescribedMedicationsNRNR NR NR NR NRNR NR Overall: NRG1:Dyslipidemia: 44%Hypertension: 31%Gastroesophagealreflux disease: 22%G2: NROverall: NRG1: 9.2 (3.2)G2: NROther PatientClinicalCharacteristicsNRNRD-61


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsSidel et al., G1: Home visits1990 50 by pharmacistsand, whenneeded,consultations withphysicians toidentify andcorrect problemsassociated withmedication use.G2: Standard carewithout any visitsor informationprovided to G1.RCT:parallel,notclusteredIncome (%)Overall: NRUnder $5000G1: 23.3G2: 22.2$5000-$15000G1: 61.0G2: 63.3>$15000G1: 15.9G2: 14.4Education: % with 9 or moreyearsOverall: NRG1: 62.2G2: 54.8% with Self-Assessed HealthFair or PoorOverall: NRG1: 44.0G2: 42.7% with Problems withActivities of Daily LivingOverall: NRG1: 33.0G2: 34.6% with Symptoms ofDepressionOverall: NRG1: 10.8G2: 22.6% with Cognitive ImpairmentOverall: NRG1: 15.4G2: 21.4Measure of Co-MorbidityNumber of medicalconditions (%)Overall: NRNoneG1: 3.3G2: 2.91-3G1: 58.2G2: 70.24 or moreG1: 38.5G2: 29.9DiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsNR Overall: 65.3%(mean 2.4, range 1-10)G1: NRG2: NROther PatientClinicalCharacteristicsNRD-62


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudyDesignOther BaselineCharacteristicsInterventionsand ComparatorDescriptionsG1: MTM servicesStaresinic et al.,Cohort Dual eligible (%)2007 51 provided as partof a Medicare PartD MTM programby an MTMCoordinator (nonclinicalstaff) anda pharmacistG2: Usual careprovided to MTMeligibleenrolleeswho chose not toparticipateG1: 6G2: 25Taylor, Byrd, G1:and Krueger,2003 52Pharmaceuticalcare provided bypharmacistsG2: Standard carewithout advice orrecommendationsgiven to patientsor physiciansRCT:parallel,notclusteredMedian years of education(Range)Overall: NRG1: 12 (4-16)G2: 12 (8-16)No insurance coverage forRx medicationsOverall: 17%G1: NRG2: NRMarital status: % marriedOverall: NRG1: 75.8G2: 72.2Measure of Co-MorbidityNRNRDiagnosedConditions orDiseases (%)Hypertension/CHFOverall: 96.1G1: 96.5G2: 96.0HyperlipidemiaOverall: 70.7G1: 75.9G2: 69.8DiabetesOverall: 51.2G1: 51.4G2: 51.1Hypertension:Overall: 51Dyslipidemia:Overall: 40Diabetes Mellitus:Overall: 27Baseline Numberof PrescribedMedicationsNRMean N ofmedications (SD)Overall: NRG1: 6.3 (2.2)G2: 5.7 (1.7)Other PatientClinicalCharacteristicsNRNRD-63


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudy Design Other BaselineCharacteristicsInterventionsand ComparatorDescriptionsTouchette et al., G1: MTM basic2012 53 (comprehensivemedication reviewand DRPassessment)G2: MTMenhanced (MTMplus 2-pageclinical summaryabstracted frompatient's medicalchart)G3: Usual care,consisting ofmedicationcounseling perclinic’s normalroutine but noformal MTM froma studypharmacistRCT: parallel,not clusteredNRMeasure of Co-MorbidityNumber ofcomorbiditiesOverall: 4.9 (1.6)G1: 5.0 (1.6)G2: 5.0 (1.6)G3: 4.9 (1.6)DiagnosedConditions orDiseases (%)HypertensionOverall: 90.9G1: 89.6G2: 90.8G3: 92.3DyslipidemiaOverall: 77.7G1: 76.3G2: 80.7G3: 76.0ArthritisOverall: 70.2G1: 68.2G2: 73.4G3: 68.8Baseline Numberof PrescribedMedicationsMean (SD)Overall: 8.0 (2.4)G1: 8.2 (2.6)G2: 7.7 (2.3)G3: 8.0 (2.3)Other PatientClinicalCharacteristicsNRD-64


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityVolume et al.,2001 54 ; Kassamet al., 2001 55PREP(PharmaceuticalCare Researchand EducationProject)G1:Comprehensivepharmaceuticalcare servicesusing a nine-stepprocess asdefined by Heplerand Strandprovided bycommunitypharmacists.G2: Traditionalpharmacy careRCT: clusterrandomizedAll Overall %’s NREducation (%)Some high schoolG1: 46G2: 50Completed high schoolG1: 17G2: 18Some trade school/collegeG1: 19G2: 17Completed collegeG1: 17G2: 14Annual income (%, CAD)< $20,000G1: 40G2: 40$20,000 - $39,000G1: 40G2: 43$40,000 - $59,000G1: 11G2: 11≥$60,000G1: 8G2: 5Living situation (%)Live aloneG1: 34G2: 29Live with spouse/partnerG1: 57G2: 61Live with other relativeG1: 7G2: 6Live with unrelated personG1: 2G2: 2Mean number ofconditions (SD)G1: 3.3 (1.7)based on studyinterview and 10(4.8) based ondata collected bytreatmentpharmacist.G2: NRDiagnosedConditions orDiseases (%)NRBaseline Numberof PrescribedMedicationsMean (SD)Overall: NRG1: 4.7 (2.8)G2: 3.9 (2.5)p < 0.05Other PatientClinicalCharacteristicsNRD-65


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudy Design Other BaselineCharacteristicsInterventionsand ComparatorDescriptionsWelch et al., G1: MTM program2009 56 provided to homebasedbeneficiaries aspart of a MedicarePart D MTMprogramG2: No-MTMcontrol group(voluntary opt-out)Measure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsOther PatientClinicalCharacteristicsCohort NR NR NR NR Mean ChronicDisease Score (SD)(ranges from 0-35,with larger scoresindicating increasingburden of chronicdiseases undertreatment)Overall: NRG1: 8.8 (3.1)G2: 8.2 (3.5)p: 0.016NOTE: Differencerepresents, onaverage, less thanone additionalchronic disease perpatientMedian (IQR)baseline medicationcost ($)G1: 3149 (2378 to4806)G2: 3186 (2363 to5123)Mean baselinemedication cost ($)(no SD reported)G1: 4465G2: 5197p: 0.525D-66


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameStudy Design Other BaselineCharacteristicsInterventionsand ComparatorDescriptionsWilliams et al.,2004 57 G1: Medicationreview andoptimization ofpatient'smedicationregimenconducted by aninterdisciplinarymedicationadjustment teamin addition tousual medicalcare and "Boundfor Health"booklet.G2: Usual medicalcare plusprovision of"Bound forHealth" bookletRCT: parallel,not clusteredEducation (%)Had not completedhigh schoolG1: 33.3G2: 32.5High school or somecollegeG1: 25.4G2: 19.5Completed collegeG1: 41.3G2: 48.1Marital status (%)MarriedG1: 47.6G2: 53.2Living AloneG1: 38.1G2: 33.8Measure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsNR NR G1: 6.6 (1.8) G2: 7.7(2.3)Other PatientClinicalCharacteristicsBaseline number ofnon-prescriptiondrugsG1: 5.1 (3.1)G2: 4.6 (2.5)D-67


Table D2. Other patient-level and clinical characteristics (continued)Author, YearTrial NameInterventionsand ComparatorDescriptionsStudy Design Other BaselineCharacteristicsMeasure of Co-MorbidityWinston andLin, 2009 58G1: MTMprovided in acommunitypharmacy (i.e.,care in face-tofacemeetings orby telephone) aspart of a MedicarePart D MTMprogramG2: MTMprovided bypharmaciststaffedcallcenters as part ofa Medicare Part DMTM programG3: Educationalmailings (i.e.,mailed lettercontaining patientspecificmedication relatedinformation,personalmedicationrecord, and tips tosave money onprescriptions)DiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsCohort NR NR NR Overall: NRG1: 9.8 (3.2)G2: 9.8 (2.9)G3: 9.7 (2.9)Other PatientClinicalCharacteristicsNRD-68


Table D2. Other patient-level and clinical characteristics (continued)Author, YearStudy Design Other BaselineTrial NameCharacteristicsInterventionsand ComparatorDescriptionsWitry, Doucette, G1: PCMand Gainer, provided by2011 59 communitypharmacists toIowa MedicaidenrolleesG2: PCMprovided bycommunitypharmacists topatients withprivate individualgroupinsuranceWittayanukorn, G1: Intervention2013 60 group: Pharmacistprovided face-tofaceMTMservices for 30-60minutes perencounter, notalways including afollow-up visitG2: Controlgroup: Patientswho did notreceive MTMservices(economicanalyses only)Measure of Co-MorbidityDiagnosedConditions orDiseases (%)Baseline Numberof PrescribedMedicationsCohort NR NR NR Mean (SD)Overall: NRG1: 7.9 (3.8)G2: 4.7 (2.2)p:


Table D2. Other patient-level and clinical characteristics (continued)Author, Year Interventions Study Other BaselineMeasure of Co-Trial Name and Comparator DesignDescriptionsCharacteristicsMorbidityYamada 2012 61G1: Kaiser-Permanente MTMenrolled patientsG2: Kaiserpatients enrolledin Medicare partD, but not in MTMprogram matchedto control on age,gender, regionand DCG riskCohort NR Mean Dx CG score:Overall: 1.8 (1.3)Mean Charlsoncomorbidity scoreG1: 2.1 (2.7)G2: 1.3 (2.0)p< 0.001Hospitalization rateG1: 27%G2: 22%p< 0.001DiagnosedConditions orDiseases (%)% with CHF:G1: 13G2: 7p< 0.001% with ESRD:G1: 20%G2: 13%p< 0.001% with DiabetesG1: 30%G2: 15%p< 0.001Baseline Numberof PrescribedMedicationsOverall: NRG1: NRG2: NROther PatientClinicalCharacteristicsDrug Costs:G1: $4,220G2: $921p


Table D3. Number of study participantsAuthor, Year Interventions andTrial Name Comparator DescriptionsBernsten, 2001 1 ;Sturgess, 2003 2Blakey, 2000 3Brummel, 2013 4: ;Soliman, 2013 5 ;Ramalho deOliveira, 2010 6Carter 1997 7 ,Barnette 1996 8G1: Structured communitypharmacy-basedpharmaceutical care programG2: Usual communitypharmacy servicesG1: Pharmacist evaluation plususual medical careG2: Usual medical careG1: Fairview PharmacyServices' MTM program (opt-in)G2: control group (did not optin)G1: Pharmaceutical careprovided by pharmacists withinan interdisciplinary practicemodel. Patient education(lifestyle, risk factormodifications, and drugtherapy) was standardized.G2: Usual careChrischilles, 2004 9 G1: PCM provided bypharmacistsG2: Did not receive PCMservicesStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)RCT: clusterrandomizedPooled sampleOverall: 2,454G1: 1,290G2: 1,164NRCT Overall: 178G1: 106G2: 72Cohort Overall: 248G1: 127G2:121CohortOverall: NRG1: NRG2: NRCohort Overall: 2,211G1: 524G2: 1,687Overall: 2,454G1: 1,290G2: 1,164Overall: 178G1: 106G2: 72Overall: 224G1: 121G2: 103Overall: 55G1: 29G2: 26Overall: 2,211G1: 524G2: 1,687N CompletersOverall: 1,340G1: 704G2: 636Overall: 178G1: 106G2: 72Overall: 224G1: 121G2: 103Overall: 51G1: 25G2: 26Overall: 2,211G1: 524G2:1,687N AnalyzedOverall: 1,340G1: 704G2: 636Overall: 178G1: 106G2: 72Overall: 224G1: 121G2: 103Overall: 51G1: 25G2: 26Overall: 2,211G1: 524G2: 1,687D-71


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)G1: MTM services designed byChristensen,NRCT2007 10 a health plan for itsbeneficiaries and provided byeither community pharmacistsor medical clinic-basedpharmacists.G2: Patients from samecounties as G1 who did notreceive intervention (controlgroup 1)G3: Patients from a differentcounty than G1 who did notreceive intervention (controlgroup 2)Clifford, 2002 11 G1: Pharmaceutical care RCT: parallel,provided by a clinicalnot clusteredpharmacist , which included acomprehensive review relatingto pharmacotherapy anddiabetes, use of proprietary andnon-proprietary medications,such as complementarymedicines, and identification ofdrug therapy problems.G2: Standard outpatient carefor diabetesFischer, 2000 12G1: Pharmaceutical care basedon Encara Practice Systemprovided by onsite healthmaintenance organization staffpharmacistsG2: Standard communitypharmacy practiceG3: Patients at eligible clinicswho declined to receiveintervention but were includedin some analyses.Overall: NRG1: 1,000G2: NRNRCT Overall: 1,051G1: NRG2: NROverall: 1,639G1: 80G2: 689G3: 870NR Overall: 73G1: 48G2: 25Overall: 748G1: 244G2: 504N CompletersOverall: 1,589G1: 30G2: 689G3: 870N AnalyzedOverall: 1,626G1: 67G2: 689G3: 870NR Overall: 73G1: 48G2: 25Overall: 578G1: 210G2: 368Overall: 578G1: 210G2: 368D-72


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Fischer, 2002 13Fox, 2009 14Gattis, 1999 15G1: Pharmaceutical care basedon Encara Practice Systemprovided by pharmacists.Pharmacist-physiciancommunication aboutpharmacist-identified DTPs.G2: Usual care with noadditional interventionsG1: Florida Health Care Plans(FHCP) MTM program,consisting of a medicationtherapy review and evaluationby a clinical pharmacist thatwas documented and sent tothe patient's physician throughhealth plan reviewG2: Opt-out from MTM programG1: Clinical pharmacy services,including an assessment ofprescribed regimen,compliance, and adverseeffects, and symptoms andresponse to therapy. Providingpatient education about thepurpose of each drug andreinforcing adherence. Detailedwritten information was alsoprovided to patients.G2: Usual medical careNRCTOverall before death,disenrollment, ordiscontinuation ofpharmacy benefits:1,070G1: 553G2: 517Overall after death,disenrollment, ordiscontinuation ofpharmacy benefits:921G1: 477G2: 444Cohort Overall: 311G1: 255G2: 56RCT: parallel,not clusteredOverall: 192G1: NAG2: NAOverall: 675 (enrolled+ control) or 921(intention-to-treat)G1: 231 (enrolled) or477 (intention-to-treat:enrolled +refused)G2: 444Overall: 311G1: 255G2: 56Overall: 181G1: 90G2: 91N CompletersOverall: 675 (enrolled+ control) or 921(intention-to-treat)G1: 231 (enrolled) or477 (intention-to-treat:enrolled +refused)G2: 444N AnalyzedOverall: 675 (enrolled+ control) or 921(intention-to-treat)G1: 231 (enrolled) or477 (intention-to-treat:enrolled +refused)G2: 444NR Overall: 311G1: 255G2: 56Overall: 181G1: 90G2: 91Overall: 181G1: 90G2: 91D-73


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Grymonpre, G1: Comprehensive drug2001 16 therapy review, then issuesaddressed with the client and/orthe client's physician, withfollow-up as required.G2: Comprehensive drugtherapy review only with referralto usual pharmacistHanlon, 1996 17Cowper, 1998 18Hirsch, 2011 19Hirsch, 2009 20Rosenquist,2010 21 (methods)G1: Pharmaceutical careprovided by a clinicalpharmacistG2: Usual care in the GeneralMedicine ClinicG1: Patients served at nonpilotpharmaciesG2: Patients served at pilotpharmaciesRCT: parallel,not clusteredRCT: parallel,not clusteredOverall: 190G1: NRG2: NROverall: 228G1: NRG2: NRCohort Overall: 2,234G1: 132G2: 2,102Overall: 131G1: 69G2: 66Overall: 208G1: 105G2: 103Overall: 2,234G1: 132G2: 2,102N CompletersOverall: 114G1: 56G2: 58Overall: 172G1: 88G2: 84Overall: 2,234G1: 628G2: 1,606N AnalyzedOverall: 131G1: 69G2: 66Overall: 208G1: 105G2: 103Overall: 2,2342005G1: 439G2: 1,7952006G1: 617G2: 1,617Isetts, 2008 22G1: MTM services provided bystaff pharmacists, including theestablishment of goals oftherapy, in collaboration withprimary care providers.G2: Usual medical care withoutMTMCohortOverall: NRG1: 2,834G2: NROverall: NRG1: 285G2: NROverall: NRG1: NRG2: NR2007G1: 628G2: 1,606For goals of drugtherapy and numberof DTPs resolvedOverall: 541G1: 285G2: 256For HEDIS outcomesOverall: 154G1: 128G2: 126D-74


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Jameson, 1995 23 G1: Pharmacotherapyconsultation and followupprovided by clinical ambulatorycare pharmacist.G2: Standard office-basedprimary care.Jeong, 2007 24 G1: Kaiser Permanente 2006Jeong, 2009 25 pharmacist-managed MTMPG2: Patients without MedicarePart D as their primary drugbenefit and likely to incur drugcosts greater than or equal to$4000 per year with a similardisease burden.Jeong 26Jeong, 2012 27Krska, 2001 28G1: Kaiser-Permanente MTMprogram participants (2010)G2: Kaiser-Permanentepatients eligible for MTM, butwho declined enrollment ordisenrolled with a PCP visitduring first half of 2010G3: Kaiser-Permanentepatients eligible for MTM, butwho declined enrollment ordisenrolled without a PCP visitduring first half of 2010G1: Medication reviews led byclinically-trained pharmacists.G2: Usual care involvinginterviews and identification ofpharmaceutical care issues butwith no pharmaceutical careplan implemented.RCT: parallel,not clusteredCohort Overall: 5,031G1: 2,780G2:2,251Cohort Overall: 39, 680G1: NAG2: NAG3: NARCT: parallel,not clusteredNR Overall: 64G1: 34G2: 30Overall: 420G1: NAG2: NAOverall: 5,031G1: 2,780G2: 2,251Overall: 39,680G1: 23,638G2: 14,232G3:1,810Overall: 381G1: 192G2: 189N CompletersOverall: 56G1: 27G2: 29Overall: 5,031G1: 2,,780G2: 2251Overall: 39,680G1: 23,638G2: 14,232G3:1,810Overall: 332G1: 168G2: 164N AnalyzedOverall: 56G1: 27G2: 29For HbA1c analysisOverall: 2,464G1: 1,323G2: 1,141For LDL analysisOverall: 2,838G1: 1,515G2: 1,323For BP analysisOverall: 2,283G1: 1,301G2: 982Overall: 39,680G1: 23,638G2: 14,232G3:1,810Overall: 332G1: 168G2: 164D-75


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Malone, 2000 29 ;Ellis, 2000 30(interventions);Malone, 2001 31(detailed QOLoutcomes);Ellis, 2000 32(dyslipidemiasubgroupintermediate andutilizationoutcomes);Marques, 2013 33G1: Pharmaceutical care RCT: parallel,provided by clinical pharmacists not clusteredpracticing according to scope ofpractice within their respectivehealth care facilities.G2: Usual care withoutpharmaceutical careG1: Intervention group: Dadermethod pharmacotherapyfollow-up intervention monthlyover 3-month follow-up periodG2: Control group: monthlypharmacist visits withoutpharmacotherapy follow-upinterventionRCT: parallel,not clusteredNR Overall: 1,054G1: 523G2: 531Overall: 58G1: NAG2: NAOverall: 58G1: 31G2: 27N CompletersOverall: 931G1: 447G2: 484Overall: 48G1: 26G2: 22N AnalyzedOverall: 1,043G1: 523G2: 531Overall: 48G1: 26G2: 22D-76


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions and ComparatorDescriptionsStudy Design N EligibleMarrufo, 2013 34 ,Perlroth, 2013 35CHFG1: enrolled in Medicare PDPreceiving MTM with a CMRG2: enrolled in PDP receivingMTM, no CMRG3: enrolled in MA-PD receivingMTM with CMRG4: enrolled in MA-PD, receivingMTM, no CMRCOPDG5: enrolled in Medicare PDPreceiving MTM with a CMRG6: enrolled in PDP receivingMTM, no CMRG7: enrolled in MA-PD receivingMTM with CMRG8: enrolled in MA-PD, receivingMTM, no CMRDiabetesG9: enrolled in Medicare PDPreceiving MTM with a CMRG10: enrolled in PDP receivingMTM, no CMRG11: enrolled in PA-PD receivingMTM with CMRG12: enrolled in MA-PD, receivingMTM, no CMRComparison - CHFG13: enrolled in PDP, usual careG14: enrolled in MA-PD, usualcareComparison - COPDG15: enrolled in PDP, usual careG16: enrolled in MA-PD, usualcareComparison - DiabetesG17: enrolled in PDP, usual careG18: enrolled in MA-PD, usualcareN Randomized(trials) or N Enrolled(observationalstudies)N CompletersN AnalyzedCohort NR NR NR G1: 12,658G2: 103,080G3: 11,260G4: 62,983G5: 16,372G6: 110,042G7: 10,575G8: 64,637G9: 16,545G10: 149,803G11: 13,527G12: 95,299G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912D-77


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)McDonough,2005 36 G1: Pharmaceutical careprovided by communitypharmacists. Drug therapymonitoring focused on 5 drugtherapy problems:appropriateness of does, properregimen, potential interactions,nonadherence, and adverseeffects. Patient education alsoprovided.G2: Usual careRCT: clusterrandomizedMoczygemba, G1: Opt-in telephone-based2011 37MTM program, in which MTMMoczygemba, services provided by clinical2008 38pharmacists or a managed careMoczygemba, pharmacy resident based on2012 39Moore, 2013 40the American PharmacistsAssociation and NationalAssociation of Chain DrugStores Foundation MTMframework.G2: No-MTM control groupG1: MTM program (opt-in)G2: control group (refusers)Pai, 2009 41 ; Pai,2009 42 G1: Pharmaceutical careincluding drug therapy reviewsconducted by a nephrologytrainedclinical pharmacist withthe patient. Also includedpatient and health care providereducation.G2: Standard of care,consisting of brief therapyreviews conducted by a nurseRCT: clusterrandomizedOverall: 163G1: NRG2: NRCohort Overall: 1,971G1: 95G2: 1,876Cohort Overall: 13,092G1: 2,966G2: 10,126Overall: 96G1: 70G2: 26Overall: 132G1: 72G2: 60Overall: 8,723G1: 2,260G2: 6,463NR Year 1Overall: 104G1: 57G2: 47Year 2Overall: 107G1: 61G2: 46N CompletersOverall: 80G1: 61G2: 19N AnalyzedOverall: 80G1: 61G2: 19NR Overall: 120G1: 60G2: 60Overall: 8,723G1: 2,260G2: 6,463Year 1Overall: 80G1: 44G2: 36Year 2Overall: 46G1: 24G2: 22Overall: 4,500G1: 2,250G2: 2,250Year 1Overall: 80G1: 44G2: 36Year 2Overall: 46G1: 24G2: 22D-78


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Park, 1996 43Pindolia, 2009 44Planas, 2009 45G1: Comprehensivepharmaceutical servicesincluding drug therapymonitoring and patienteducation provided by acommunity pharmacy resident.G2: Usual careG1: Telephone-based MTMservices provided as part of aMedicare Part D MTM programby pharmacy care managementclinical pharmacists(acceptors).G2: Usual medical care(refusers)G1: MTM services provided bycommunity pharmacists. Alsoincluded patient education ondiet and lifestyle modificationsto lower blood pressure.G2: No MTM received, but onlyinformed of blood pressuregoals for patients with diabetesRCT: parallel,not clusteredOverall: NRG1: NRG2: NRCohort Overall: 2,696G1: NAG2: NARCT: parallel,not clustered2006Overall: 1,388G1: NAG2: NA2007Overall: 1,308G1: NAG2: NAOverall: 52G1: 32G2: 20Overall: 64G1: 32G2: 322006Overall: 1,388G1: 307G2: 1,0812007Overall: 1,308G1: 228G2: 1,080Overall: 52G1: 32G2: 20N CompletersOverall: 53G1: 27G2: 26N AnalyzedOverall: 53G1: 27G2: 26NR 2006Overall: 1,373G1: 292G2: 1,081Overall: 33G1: 20G2: 132007Overall: 1,308G1: 228G2: 1,080Overall: 40G1: 25G2: 15D-79


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Roughead, 2009 46Sellors, 2003 47Sellors, 2003 47Shimp, 2012 49Sidel, 1990 50G1: Home Medication Reviews Cohort NR Overall: 5,717(HMR), a collaborative model ofpharmaceutical care, conductedby accredited pharmacists.G2: No medication reviewreceivedG1: 273G2: 5,444G1: Clinical pharmacist RCT: clusterrandomizedconsultations provided to familyphysicians and their patients bycommunity pharmacists.G2: Usual care for familyphysicians and their patientsfrom matched postal codesG1: PharmaceuticalconsultationG2: Usual careG1: MTM program forUniversity of Michiganbeneficiaries, entitled FOMG2: Usual care (not described)RCT: parallel,not clusteredRCT: parallel,not clusteredG1: Home visits by pharmacists RCT: parallel,and, when needed,not clusteredconsultations with physicians toidentify and correct problemsassociated with medication use.G2: Standard care without anyvisits or information provided toG1.Overall: 1,279G1: NRG2: NROverall: 191G1: NRG2: NROverall: 1,862G1: NRG2: NROverall: 2,540G1: NRG2: NROverall: 889G1: 431G2: 458Overall: 132G1: 66G2: 66Overall: 133G1: NRG2: NROverall: 284G1: 141G2: 143N CompletersN AnalyzedNA Overall: 5,717G1: 273G2: 5,444Overall: NRG1: 379G2: 409Overall: 121G1: 61G2: 60Overall: 128G1: NRG2: NROverall: NRG1: 113G2: 104NROverall: 121G1: 61G2: 60Overall: NRG1: NRG2: NROverall: NRG1: 92G2: 104D-80


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Staresinic, 2007 51Taylor, 2003 52Touchette, 2012 53Volume, 2001,PREP(PharmaceuticalCare Researchand EducationProject), #2579and #2631KassamG1: MTM services provided aspart of a Medicare Part D MTMprogram by an MTMCoordinator (non-clinical staff)and a pharmacistG2: Usual care provided toMTM-eligible enrollees whochose not to participateG1: Pharmaceutical careprovided by pharmacistsG2: Standard care withoutadvice or recommendationsgiven to patients or physiciansCohort Overall: 1,890G1: NAG2: NARCT: parallel,not clusteredG1: MTM basic (comprehensive RCT: parallel,medication review and DRP not clusteredassessment)G2: MTM enhanced (MTM plus2-page clinical summaryabstracted from patient'smedical chart)G3: Usual care, consisting ofmedication counseling perclinic’s normal routine but noformal MTM from a studypharmacistG1: Comprehensivepharmaceutical care servicesusing a nine-step process asdefined by Hepler and Strandprovided by communitypharmacists.G2: Traditional pharmacy careRCT: clusterrandomizedOverall: NRG1: NRG2: NROverall: 1,941G1: NAG2: NAOverall:Approximately 960G1:NRG2: NROverall: 1,826G1: 282G2: 1,544Overall: 81G1: NRG2: NROverall: 637G1: 211G2: 218G3: 208Overall: 363G1: NRG2: NRN CompletersOverall: 1,682G1: 138G2: 1,544Overall: 69G1: 33G2: 36Overall: 556G1: 183G2: 190G3: 183Overall: 292G1: NRG2: NRN AnalyzedOverall: 1,826G1: 282G2: 1,544Overall: 69G1: 33G2: 36Overall: 637G1: 211G2: 218G3: 208Overall: 292G1: 159G2: 204D-81


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Welch, 2009 56Williams, 2004 57Winston, 2009 58G1: MTM program provided tohome-based beneficiaries aspart of a Medicare Part D MTMprogramG2: No-MTM control group(voluntary opt-out)G1: Medication review and RCT: parallel,optimization of patient's not clusteredmedication regimen conductedby an interdisciplinarymedication adjustment team inaddition to usual medical careand "Bound for Health" booklet.G2: Usual medical care plusprovision of "Bound for Health"bookletG1: MTM provided in acommunity pharmacy (i.e., carein face-to-face meetings or bytelephone) as part of aMedicare Part D MTM programG2: MTM provided bypharmacist-staffed call centersas part of a Medicare Part DMTM programG3: Educational mailings (i.e.,mailed letter containing patientspecificmedication relatedinformation, personalmedication record, and tips tosave money on prescriptions)Cohort Overall: 1,231G1: NRG2: NROverall: 144G1: NAG2: NACohort Overall: 101,846G1: 33,954G2: 3,961G3: 63,931Overall: 904G1: 539G2: 365Overall: 140G1: 63G2: 77Overall: 95,736G1: 31,347G2: 3,787G3: 60,602N CompletersOverall: 795G1: 459G2: 336Overall: 133G1: 57G2: 76Overall: NRG1: NRG2: NRN AnalyzedOverall: 795G1: 459G2: 336Overall: 133G1: 57G2: 76Overall: 73,793G1: 21,336G2: 3,436G3: 49,021D-82


Table D3. Number of study participants (continued)Author, YearTrial NameInterventions andComparator DescriptionsStudy Design N Eligible N Randomized(trials) or N Enrolled(observationalstudies)Witry, 2011 59G1: PCM provided by Cohortcommunity pharmacists to IowaMedicaid enrolleesG2: PCM provided bycommunity pharmacists topatients with private individualgroupinsuranceWittayanukorn, G1: Intervention group:2013 60 Pharmacist provided face-tofaceMTM services for 30-60minutes per encounter, notalways including a follow-upvisitG2: Control group: Patientswho did not receive MTMservices (economic analysesonly)Yamada 2012 61 G1: Kaiser-Permanente MTMenrolled patientsG2: Kaiser patients enrolled inMedicare part D, but not inMTM program matched tocontrol on age, gender, regionand DCG riskOverall: NRG1: NRG2: 250Cohort Overall: 3,233G1: NAG2: NACohort Overall: 172,534G1: 34,352G2: 138,182Overall: 615G1: 524G2: 91Overall: 125G1: 63G2: 62Overall: 172,534G1: 34,352G2: 138,182N CompletersOverall: 615G1: 524G2: 91Overall: 125G1: 63G2: 62Overall: 172,534G1: 34,352G2: 138,182N AnalyzedOverall: 615G1: 91G2: 524Overall: 125G1: 63G2: 62Overall: 172,534G1: 34,352G2: 138,182Abbreviations: BP = blood pressure; CHF = chronic heart failure; COPD = chronic obstructive pulmonary disease; DCG = diagnostic cost group; DTP = drug therapy problem;FHCP = Florida Health Care Plans; FOM = focus on medicine; G = group; HbA1c = hemoglobin A1c; HEDIS = Healthcare Effectiveness Data and Information Set; HMR = homemedication review; LDL = low- density lipoprotein; MA-PD = Medicare Advantage Part D; MTM = medication therapy management; N = number; NA = not applicable; NR = notreported; NRCT = non-randomized controlled trial; PCM = pharmaceutical case management; PCP = primary care provider; PREP = Pharmaceutical Care Research and EducationProject; QOL = quality of life; RCT = randomized controlled trialD-83


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studiesAuthor, YearIntervention and Level of Integration withUsual CareState (Province) orCountryBernsten et al., 2001 1 ;Sturgess et al., 2003 27 European countries:Denmark, Germany, TheNetherlands, NorthernIreland, Portugal, Republicof Ireland, and SwedenBlakey et al., 2000 3Georgia, USBrummel et al., 2013 4Soliman et al., 2013 5Ramalho de Oliveira et al.,2010 6Minnesota, USIntervention: Structured pharmaceutical careprovided by community pharmacists.Level of Integration with Usual Care:Pharmacists were encouraged to use thepatient’s GP to obtain information, but specificdetails regarding pharmacist access to clinicalinformation was NR. Rationalizing andsimplifying drug regimens in collaboration withthe patient’s general practitioner was structuredusing drug use profiles, however specific detailsregarding the communication betweenpharmacist and physicians was NR.Intervention: Interdisciplinary team providedgeriatric medical care that included a clinicalpharmacistLevel of Integration with Usual Care:Pharmacists were members of theinterdisciplinary care team.Intervention: Standardized MTM programprovided by pharmacists.Level of Integration with Usual Care: MTMpharmacists staff medical clinics and havecollaborative practice agreements to initiate,modify, or discontinue drug therapy or orderlaboratory tests.Method of IdentifyingPatients for Receipt ofMTM ServicesPersonal recruitment bypharmacists withinpharmacy, or via GPrecords or pharmacyrecordsPharmacist review ofscheduled patients,identification during careteam meetings, providerreferral for services.Patients “opt in” to MTMthrough direct referral,mailed leters, andtelephone outreach.Setting, Mode ofDelivery, Frequency andInterval of FollowupHealth Care System andReimbursement ContextSetting: Community Health care systems variedpharmacies, but also by country, but all featuredincluded some home visits. single payer systems.Mode of delivery: NRFrequency and interval offollow-up: NRSetting: Outpatientgeriatric medicine clinic.Mode of delivery: Face-tofaceFrequency and interval offollow-up: initialconsultation and at leastone follow-up at 3 months.Setting: Outpatient medicalclinicsMode of delivery: Face-to-FaceFrequency and interval offollow-up: No specificfrequency or interval bydesign, median number ofvisits per year was 4.Reimbursementcharacteristics: NRSingle Veterans HealthAdministration MedicalCenterReimbursementcharacteristics: NRClinics were part of a large,integrated health systemwith its own pharmacyservicesReimbursementcharacteristics: NRD-84


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryChrischilles et al., 2004 9Iowa, USChristensen et al., 2007 10North Carolina, USIntervention and Level of Integration withUsual CareIntervention: Pharmaceutical case managementprovided by pharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in medical recordNR. Pharmacist written communication withphysicians about problems identified. Acollaboratively determined action plan can beimplemented by the pharmacist withoutrequiring a patient visit to a physician.Intervention: Medication therapy managementservices designed by a health plan for itsbeneficiaries and provided by either communitypharmacists or medical clinic-basedpharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord NR. Pharmacist contacted prescribingphysicians to discuss drug therapy problemsand implemented any resulting approved actionplan.Method of IdentifyingPatients for Receipt ofMTM ServicesClaims data or pharmacyprescription profile recordsEligible patients identifiedthrough claims data, thenrecruited through a lettersent inviting them toparticipate.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Communitypharmacy.Mode of delivery: Face-tofaceFrequency and interval offollow-up: Initialconsultations with followupcontacts as needed androutine re-assessmentsevery 6 months by design.Setting: Some patientsreceived services withintheir medical clinic, somereceived services within acommunity pharmacysetting.Mode of delivery: Face-toface and telephoneFrequency and interval offollow-up: study designedas one initial visit and onefollow-up, 37.5% ofenrolled patients receivedfollow-up contact.Health Care System andReimbursement ContextHealth plan intervention thatincluded a payment reformto allow for reimbursementof multiple participatingpharmacies and providersacross different systems.Reimbursementcharacteristics: provided asa Medicaid benefit usingstate and federal matchingfunds.Health plan interventioninvolving multiple healthsystems.Reimbursementcharacteristics: Pharmacistscompensated through studyrelatedfunding (e.g., grant).D-85


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryFischer et al., 2000 12Midwest, USFischer et al., 2002 13Minnesota, USIntervention and Level of Integration withUsual CareIntervention: Pharmaceutical care based on theEncara Practice System provided by onsitehealth maintenance organization staffpharmacists.Level of Integration with Usual Care:Pharmacist access to clinical information in themedical record NR. Information to andconsultation with physicians on behalf ofpatients mentioned but specific operationaldetails NR.Intervention: Pharmaceutical care based on theEncara Practice System provided bypharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord NR. Communication with the patient’sphysician about drug therapy problemsidentified, but specific operational details NR.Method of IdentifyingPatients for Receipt ofMTM ServicesClaims data or pharmacyprescription profile recordsto identify eligibleparticipants who were theninvited by letter.Claims data or pharmacyprescription profile recordsto target eligibleparticipants.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Pharmacieslocated within clinicsMode of delivery: Face-tofaceFrequency and interval offollow-up: NRSetting: Pharmacieslocated within clinics andfree-standing pharmaciesMode of delivery: Face-tofaceFrequency and interval offollow-up: With eachprescription refill (asdesigned); actualfrequency and interval NR.Health Care System andReimbursement ContextHealth maintenanceorganization with clinics andon-site pharmacies.Reimbursement for services:NRHealth maintenanceorganization clinics and freestandingpharmacies.Reimbursementcharacteristics: NRD-86


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryFox et al., 2009 14Florida, USGrymonpre et al., 2001 16Manitoba, CanadaIntervention and Level of Integration withUsual CareIntervention: MTM services provided by staffclinical pharmacist as part of a Medicare Part DMTM program.Level of Integration with Usual Care: Pharmacisthad access to clinical information in the medicalrecord, including laboratory data. Pharmacistdocumented findings on a form, which was sentto the patient’s physician.Method of IdentifyingPatients for Receipt ofMTM ServicesClaims data or pharmacyprescription profile recordsto target eligibleparticipants.Intervention: Pharmaceutical care provided by Self-referral or referral by ateam of non-pharmacist staff or volunteers, a BS home health servicelevel pharmacist, and supervision by pharmacist program.with geriatric medicine experience.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord NR. Pharmacist recommendations sentvia letter to prescribing physicians, who may nothave been known or geographically close.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Health planpharmacy, unclear whethera single centralized centeror outpatient clinic-basedpharmacies used.Mode of delivery: Primarilytelephone, supplementedby mailed writtenmaterials.Frequency and interval offollow-up: Initialconsultation, and up to 3follow-up contacts if a drugtherapy problem identifiedor based on clinical need.Setting: Patient’s home ora private officeMode of delivery: Face-tofaceFrequency and interval offollow-up: Initialconsultation with follow-upas needed (as designed);actual frequency andinterval NR.Health Care System andReimbursement ContextMixed-staff model healthmaintenance organizationthat combines pharmacistservices, primary care, andspecialty medical care with aMedicare Advantage Part DPlan.Reimbursementcharacteristics: MedicarePart D drug benefitCommunity-basedinterdisciplinary health clinic.Reimbursementcharacteristics: NRD-87


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryHanlon et al., 1996 17North Carolina, USIntervention and Level of Integration withUsual CareIntervention: Pharmaceutical care provided by aclinical pharmacistLevel of Integration with Usual Care:Pharmacists had access to clinical informationin medical record. Pharmacist recommendationswere presented orally and in writing to thepatient’s primary physician, pharmacistreinforced and amplified the primary physician’sinstructions.Method of IdentifyingPatients for Receipt ofMTM ServicesComputerized and manualchart audits to identifyeligible subjectsSetting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Outpatient medicalclinicMode of delivery: Face-tofaceFrequency and Interval ofFollow-up: NRHealth Care System andReimbursement ContextSingle Veterans HealthAdministration MedicalCenterReimbursementcharacteristics: NRD-88


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryIsetts et al., 2008 22Minnesota, USJameson, VanNoord, andVanderwoud, 1995 23Michigan, USIntervention and Level of Integration withUsual CareIntervention: Medication therapy managementservices provided by staff pharmacists, includingthe establishment of goals of therapy.Level of Integration with Usual Care:Pharmacists had access to clinical informationin medical record. Pharmacist urgentlyconsulted with primary care provider forpotentially harmful drug therapy problems, butdetails regarding routine communication wereNR.Intervention: Pharmacotherapy consultation andfollowup provided by clinical ambulatory carepharmacist.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord was NR. Pharmacist met with treatingphysician to discuss findings and new regimenwas developed collaboratively with thephysician.Method of IdentifyingPatients for Receipt ofMTM ServicesClaims data or pharmacyprescription profile recordsused to identify eligibleparticipants who were theninvited by letter andprovider referral.Medical records of patientsseen in an outpatient clinicwere randomly screenedfor risks of adversemedication outcomes.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Outpatient medicalclinicsMode of delivery: Face-tofaceFrequency and Interval ofFollow-up: NR (at least 1follow-up visit was requiredfor inclusion in theevaluation)Setting: Outpatient medicalclinicMode of Delivery: Face-tofaceand telephoneFrequency and interval ofFollow-up: 1 initial visit and1 follow-up visit 1 monthlater (by design); actualfrequency and interval offollow-up NR.Health Care System andReimbursement ContextIntegrated health systemwith an establishedpharmaceutical careprogram involvingpharmacist certification inpharmaceutical care and aspecific pharmaceutical caredocumentation system.Reimbursementcharacteristics: Costs ofproviding services werefunded through researchgrants, demonstrationprojects, third-party payerpilot programs, and internalhealth system funding.Single family health clinicthat was part of a familymedicine residency program.Reimbursementcharacteristics: NRD-89


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryJeong et al., 2009 25California, USJeong et al., 2012 26Jeong et al., 2012 27California, USKrska et al., 2001 28United KingdomIntervention and Level of Integration withUsual CareIntervention: MTM services provided by clinicalambulatory care pharmacists and health caresupport staff.Level of Integration with Usual Care: Pharmacistwork under collaborative practice agreements toinitiate, titrate, and discontinue medications andorder laboratory tests. Patient data andinterventions entered into an internal database(not clear if this accessible to care team).Intervention: MTM services provided by clinicalambulatory care pharmacists and support staffLevel of Integration with Usual Care: Pharmacistwork under collaborative practice agreements toinitiate, titrate, and discontinue medications andorder laboratory tests.Intervention: Medication reviews led by clinicallytrained pharmacists.Level of Integration with Usual Care: Pharmacisthad access to medical notes and practicecomputer records. Copies of the pharmaceuticalcare plan developed by the pharmacist wereinserted into the patient’s medical record andgiven to the patients’ GP, who was asked toindicate level of agreement withrecommendations.Method of IdentifyingPatients for Receipt ofMTM ServicesNRNRProvider referral requiredbut enrollment limited to 70patients from eachparticipating practice;selection process unclearSetting, Mode ofDelivery, Frequency andInterval of FollowupSetting: CentralizedpharmacyMode of delivery:TelephoneFrequency and interval ofFollow-up: Initialconsultation and twoadditional follow-ups,interval NR.Setting: CentralizedpharmacyMode of delivery:TelephoneFrequency and interval offollow-up: NRSetting: Home visitsMode of Delivery: Initialconsultation was face-toface;follow-up consultationNRFrequency and Interval ofFollow-up: Two contacts, 3months apart as designed;actual frequency andinterval NR.Health Care System andReimbursement ContextIntegrated health caredelivery system providingMTM services to eligiblebeneficiariesReimbursementcharacteristics: PresumablyMedicare Part D drugbenefit.Integrated health caredelivery system providingMTM services to eligiblebeneficiaries.Reimbursementcharacteristics: PresumablyMedicare Part D drugbenefit.General medicine clinics thatwere part of a single payerhealth care system.Reimbursementcharacteristics: NRD-90


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryMalone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32(dyslipidemia subgroupintermediate and utilizationoutcomes)Multiple states, USMoczygemba et al., 2011 37Moczygemba et al., 2008 38Texas, USIntervention and Level of Integration withUsual CareIntervention: Pharmaceutical care provided byclinical pharmacists practicing according toscope of practice within their respective healthcare facilities.Level of Integration with Usual Care: Pharmacisthad access to medical record information.Pharmacist communication with primary carephysician or other prescribers NR.Intervention: Medication therapy managementservices provided by clinical pharmacists or amanaged care pharmacy resident based on theAmerican Pharmacists Association and NationalAssociation of Chain Drug Stores FoundationMTM framework.Level of Integration with Usual Care:Pharmacist had access to medical recordinformation, including laboratory information.Patients are encouraged to share the PersonalMedication Record and Medication Action Plandeveloped in the course of MTM consultationwith their health care providers and patients arerequested to contact their physicians regardingpharmacist’s recommendations. Copies are keptin the patient’s MTM file, but no MTMdocumentation goes back to the patient’smedical record.Method of IdentifyingPatients for Receipt ofMTM ServicesPharmacy prescriptionrecords to identify patientsat high risk for drug-relatedproblems.Eligible patients areidentified quarterly andmailed MTM Programinformation and instructionsfor opting in to theprogram.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Outpatient medicalclinicMode of Delivery: Face-toface(76.6% of contacts)and telephone (23.4%)Frequency and Interval ofFollow-Up: At least 3 visitsover 12 months asdesigned. Actualfrequency: mean (SD)number of visits was 3.5(2.3). 27.7% did notcomplete the minimumnumber of visits (3) asdesignedSetting: centralized MTMprogramMode of Delivery:TelephoneFrequency and Interval ofFollow-Up: one initialconsultation by design withfollow-up scheduled asneeded. Actual frequencyand interval of follow-upNR.Health Care System andReimbursement ContextMultiple Veterans HealthAdministration MedicalCenters with establishedambulatory clinicalpharmacy servicesReimbursementcharacteristics: Servicesprovided as part of patient’sVHA health care benefitsHealth plan interventionprovided by a regionalMedicare Part D MTMProvider.Reimbursementcharacteristics: MedicarePart D drug benefit.D-91


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryMoore et al., 2013 40Multiple states, USPerlroth et al., 2013 35Multiple states, USPindolia et al., 2009 44Michigan, USIntervention and Level of Integration withUsual CareIntervention: Medication therapy managementservices provided by a large pharmacy benefitsmanagement company.Level of Integration with Usual Care: Patientshad to send validated lab data to pharmacistbefore initial appointment. With patientauthorization, pharmacist faxedrecommendations to prescribing provider(unless patient was referred by case or diseasemanager).Intervention: Medicare Part D MTM ProgramsLevel of Integration with Usual Care: Varies byprogram sponor.Intervention: Medication therapy managementservices provided as part of a Medicare Part DMTM program by pharmacy care managementclinical pharmacists.Level of Integration with Usual Care:Pharmacists had access to clinical informationin the medical record. Communications withphysicians were by telephone, face-to-face, ore-mail.Method of IdentifyingPatients for Receipt ofMTM ServicesPatients with claims datashowing more than 14pharmacy claims within 4months or absence ofclaims for recommendedtherapy or evidence ofconflicting therapy weresent a letter offeringenrollment.Varies by programsponsor.Monthly query of clinicalcare management systemsfor eligible patients withsubsequent letter mailedand follow-up phone call toenroll patients.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: CentralizedpharmacyMode of delivery:TelephoneFrequency and interval offollow-up: Initialconsultation with at leasttwo follow-up contactswithin 12 months.Setting: Varies by programsponsorMode of delivery: Varies byprogram sponsorFrequency and interval offollow-up: Varies byprogram sponsorSetting: Integratedhealthcare delivery systemMode of Delivery:telephoneFrequency and Interval ofFollow-Up: NRHealth Care System andReimbursement ContextHealth plan intervention witha pharmacy benefitsmanagement company.Reimbursementcharacteristics: NRVaries by program sponsorReimbursementcharacteristics: MedicarePart DHealth plan interventionwithin an Integrated healthsystem with an establishedpharmaceutical careprogram.Reimbursementcharacteristics: MedicarePart D drug benefit.D-92


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountrySellors et al., 2003 47Ontario, CanadaC. Sellors et al., 2003 48Ontario, CanadaShimp et al., 2012 49Michigan, USIntervention and Level of Integration withUsual CareIntervention: Clinical pharmacy consultationsprovided to elderly patients by pharmacists.Level of Integration with Usual Care:Pharmacist access to clinical information inmedical record was NR. Pharmacists provided aconsultation letter to physician andsubsequently met with physician to review theletter. They met again in 3 months to discussprogress in implementing recommendations.Intervention: Clinical pharmacy consultationsprovided to elderly patients by pharmacist.Level of Integration with Usual Care: Pharmacistreviewed information in patient’s medical record.Pharmacist met with provider to discuss writtenrecommendations.Intervention: Medication therapy managementservices provided by pharmacists as part of anemployee health program.Level of Integration with Usual Care: Pharmacistmay have had access to clinical information inmedical records for some enrolled patients. Withpatient approval, pharmacist recommendationswere communicated to providers via the EHR.Method of IdentifyingPatients for Receipt ofMTM ServicesAbout 20 randomly choseneligible senior citizens perpractice were recruited bythe office staff of thepractice, selection processfor recruitment notreported.Patients identified duringoffice visits or throughreview of practice rosters.Pharmacy benefitsmanager identified eligibleindividuals on a quarterlybasis. From this list,individuals were randomlyselected for an invitation toparticipate.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Outpatient medicalclinicMode of Delivery: Face-tofaceand telephoneFrequency and Interval ofFollow-Up: Initial contactplus 2 follow-up contacts at1 and 3 months asdesigned. Actual frequencyand interval of contact NR.Setting: Outpatient medicalclinicMode of delivery: Face-tofaceand telephoneFrequency and interval offollow-up: Initialconsultation, follow-up at 2weeks, then monthly for 6monthsSetting: employer-basedMode of delivery:Telephone and face-tofaceFrequency and interval offollow-up: Initial historytaken by study staff byphone, with two follow-upvisits face-to-face withpharmacist, interval NRHealth Care System andReimbursement ContextFamily medicine practiceswithin a single-payer healthcare system.Reimbursementcharacteristics: NRFamily medicine practiceswithin a single-payer healthcare system.Reimbursementcharacteristics: NRLarge university and itsassociated pharmacybenefits managementorganization.Reimbursementcharacteristics: NRD-93


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountrySidel et al., 1990 50New York, USStaresinic, 2007 51Wisconsin, USTaylor, Byrd, and Krueger,2003 52Alabama, USIntervention and Level of Integration withUsual CareIntervention: Home visits by pharmacists toidentify and correct problems associated withmedication use.Level of Integration with Usual Care: Pharmacistaccess to clinical information in medical recordwas NR. No information about communicationwith providers was reported.Intervention: Medication therapy managementservices provided as part of a Medicare Part DMTM program by an MTM Coordinator (nonclinicalstaff) and a pharmacist.Level of Integration with Usual Care:Pharmacists request lab data from participants;access to clinical information in medical recordswas NR. Pharmacists send a tailored letter byfax to each of the patient’s health careproviders.Intervention: Pharmaceutical care provided bypharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in medical recordwas NR, but visits with pharmacist occurred 20minutes before seeing the physician in the sameclinic. Recommendations to physicians werecommunicated through discussions or progressnotes.Method of IdentifyingPatients for Receipt ofMTM ServicesStudy population identifiedfrom a combination of thefollowing: Medicarerecipients living in theregion, Senior Centers,houses of worship, Mealson-Wheels,hospitaladmissions records andvoter registration rolls.Pharmacy claims basedalgorithm identifies eligiblebeneficiaries with invitationletters mailed within 2weeks of identifyingeligibility.Patients were identified bythe participatingpharmacists throughmanual evaluation of clinicmedical records and reviewof computerized medicalrecords in physicianoffices.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Community settingMode of delivery: homevisits and telephoneHealth Care System andReimbursement ContextImplemented outside thehealth care system througha multidisciplinary researchprogram on aging.Frequency and Interval of ReimbursementFollow-Up: 2 visits over a characteristics: NR12 month period asdesigned. Actual frequencyand interval of follow-upNR.Setting: centralized MTMprogramMode of Delivery:telephoneFrequency and Interval ofFollow-Up: One initialcontact and one follow-upcontact at 3 months asdesigned. Actual frequencyand interval of follow-upNR.Setting: Outpatient medicalclinicMode of Delivery: Face-tofaceFrequency and Interval ofFollow-Up: Before eachscheduled physician visitby design. Actualfrequency and follow-upinterval NR.Health plan interventionprovided by a regionalMedicare Part D MTMProviderReimbursementcharacteristics: MedicarePart D drug benefit.Three community-basedfamily medicine clinicsaffiliated with an academicmedical center.Reimbursementcharacteristics: NRD-94


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryTouchette et al., 2012 53Multiple States, USVolume et al., 2001 54 andKassam et al., 2001 55PREPAlberta, CanadaIntervention and Level of Integration withUsual CareIntervention: Patient-safety focused medicationtherapy management services provided bypharmacists.Level of Integration with Usual Care: Twoversions of the intervention were evaluated. Inthe enhanced version, pharmacists wereprovided with a clinical summary excerpted fromthe patient’s medical record. No such summarywas provided to pharmacists in the basicversion. Drug therapy problems werecommunicated to physicians via fax, excepturgent issues were communicated by telephone.Intervention: Pharmaceutical care using a ninestepprocess as defined by Hepler and Strandprovided by community pharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord NR. Details regarding communicationwith physicians regarding drug therapyproblems NR.Method of IdentifyingPatients for Receipt ofMTM ServicesAdministrative andpharmacy databases andprovider referral used forinitial eligibilitydetermination followed upwith letter or phone call orin clinic forrecruitment/enrollmentPharmacies evaluated 60consecutive patients duringone week for eligibility.Eligible patients wereasked in person or byphone about interest inparticipating.Setting, Mode ofDelivery, Frequency andInterval of FollowupSetting: Outpatient medicalclinicMode of Delivery: Face-tofaceFrequency and Interval ofFollow-up: 1 initial contactand a follow-up contact at3 months as designed.Actual: 89.9% completed 1contact and 75.7%completed 2 contacts inthe enhanced MTM arm,88.6% and 73.8%completed the first andsecond contactsrespectively in the basicMTM arm.Setting: communitypharmacyMode of Delivery: Face-tofaceFrequency and Interval ofFollow-Up: Initial contactplus frequent follow-up atunspecified intervals asdesigned. Actual frequencyand interval of follow-upNR.Health Care System andReimbursement ContextThree academic medicalcenters in three differentstates.Reimbursementcharacteristics: NRCommunity-pharmacyintervention within a non-USsingle-payer healthcaresystem.Reimbursementcharacteristics: NRD-95


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryWelch et al., 2009 56Colorado, USWilliams et al., 2004 57North Carolina, USWinston and Lin, 2009 58Multiple States, USIntervention and Level of Integration withUsual CareIntervention: Medication therapy managementservices provided by clinical pharmacists as partof a Medicare Part D MTM program.Level of Integration with Usual Care:Pharmacists had access to clinical informationin the medical record. Pharmacists forwardedcopies of consultation notes to providers andalso placed a copy in the patient’s medicalrecord.Intervention: Medication review and optimizationprovided by a consulting pharmacist.Level of Integration with Usual Care: Pharmacisthad access to clinical information in the medicalrecord. A MAT comprised of a physician, nurse,and consultant pharmacist met to discusspharmacy recommendations.Intervention: Medication therapy managementservices provided by either communitypharmacists or call center pharmacists as partof a Medicare Part D MTM Program.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord was NR. Pharmacist contactedprescribers on behalf of the patients by phone orfax for medication adjustments related to cost orsafety.Method of IdentifyingPatients for Receipt ofMTM ServicesSetting, Mode ofDelivery, Frequency andInterval of Follow-upMedicare beneficiaries Setting: centralized MTMidentified as eligible using a programcomputerized system.Mode of Delivery:TelephonePatients were recruitedfrom practices and throughcommunity print and radioadvertisements and massmailings, and presentationsto community groups.Health plan used pharmacyprescription profile recordsidentify eligible patients.Information on eligiblepatients wascommunicated topharmacies by fax or email.D-96Frequency and interval ofFollow-Up: Initial consultwith follow-up dependingon clinical situation asdesigned. Actual frequencyand interval of follow-upNR.Setting: Outpatient medicalclinicMode of Delivery: Face-tofaceFrequency and Interval ofFollow-Up: Initial contactwith follow-up contact asneeded as designed.Actual frequency andinterval of follow-upcontact NR.Setting: communitypharmacy and centralizedpharmacy call centerMethod of Delivery: Faceto-faceor telephoneFrequency and Interval ofFollow-up: NRHealth Care System andReimbursement ContextGroup-model healthmaintenance organizationusing a centralized clinicalpharmacy call center.Reimbursementcharacteristics: MedicarePart D drug benefit.General medicine clinic ofan academic medical center.Reimbursementcharacteristics: NRHealth plan interventionprovided by a nationalMedicare Part D MTMprovider.Reimbursementcharacteristics: MedicarePart D drug benefit


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryWitry, Doucette, andGainer, 2011 59Iowa, USWittayanukorn et al.,2013 60Southeast USIntervention and Level of Integration withUsual CareIntervention: Pharmaceutical case managementprovided by community pharmacists.Level of integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord was NR. Pharmacists faxed a one-pagesummary of findings to physician.Intervention: Medication therapy managementprovided by pharmacists.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord was NR. Provider communication andfollow-up NR.Method of IdentifyingPatients for Receipt ofMTM ServicesHealth plan used pharmacyprescription profile recordsidentify eligible patients.This was sent to eachpharmacy. The health planalso sent letters to informeligible patients about thePCM service benefit, andan article about PCMappeared in the health plannewsletter. Pharmaciesalso sent letters andtelephoned eligiblepatients.Patients were identified bypharmacists, referred byproviders, or self-referred.Setting, Mode ofDelivery, Frequency andInterval of Follow-upSetting: communitypharmacyMode of Delivery: Face-tofaceand telephoneFrequency and Interval ofFollow-Up: Initial contactwith additional follow-upcontacts as needed asdesigned. Actual frequencyand interval of follow-up:46% received 1 contact,24% received 2 contacts,16% received 3contacts,13% received 4 ormore contactsSetting: Pharmaceuticalcare center for employeesof a self-insured employerMode of delivery: Face-tofaceFrequency and interval offollow-up: NRHealth Care System andReimbursement ContextHealth plan interventionexecuted by pharmacies thathad previously participatedin a similar interventionsponsored by Medicaid.Reimbursementcharacteristics: Participatingpharmacies werereimbursed for servicesprovided by study grant.Employer-based interventionwithin a university setting.Reimbursementcharacteristics: Self-insuredemployer paid for cost ofMTM servicesD-97


Table D4. Key Question 1: Components and features of medication therapy management interventions: Broadly focused studies(continued)Author, YearState (Province) orCountryYamada et al., 2012 61California, USIntervention and Level of Integration withUsual CareIntervention: MTM services provided by clinicalambulatory care pharmacists and health caresupport staff.Level of Integration with Usual Care: NR, butother evaluations of this MTM program reportedthat pharmacist work under collaborativepractice agreements to initiate, titrate, anddiscontinue medications and order laboratorytests. Patient data and interventions entered intoan internal database (not clear if this accessibleto care team).Method of IdentifyingPatients for Receipt ofMTM ServicesNRSetting, Mode ofDelivery, Frequency andInterval of Follow-upSetting: CentralizedpharmacyMode of delivery:TelephoneFrequency and interval offollow-up: NRHealth Care System andReimbursement ContextIntegrated health caredelivery system providingMTM services to eligiblebeneficiariesReimbursementcharacteristics: PresumablyMedicare Part D drugbenefit.Abbreviations: GP = general practitioner; MAT = medication adjustment team; MTM = Medication Therapy Management; NR = not reported; PCM = pharmaceutical casemanagement; PREP = Pharmaceutical Care Research and Education Project; QOL = quality of life; SD = standard deviation; US = United States; VHA = Veterans HealthAdministration.D-98


Table D5. Key Question 1: Components and features of medication therapy management interventions: Narrow focused studiesAuthor, Year Special Focus, Intervention, and IntegrationHealth Care System andwith Usual CareReimbursement ContextsCountry/ RegionCarter et al., 1997 7 ;Barnette, Murphy,and Carter, 1996 8Illinois, USClifford et al., 2002 11AustraliaFocus: HypertensionIntervention: Pharmaceutical care provided bypharmacists within an interdisciplinary practicemodel. Patient education (lifestyle, risk factormodifications, and drug therapy) wasstandardized.Level of Integration with Usual Care: Thepharmacist had access to patients’ medicalrecords, diagnostic data, and laboratory data, andhad face-to-face interaction with the clinicphysicians and nurses.Focus: DiabetesIntervention: Pharmaceutical care provided by aclinical pharmacist , which included acomprehensive review relating topharmacotherapy and diabetes, use of proprietaryand non-proprietary medications, such ascomplementary medicines, and identification ofdrug therapy problems.Level of Integration with Usual Care: Pharmacisthad access to patient’s case notes.Pharmaceutical care was provided in cooperationwith the patient’s diabetes physicians and otherdiabetes health team members.Method of IdentifyingPatients for Receipt ofMTM ServicesPatients were identifiedthrough a computerizedprofile review, details NR.Medical records werescreened for eligiblepatients. Eligible patientswere telephoned about theirwillingness to participate.Setting, Mode ofDelivery, Frequencyand interval of followup(as reported).Setting: Outpatientprimary care clinicMode of delivery: Faceto-faceFrequency and Intervalof Follow-up: Monthlycontacts for 6 months asdesigned. Actualfrequency and intervalof follow-up NR.Setting: Outpatienthospital diabetes clinicMode of Delivery: Faceto-faceFrequency and Intervalof Follow-Up: Initial visitfollowed by follow-upvisits at 6 week intervalsfor 6 months asdesigned. Actualfrequency and intervalof follow-up NR.Rural medical clinic co-locatedin same building as a privatelyowned pharmacy.Reimbursement characteristics:NRNon-US, single payer healthcare systemReimbursement characteristics:NRD-99


Table D5. Key Question 1: Components and features of medication therapy management interventions: narrow focused studies(continued)Author, YearCountry/ RegionSpecial Focus, Intervention, and Integrationwith Usual CareMethod of IdentifyingPatients for Receipt ofMTM ServicesSetting, Mode ofDelivery, Frequencyand interval of followup(as reported).Health Care System andReimbursement ContextsGattis et al., 1999 15North Carolina, USHirsch et al., 2011 19Hirsch et al., 2009 20California, USFocus: Chronic heart failureIntervention: Clinical pharmacy services, includingan assessment of prescribed regimen,compliance, and adverse effects, and symptomsand response to therapy. Providing patienteducation about the purpose of each drug andreinforcing adherence. Detailed writteninformation was also provided to patients.Level of integration with usual care: Pharmacisthad access to patient medical records andverbally recommendations regarding optimizationof therapy with attending physician.Focus: HIV/AIDSIntervention: Medication therapy managementservices provided by pharmacist.Level of Integration with Usual Care: Pharmacistaccess to clinical information in the medicalrecord was NR. Follow-up and communicationwith providers was at the discretion of thepharmacy.Patients seen in a generalcardiology clinic meetinginclusion criteria wererecruited for enrollment.Patients filling prescriptionsat participating pharmacieswere eligible to receiveMTM services, but actualselection criteria were left todiscretion of pharmacy.Setting: Outpatientcardiology clinicMode of Delivery: Initial Reimbursement characteristics:visit was face-to-face NRand follow-up visits wereby telephoneFrequency and Intervalof Follow-Up: 3 visits,baseline, two weeks,and 24 weeks bydesign. Actualfrequency and intervalof follow-up NR.Setting: SpecialtyHIV/AIDS communitypharmacies (90 percentor more of populationserved are individualswith HIV/AIDS.)Mode of delivery: Faceto-FaceFrequency and intervalof follow-up: NRSingle clinic within an academicmedical center.Plan-level intervention providedby state Medicaid agency.Reimbursement characteristics:Compensation provided toparticipating pharmacies on aper prescription basis by thestate Medicaid agency.D-100


Table D5. Key Question 1: Components and features of medication therapy management interventions: narrow focused studies(continued)Author, YearCountry/ RegionSpecial Focus, Intervention, and Integrationwith Usual CareMethod of IdentifyingPatients for Receipt ofMTM ServicesSetting, Mode ofDelivery, Frequencyand interval of followup(as reported).Health Care System andReimbursement ContextsMarques et al.,2013 33BrazilMcDonough et al.,2005 36Iowa, USFocus: Depression and anxietyIntervention: Dader method of pharmacotherapyfollow-up provided by pharmacist. Includesestablishing a plan of action for medications witha goal of enhancing or preserving health.Level of Integration with Usual Care: Interventionswith referring psychiatrist were provided asneeded, but specific details NR.Psychiatrist referral to thepharmaceutical assistanceservice.Focus: Glucocorticoid-induced osteoporosis Claims data or pharmacyprescription profile recordsIntervention: Pharmaceutical care provided by used to identify eligiblecommunity pharmacists. Drug therapy monitoring patients who were thenfocused on 5 drug therapy problems:contacted by mail orappropriateness of does, proper regimen, telephone to participate.potential interactions, nonadherence, and adverseeffects. Patient education was also provided.Level of Integration With Usual Care: Pharmacistaccess to patient medical records NR. Astandardized physician communication form wasused by pharmacists to communicate informationto prescribing physicians.Setting: Outpatientpsychiatry andpsychology clinicsSpecialty clinics of an academicmedical center in a single payersystemMode of Delivery: Face Reimbursement characteristics:to face in patient’s home NRFrequency and Intervalof Follow-up: Initial visitand at least one followevery30 days for 3months (or sooner ifneeded)Setting: communitypharmacyMode of Delivery: NRFrequency and Intervalof Follow-Up: NRNetwork of independent andretail chain pharmacies. Somepharmacies located within aclinic, while others arefreestanding.Reimbursement characteristics:Pharmacists were reimbursedusing a web-based claimssystem, but entity providingreimbursement was NR.D-101


Table D5. Key Question 1: Components and features of medication therapy management interventions: narrow focused studies(continued)Author, YearCountry/ RegionSpecial Focus, Intervention, and Integrationwith Usual CareMethod of IdentifyingPatients for Receipt ofMTM ServicesSetting, Mode ofDelivery, Frequencyand interval of followup(as reported).Health Care System andReimbursement ContextsPai et al., 2009 41 ;Pai et al., 2009 42New Mexico, USFocus: HemodialysisIntervention: Pharmaceutical care including drugtherapy reviews conducted by a nephrologytrainedclinical pharmacist with the patient. Alsoincluded patient and health care providereducation.Level of Integration with Usual Care: Thepharmacist had access to patient’s medical recordand laboratory data. The pharmacists providedcognitive services during weekly rounds andduring monthly formal reviews of the patients withthe multidisciplinary health care team.Patients on stablehemodialysis regimen forthe previous 3 months wereapproached forparticipation.Setting: Outpatienthemodialysis clinicMode of Delivery: Faceto-faceFrequency and Intervalof Follow-Up: Every 8weeks for two years bydesign. Actualfrequency and intervalof follow-up NR.University-affiliated outpatientdialysis clinicReimbursement characteristics :NRD-102


Table D5. Key Question 1: Components and features of medication therapy management interventions: narrow focused studies(continued)Author, YearCountry/ RegionSpecial Focus, Intervention, and Integrationwith Usual CareMethod of IdentifyingPatients for Receipt ofMTM ServicesSetting, Mode ofDelivery, Frequencyand interval of followup(as reported).Health Care System andReimbursement ContextsPark et al., 1996 43Wisconsin andIllinois, USPlanas et al., 2009 45Oklahoma, USFocus: HypertensionIntervention: Comprehensive pharmaceutical careincluding drug therapy monitoring and patienteducation provided by a community pharmacyresident.Level of Integration with Usual Care: Pharmacistsaccess to patient medical records, or labs or vitalsigns from clinic was NR. Communication withprovider was via fax or mail after each pharmacistvisit, unless urgency required telephonecommunication.Focus: Patients with both hypertension anddiabetesIntervention: Medication therapy managementservices provided by community pharmacists.Also included patient education on diet andlifestyle modifications to lower blood pressure.Level of integration with usual care: Pharmacistaccess to patient medical records, or labs, or vitalsigns from clinic was NR. Providers werecontacted via fax or telephone when drug therapyproblems were identified in order to makerecommendations.Claims data or pharmacyprescription profile recordsThree methods were used:managed care organizationidentification of patients withuncontrolled diabetesthrough lab data screening,screening for uncontrolleddiabetes at a health fair foremployees sponsored bythe managed careorganization, providerreferral of patients withuncontrolled diabetes.Setting: CommunitypharmacyMode of Delivery: Faceto-faceFrequency and Intervalof Follow-up: 4 visitsscheduled 1 monthapart by design. Actualfrequency and intervalof follow-up NR.Setting: CommunitypharmacyMode of Delivery: Faceto-faceFrequency and Intervalof Follow-up: Monthlyvisits for 9 months bydesign. Actualfrequency and intervalof follow-up NR.Chain pharmacy with communitypharmacy residents.Reimbursement characteristics:NRServices provided through acollaboration between amanaged care organization anda regional retail chain pharmacy.Reimbursement characteristics:NRD-103


Table D5. Key Question 1: Components and features of medication therapy management interventions: narrow focused studies(continued)Author, YearCountry/ RegionRoughead et al.,2009 46AustraliaSpecial Focus, Intervention, and Integrationwith Usual CareFocus: Chronic heart failureIntervention: HMR, a collaborative model ofpharmaceutical care. HMRs are conducted byaccredited pharmacists.Level of Integration with Usual Care: NR for thisstudy specifically, but the HMR model is that theGP provides pharmacist with diagnosis, currentmedications, relevant test results and medicalhistory. Pharmacist conducts the HMR andsubmits a written and verbal report to the GP forassistance in developing or revising amanagement plan.Method of IdentifyingPatients for Receipt ofMTM ServicesHMRs are conducted uponrequest of a provider.Claims data or pharmacyprescription profile recordsSetting, Mode ofDelivery, Frequencyand interval of followup(as reported).Setting: Outpatient clinicand home visitsMode of Delivery: Faceto-faceFrequency and Intervalof Follow-Up: NRHealth Care System andReimbursement ContextsNon-US, single payer healthcare systemReimbursement characteristics:services reimbursed throughpayer’s home medicine reviewbenefit.Abbreviations: CHF = chronic heart failure; GP = general practitioner; HIV/AIDS= human immunodeficiency virus/acquired immune deficiency syndrome HMR = homemedication review; MTM = medication therapy management; NR = not reported; US = United States.D-104


Table D6. Anticoagulation: Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsTaylor, Byrd, and Krueger,2003RCT/MediumG1: Pharmaceutical careG2: Standard careAbbreviations: G = group; INR= international normalized ratio.G1: 4G2: 6(Was only assessedamong patients onanticoagulation)Percent of patients at goal INR(goal was INR 2-3) at 12 monthsG1: 100G2: 16.7P=0.048D-105


Table D7. Hemoglobin A1c: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasClifford et al. 2002 11RCT/MediumG1: Pharmaceutical careG2: Standard careG1: 48G2: 25Mean (SD) HbA1c at six months.Baseline:G1: 8.4 (1.4)G2: 8.5 (1.6)p: NS6 monthsG1: 8.2 (1.5)G2: 8.1 (1.6)Taylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careG2: Standard careG1: 13G2: 16(Study includedmore subjects, butthis outcome wasonly assessedamong patients withdiabetes within eachstudy arm)Percent with HbA1c at goal(defined as less than or equal to7.5 percent) at baseline and at 12months.Calculated mean difference:-0.20, 95% CI, -0.93 to 0.53 (assuming prepostcorrelation of 0.5)p=0. 590Baseline:G1: 23.1G2: 56.3p=0.071Calculated OR: 0.295% CI, 0.05 to 1.19Follow-upG1: 100G2: 26.7p=0.001Calculated OR: 56.595% CI, 2.81 to 1,133.91p=0.008D-106


Table D7. Hemoglobin A1c: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasBrummel et al. 2013 4 , G1: Opted into clinic-based MTMSoliman, 2013 et al. 5 programRamalho de Oliveira et al., G2: Usual medical care (opted2010 6out of MTM program)Cohort study/MediumG1: 121G2: 103Percentage with HbA1c at goal(defined as less than 7) after 12months of demonstrationPercentage with HbA1c at goal(defined as less than 7) after 24months (i.e., 12 months after endof demonstration study)Baseline:G1: 43.80G2: 63.11p=0.6212 months (during demonstration)G1: 73.55G2: 72.82p=0.9Calculated OR: 1.038 95% CI 0.574 to1.879, p=0.901 (unadjusted)Adjusted difference-in-difference coefficient:2.44 95% CI 1.22 to 4.86, p=0.0124 months (12 months post-demonstration)G1: 42.15G2: 59.22p=0.01Calculated OR: 0.502 95% CI 0.294 to0.855, p=0.011 (unadjusted)Adjusted difference-in-difference coefficient:NRD-107


Table D7. Hemoglobin A1c: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasJeong et al., 2009 25Cohort/MediumG1: Participants in Part DMedicare MTM programG1: 1,323G2: 1,141Mean (SD) and mean change(SD) in HbA1c at 6 monthsG2: Control subjects without Part (Study includedD Medicare as their primary drug more subjects butbenefit but otherwise similar tointervention subjects.this outcome wasassessed amongonly patients withdiabetes within eachstudy arm)Percentage with HbA1c less than7 percent at 6 monthsBaselineG1: 6.74 (1.13)G2: 6.84(1.21)6 monthsG1: 6.76 (1.08)G2: 6.88 (1.23)Mean ChangeG1: 0.02 (0.96)G2: 0.04 (1.09)p= NSCalculated mean difference: -0.020 (0.041)95% CI: -0.101 to 0.061p=0.628Baseline:G1: 66G2: 63p: NS6 months:G1:65G2:62p=NSPindolia et al., 2009 44Cohort study/HighG1: Opted in to a telephonebased MTM ProgramG2: Usual medical care (optedout of MTM program)G1: NRG2: NR(Was only assessedamong patients withDM in each studyarm and N for thisoutcome was notreported)Change in percent of patients withHbA1c less than 7 percent at 6monthsCalculated OR:1.142 95% CI: 0.969 to 1.347p=0.114G1: +3G2: +7Between-group p: inferred to be NS, exact pNRWithin-group p: NRAbbreviations: CI = confidence interval; DM = diabetes mellitus; G = group; HbA1C = hemoglobin A1C or glycosylated hemoglobin; MTM = Medication Therapy Management;N = number; NR = not reported; NS = not significant; OR = odds ratio; RCT= randomized controlled trial; SD = standard deviation; vs. = versusD-108


Table D8. LDL cholesterol: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTaylor, Byrd, andKrueger, 2003 52RCT/MediumBrummel et al. 2013 4 ,Soliman, 2013 et al. 5Ramalho de Oliveira etal., 2010 6Cohort study/MediumG1: Pharmaceutical careG2: Standard careG1: Opted into clinic-basedMTM programG2: Usual medical care (optedout of MTM program)BaselineG1: 19G2: 19(Was only assessedamong patients withdyslipidemia in eachstudy arm)Follow-up (N inferredfrom percent in results)G1: 18G2: 17G1: 121G2: 103Percent of patients at LDL-C goalbased on ATPIII criteria at 12months.Percentage with LDL-C at goal(defined as less than 100 mg/dl)after 12 months of demonstrationPercentage with LDL-C at goal(defined as less than 100 mg/dl)after 24 months of demonstration(i.e., 12 months after end ofdemonstration)Baseline:G1: 10.5G2: 15.8Calculated OR: 0.6; 95% CI, 0.09 to4.25;p=0.631Follow-upG1: 77.8G2: 5.9Calculated OR: 56.0; 95% CI, 5.58 to561.75;p: 0.001Baseline:G1: 63.64G2: 65.05p=0.8212 months:G1: 83.47G2: 73.79p=0.07Unadjusted calculated OR: 1.794 95% CI0.936 to 3.438, p=0.078Adjusted difference in differencecoefficient: 1.95, 95% CI 0.81, 4.84,p=0.1324 months:G1: 79.34G2: 73.79p=0.32Unadjusted calculated OR: 1.362 95% CI0.733 to 2.540, p=0.328Adjusted difference-in-differencecoefficient: NRD-109


Table D8. LDL cholesterol: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasJeong et al., 2009 25Cohort/MediumIsetts et al., 2008 22Cohort study/HighPindolia et al., 2009 44Cohort study/HighG1: Participants in Part D G1: 1,515Medicare MTM program (opted G2: 1,323in to program)G2: Control subjects withoutPart D Medicare as their primarydrug benefit but otherwisesimilar to intervention subjects.G1: MTM services provided byhealth plan in existing medicalcare clinics in collaboration withprimary care providers.G2: Usual medical care withoutMTMG1: Opted in to a telephonebased MTM ProgramG2: Usual medical care (optedout of MTM program)(Study included moresubjects but thisoutcome was assessedamong only patientswith diabetes or CADwithin each study arm)G1: 128G2: 126G1: NRG2: NR(Was only assessedamong patients withcoronary artery diseasein each study arm)Mean change (SD) in LDLcholesterol at 6 monthsPercentage with LDL cholesterolat goal (Less than 100 mg/dl)Percent of patients meetingHEDIS measures related tocholesterol control aftercardiovascular event at 12months.Change in percent of patientswith LDL-C less than 100 mg/dlat 6 months.G1: -5.4 (26.2)G2: -1.3(25.8)Calculated mean difference: -4.195% CI -6.019 to -2.181 p< 0.001BaselineG1: 74G2: 756 months:G1: 82G2: 76Calculated OR at 6 months:1.392 95% CI: 1.160 to 1.670p


Table D8. LDL cholesterol: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasFox et al. 2009 14Cohort study/HighG1: MTM program, providedthrough a health planG2: Usual medical care (eligiblebut opt-out from MTM program)G1: 255G2: 56G1: 215G2: 46Percent of patients with diabeteswith LDL-C less than 100 mg/dlat 12 to 24 months.Mean (SD) LDL-C at 12 to 24months.G1: 69G2: 50Calculated OR: 2.2; 95% CI, 1.24 to 4.01;p=0.008G1: 83.4 (31.1)G2: 90.8 (31.0)Calculated mean difference: -7.4,95% CI, -17.30 to 2.50p=0.33 as reported by study authors,p=0.143 as calculatedAbbreviations: ATPIII=Adult Treatment Panel III (Expert Panel on Detection, Evaluation; and Treatment of High Blood Cholesterol); CAD = coronary artery disease; CI =confidence interval; G = group; HEDIS= Healthcare Effectiveness Data and Information Set; LDL = low density lipoprotein; LDL-C= low density lipoprotein cholesterol; mg/dl =milligrams per deciliter; MTM = Medication Therapy Management; N = number; NR = not reported; OR = odds ratio; RCT= randomized controlled trial; SD = standard deviation;vs. = versus.D-111


Table D9. Blood pressure: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTaylor, Byrd, andKrueger, 2003 52RCT/MediumG1: Pharmaceutical careG2: Standard careG1: 24G2: 29(Was only assessedamong patients withHTN in each studyarm)Percent of patients with SBP andDBP at goal at 12 months.Baseline:G1: 12.5G2: 31.0p=0.109Follow-up:G1: 91.7G2: 27.6p=0.001Calculated OR: 28.875; 95% CI, 5.486 to151.993; p


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPark et al. 1996 43RCT/HighG1: Community-pharmacybased pharmaceutical careprogramG2: Usual careG1: 23G2: 26Mean (SD) SBP (mm Hg) at fourmonths.Baseline:G1: 155.5 (21.1)G2: 147.9 (19.6)p:NS (between-group difference)Follow up:G1: 143.2 (11.5) (p


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPlanas et al. 2009 45RCT/HighG1: Community pharmacybasedhypertension MTM G2: 15G1: 25program for patients withdiabetesG2: Control group (BP recorded,informed of BP goals at threetime during study)Mean (SD) change in SBP (mmHg) at nine monthsPercent of patients at BP goal atnine months.G1: -17.3G2: 2.7Between-group difference (95% CI): -20.0(-32.7 to -7.4)p: 0.003BaselineG1: 16.0G2: 20.0Calculated p: 0.7149 monthsG1: 48.0G2: 6.7p: 0.007OR for intervention group OR : 12.9 (1.5 to 113.8)participant achieving BP goal p: 0.021relative to control group. (95% CI)D-114


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasBrummel et al. 2013 4 ,Soliman, 2013 et al. 5Ramalho de Oliveira etal., 2010 6Cohort study/MediumG1: Opted into clinic-basedMTM programG2: Usual medical care (optedout of MTM program)G1: 121G2: 103Percentage achieving BP goal(defined as less than 130/80)after 12 months of demonstrationand after 24 months ofdemonstration (i.e., 12 monthsafter end of demonstration)BaselineG1: 66.12G2: 61.17p=0.4412 monthsG1: 71.07G2: 72.82p=0.77Unadjusted calculated OR: 0.917 95% CI0.511 to 1.647, p=0.773Adjusted difference in differencecoefficient:0.73, 95% CI 0.32 to 1.65,p=0.4524 monthsG1: 76.03G2: 69.9p=0.3Unadjusted calculated OR: 1.366 95% CI0.755 to 2.471, p=0.303Adjusted difference-in-differencecoefficient: NRD-115


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasJeong et al., 2009 25Cohort/MediumG1: Participants in Part D G1: 1,301Medicare MTM program G2: 982G2: Control subjects withoutPart D Medicare as their primarydrug benefit but otherwisesimilar to intervention subjects.(Study included moresubjects but thisoutcome was assessedamong only patientswith diabetes and HTNwithin each study arm)G1: 1101G2: 895(Study included moresubjects but thisoutcome was assessedamong only patientswith HTN but withoutDM within each studyarm)Percentage with BP control(defined as < 130/80 mmHg)BaselineG1: 48G2: 436 monthsG1: 48G2: 49p=NS (adjusted for baseline BP status)Calculated OR: 0.953, 95% CI 0.808 to1.125, p=0.571BaselineG1:75G2: 706 monthsG1: 73G2: 76p=NS (adjusted for baseline BP status)Calculated OR: 0.898, 95% CI 0.733 to1.099, p=0.296D-116


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al., 1997 7Barnette, Murphy, andCarter, 1996 8Cohort study/HighG1: Pharmacy-basedpharmaceutical careG2: Usual medical careG1: 25G2: 26Mean (SD) SBP (mm Hg) at 6monthsBaselineG1: 151 (21)G2: 145 (19)p: 0.29Mean (SD) DBP (mm Hg) at 6months.Follow-upG1: 140 (14)G2: 143 (20)Calculated Mean Difference: -9.0095% CI, -19.451 to 1.451; p=0.09558BaselineG1: 82 (9)G2: 80 (9)p: NSPercent with blood pressurecontrolFollow-upG1: 80 (8)G2: 79 (10)Calculated Mean Difference, -1.00;95% CI, -5.977 to 3.977; p=0.694Baseline:G1: 52G2: 54Calculated p=0.90Follow-up:G1: 68G2: 58Calculated OR: 1.558;95% CI, 0.496 to 4.898; p=0.448D-117


Table D9. Blood pressure: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasIsetts et al., 2008 22Cohort study/HighG1: MTM services provided byhealth plan in existing medicalcare clinics in collaboration withprimary care providers.G2: Usual medical care withoutMTMG1: 128G2: 126Percent of patients meetingHEDIS measures related tohypertension management at 12months.G1: 71G2: 59Calculated OR: 1.728;95% CI, 1.026 to 2.911; p=0.04Abbreviations: BP = blood pressure; CI = confidence interval; DBP = diastolic blood pressure; DM = diabetes mellitus; G = group; HEDIS = Healthcare Effectiveness Data andInformation Set; HTN = hypertension; mm Hg = millimeter mercury; MTM = Medication Therapy Management; NR= not reported; NS = not sufficient; OR = odds ratio; Q =quarter; RCT= randomized controlled trial; SBP = systolic blood pressure; SD = standard deviation; SMD = standardized mean differenceD-118


Table D10. Drug therapy problems identified: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasKrska et al., 2001 28RCT/HighWelch et al., 2009 56Cohort/HighG1: Pharmacist-led medicationreviewG2: Usual care includingidentification of pharmaceuticalcare issues, but no planG1: MTM program provided tohome-based beneficiariesG2: No-MTM control group(voluntary opt-out)G1: 168G2: 164G1: 459G2: 123Drug therapy problems identifiedfor each study arm at 3 monthsAt least 1 potential drug therapyproblem during MTM processAbbreviations: G = group; MTM = Medication Therapy Management; NA = not applicable; RCT= randomized controlled trial.G1: 1206G2: 1380G1: 89.8%G2: 83.7%Calculated p=0.062D-119


Table D11. Drug therapy problems resolved: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasKrska et al., 2001 28RCT/HighBernsten et al., 2001 1,2RCT/HighMoczygemba et al., 2011 37Moczygemba et al., 2008 38Cohort/MediumG1: Pharmacist-led medicationreviewG2: Usual care includingidentification of pharmaceuticalcare issues, but no planG1: Structured communitypharmacy-basedpharmaceutical care programG2: Usual community pharmacyservicesG1: Opt-in telephone MTMprogramG2: No-MTM control groupG1: 168G2: 164BaselineG1: 1290G2: 11646 monthsG1: 1024G2: 95312 monthsG1: 863G2: 76418 monthsG1: 704G2: 636G1: 60G2: 60Drug therapy problems wholly orpartially resolved at 3 monthsNumber of changes in therapy atbaseline, 6, 12, and 18 monthsMedication and health-relatedproblems identified at baselineand 6 monthsG1: 998G2: 569BaselineG1: 1.1 (1.3)G2: 0.9 (1.2)p:


Table D11. Drug therapy problems resolved: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35aCohort/MediumCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD,receiving MTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD,receiving MTM with CMRG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Remove drug-drug interactionwithin 365 days after date ofMTM enrollment (forinterventions) or randomlyassigneddate in 2010 (forcomparators)Odds (95% CI)Congestive heart failureG1 vs. G13: 0.87 (0.76, 1.00), p>0.05G3 vs. G14: 1.05 (0.88, 1.26), p>0.05Chronic obstructive pulmonary diseaseG5 vs. G15: 0.92 (0.79, 1.07), p>0.05G7 vs G16: 1.11 (0.89, 1.38), p>0.05DiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRComparison – congestive heartfailureG13: enrolled in PDP, usualcareG14: enrolled in MA-PD, usualcareComparison - Chronicobstructive pulmonary diseaseG15: enrolled in PDP, usualcareG16: enrolled in MA-PD, usualcareComparison - DiabetesG17: enrolled in PDP, usualcareG18: enrolled in MA-PD, usualcareD-121


Table D11. Drug therapy problems resolved: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35a(continued)Discontinue use of high risk Odds (95% CI)medications within 365 days after Congestive heart failuredate of MTM enrollment (for G1 vs. G13: 1.04 (.97, 1.11), p>0.05interventions) or randomlyassigneddate in 2010 (forG3 vs. G14: 0.93 (0.86, 1.01), p>0.05comparators)Chronic obstructive pulmonary diseaseG5 vs. G15: 1.06 (0.99, 1.13), p>0.05G7 vs. G16: 1.00 (0.92, 1.09), p>0.05Discontinue use of medicationcontraindicated for congestiveheart failure within 365 days afterdate of MTM enrollment (forinterventions) or randomlyassigneddate in 2010 (forcomparators)Odds ratio (95% CI)G1 vs. G13: 0.63 (0.58, 0.67), p


Table D12. Medication adherence: Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsAdherence Outcome Type 1: Proportion of patients adherent based on a threshold of percent of pills takenPindolia et al., 2009 44Cohort study/HighTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Telephone based MTMProgramG2: Patients eligible for MTMprogram who declinedenrollmentG1: Pharmaceutical careG2: Standard careG1: 292G2: 1081(study year 1)G1: 33G2: 36Percent of CHF patients whoare adherent to at least 75% ofACE/ARB based on 2006claims data: Measured during 6months post-MTMP enrollmentcompared with 6 months preenrollmentPercent of CHF patients whoare Adherent to at least 75% ofBeta Blocker based on 2006claims data: Measured during 6months post-MTMP enrollmentcompared with 6 months preenrollmentPercentage of patients adherentdefined as self-reported taking80% or more of medications 12months after baselinePre-testG1: 36G2: 38.5OR: 0.995% CI (0.7 to 1.2)p: 0.43Post-testG1: 40G2: 38OR: 1.195% CI (0.8 to 1.4)p=0.53Pre-testG1: 34.5G2: 33OR: 1.195% CI (0.8 to 1.4)p=0.63Post-testG1: 34G2: 30.5OR: 1.295% CI (0.9 to 1.5)p: 0.25G1: 100G2: 88.9p: 0.115Calculated OR: 9.3;95% CI, 0.5 to 179.3; p=0.140D-123


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsBrummel et al. 2013 4 , Soliman,2013 et al. 5 Ramalho deOliveira et al., 2010 6Cohort study/MediumG1: Opted into clinic-basedMTM programG2: Usual medical care (optedout of MTM program)G1: 121G2: 103Percentage of patients adherentto aspirin (from pharmacyclaims data) at baseline (beforeMTM), 12-month (during MTMdemonstration project), and 24-month (1 year postdemonstration)Before MTMG1: 97.52%G2: 93.29%P: 0.11BaselineCalculated OR 2.828, 95% CI 0.710 to11.259, p=0.14)During MTM Demonstration ProjectG1: 100%G2: 98.06%P: 0.1212 MonthCalculated OR 5.981 (95% CI 0.284 to126.030, p=0.250)1 Year Post-Demonstration ProjectG1: 99.17%G2: 99.03%P: 0.924 MonthCalculated OR 1.17 (95% CI 0.072 to18.903, p=0.912)D-124


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsAdherence Outcome Type 2: Absolute measure of adherence as percent of prescribed doses takenPerlroth et al., 2013 35 a Congestive heart failureG1: enrolled in PDP receivingMTM with CMRCohort/MediumG3: enrolled in MA-PD,receiving MTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD,receiving MTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRComparison—congestive heartfailureG13: enrolled in PDP, usualcareG14: enrolled in MA-PD, usualcareG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Percentage of patientsachieving adherence (> 80% ofprescribed pills taken) tovarious specified medications365 days after date of MTMenrollment (for interventions) orrandomly-assigned date in 2010(for comparators)CHFAdherent to any evidence-basedmedicine (EBM) for CHFG1 vs. G13: 1.28 a (1.19, 1.37)G3 vs. G14: 1.40 a (1.29, 1.52)p


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsPerlroth et al., 2013 35a(continued)G18: enrolled in MA-PD, usualcareAdherent to DPP-IV inhibitors medicationG9 vs. G17: 1.32 a (1.12, 1.55)G11 vs. G18: 1.19 (.96, 1.48)p


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsPlanas, et al 2009 45RCT/HighPark et al., 1996 43RCT/HighG1: Collaborative home-basedmedication reviewG2: No medication reviewreceivedG1: Comprehensivepharmaceutical careG2: Usual careParticipantsG1: 25G2: 15Visit 1G1: 7G2: 5Visit 2G1: 21G2: 23Visit 3G1: 23G2: 20Visit 4G1: 21G2: 22Percent mean adherence(percent of prescribed dosestaken) to antihypertensivemedicationMeasured twice (9 monthsbefore and 9 months afterbaseline visit) and continuouslyusing medication acquisitionmethod, in which days' supplyof medication compared withdates medication filled usingpharmacy refill data.Mean percent compliance(percent of prescribed pillstaken) from pharmacist report ofpill counts4 month timeframe9 months before baseline, % (95% CI)G1: 80.5 (74.9 to 86.0)G2: 79.5 (71.0 to 88.1)9 months after baseline, % (95% CI)G1: 87.5 (82.1 to 93.0)G2: 78.8 (69.7 to 87.9)p: 0.0712Calculated standardized difference inmeans from Baseline to 9 months: 0.295% CI (-0.4 to 0.9)p: 0.46Baseline/Visit 1G1: 87.4 (0.9)G2: 87.8 (13.7)Visit 2G1: 96.7 (4)G2: 86.0 (20.7)p=0.025Visit 3G1: 97.2 (4.4)G2: 86.7 (23.1)p=0.037Visit 4G1: 86.8 (28.7)G2: 89.1 (21.8)Calculated standardized difference inmeans for change from baseline toVisit 4: -0.195% CI (-0.7 to 0.5)p=0.77D-127


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsMoore, 2013 40Cohort/MediumG1: MTM program (opt-in)G2: control group (refusers)G1: 2,250G2: 2,250Medication possession ratio bymedication type from pharmacydata and medical claims datafrom 365 days preceding thepatient’s program invitation datetos 365 days following patient’sprogram invitation dateAsthma MPR at baseline [% (SE)]G1: 56.8 (1.544)G2: 53.8 (1.636)p: 0.180mean change in Asthma MPR [% (SE)]G1: 2.33 (1.22)G2: 1.71 (1.38)p: 0.739Asthma MPR (%)Calculated mean difference = 0.62; 95%CI: -2.988 to 4.228; p = 0.736Depression MPR at baseline [% (SE)]G1: 72.0 (1.064)G2: 74.5 (0.946)p: 0.075mean change in Depression MPR [%(SE)]G1: 1.23 (1.06)G2: 0.07 (0.98)p: 0.420Depression MPR (%)Calculated mean difference = 1.16; 95%CI: -1.667 to 3.987; p = 0.421Diabetes MPR at baseline [% (SE)]G1: 76.0 (1.031)G2: 73.7 (0.984)p: 0.108mean change in Diabetes MPR [% (SE)]G1: 1.64 (1.01)G2: -0.73 (1.08)p: 0.112D-128


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsMoore, 2013 40(continued)Diabetes MPR (%)Calculated mean difference = 2.37; 95%CI: -0.549 to 5.289; p = 0.112Dyslipidemia MPR at baseline [% (SE)]G1: 80.9 (0.581)G2: 80.0 (0.639)p: 0.263mean change in Dyslipidemia MPR [%(SE)]G1: 2.10 (0.66)G2: -2.61 (0.76)p


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsAdherence Outcome Type 3: Self-reported Adherence using Morisky ScaleBernsten, 2001 1 ;Sturgess, 2003 2RCT/ High (pooled data)G1: Structured communitypharmacy-basedpharmaceutical care programG2: Normal pharmaceuticalUsual community pharmacyservicesPooled sample(excluding TheNetherlands becauseno baselineadherence datacollected)BaselineG1: 867G2: 7486 monthsG1: NRG2: NR12 monthsG1: NRG2: NR18 monthsG1: 792G2: 758Medication adherence: selfreportedas assessed byMorisky Scale(Note: Percent of participantswho we adherent defined aspatients responded that they“never” experienced anyaspects of non-compliance onthe 4-item 4-point scale)Pooled sample (percent adherent)BaselineG1: 33.9G2: 38.6OR: 0.8Calculated 95% CI (0.7 to 1.0)p: 0.0496 monthsG1: 38.5G2: 36.6p: NR12 monthsG1: 43.8G2: 37.3p: NR18 monthsG1: 38.2G2: 39.4OR: 1.195% CI (0.9 to 1.3)p=0.440101Percent changing from nonadherent toadherent over 18 monthsG1: 15.2G2: 12.2p: 0.028D-130


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsVolume et al., 2001 54 (PREP);Kassam et al., 2001 55RCT/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careT1: N=363G1: 159G2: 204T2: N=317T3: N=292Self-reported adherence usingthe 4-item 2-point MoriskyScale where summary score is0-4 with lower scores beingbetter adherenceMean Adherence Scale ScoreBaseline:G1: 0.5 (0.8)G2: 0.6 (0.9)p: NSEstimated by groupbased on overallretentionG1: 127G2: 163Time 1 (Baseline), Time 2 (midpoint,6 to 7 months afterintervention) and Time 3 (12 to13 months after intervention)Calculated standardized difference inmeans: -0.195% CI (-0.3 to 0.1)p=0.2089576-month followup:G1: 0.5 (0.7)G2: 0.6 (0.8)p:NS12-month followup:G1: 0.6 (0.8)G2: 0.5 (0.7)p: NSCalculated standardized difference inmeans: -0.1395% CI (-0.11 to 0.36)p=0.289285D-131


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsJameson, VanNoord, andVanderwoud, 1995 23RCT/High (medium for studyoverall by high for adherencedue to poor measure)G1: Consultation with a clinicalpharmacist within a primarycare office.G2: Standard medical care atthe primary care office.G1: 27G2: 29Self-reported composite“understanding andcompliance” 0-12 score atbaseline and 6 months(no further information onmeasure used)Change in self-reportedcomposite score over 6 monthswith negative scorerepresenting improvementBaseline Means Scale Score (SD notreported)G1: 2.3G2: 2.3p: NS6 monthsG1: 0.6G2: 2.1p: NSG1: -1.6G2: -0.2p: NSD-132


Table D12. Medication adherence: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsMiscellaneous Adherence OutcomesHanlon et al., 1996 17G1: Usual care, plus clinicalRCT/Medium (low for study pharmacist care.overall but medium foradherence due to lack of G2: Usual care in the GMCinformation about and precisionof adherence measure)Sidel et al., 1990 50RCT/MediumG1: received at least 2pharmacist visits involvingmedication review, patientspecific education andcounseling; follow up patientphone calls and contact ofphysicians as neededG2: only contacted for tocomplete the survey.G1: 86G2: 83G1: 92G2: 104Self-report MedicationCompliance with 12 month timeframe, assessed by determiningwhether the way patient saidthey took the medicine (in termsof number of pills and dailyfrequency) matched how it wasprescribed. Compliance wasdefined as the proportion ofmedications for which thepatients’ response agreed withthe directionsMedication-taking BehaviorSubscore in change frombaseline to 6 month follow-up(negative scores indicateimprovement= decreased risk)Change in normative score forRemembering to take Medicineat 6 monthsBaseline:G1: 73%G2: 74%OR: 0.9595% CI (0.48 to 1.88)p: 0.8812 Months Follow-upG1: 77.4%G2: 76.1%p: 0.88G1: -3.47G2: -4.38p< .001 for within group differencesp: 0.52 for between group differencesG1: 0.09G2: -0.19p: 0.52a Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement for CMR. We have excluded groups 2, 4, 6, 8, 10, and12 from our summary tables.Abbreviations: ACE/ARB = Angiotensin-Converting Enzyme/Angiotensin II Receptor Blockers; CMR = comprehensive medication review; CHF = Cardiovascular Heart Failure;CI = confidence interval; G = group; GMC = General Medicine Clinic; MA-PD = Medicare Advantage Part D Plan; MPR = medication possession ratio; MTM = MedicationTherapy Management; MTMP = Medication Therapy Management Program; NR = not reported; NS = not sufficient; OR = odds ratio; PDP = Medicare Part D Plan; PREP =Pharmaceutical Care Research and Education Project; RCT= randomized controlled trial; SD = standard deviation; SWHP = Scott & White Health Plan; T = time.D-133


Table D13. Medication appropriateness scales: Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: 105G2: 103Covariate-adjusted MedicationAppropriateness IndexAssessed at baseline, 3, 12months by blinded researchpharmacistBaselineG1: 17.7 (0.6)G2: 17.6 (0.6)3 monthsG1: 13.4 (0.6)G2: 16.5 (0.6)95% CI: NRp:


Table D13. Medication appropriateness scales: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsCarter et al., 1997 7 ;Barnette, Murphy, andCarter, 1996 8Cohort study/HighG1: Pharmacy-basedpharmaceutical careG2: Usual medical careG1: 25G2: 26Appropriateness of BP regimenA blinded review panel of threeevaluated cases in random orderon a visual analog scale, usingmedical records. Theinvestigators averaged andconverted scores to a numericalvalue by measuring the distancefrom the best option. Scorearranged from 0-16.2. Higherscores are more positive.Appropriateness of daily dosageBP RegimenBaselineG1: 8.7 (4.7)G2: 10.3 (4.8)Follow-upG1: 10.9 (4.5)G2: 10.1 (5.2)p for change scores NRAppropriateness of daily dosageBaselineG1: 11.6 (4.5)G2: 12.6 (4.5)Appropriateness of dosingintervalFollow-upG1: 13.4 (3.7)G2: 13.2 (4.1)p for change scores NRAppropriateness of dosing intervalBaselineG1: 13.8 (4.3)G2: 13.4 (4.6)Follow-upG1: 15.1 (2.3)G2: 13.8 (4.1)p for change scores NRAbbreviations: BP = blood pressure; CI = confidence interval; G = group; GMC = General Medicine Clinic; NR = not reported; RCT = randomized controlled trialD-135


Table D14. Medication appropriateness for individual medications: Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsMcDonough et al.,2005 36cluster-randomizedRCT/MediumG1: Pharmaceuticalcare provided bypharmacist in acommunity pharmacyG2: Usual careBaselineG1: 70G2: 26Follow-upG1: 61G2: 19Nine Month Follow-upPercentage of patients (at risk forglucocorticoid-inducedosteoporosis) on bisphosphonatedrug therapyPercentage of patients (at risk forglucocorticoid-inducedosteoporosis) on estrogen drugtherapyPercentage of patients (at risk forglucocorticoid-inducedosteoporosis) taking calciumsupplementsBaselineG1: 17.1G2: 0p:


Table D14. Medication appropriateness for individual medications: Summary of results (continued)StudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsGattis et al., 1999 15RCT/MediumG1: ClinicalpharmacistinterventionG2: Usual medicalcareG1: 90G2: 91G1: 12G2: 196 month follow-upPercent receiving an ACEI atfollow-upFraction of target ACEI dose atfollow upOf those NOT on an ACEI atfollow-up, percentage receivingalternative drug therapyG1: 87G2: 79p: 0.18G1: 1 (25%: 0.5, 75%: 1)G2: 0.5 (25% 0.188, 75%: 1)95% CI: NRp: < 0.001G1: 75G2: 26p: 0.02Abbreviations: ACEI = Angiotensin-Converting Enzyme Inhibitors; CI = confidence interval; G = group; NR = not reported; NS = not sufficient; RCT = randomized controlledtrialD-137


Table D15. Medication Appropriateness Index Item 1 (Is there an indication for the drug?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 10.5G2: 12.43 monthsG1: 8.1G2: 10.5p: NR12 monthsG1: 6.0G2: 9.7p: NRBaselineG1: 33.3G2: 46.812 monthsG1: 16.1G2: 48.212 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-138


Table D16. Medication Appropriateness Index Item 2 (Is the medication effective for the condition?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 4.5G2: 4.93 monthsG1: 3.6G2: 4.9p: NR12 monthsG1: 3.4G2: 4.9p: NRBaselineG1: 29.1G2: 44.912 monthsG1: 13.6G2: 44.612 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-139


Table D17. Medication Appropriateness Index Item 3 (Is the dosage correct?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 17.4G2: 17.33 monthsG1: 13.1G2: 18.2p: NR12 monthsG1: 15.0G2: 20.4p: NRBaselineG1: 63.3G2: 62.312 monthsG1: 12.9G2: 63.812 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-140


Table D18. Medication Appropriateness Index Item 4 (Are the directions correct?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 32.7G2: 32.23 monthsG1: 28.1G2: 32.6p: NR12 monthsG1: 27.5G2: 29.9p: NRBaselineG1: 70.5G2: 64.312 monthsG1: 29.7G2: 56.712 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-141


Table D19. Medication Appropriateness Index Item 5 (Are the directions practical?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 20.7G2: 20.03 monthsG1: 15.8G2: 18.9p: NR12 monthsG1: 15.3G2: 21.2p: NRBaselineG1: 61.0G2: 57.012 monthsG1: 29.7G2: 56.712 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-142


Table D20. Medication Appropriateness Index Item 6 (Are there clinically significant drug-drug interactions?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 0G2: 03 monthsG1: 0G2: 0.1p: NR12 monthsG1: 0G2: 0.1BaselineG1: 22.9G2: 17.912 monthsG1: 5.8G2: 22.812 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-143


Table D21. Medication Appropriateness Index Item 7 (Are there clinically significant drug-disease interactions?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 1.9G2: 1.03 monthsG1: 2.0G2: 0.7p: NR12 monthsG1: 1.9G2: 1.1p: NRBaselineG1: 18.6G2: 21.312 monthsG1: 9.0G2: 19.612 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-144


Table D22. Medication Appropriateness Index Item 8 (Is there unnecessary duplication with other drugs?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847Percent PrescriptionInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistBaselineG1: 4.9G2: 6.43 monthsG1: 3.0G2: 5.9p: NR12 monthsG1: 4.9G2: 8.2p: NRTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careG2: Standard careN participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 11.9G2: 6.812 monthsG1: 4.5G2: 7.612 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-145


Table D23. Medication Appropriateness Index Item 9 (Is the duration of therapy acceptable?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: Pharmaceutical careG2: Standard careN participantsG1: 105G2: 103Number ofprescriptions:BaselineG1: 798G2: 84612 monthsG1: 734G2: 847N participantsG1: 33G2: 36Number ofprescriptions:BaselineG1: 210G2: 207Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistPercent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistBaselineG1: 15.4G2: 17.53 monthsG1: 11.8G2: 14.9p: NR12 monthsG1: 10.1G2: 14.9p: NRBaselineG1: 35.2G2: 48.812 monthsG1: 18.1G2: 49.112 monthsG1: 155G2: 224Abbreviations: G = group; GMC = General Medicine Clinic; N = number; NR = not reported; RCT = randomized controlled trialD-146


Table D24. Medication Appropriateness Index Item 10 (Is this drug the least expensive alternative compared with others of equalutility?): Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsHanlon et al., 1996 17RCT/LowG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GMCG1: 105G2: 103Percent PrescriptionsInappropriateAssessed at baseline, 3, 12months by blinded researchpharmacistBaselineG1: 29.2G2: 30.33 monthsG1: 25.6G2: 27.7p: NRTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careG2: Standard careBaselineG1: 210G2: 20712 monthsG1: 155G2: 224Percent PrescriptionsInappropriateAssessed at baseline, 12 monthsby blinded research pharmacistAbbreviations: G = group; GMC = General Medicine Clinic; NR = not reported; RCT = randomized controlled trial12 monthsG1: 25.3G2: 28.2p: NRBaselineG1: 50.0G2: 62.312 monthsG1: 38.7G2: 60.3D-147


Table D25. Medication dosing: Summary of resultsStudyDesign/Risk of BiasStudy Arms N Analyzed Outcome and Time Period ResultsJameson, VanNoord, andVanderwoud, 1995 23RCT/MediumG1: Consultation with a clinicalpharmacist within a primarycare office.G2: Standard medical care atthe primary care office.G1: 27G2: 29Change in number of doses per day at6 months follow up.Abbreviations: G = group; mg/kg = milligram/kilogram; MTM = Medication Therapy Management; RCT = randomized controlled trialG1: - 1.6G2: 2.2p: 0.007D-148


Table D26. Adverse events: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTouchette et al., 2012 53RCT/LowG1: Basic MTM services (withmedication information gleanedthrough patient interview)G2: Enhanced MTM services(pharmacist provided with 2page clinical summary frompatient medical record).G3: Usual pharmacy careG1: 211G2: 218G3: 208Percent of patients with an ADEbetween 0 and 3 months and ORPercent of patients with an ADEbetween 3 and 6 months and ORMean number (SD) of ADEs perpatient between 0 and 3 monthsG1: 42.2G2: 27.9G3: 33.7G1 vs. G3: OR: 1.6p=0.078G2 vs. G3: OR: 0.7p=0.278G1: 36.1G2: 31.1G3: 34.4G1 vs. G3: OR: 1.1p=0.717G2 vs. G3: OR: 0.9p=0.672G1: 0.750 (1.113)G2: 0.547 (1.184)G3: 0.559 (1.202)G1 vs. G3: Calculated Mean Difference,0.191;95% CI, -0.031 to 0.413p=0.091Mean number (SD) of ADEs perpatient between 3 and 6 monthsG2 vs. G3: Calculated Mean Difference,-0.012;95% CI, -0.239 to 0.215p=0.917G1: 0.814 (1.421)G2: 0.530 (0.894)G3: 0.468 (0.820)G1 vs. G3: Calculated Mean Difference,0.284;95% CI, 0.056 to 0.512p=0.014G2 vs. G3: Calculated Mean Difference,-0.062; 95% CI, -0.225 to 0.101p=0.455D-149


Table D26. Adverse events: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17RCT/MediumTaylor, Byrd, and Krueger,2003 52RCT/High aG1: Clinical pharmacist carewithin a general medicine clinic.G2: Usual careG1: Pharmaceutical careG2: Standard careG1: 86G2: 83G1: 33G2: 36Percent with an ADE at 12monthsPercent of patients with at leastone medication misadventure at12 monthsG1: 30.2G2: 40.0p=0.19Calculated OR: 0.6;95% CI, 0.37 to 1.15p=0.14G1: 2.8 b (N=4)G2: 3.0 b (N=3)Calculated OR based on reported percent:0.93;95% CI, 0.056 to 15.603p: 0.0961Jameson, VanNoord, andVanderwoud, 1995 23RCT/High aFischer et al., 2000 10NRCT/High aG1: Consultation with a clinical G1: 27pharmacist within a primary care G2: 29office.G2: Standard medical care at theprimary care office.G1: Comprehensive drugtherapy management programG2: Standard communitypharmacy practiceG1: 201G2: 368Change in mean medication sideeffect score at six months.OR for likelihood of reporting sideeffects or problems due toprescription medication (95% CI)Calculated OR based on reported N: 1.5(95% CI, 0.31 to 7.34), p= 0.606G1: -3.7G2: -1.9p: NS and unable to calculate.1.8 (1.20 to 2.80)a This study was rated medium risk of bias overall, but due to measurement bias with this specific outcome, it was considered high risk of bias for this outcome.b The percent reported by authors cannot be generated based on the reported N and the reported number of events.Abbreviations: ADE = adverse drug event; CI = confidence interval; G = group; MTM = Medication Therapy Management; N = number; NRCT = nonrandomized controlled trial;NS = not significant; NS = not sufficient; OR = odds ratio; RCT = randomized controlled trial; SD = standard deviation; vs. = versus.D-150


Table D27. Cognitive, affective, and physical functioning: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasWilliams, 2004 57RCT/MediumG1: Modification of patient'smedication regimen by aninterdisciplinary medicationadjustment teamG2: Usual medical careG1: 57G2: 76Baseline and 6 weeksMean (SD)Timed manual performance(seconds)Physical performance test(seconds)Baseline 6 weeksG1: 9.8 (4.2) 9.3 (4.0)G2: 9.6 (3.81) 9.0 (4.1)Calculated mean differences at 6 weeks(unadjusted for baseline differences):0.300; 95% CI, -1.093 to 1.693; p=0.673G1: 57.2(28.59) 59.6(31.6)G2: 57.2(28.88) 56.3(27.5)Calculated mean differences at 6 weeks(unadjusted for baseline differences):-3.300; 95% CI, -13.370 to 6.770; p=0.52Functional reach (inches) G1: 11.5(3.0) 11.3(3.3)G2: 11.2(3.1) 11.3(3.0)Calculated mean differences at 6 weeks(unadjusted for baseline differences):0.00; 95% CI, -1.076 to 1.076; p=1.0Digit Span (WAIS) G1: 12.8(4.72) 13.3(4.3)G2:12.8(3.82) 13.1(4.3)Calculated mean differences at 6 weeks(unadjusted for baseline differences):0.200; 95% CI, -1.277 to 1.677; p=0.791Digit Symbol (WAIS) G1: 33.9(15.8) 33.1(15.9)G2: 30.4(14.3) 33.1(14.2)Calculated mean differences at 6 weeks(unadjusted for baseline differences):0.00; 95% CI, -5.134 to 5.134; p=1.0Randt Memory Test G1: 9.7(3.7) 9.6(3.5)G2:10.1(3.75) 9.6(3.4)Calculated mean differences at 6 weeks(unadjusted for baseline differences):0.00; 95% CI, -1.182 to 1.182; p-1.00D-151


Table D27. Cognitive, affective, and physical functioning: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasWilliams, 2004 57RCT/Medium(continued)Marques et al., 2013 33RCT/MediumG1: Modification of patient'smedication regimen by aninterdisciplinary medicationadjustment teamG2: Usual medical care(continued)G1: Dader methodpharmacotherapy interventionG2: Usual careG1: 57G2: 76(continued)G1: 22G2: 26G1: 22G2: 26G1: 22G2: 26G1: 5G2: 5G1:13G2:13CES-D score G1: 12.4(7.4) 11.9(7.9)G2:11.9(8.3) 10.8(7.9)Calculated mean differences at 6 weeks(unadjusted for baseline differences):-1.10 95% CI -3.813 to 1.613, p=0.427Self-rating AnxietyScale scoreBeck Depression Inventory Scoreat baseline and three months:Beck Anxiety Inventory Score atbaseline and three months:Percentage with DepressionRemission(defined as BDI < 11)Percentage with severedepression improvementPercentage with moderatedepression improvementG1: 14.2(7.8) 13.2(6.5)G2: 14.4(6.4) 13.1(6.8)Calculated mean differences at 6 weeks(unadjusted for baseline differences):-0.100 95% CI -2.392 to 2.192, p=0.932Baseline , 3 months, Mean changeG1:28, 14.5, -13.5G2: 23, 20.5, -2.595% CI: NRp: 0.0275Baseline, 3 months , Mean changeG1: 29, 16, -13.0G2: 24, 20,.5 -3.095% CI: NRp: 0.0194G1: 30.4G2: 15.3Calculated OR, 2.406; 95% CI, 0.601 to9.632, p=0.215G1: 80.0G2: 60.0Calculated OR, 2.667; 95% CI, 0.158 to45.141, p=0.497G1:53.8G2: 7.7Calculated OR, 14.00; 95% CI, 1.385 to141.485, p=0.025Abbreviations: BDI= Beck Depression Inventory; CES-D= Center for Epidemiological Studies-Depression Scale; CI= confidence interval; G = group; NR = not reported; OR=odds ratio; SD= standard deviation; RCT = randomized controlled trial; WAIS= Wechsler Adult Intelligence ScaleD-152


Table D28. All-cause mortality: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasGattis et al., 1992 15RCT/MediumWelch et al., 2009 56Cohort study/MediumYamada et al., 2012 61Cohort study/MediumG1: Clinical pharmacistintervention in addition to usualmedical careG2: Usual medical careG1: MTM program provided tohome-based beneficiariesG2: No-MTM control group (optout)G1: MTM enrolled patientsG2: Eligible MTM patients notenrolled but matched on age,gender, region and DCG riskG1: 90G2: 91G1: 459G2: 336G1: 34,352G2: 138,182Percent died and OR for all-causemortality within 6 months(95% CI)Percent died and adjusted OR forall-cause mortality, within 6months (adjusted for age, sex,chronic disease score, specificbaseline utilization)(95% CI)Adjusted HR for all-causemortality within 1 to 4 years(adjusted for age, sex, Charlson,CHF, ESRD)G1: 3.3%G2: 5.5%OR: 0.59 (0.12 to 2.49)p=0.48G1: 4.1%G2: 7.4%Adjusted OR: 0.5 (0.3 to 0.9)p=0.044Adjusted HR: 0.9295% CI, 0.87 to 0.96p< 0.001Abbreviations: CHF=congestive heart failure; CI = confidence interval; DCG=diagnostic cost group (a measure of health care use and comorbidity), ESRD=end-stage renaldisease; G = group; HR=hazard ratio; MTM = Medication Therapy Management; OR = odds ratio; RCT = randomized controlled trial;D-153


Table D29. Gastrointestinal bleeding events: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPindolia et al., 2009 44Cohort study/HighG1: Opted in to a telephonebased MTM ProgramG2: Usual medical care (optedout of MTM program)G1: NRChange in percent of patientsG2: NRgastrointestinal bleeding at 6(Was only assessed months.among patients witharthritis in eachstudy arm and N forthis outcome wasnot reported)Abbreviations: CI = confidence interval; G = group; MTM = Medication Therapy Management.G1: -60%G2: 0 %Between-group p: 0.001D-154


Table D30. Self-reported health status: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: 447G2: 484SF-36 Physical FunctioningDomain (change from baseline)6-Month Follow-upG1: -4.9 (1.0 SE)G2: -3.4 (0.9 SE)12-Month Follow-upG1: -5.3 (1.0 SE)G2: -6.1 (1.0 SE)p=0.412Taylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 33G2: 36SF-36 Physical FunctioningDomainBaselineG1: 62.0 (29.4)G2: 61.9 (24.3)12-Month Follow-upG1: 68.6 (24.0)G2: 56.1 (27.5)Hanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 Physical FunctioningDomainp: NSBaseline:G1: 48.0 (2.7)G2: 45.3 (2.7)12-Month Follow-upG1: 44.1 (2.0)G2: 42.2 (2.0)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636SF-36 Physical FunctioningDomain (Change betweenBaseline and 18-Month Follow-Up)p= 0.99G1: -1.0G2: -0.7p: NSD-155


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Physical FunctioningDomainBaselineG1: 61.5G2: 66.56-Month Follow-upG1: 70.7G2: 67.7Park et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Physical FunctioningDomainp=NRBaselineG1: 77.0 (26.1)G2: 66.3 (29.1)4-Month Follow-upG1: 77.8 (30.4)G2: 70.2 (29.2)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)SF-36 Physical FunctioningDomainSF-36 Physical FunctioningDomainp=NSBaselineG1: 55.6 (95% CI, 55.5 to 56.0)G2: 54.2 (95% CI, 48.0 to 54.4)5-Month Follow-upG1: 55.0 (95% CI, 54.6 to 55.3)G2: 55.0 (95% CI, 54.8 to 55.2)p: 0.93G1: NRG2: NRp: NSD-156


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: 447G2: 484SF-36 Role Physical Domain(change from baseline)6-Month Follow-upG1: -3.5 (1.8 SE)G2: -4.3 (2.1 SE)12-Month Follow-upG1: -4.3 (2.0 SE)G2: -8.2 (2.00 SE)Taylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 33G2: 36SF-36 Role Physical Domainp=0.245BaselineG1: 50.8 (42.2)G2: 47.9 (42.8)12-Month Follow-upG1: 68.2 (42.1)G2: 52.8 (42.2)95% CI: NRHanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 Role Physical Domainp=NSBaseline:G1: 38.3 (3.2)G2: 36.5 (3.2)12-Month Follow-upG1: 38.6 (3.2)G2: 32.3 (3.7)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636SF-36 Role Physical Domain(Change between Baseline and18-Month Follow-Up)p=0.99G1: -1.1G2: -0.3p=NSD-157


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ,Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Role Physical DomainBaselineG1: 54.3G2: 63.56-Month Follow-upG1: 74.0G2: 62.5Park et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Role Physical Domainp=NRBaselineG1: 85.9 (30.0)G2: 77.9 (31.1)4-Month Follow-upG1: 85.2 (31.5)G2: 73.1 (40.6)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)SF-36 Role Physical DomainSF-36 Role Physical Domainp=NSBaselineG1: 53.8 (95% CI, 53.1 to 54.6)G2: 55.0 (54.5 to 55.5)5-Month Follow-upG1: 48.5 (95% CI, 47.8 to 49.3)G2: 52.1 (95% CI, 41.6 to 42.6)p: 0.65G1: NRG2: NRp: NSD-158


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: 447G2: 484SF-36 Bodily Pain Domain(change from baseline)6-Month Follow-upG1: -0.8 (1.0 SE)G2: -3.3 (0.9 SE)12-Month Follow-upG1: -0.3 (1.0 SE)G2: -4.8 (1.0 SE)Taylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 33G2: 36SF-36 Bodily Pain Domainp=0.004BaselineG1: 60.0 (27.0)G2: 65.4 (23.0)12-Month Follow-upG1: 68.5 (22.3)G2: 63.1 (25.8)Hanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 Bodily Pain Domainp=NSBaselineG1: 45.0 (2.8)G2: 42.2 (2.8)12-Month Follow-upG1: 43.6 (2.7)G2: 41.7 (2.7)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636SF-36 Bodily Pain Domain(Change between Baseline and18-Month Follow-Up)p=0.99G1: -0.06G2: +0.53p=NSD-159


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997, 7Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Bodily Pain DomainBaselineG1: 58.4G2: 76.76-Month Follow-upG1: 71.1G2: 74.7Park et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Bodily Painp=NRBaselineG1: 77.4 (19.0)G2: 73.1 (21.3)4-Month Follow-upG1: 80.5 (22.9)G2: 73.7 (19.0)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)SF-36 Bodily PainSF-36 Bodily Painp=NSBaselineG1: 60.5 (95% CI, 60.2 to 60.8)G2: 60.8 (95% CI, 60.6 to 61.0)5-Month Follow-upG1: 56.6 (95% CI, 56.4 to 56.8)G2: 59.0 (95% CI, 58.8 to 59.2)p: 0.65G1: NRG2: NRp: NSD-160


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 447G2: 484G1: 33G2: 36SF-36 General Health PerceptionDomain (change from baseline)SF-36 General Health PerceptionDomain6-Month Follow-upG1: -1.6 (0.8 SE)G2: -2.2 (0.7 SE)12-Month Follow-upG1: -2.4 (0.8 SE)G2: -5.3 (0.8 SE)95% CI: NRp=0.026BaselineG1: 50.8 (19.5)G2: 49.9 (19.8)12-Month Follow-upG1: 57.0 (19.6)G2: 50.1 (15.9)Hanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 General Health PerceptionDomainp: NSBaselineG1: 34.9 (2.1)G2: 34.2 (2.1)12-Month Follow-upG1: 37.4 (1.6)G2: 35.2 (1.7)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636SF-36 General Health PerceptionDomain (Change betweenBaseline and 18-Month Follow-Up)p=0.99G1: +0.28G2: -0.66p: NSD-161


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 General Health PerceptionDomainBaselineG1: 58.2G2: 61.26-Month Follow-upG1: 58.7G2: 64.0Park et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 General Health PerceptionDomainp=NRBaselineG1: 67.8 (18.7)G2: 59.5 (15.1)4-Month Follow-upG1: 72.3 (13.1)G2: 64.7 (19.0)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)SF-36 General Health PerceptionDomainSF-36 General Health PerceptionDomainp: NSBaselineG1: 62.2 (95% CI, 61.9 to 62.6)G2: 65.0 (95% CI, 64.8 to 65.2)5-Month Follow-upG1: 60.5 (95% CI, 60.3 to 60.7)G2: 60.8 (95% CI, 60.6 to 61.0)p: 0.17G1: NRG2: NRp: NSD-162


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPark et al., 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Social FunctioningDomainBaselineG1: 88.6 (16.8)G2: 81.3 (18.5)4-Month Follow-upG1: 90.2 (15.5)G2: 81.0 (19.1)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.G1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)G1: 447G2: 484SF-36 Social FunctioningDomainSF-36 Social FunctioningDomainSF-36 Role Emotional Domain(change from baseline)p: NSBaselineG1: 79.2 (95% CI, 79.0 to 79.4)G2: 81.9 (95% CI, 81.8 to 82.0)5-Month Follow-upG1: 75.4 (95% CI, 75.1 to 75.8)G2: 77.5 (95% CI, 77.3 to 77.7)p: 0.34G1: NRG2: NRp: NS6-Month Follow-upG1: -2.6 (2.2 SE)G2: -3.4 (1.9 SE)12-Month Follow-upG1: -0.3 (2.3 SE)G2: -7.4 (2.3 SE)p=0.065D-163


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 33G2: 36SF-36 Role Emotional DomainBaselineG1: 59.6 (44.7)G2: 69.4 (45.3)12-Month Follow-upG1: 82.8 (36.4)G2: 65.8 (45.4)Hanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 Role Emotional Domainp: NSBaseline:G1: 73.0 (4.1)G2: 68.1 (4.1)12-Month Follow-upG1: 66.4 (1.8)G2: 67.0 (3.9)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighCarter et al.,1997 7 ,Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636G1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Role Emotional Domain(Change between Baseline and18-Month Follow-Up)SF-36 Role Emotional Domainp=0.99G1: +0.2G2: -2.9p: NSBaselineG1: 50.0G2: 69.46-Month Follow-upG1: 63.9G2: 65.3p=NRD-164


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPark et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Role Emotional DomainBaselineG1: 88.4 (25.8)G2: 88.5 (28.2)4-Month Follow-upG1: 92.8 (24.5)G2: 78.2 (29.7)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1Pharmacists conducted faceto-facemedication reviews with G1: 379Baselinethe patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.G1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)G1: 447G2: 484SF-36 Role Emotional DomainSF-36 Role Emotional DomainSF-36 Mental Health Domain(change from baseline)p: NSBaselineG1: 71.8 (95% CI, 70.9 to 72.7)G2: 74.9 (95% CI, 74.5 to 75.2)5-Month Follow-upG1: 66.4 (95% CI, 65.7 to 67.0)G2: 72.7 (95% CI, 72.1 to 73.2)p: 0.80G1: NRG2: NRp: NS6-Month Follow-UpG1: -0.5 (0.8 SE)G2: -1.4 (0.7 SE)12-Month Follow-upG1: 0.1 (0.8 SE)G2: -2.3 (0.8 SE)p=0.029D-165


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTaylor, Byrd, and Krueger,2003 52RCT/MediumG1: Pharmaceutical careprovided by pharmacist inconjunction with an outpatientphysician office visitG2: Standard care.G1: 33G2: 36SF-36 Mental Health DomainBaselineG1: 72.0 (17.4)G2: 69.0 (18.6)12-Month Follow-upG1: 73.1 (21.2)G2: 72.3 (17.1)Hanlon et al., 1996 17RCT/LowG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: 86G2: 83SF-36 Mental Health Domainp=NSBaseline:G1: 61.0 (2.5)G2: 63.5 (2.5)12-Month Follow-upG1: 61.1 (1.8)G2: 60.4 (1.8)Bernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices18 monthsG1: 704G2: 636G1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Mental Health Domain(Change between Baseline and18-Month Follow-Up)SF-36 Mental Health Domainp=0.99G1: -0.8G2: -1.3p=NSBaselineG1: 73.4G2: 75.56-Month Follow-upG1: 71.0G2: 75.7p: NRD-166


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPark et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Mental Health DomainBaselineG1: 77.0 (14.6)G2: 73.1 (21.3)4-Month Follow-UpG1: 80.2 (14.6)G2: 73.7 (19.0)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumBernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.G1: Structured communitypharmacy-basedpharmaceutical care programG2: Usual community pharmacyservicesBaselineG1: 168G2: 164(Not clear if allwere included inanalyses)BaselineG1: 1290G2: 116418 monthsG1: 704G2: 636SF-36 Mental Health DomainSF-36 Mental Health DomainSF-36 Mental Health Domain(Change between Baseline and18-Month Follow-Up)p=NSBaselineG1: 75.2 (95% CI, 75.1 to 75.3)G2: 76.7 (95% CI, 75.8 to 77.6)5 Month Follow-UpG1: 74.2 (95% CI, 74.0 to 74.3)G2: 74.7 (95% CI: 74.7 to 74.8)p: 0.49Baseline and 3-Month Follow-UpG1: NRG2: NRp: NSG1: -0.8G2: -1.3p=NSD-167


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26SF-36 Mental Health DomainBaselineG1: 73.4G2: 75.56-Month Follow-UpG1: 71.0G2: 75.7Park et al, 1996 43RCT/HighG1: comprehensivepharmaceutical careG2: usual careG1: 23G2: 26SF-36 Mental Health Domainp:NRBaselineG1: 77.0 (14.6)G2: 73.1 (21.3)4-Month Follow-UpG1: 80.2 (14.6)G2: 73.7 (19.0)Sellors et al., 2003 47RCT-Clusterrandomized/MediumKrska et al, 2001 28RCT/MediumG1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacist-led medicationreviewG2: Usual care involvinginterviews and identification ofPCIs but with nopharmaceutical care planimplemented.BaselineG1: 168G2: 164(Not clear if allwere included inanalyses)SF-36 Mental Health DomainSF-36 Mental Health Domainp=NSBaselineG1: 75.2 (95% CI, 75.1 to 75.3)G2: 76.7 (95% CI, 75.8 to 77.6)5 Month Follow-UpG1: 74.2 (95% CI, 74.0 to 74.3)G2: 74.7 95% CI, (74.7 to 74.8)p: 0.49G1: NRG2: NRp: NSD-168


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumTriller and Hamilton, 2007 62RCT/MediumSellors et al., 2003 47RCT-Clusterrandomized/MediumSellors et al., 2003 47RCT-Clusterrandomized/MediumG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: Visiting nurse associationhome visit services pluscomprehensive pharmaceuticalcare servicesG2: Visiting nurse associationhome visit servicesG1: 447G2: 484G1: 77G2: 77G1: Pharmacists conducted Baselineface-to-face medication reviews G1: 379with the patients and then gave G2: 409written recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.G1: Pharmacists conductedface-to-face medication reviewswith the patients and then gavewritten recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.BaselineG1: 379G2: 409SF-36 Change in Health (changefrom baseline)SF -12 assessed at 30, 90, and180 day follow upsSF-36 Question 1: Overall HealthRatingSF-36 Physical Component6-Month Follow-UpG1: -1.1 (1.3)G2: -4.8 (1.3)12-Month Follow-UpG1: -2.4 (1.5 SE)G2: -6.3 (1.3 SE)95% CI: NRp=0.004Values not reported, but results state thatvalues did not significantly differ betweenthe two groups.BaselineG1: 3.3 (95% CI, 3.3 to 3.3)G2: 3.4 (95% CI, 3.3 to 3.4)5-Month Follow-UpG1: 3.2 (95% CI, 3.2 to 3.3)G2: 3.2 (95% CI, 3.2 to 3.3)p: 0.35BaselineG1: 39.1 (95% CI, 37.2 to 41.0)G2: 38.9 (95% CI, 37.7 to 40.1)5-Month Follow-UpG1: 37.9 (95% CI, 36.6 to 39.2)G2: 38.4 (95% CI, 37.2 to 39.7)p: 0.30D-169


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001, 54,55RCT-ClusterRandomized/MediumSellors et al., 2003 47RCT-Clusterrandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careG1: Pharmacists conductedface-to-face medication reviewswith the patients and then gavewritten recommendations to thephysicians to resolve any drugrelatedproblems.G2: Usual Care for FamilyPhysicians and their Patientsfrom matched postal codes.BaselineN = 363G1: 159G2: 2046 Month Follow-UpN = 317G1: NRG2: NR12 Month Follow-UpN = 292G1: NRG2: NRBaselineG1: 379G2: 409SF-36 Physical ComponentSF-36 Mental ComponentBaselineG1: 38.4 (12.7)G2: 40.1 (11.9)6 Month Follow-UpG1: 38.0 (11.9)G2: 39.2 (11.6)12 Month Follow-UpG1: 36.9 (11.6)G2: 38.4 (11.4)p= NS (Between group comparisons atfollow-up assessments)BaselineG1: 52.2 (95% CI, 50.8 to 53.5)G2: 53.4 (95% CI, 52.6 to 54.3)5-Month Follow-UpG1: 51.0 (95% CI, 49.7 to 52.4)G2: 52.2 (95% CI, 51.2 to 53.2)p: 0.65D-170


Table D30. Self-reported health status: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 , 54,55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 204SF-36 Mental ComponentBaselineG1: 55.1 (8.7)G2: 53.2 (9.3)6 Month Follow-UpN = 317G1: NRG2: NR6 Month Follow-UpG1: 55.9 (9.1)G2: 54.4 (9.3)Williams et al., 2004 57RCT/Medium12 Month Follow-UpN = 292G1: NRG2: NRG1: Modification of patient's G1: 57medication regimen by an G2: 76interdisciplinary team in additionto usual care and "Bound forHealth" booklet.G2: Usual care plus provision of"Bound for Health" bookletSF-36 Overall Score12 Month Follow-UpG1: 56.1 (8.3)G2: 54.6 (8.7)p= NS (Between group comparisons atfollow-up assessments)Baseline:G1: 61.8 (17.8)G2: 63.3 (16.5)6-Week Follow-Up:G1: 65.5 (18.9)G2: 65.7 (17.0)p=NSAbbreviations: CI = confidence interval; G = group; N = number; NR = not reported; NS = not sufficient; PCIs = pharmaceutical care issues; RCT= randomized controlled trial;SE = standard error; SF-36 = multi-purpose, short-form health survey with only 36 questions; VA = Veteran’s AdministrationD-171


Table D31. Condition-specific quality of life: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPai et al., 2009 41 ;Pai et al., 2009 42RCT/HighG1: Pharmaceutical care,consisting of one-on-one care,with in-depth drug therapyreviews conducted by a clinicalpharmacistG2: Standard of care,consisting of brief therapyreviews conducted by a nurseBaselineG1: 61G2: 44Year 1:G1: 44G2: 36Year 2:G1: 24G2: 32Renal Quality of Life Profile(Increased score indicatesworsening of HRQOL, maximumscore=172)Total ScoreBaselineG1: 71.9 (40)G2: 74.5 (33.5)Year 1G1: 71.4 (33.6)G2: 87.5 (30.4)Year 2G1: 56.5 (32.6)G2: 68.8 (35.8)Clifford et al., 2002 11RCT/MediumG1: Collaborativepharmaceutical care programG2: Standard outpatient carefor diabetesG1: 48G2: 25Diabetes Quality of LifeinstrumentScale of 1-5, with higher scoresindicating greater dissatisfaction,worry, or impact of diabetesp0.15D-172


Table D32. Patient satisfaction: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17RCT/LowHanlon et al., 1996 17RCT/LowMalone et al., 2000 29 ;Ellis et al., 2000 30 ;Malone et al, 2001 31 ;Ellis et al., 2000 32RCT/MediumG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: Usual care at outpatientclinic, plus clinical pharmacistcare.G2: Usual care at outpatientclinicG1: Pharmaceutical careprovided by clinical pharmacistswithin ambulatory VA clinicsG2: Usual care (i.e., nopharmaceutical care)G1: 86G2: 83G1: 86G2: 83G1: 447G2: 484General health care satisfactionat 12-Month Follow Up(Higher scores indicate greaterdissatisfaction)Pharmacy-related health caresatisfaction at 12-month Follow-Up(Higher scores indicate greaterdissatisfaction)Patient satisfaction with primaryhealth care provider(Higher scores indicate greatersatisfaction)G1: 1.5 (0.7)G2: 1.6 (0.8)p=0.70G1: 5.2 (1.5)G2: 5.4 (1.7)p=0.52G1:Baseline 51.9 (7.5)Time 2: 51.7 (7.3)G2:Baseline 51.9 (7.5)Time 2: NRBernsten et al., 2001 1 ;Sturgess et al., 2003 2RCT, Cluster-Randomized/HighG1: Structured community Baselinepharmacy-basedG1: 1290pharmaceutical care program G2: 1164G2: Usual community pharmacyservices6 monthsG1: 1024G2: 95312 monthsG1: 863G2: 76418 monthsG1: 704G2: 636% rating pharmacy servicesprovided as "excellent"p=NSBaselineG1: 66.2G2: 68.2p: NR6 monthsG1: 72.8G2: 63.7p:


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26Percent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100“I am very satisfied with the G2: 96pharmacy services I receive,” p: 0.065collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100"Overall, the program provided a G2: 80valuable service to me," p: 0.0018collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100"The quality of information G2: 88provided to me by thep: 0.012pharmacist was excellent,"collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100"My participation in this program G2: 83helped me to understand high p: 0.011blood pressure better," collectedat 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 96"The area was private enough G2: 96for me to feel comfortable talking p: 0.036about my high blood pressure,"collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100"I felt comfortable talking with the G2: 96pharmacist about my health p: 0.052problems," collected at 6 monthsD-174


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighCarter et al.,1997 7 ;Barnette, Murphy, and Carter,1996 8Cohort/HighG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: Pharmaceutical careG2: Usual care with patientsseen by pharmacists who didnot participate in the intensiveskills development programG1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26G1: 25G2: 26Percent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 100"I am confident the pharmacist is G2: 92able to help me control my high p: 0.340blood pressure," collected at 6monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 87"I am confident the information G2: 83provided by the pharmacist to p: 0.325the physician improved myhealth care," collected at 6monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 9"There are things about the high G2: 0blood pressure program that p: 0.157could be better," collected at 6monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 95"I am very willing to continue to G2: 88see the pharmacist for help with p: 0.459my high blood pressure control,”collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 77"I think the pharmacist should G2: 75provide this type of service for p: 0.890everyone," collected at 6 monthsPercent of patients agreeing or 6-Month Follow-Upstrongly agreeing with statement, G1: 91"I think the pharmacist should be G2: 82paid for this type of service," p: 0.379collected at 6 monthsD-175


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 2046 Month Follow-UpN = 317G1: NRG2: NR12 Month Follow-UpN = 292G1: NRG2: NRBaselineN = 363G1: 159G2: 2046 Month Follow-UpN = 317G1: NRG2: NR12 Month Follow-UpN = 292G1: NRG2: NRGeneral satisfaction(Higher numbers reflect greaterdissatisfaction)Interpersonal skills(Higher numbers reflect greaterdissatisfaction)BaselineG1: 1.59 (0.77)G2: 1.56 (0.73)6 Month Follow-UpG1: 1.51 (0.84)G2: 1.57 (0.72)12 Month Follow-UpG1: 1.53 (0.77)G2: 1.62 (0.88)p= NS for all between-group differencesBaselineG1: 1.36 (0.48)G2: 1.37 (0.536 Month Follow-UpG1: 1.37 (0.59)G2: 1.35 (0.57)12 Month Follow-UpG1: 1.31 (0.50)G2: 1.45 (0.72)p= NS for all between-group differencesD-176


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 ;Kassam et al., 2001 5555RCT-ClusterRandomized/MediumVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NRBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NREvaluation and goal setting(Higher numbers reflect greaterdissatisfaction)Trust(Higher numbers reflect greaterdissatisfaction)BaselineG1: 2.58 (1.12)G2: 2.74 (1.09)6-Month Follow-UpG1: 2.46 (0.98)G2: 2.98 (1.24)12-Month Follow-UpG1: 2.49 (1.10)G2: 2.90 (1.08)p


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 ; Kassamet al., 2001 55RCT-ClusterRandomized/MediumVolume et al., 2001 54 ; Kassamet al., 2001 55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NRBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NRHelping patients(Higher numbers reflect greaterdissatisfaction)Explanation(Higher numbers reflect greaterdissatisfaction)BaselineG1: 2.25 (1.31)G2: 2.22 (1.14)6-Month Follow-UpG1: 1.98 (1.17)G2: 2.23 (1.15)12-Month Follow-UpG1: 2.07 (1.22)G2: 2.37 (1.21)p= NS for all between-group differencesBaselineG1: 1.34 (0.55)G2: 1.34 (0.63)6-Month Follow-UpG1: 1.39 (0.67)G2: 1.30 (0.56)12-Month Follow-UpG1: 1.38 (0.73)G2: 1.35 (0.61)p= NS for all between-group differencesD-178


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NRBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NR12-Month Follow-UpN = 292G1: NRG2: NRPharmacy finances(Higher numbers reflect greaterdissatisfaction)Drug plan finances(Higher numbers reflect greaterdissatisfaction)BaselineG1: 3.08 (1.82)G2: 2.85 (1.80)6-Month Follow-UpG1: 2.89 (1.89)G2: 2.86 (1.75)12-Month Follow-UpG1: 3.08 (1.80)G2: 3.16 (1.88)p= NS for all between-group differencesBaselineG1: 3.31 (1.70)G2: 3.41 (1.75)6-Month Follow-UpG1: 3.45 (1.96)G2: 3.39 (1.83)12-Month Follow-UpG1: 3.65 (1.67)G2: 3.56 (1.83)p= NS for all between-group differencesD-179


Table D32. Patient satisfaction: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasVolume et al., 2001 54 ;Kassam et al., 2001 55RCT-ClusterRandomized/MediumG1: Comprehensivepharmaceutical care servicesG2: Traditional pharmacy careBaselineN = 363G1: 159G2: 2046-Month Follow-UpN = 317G1: NRG2: NRCommunicates with doctor(Higher numbers reflect greaterdissatisfaction)BaselineG1: 1.50 (0.77)G2: 1.60 (0.89)6-Month Follow-UpG1: 1.36 (0.63)G2: 1.72 (1.00)12-Month Follow-UpG1: 1.36 (0.65)G2: 1.74 (0.97)12-Month Follow-UpN = 292G1: NRG2: NRp


Table D33. Use of generic medications: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35aCohort/MediumCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD,receiving MTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD,receiving MTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRComparison—congestive heartfailureG13: enrolled in PDP, usualcareG14: enrolled in MA-PD, usualcareG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Generic substitution ratio within365 days after date of MTMenrollment (for interventions) orrandomly-assigned date in 2010(for comparators)Odds (95% CI)For CHF/COPD/diabetes drugsCongestive heart failureG1 vs. G13: 0.001 (-0.000, 0.002), p>0.05G3 vs. G14: 0.005 (0.003, 0.006), p0.05G7 vs. G16: 0.000 (-0.002, 0.002), p>0.05DiabetesG9 vs. G17: -0.000 (-0.000, 0.000), p>0.05G11 vs. G18: 0.000 (-0.000, 0.000),p>0.05For non-CHF/COPD/diabetes drugsCongestive heart failureG1 vs., G13: 0.000 (-.002, 0.002), p>0.05G3 vs., G14: -0.010 (-0.013,-0.008), p0.05G7 vs., G16: 0.006 (0.003, 0.009), p0.05G11 vs., G18: -0.002 (-0.003,-0.001), p


Table D33. Use of generic medications: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPindolia et al., 2009 44Cohort/HighWinston and Lin, 2009 58Cohort/HighG1: Telephone based MTMprogram (acceptors)G2: Usual medical care (optout)G1: Community pharmacy MTMG2: Pharmacist-staffed callcenter-based MTMG3: Educational mailingsG1: 292G2: 1,081G1: 21,336G2: 3,436G3: 49,021Increase in the overall use ofgeneric drugsWeighted generic substitutionratio: 30-day equivalent claimsdivided by total number of claimsG1: 6%G2: 3%p not calculated because baselinepercentages not providedPre-MTM (Jan 1 2007-April 30, 2007)G1: 60.1 (29.8)G2: 58.6 (25.7)G3: 58.7 (27.6)p: NRPost-MTM (Jan 1 2008-April 30, 2008)G1: 65.7 (32.5)G2: 64.6 (30.5)G3: 63.5 (32.2)p: NRCalculated SMD for G1 vs. G3: -0.04 (95%CI, -0.06 to -0.02; p


Table D34. Patient co-payments: Summary of resultsStudyStudy Arms N Analyzed Prescription Costs to Patients ResultsDesign/Risk of BiasShimp et al., 2012 49RCT/MediumG1: MTM program for Universityof Michigan beneficiaries,entitled Focus on MedicinesG2: Usual care (not described)G1: NRG2: NRAnnualized prescription drugcosts for patient-paid amount12 months before first visitG1: 1,334 ± 593G2: 1,293 ± 68095% CI: NRp: NR at baseline12 months after second visitG1: 1,100 ± 645G2: 1,123 ± 64395% CI: NRp: NR at followupChristensen et al., 2007 10NRCT/MediumPindolia et al., 2009 44Cohort/HighG1: Patients receivingpharmacist-provided MTMservicesG2: Patients from samecounties as G1 who did notreceive intervention (controlgroup 1)G3: Patients from a differentcounty than G1 who did notreceive intervention (controlgroup 2)G1: Telephone based MTMprogram (acceptors)G2: Usual medical care (optout)G1: 67G2: 669G3: 870G1: 292G2: 1081Mean difference in patient copaymentfor prescriptions over 6months in $ (SD)Mean out-of-pocket prescriptioncosts per health plan member in$ (assumed per year, as NR instudy) (SD)Mean difference with variance betweenarms not calculable without N, reported Pfor G1 from baseline to followup: 0.004p for G2 from baseline to followup: 0.062G1: 34.3 (263.6)G2: -54.3 (253.9)Calculated SMD for G1 vs. G2, assumingcorrelation between baseline and followupof 0.5 = -0.3; 95% CI, -0.5 to -0.04(p=0.025)G3: -46.1 (282.9)Calculated SMD for G1 vs. G3, assumingcorrelation between baseline and followupof 0.5 = -0.2; 95% CI, -0.6 to -0.1(p=0.007)2006G1: 1513 (1171)G2: 1183 (1084)2007G1: 1571 (1163)G2: 1164 (1201)Calculated SMD, assuming correlationbetween baseline and followup of 0.5= -0.1; 95% CI, -0.2 to 0.1 (p=0.328)D-183


Table D34. Patient co-payments: Summary of resultsStudyStudy Arms N Analyzed Prescription Costs to Patients ResultsDesign/Risk of BiasFox et al., 2009 14Cohort/HighG1: MTM program (acceptors)G2: Opt-out from MTM programG1: 247G2: 50Mean difference in Medicare PartD medication copayment costsper patient per monthMean difference in all medicationcopayments (Medicare Part Dand not Part D) costs per patientper monthG1: 7.4 (76.0)G2: 11.3 (43.8)p: 0.62G1: 5.2 (80.5)G2: 6.9 (37.5)p: 0.82Abbreviations: CI: confidence interval; G = group; MTM = Medication Therapy Management; NR = not reported; NRCT = nonrandomized controlled trial; RCT= randomizedcontrolled trial; SD = standard deviation; SMD: standardized mean difference; vs. = versusD-184


Table D35. Total expenditures on medications by health plans: Summary of resultsStudyStudy Arms N Analyzed Prescription Costs to HealthDesign/Risk of BiasPlansResultsJameson, VanNoord, andVanderwoud, 1995 23RCT/MediumSellors et al., 2003 47RCT/MediumSellors et al., 2001 48RCT/MediumG1: PharmacotherapyconsultationG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: 27G2: 29G1: 379G2: 409G1: 61G2: 60Change in cost (USD) of G1: -130prescription drugs over 6 months, G2: 163based on maximum allowablecost for Medicaid reimbursement Calculated mean difference: -293, 95% CI,-501.5 to -84.5p< 0.01Mean daily medication costs perpatient to the Ontario DrugBenefit Program (assumed CAD)at 5 monthsMean daily medication costs tothe Ontario Drug BenefitProgram (assumed CAD) at 6monthsG1: 3.6G2: 3.8Calculated mean difference: 0.19, 95% CI,-1.5 to 1.1p: 0.78G1: 3.28, 95% CI: 2.64 to 3.92G2: 3.76, 95% CI: 3.76 to 4.45Calculated mean difference in CAD (95%CI) = -0.48(-1.44 to 0.48)p=0.33Christensen et al., 2007 10NRCT/MediumG1: Patients receivingpharmacist-provided MTMservicesG2: Patients from samecounties as G1 who did notreceive intervention (controlgroup 1)G3: Patients from a differentcounty than G1 who did notreceive intervention (controlgroup 2)G1: 67G2: 669G3: 870Calculated mean difference over 6months=-0.48*30*6=-86.4Mean difference in amount G1: -90.1 (793.0)insurer paid for prescriptions over G2: -35.4 (939.5)6 months in USD (SD)G3: -97.3 (907.4)Calculated mean difference for G1 vs. G2,assuming correlation between baselineand followup of 0.5= -54.7,95% CI, -287.6 to 178.2 (p=0.645)Calculated mean difference for G1 vs. G3,assuming correlation between baselineand followup of 0.5 = -7.2;95% CI, -230.8 to 216.4 (p=0.950)D-185


Table D35. Total expenditures on medications by health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to HealthDesign/Risk of BiasPlansResultsMoczygemba et al., 2011 37Moczygemba et al., 2008 38Moczygemba et al., 2008 38Cohort/MediumG1: MTM-eligible patients whoopting-in to a telephone MTMprogram (acceptors)G2: MTM- eligible patients whodid not opt-in to the MTMprogram (opt-out)G1: 60G2: 60Mean Part D drug costs in USD(based on prescription claimrecords, excludes non-Part Ddrug costs) (SD) at baseline and6 months and 12 monthsBaselineG1: $2289 ($887)G2: $2131 ($1273)p: NRFollow up at 6 monthsG1: $2311 ($1148)G2: $2429 ($1697)Adjusted p: 0.80Calculated mean difference: -276.0, 95%CI, -751.3 to 199.3, p: 0.2612 monthsG1: $3,938 ($1,022)G2: $4,842 ($3,405)Unadjusted P= 0.03Adjusted p: NSMoore et al., 2013 40Cohort/MediumG1: MTM program (opt-in)G2: control group (refusers)G1: 2,250G2: 2,250Total plan-paid costs for alldispensed medications in preandpost-periods 1 year beforeinvitation to MTM program and 1year afterReported mean difference:-USD 80095% CI NR, p=0.03 for t-test, but nosignificant predictors whensociodemographic, health-related, and usevariables were controlled for in the multipleregression analysisBaselineG1: $4,853 (122.77)G2: $5,081 (151.77)p: 0.242mean change in total plan-paid pharmacycosts [Mean (SE)]G1: $327 (85.65)G2: -$98 (86.69)p< 0.001Calculated mean difference in USD (95%CI): 425 (109.79 to 12,054.24)p: < 0.001D-186


Table D35. Total expenditures on medications by health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to HealthDesign/Risk of BiasPlansResultsPerlroth et al., 2013 35aCohort/MediumCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD,receiving MTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD,receiving MTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Total payments recorded on PartD claims for all prescriptionmedications not used fortreatment of condition specific toMTM eligibility (CHF, COPD, ordiabetes) within 365 days afterdate of MTM enrollment (forinterventions) or randomlyassigneddate in 2010 (forcomparators)Adjusted costs in USD (95% CI)Congestive heart failure cohortG1 vs. G13: 87.05 (7.33, 166.78), p


Table D35. Total expenditures on medications by health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to HealthDesign/Risk of BiasPlansResultsWittayanukorn et al., 2013 60Cohort/MediumG1: Pharmacist provided faceto-faceMTM services for 30–60minutes per encounter, notalways including a followup visitG2: Patients who did notreceive MTM servicesG1: 63G2: 62Mean pharmacy expenditures, 6 months pre-MTMduring the 6 months prior to the G1: IG: 126.8 (20.3)initial MTM visit and costs during G2: CG: 133.7 (23.8)the 6 months after the initial MTM 95% CI: NRvisit, in USD:


Table D35. Total expenditures on medications by health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to HealthDesign/Risk of BiasPlansResultsHirsch et al., 2011 19Hirsch et al., 2009 20Cohort/HighG1: Patients served at pilotpharmaciesG2: Patients served at nonpilotpharmacies2005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Paid claims amount for allprescription medications, ARTmedications, and non-ARTmedications (total cost-cost ofART medications)Mean USD (SE)2005G1: 26,797 (703)G2: 22,544 (290)p


Table D36. Total expenditures on medications by patients and health plans: Summary of resultsStudyStudy Arms N Analyzed Prescription Costs ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowKrska et al., 2001 28RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist-led medicationreviewG2: Usual care includingidentification of pharmaceuticalcare issues, but no planG1: 105G2: 103G1: 168G2: 164Price to the VA for the agent,plus the average cost of fillingprescriptions (time period NR)(25th-75th percentile)Average monthly costs ofprescribed medication per patientin British? pounds (SD) at 3months (calculated usinginformation from patient on actualuse)Mean cost in USDG1: 1,006 (574–1,285)G2: 1,096 (566–1,456)95% CI: NRp: NS at 0.05 level, specifics NRBaseline:G1: 39.3 (29.1)G2: 42.8 (33.5)FollowupG1: 38.8 (29.6)G2: 42.6 (31.8)Malone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31 ;Ellis et al., 2000 32RCT/MediumSellors et al., 2003 47RCT/MediumSellors et al., 2001 48RCT/MediumG1: Pharmaceutical careG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: 523G2: 531G1: 379G2: 409G1: 61G2: 60Mean change in annual drugcosts in USD (calculated fromDenver VAMC pharmacydepartment, individual sites, orthe VA Pharmacy BenefitsManagement group)Mean daily medication costs perpatient at 5 months (assumedCAD)Mean daily medication costs(assumed CAD) at 6 monthsCalculated mean difference: -0.2, 95% CI,-6.7 to 6.5)p=0.956G1: +203G2: +140Calculated mean difference: 63,95% CI, -5.1 to 131.1; p: 0.07G1: 5.01G2: 4.82Calculated mean difference: 0.2,95% CI, -0.8 to 1.2; p=0.72G1: CAD 3.85 (2.77)G2: CAD 4.26 (2.78)95% CI: NRP: 0.43Calculated mean difference:-0.41 CAD (-1.40 to 0.58)p: 0.42D-190


Table D36. Total expenditures on medications by patients and health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs ResultsDesign/Risk of BiasWilliams et al., 2004 57RCT/MediumJeong et al., 2009 25Cohort/MediumG1: Modification of patient'smedication regimen by aninterdisciplinary medicationadjustment teamG2: Usual medical careG1: Participants in Part DMedicare MTM program (optedin to MTM program)G2: Control subjects withoutPart D Medicare as theirprimary drug benefit butotherwise similar to interventionsubjects.G1: 57G2: 76G1: 2,780G2: 2,251Average monthly wholesale price(USD) of prescription and nonprescriptiondrugs in USDG1: -26.92G2: -0.68Reported mean difference: -20.2,95% CI, 5.8 to 34.5p: 0.006Full retail cost of medication had Prethe patient not had insurance G1:$3572 (3464)coverage.at 6 months before and G2: $3438(2840)6 months after enrollment in USD 95% CI:P:Post:G1: $3458 (3968)G2:$3888 (3388)Change:G1: -$114 (2893)G2: +449(3340)p< 0.001Yamada et al., 2012 61Cohort study/MediumG1: MTM enrolled patientsG2: Eligible MTM patients notenrolled but matched on age,gender, region and DCG riskG1: 34,352G2: 138,182Change in annual prescriptioncostsCalculated mean difference in USD (95%CI): -563 (-735.33 to -390.67)p


Table D36. Total expenditures on medications by patients and health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs ResultsDesign/Risk of BiasPai et al., 2009 41 ;Pai et al., 2009 42RCT/HighG1: Pharmaceutical careG2: Usual careG1: NRG2: NRMean drug costs in USD(calculated from averagewholesale price) over 2 yearsBaseline:G1: 430 (197)G2: 451 (267)Fox et al., 2009 14Cohort/HighPindolia et al., 2009 44Cohort/HighG1: MTM program (acceptors)G2: Opt-out from MTM program(opt-out)G1: Telephone based MTMprogram (acceptors)G2: Usual medical care (optout)G1: 247G2: 50G1: 292G2: 1081Mean difference in annualMedicare Part D drug cost inUSD (patientcopay + insurance planmedication costs + dispensingfee)Total annual prescription drugcost per health plan member inUSDFollowup:Pharmaceutical care reduced mean drugcosts by $6.21 compared with thestandard of care group, p=NS, no absolutecosts or other details reportedG1: -76.7 (350.8)G2: -49.0 (92.8)Calculated mean difference: -27.8,95% CI, -125.8 to 26.6p: 0.57Pre-enrollment ($) (January-June 2006)G1: 576.3 (394.3)G2: 468.1 (335.9)Post-enrollment ($) (July-December 2006)G1: 480.7 (404.3)G2: 434.7 (421.4)Staresinic et al., 2007 51Cohort/HighG1: MTP program (acceptors)G2: Usual care (opt-out)G1: 282G2: 1544Total prescription cost per MTMPbeneficiary per month in USD(gross drug cost=ingredient costpaid + dispensing fee + salestax/member months in part Dcontract)Calculated mean difference: -62.2,95% CI, -112.5 to -12.0; p=0.015Participants spent less on prescriptionmedications on average (described as permember per month drug spending) thannon-participants. Figure provided suggestsa decrease spend of between 100 and 150in the intervention group, but exactnumbers not reported.D-192


Table D36. Total expenditures on medications by patients and health plans: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs ResultsDesign/Risk of BiasWelch et al., 2009 56Cohort/HighWinston and Lin, 2009 58Cohort/HighG1: MTM program provided tohome-based beneficiariesG2: No-MTM control group(voluntary opt-out)G1: Community pharmacy MTMG2: Pharmacist-staffed callcenter-based MTMG3: Educational mailingsG1: 459G2: 336G1: 21,336G2: 3,436G3: 49,021Mean change in medication costsper day in USD at 6 months. NoSD reported. (from data on studybeneficiaries’ purchasesof ambulatory prescriptionmedications)Mean percent increase inmedication costs per day in USDat 6 months (no SD reported)Mean (SD) drug cost per patientper month in USD after 8 monthsof services (based drug claimsprocessing data, total allowedcharges, including ingredient costpaid, dispensing fee, and salestax, prior to subtracting anypatient cost-sharing amounts)G1: 0.3G2: -3.3Adjusted p: 0.203NOTE: Age, sex, chronic disease score,and preperiod drug cost included inmultivariate regression modeling foradjusted PG1: 49.7G2: 39.9p: 0.006Adjusted OR (95% CI): 1.4 (1.1 to 1.9)NOTE: Model adjusted for age, sex,chronic disease score, and baselinemedication costPre-MTM period (Jan 1 2007-April 30,2007)G1: 669 (461)G2: 676 (463)G3: 698 (513)Post-MTM period (Jan 1 2008-April 30,2008)G1: 634 (512)G2: 661 (494)G3: 698 (556)Calculated mean difference for G1 vs. G3:-35.0, 95% CI, -43.4 to -26.6; p


Table D37. Medication and other costs: Summary of resultsStudyStudy Arms N Analyzed Prescription Costs to Patients ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowSellors et al., 2003 47RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: 105G2: 103G1: 379G2: 409Total cost = drug outlays +average per diem cost ofinpatient care based on annualoutput and expenditure data forbed sections in the costdistribution report + costs ofsurgery based on relative valueweights, and VA costs perrelative value weight + healthservices valued using 1991estimates of VAMC unit costs;costs for non-VA hospital carewere imputed using logicunderlying VA cost methodology(time period NR) (25th-75thpercentile)Mean cost (SE) of health careresources per patient, includingall hospital stays at 5 months(CAD assumed)Mean cost (SE) of health careresources per patient, includingonly drug-related hospital staysat 5 months (CAD assumed)Mean cost in USDG1: 7,873 (1,622–6,649)G2: 5,926 (1,608–7,036)95% CI: NRp: NS at 0.05 level, specifics NRG1: 1894.1 (200.7)G2: 1644.7 (220.8)p=0.83Calculated mean difference: 249.4,95% CI, -338.4 to 837.2G1: 1281.3 (101.4)G2: 1299.4 (154.7)p=0.45Calculated mean difference: -18.1,95% CI, -386.7 to 350.5D-194


Table D37. Medication and other costs: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to Patients ResultsDesign/Risk of BiasBernsten et al., 2001 1,2RCT/HighG1: Structured communitypharmacy-basedpharmaceutical care programG2: Usual community pharmacyservicesBaselineG1: 867G2: 7486 monthsG1: NRG2: NR12 monthsG1: NRG2: NRMean total cost per patientincluding (1) cost associatedwith additional time spent bypharmacists; (2) costassociated with contacts withGPs, specialists and nurses; and(3) cost of hospitalizations anddrugsCost data not pooled and analyzed forcosts because of differing health caresystems between countries. However, nosignificant between-group differences inany country (p=NS)Fischer et al., 2002 13NRCT/MediumMoore et al., 2013 40Cohort/MediumG1: Pharmaceutical careG2: Usual careG1: MTM program (opt-in)G2: control group (refusers)18 monthsG1: NRG2: NRG1: 231G2: 444G1: 2250G2: 2250Change in total charges (in USD)for inpatient care, outpatientcare, and pharmacyTotal plan-paid costs for alldispensed medications + totalplan-paid costs for all coveredmedical services 1 year beforeinvitation to MTM program and 1year afterG1: -900G2: -200095% CI: NRp: NS, no details reportedCalculated mean difference: 1100.Total plan-paid health care costs atbaseline (SE)G1: $9,456 (372.05)G2: $9,499 (375.26)P: 0.935Mean change in total plan-paid health carecosts [Mean (SE)]G1: -$977 (357.16)G2: $62 (396.45)P: 0.048Calculated mean difference in USD (95%CI): -1,039 (-2,084.85 to 6.849)p=0.052D-195


Table D37. Medication and other costs: Summary of results (continued)StudyStudy Arms N Analyzed Prescription Costs to Patients ResultsDesign/Risk of BiasWittayanukorn et al., 2013 60Cohort/MediumG1: Pharmacist provided faceto-faceMTM services for 30-60minutes per encounter, notalways including a followup visitG2: Patients who did notreceive MTM servicesG1: 63G2: 62Mean total expenditures(pharmacy+medical), during the6 months prior to the initial MTMvisit and costs during the 6months after the initial MTM visit,in USD6 months pre-MTMG1: IG: 481.2 (137.0)G2: CG: 291.3 (49.0)95% CI: NRP:


Table D38. Number of outpatient visits: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowKrska et al., 2001 28RCT/MediumMalone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32RCT/MediumSellors et al., 2003 47RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist-led medicationreviewG2: Usual care includingidentification of pharmaceuticalcare issues, but no planG1: Pharmaceutical careG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: received at least 2pharmacist visits involvingmedication review, patientspecific education andcounseling; follow up patientphone calls and contact ofphysicians as neededG2: only contacted for tocomplete the survey.G1: 105G2: 103G1: NRG2: NRG1: 523G2: 531G1: 379G2: 409G1: 92G2: 104Mean general medicine clinicvisits (time period NR) (25th–75thpercentile)Mean other clinic visits (timeperiod NR) (25th–75th percentile)Hospital clinic attendance, use ofsocial services or contacts withdistrict nurses and health visitorsbefore and after the pharmacistreviewMean change in number of clinicvisits (including visits with thepharmacists in the interventionarm)G1: 5.5 (3–6)G2: 5.8 (3–7)95% CI: NRp: NS at 0.05 level, specifics NRG1: 7.7 (3–10)G2: 10.9 (2–15)95% CI: NRp: NS at 0.05 level, specifics NRNo differences, details NRG1: +4.8G2: +2.8p: 0.003Number of clinic visits (SE) G1: 0.3 (0.15)G2: 0.3 (0.6)p: 0.40Change in number of ambulatoryvisits over 3 monthsG1: -1.2G2: 0.3p: 0.08D-197


Table D38. Number of outpatient visits: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasSidel, 1990 50RCT/MediumTouchette et al., 2012 53RCT/MediumG1: Patients received at least 2pharmacist visits involvingmedication review, patientspecificeducation andcounseling; followup patienttelephone calls and contact ofphysicians as neededG2: Patients contacted only tocomplete the survey.G1: 92G2: 104G1: MTM basic (comprehensive G1: 183medication review and DRP G2: 190assessment)G3: 183G2: MTM enhanced (MTM plus 2page clinical summaryabstracted from patient's medicalchart).G3: Usual careChange in number of ambulatoryvisits over past 3 months,measured at baseline and againat 36 months0-3 monthsG1: 180G2: 190G3: 1933-6 monthsG1: 183G2: 190G3: 183G1: -1.16G2: 0.2595% CI: NRP: 0.08Calculated mean difference:-1.41,95% CI: -2.98 to 0.160, p=0.0780-3 monthsG1: 2.6 (2.2)G2: 2.7 (2.3)G3: 2.6 (2.2)G1 vs. G3:(p=0.646)G2 vs. G3:(p=0.816)3-6 monthsG1: 2.2 (2.1)G2: 2.1 (2.1)G3: 2.2 (2.2)G1 vs. G3:(p=0.760)G2 vs. G3:(p=0.458)D-198


Table D38. Number of outpatient visits: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasBernsten et al., 2001 1,2RCT/HighMoore et al., 2013 40Cohort/MediumFischer et al., 2002 13NRCT/HighCarter et al., 1997 7,8Cohort/HighChrischilles et al., 2004 9Cohort/HighG1: Structured communitypharmacy-based pharmaceuticalcare programG2: Usual community pharmacyservicesG1: MTM program (opt-in)G2: control group (refusers)G1: Pharmaceutical careG2: Usual careG1: Pharmaceutical careG2: Usual careG1: PCM-eligible patients whoreceived PCM servicesG2: PCM-eligible patients whodid not receive PCM servicesBaselineG1: 1290G2: 11646 monthsG1: 1024G2: 95312 monthsG1: 863G2: 76418 monthsG1: 704G2: 636G1: 2,250G2: 2,250G1: 231G2: 444G1: 25G2: 26G1: 524G2: 1,687Mean number of contacts withprimary care providers, includinghome visits and officeappointments (SD)Change in number of physicianvisits from 365 days preceding thepatient's program invitation dateto 365 days following patient'sprogram invitation dateChanges in number of clinic visitsover 1 yearNumber of distinct dates ofservice over 6 monthsNo. of outpatient facility claims at12 monthsBaselineG1: 4.8 (8.4)G2: 4.3 (6.2)p: NS6 monthsG1: 4.0 (5.7)G2: 3.6 (4.6)p: NS12 monthsG1: 4.0 (7.0)G2: 3.5 (5.5)p: NS18 monthsG1: 4.3 (8.0)G2: 3.2 (4.0)p: NSNumber of physician visits at baseline (SE)G1: 27.7 (0.344)G2: 27.2 (0.379)P: 0.325mean change in number of physician visitsG1: -0.70 (0.27)G2: -3.18 (0.31)p< 0.001Calculated mean difference (95% CI): 2.48(1.674 to 3.286, p


Table D39. Costs of outpatient visits: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowMalone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32RCT/MediumSellors et al., 2003 47RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmaceutical careG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: 105G2: 103G1: 523G2: 531G1: 379G2: 409Annual health care costs in USDfor general medicine clinic care at1-year closeout or adjusted to a 1-year followup and weighted byactual time for censoredpatients(time period NR) (25th–75th percentile)Annual health care costs for otherclinic care in USD at 1-yearcloseout or adjusted to a 1-yearfollowup and weighted by actualtime for censored patients(timeperiod NR) (25th–75th percentile)Mean change in annual cost ofclinic visits in US $Mean cost of physician visits in in$ (assumed CAD) (SE) at 5monthsMean cost of clinic visits in in $(assumed CAD) (SE) at 5 monthsMean cost of other health careservices/visits to health careprofessionals in in $ (assumedCAD) (SE) at 5 monthsG1: 334 USD (200–366)G2: 356 USD (183–427)95% CI: NRp: NS at 0.05 level, specifics NRG1: 422 (67–500)G2: 565 (23–923)95% CI: NRp: NS at 0.05 level, specifics NRG1: +231G2: +33395% CI: NRp: 0.02G1: 204.0 (11.1)G2: 198.3 (10.4)95% CI (calculated for standardized differencein means): -0.11 to 0.12p (calculated for standardized difference inmeans): 0.71G1: 18.8 (8.1)G2: 20.9 (5.0)95% CI (calculated for standardized differencein means): -0.16 to 0.12p (calculated for standardized difference inmeans): 0.82G1: 288.30 (40.02)G2: 293.00 (55.25)95% CI (calculated for standardized differencein means): -0.145 to 0.135p (calculated for standardized difference inmeans): 0.97D-200


Table D39. Costs of outpatient visits: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasChrischilles et al., 2004 9Cohort/HighCarter et al., 1997 7,8Cohort/HighHirsch et al., 2011 19Hirsch et al., 2009 20Cohort/HighG1: PCM-eligible patients whoreceived PCM servicesG2: PCM-eligible patients whodid not receive PCM servicesG1: Pharmaceutical careG2: Usual careG1: Patients served at pilotpharmaciesG2: Patients served at nonpilotpharmaciesG1: 524G2: 1,687G1: 25G2: 262005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Outpatient facility claims at 12monthsHypertension-related charges in $(SD) at 6 monthsMean visit charges in $ (SD) at 6monthsOutpatient costs (not defined)Results NRp: 0.107G1: 122 (124)G2: 52 (65)p=0.03G1: 823 (1,123)G2: 336 (246)p=0.02Mean USD (SE)2005G1: 112 (11)G2: 44 (2)p


Table D40. Number of laboratory tests: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasSellors et al., 2003 47RCT/MediumG1: Pharmacist consultationprogramG2: Usual careMalone et al., 2000 29 ; G1: Pharmaceutical careEllis et al., 2000 30 G2: Usual care(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32RCT/MediumG1: 379G2: 409G1: 523G2: 531Mean number of laboratory testsand imaging procedures at 5monthsMean change in annual number oflaboratory testsAbbreviations: CI = confidence interval; G = group; NR = not reported; QOL = quality of life; RCT = randomized controlled trial.G1: 8.7 (0.6)G2: 8.6 (0.1)95% CI (calculated for standardizeddifference in means): -0.12 to 0.16p (calculated for standardized difference inmeans): 0.791G1: +3.1G2: +4.795% CI: NRp: 0.001D-202


Table D41. Costs of laboratory tests: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowSellors et al., 2003 47RCT/MediumMalone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32RCT/MediumHirsch et al., 2011 19Hirsch et al., 2009 20Cohort/HighG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: Pharmaceutical careG2: Usual careG1: Patients served at pilotpharmaciesG2: Patients served at nonpilotpharmaciesG1: 105G2: 103G1: 379G2: 409G1: 523G2: 5312005G1: 439G2: 1,7952006G1: 617G2: 1,6172007G1: 628G2: 1,606Annual health care costs for G1: 214 (52–318)diagnostic tests at 1-year closeout G2: 354 (55–473)or adjusted to a 1-year followup 95% CI: NRand weighted by actual time for P: NS at 0.05 level, specifics NRcensored patients, (25th–75thpercentile)Mean cost of all lab and imagingprocedures at 5 months $(assumed CAD) (SE)Mean change in annual costs forlaboratory tests in US $Costs of laboratory/x-ray services Mean USD (SE)2005G1: 389 (34)G2: 402 (17)p=0.736G1: 249.3 (20.8)G2: 243.1 (17.2)95% CI (calculated for standardizeddifference in means): -0.12 to 0.16p (calculated for standardized difference inmeans): 0.816G1: +$43G2: +$7695% CI, NRp: 0.052006G1: 387 (28)G2: 407 (18)p=0.5302007G1: 401 (29)G2: 402 (18)p=0.974Abbreviations: CAD = Canadian dollars; CI = confidence interval; G = group; N = number; NR = not reported; NS = not significant; RCT = randomized controlled trial; SE =standard error; USD = United States dollars.D-203


Table D42. ED visits: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowSellors et al., 2003 47RCT/MediumTouchette et al., 2012 53RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: MTM basic (comprehensivemedication review and DRPassessment)G2: MTM enhanced (MTM plus 2page clinical summaryabstracted from patient's medicalchart).G3: Usual careG1: 105G2: 103G1: 379G2: 409G1: 211G2: 218G3: 208Mean emergency room visits (timeperiod NR) (25th-75th percentile)Mean number of ED/urgent carevisits and ambulance use (SE) at5 monthsMean number of ED visits perparticipantG1: 1.6 (0–2)G2: 2.3 (0–3)95% CI: NRp: NS at 0.05 level, specifics NRG1: 0.2 (0.03)G2: 0.2 (0.03)0 to 3 monthsG1: 0.3 (0.6)G2: 0.2 (0.6)G3: 0.2 (0.5)G1 vs. G3:(p=0.735)G2 vs. G3:(p=0.963)Taylor, Byrd, and Krueger,2003 52RCT/HighMoore et al., 2013 40Cohort/MediumG1: Pharmaceutical care groupG2: Standard careG1: MTM program (opt-in)G2: control group (refusers)G1: 33G2: 36G1: 2,250G2: 2,250Change in no, of ED visits from 12months before baseline through12 months afterChange in number of ED visitsfrom 365 days preceding thepatient's program invitation date to365 days following patient'sprogram invitation date3 to 6 monthsG1: 0.2 (0.5)G2: 0.2 (0.6)G3: 0.4 (0.8)G1 vs. G3:(p=0.077)G2 vs. G3:(p=0.057)G1: -12G2: 0p=0.044Number of ER visits at baseline (SE)G1: 0.7 (0.027)G2: 0.8 (0.032)P: 0.016mean change in number of ER visitsG1: -0.04 (0.03)G2: -0.08 (0.03)P: 0.399Calculated mean difference (95% CI): 0.04(-0.043 to 0.123, p=0.346D-204


Table D42. ED visits: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35aCohort/MediumCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD, receivingMTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD, receivingMTM with CMRG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Odds of any all-cause emergencyroom visits within 365 days afterdate of MTM enrollment (forinterventions) or randomlyassigneddate in 2010 (forcomparators) (95% CI)For PDPCongestive heart failureG1 vs. G13: 0.94 (0.90, 0.98), p


Table D42. ED visits: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35a(continued)Odds of condition-specific For PDPemergency room visits within 365 Congestive heart failuredays after date of MTM enrollment G1 vs. G13: 1.01 (0.95, 1.07), p>0.05(for interventions) or randomlyassigneddate in 2010 (for Chronic obstructive pulmonary diseasecomparators) (95% CI)G5 vs. G15: 1.09 (1.04, 1.15), p0.05Adjusted OR: 0.9 (0.6 to 1.3)G1: 47.95G2: 51.52G3: 52.15p


Table D43. Costs of emergency department visits: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowSellors et al., 2003 47RCT/MediumChrischilles et al., 2004 9Cohort/HighPerlroth et al., 2013 35aCohort/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: PCM-eligible patients whoreceived PCM servicesG2: PCM-eligible patients whodid not receive PCM servicesCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD, receivingMTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD, receivingMTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRComparison—congestive heartfailureG13: enrolled in PDP, usual careG14: enrolled in MA-PD, usualcareG1: 105G2: 103G1: 379G2: 409G1: 524G2: 1,687G1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Annual health care costs foremergency room visits (25th-75thpercentile) in USD (95% CI)Mean cost of ED/urgent carevisits and ambulance use at 5months in $ (assumed CAD) (SE)Charges of ED claims at 12monthsChange in costs of any all-causeemergency room visits within 365days after date of MTMenrollment (for interventions) orrandomly assigned date in 2010(for comparators) in USD (95%CI)Change in costs of conditionspecificemergency room visitswithin 365 days after date of MTMenrollment (for interventions) orrandomly assigned date in 2010(for comparators) (95% CI)G1: 119 (0–146)G2: 171 (0–219)95% CI: NRp = NS at 0.05 level, specifics NRG1: 0.2 (0.03)G2: 0.2 (0.03)95% CI (calculated for standardizeddifference in means): -0.19 to 0.10p (calculated for standardized difference inmeans): 0.53Results NRp: 0.513For PDPCongestive heart failureG1 vs. G13: -12.66 (-33.61, 8.30), p>0.05Chronic obstructive pulmonary diseaseG5 vs. G15: -16.21 (-35.37, 2.96), p>0.05DiabetesG9 vs. G17: -8.76 (-23.65, 6.12), p>0.05For PDPCongestive heart failureG1 vs. G13: -3.17 (-14.59, 8.25), p>0.05Chronic obstructive pulmonary diseaseG5 vs. G15: 12.81 USD (-.14, 25.76),p>0.05DiabetesG9 vs. G17: -3.27 (-15.37, 8.84), p>0.05D-207


Table D43. Costs of emergency department visits: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35a(continued)Comparison—Chronicobstructive pulmonary diseaseG15: enrolled in PDP, usual careG16: enrolled in MA-PD, usualcareComparison—DiabetesG17: enrolled in PDP, usual careG18: enrolled in MA-PD, usualcarea Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement for CMR. We have excluded groups 2, 4, 6, 8, 10, and12 from our summary tables.a Abbreviations: CAD = Canadian dollars, CI = confidence interval; CMR = comprehensive medication review; ED = Emergency department; G = group; MA-PD = MedicareAdvantage Part D Plan; MTM = medication therapy management; NR = not reported; PCM = pharmaceutical care management; PDP = Medicare Part D Plan; SE = standard error;USD = United States dollars.D-208


Table D44. Number of hospitalizations: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowMalone et al., 2000 29 ;Ellis et al., 2000 30(interventions);Malone et al., 2001 31(detailed QOL outcomes);Ellis et al., 2000 32RCT/MediumSellors et al., 2003 47RCT/MediumTouchette et al., 2012 53RCT/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmaceutical careG2: Usual careG1: Pharmacist consultationprogramG2: Usual careG1: 105G2: 103G1: 523G2: 531G1: 379G2: 409G1: MTM basic (comprehensive Time Onemedication review and DRP G1: 180assessment)G2: 190G2: MTM enhanced (MTM plus G3: 1932 page clinical summaryabstracted from patient's medicalchart).G3: Usual careTime TwoG1: 183G2: 190G3: 183Mean hospital admission (timeperiod NR) (25th–75th percentile)Mean change in number ofhospitalizationsMean all-cause hospitalizations(SE)Mean drug-relatedhospitalizations (SE)Percent of participants with atleast one hospital visitPercent of participants with atleast one hospital visitG1: 0.7 (0–1)G2: 0.8 (0–1)95% CI: NRp= NS at 0.05 level, specifics NRG1: +0.1G2: +0.2p: 0.29G1: 0.1 (0.02)G2: 0.1 (0.02)p: 0.77G1: 0.04 (0.01)G2: 0.04 (0.01)p: 0.08Time OneG1: 13.9G2: 7.9G3: 10.4G1 vs. G3: 1.6 (p=0.350)G2 vs. G3: 0.6 (p=0.370)G2 vs. G1: 0.4 (p=0.080)Time TwoG1: 17.6G2: 12.1G3: 9.3G1 vs. G3: 2.6 (p=0.049)G2 vs. G3: 1.4 (p=0.484)G2 vs. G1: 0.3 (p=0.214)D-209


Table D44. Number of hospitalizations: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasTouchette et al., 2012 53RCT/Medium(continued)Mean number of hospital visitsper participantTime OneG1: 0.2 (0.5)G2: 0.1 (0.4)G3: 0.1 (0.4)G1 vs. G3: (p=0.265)G2 vs. G3: (p=0.619)G2 vs. G1: (p=0.109)Time TwoG1: 0.2 (0.5)G2: 0.1 (0.4)G3: 0.1 (0.4)Moore et al., 2013 40Cohort/MediumG1: MTM program (opt-in)G2: control group (refusers)G1: 2,250G2: 2,250Change in number of inpatientvisits from 365 days preceding thepatient's program invitation dateto 365 days following patient'sprogram invitation dateG1 vs. G3: (p=0.056)G2 vs. G3: (p=0.547)G2 vs. G1: (p=0.174)Number of inpatient visits at baseline (SE)G1: 0.5 (0.018)G2: 0.5 (0.016)P: 0.853mean change in number of inpatient visitsG1: -0.09 (0.02)G2: 0.12 (0.02)p< 0.001Calculated mean difference (95% CI): -0.21(-0.265 to -0.155, p


Table D44. Number of hospitalizations: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35aCohort/MediumCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD, receivingMTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD, receivingMTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRComparison—congestive heartfailureG13: enrolled in PDP, usual careG14: enrolled in MA-PD, usualcareComparison—Chronicobstructive pulmonary diseaseG15: enrolled in PDP, usual careG16: enrolled in MA-PD, usualcareComparison—DiabetesG17: enrolled in PDP, usual careG18: enrolled in MA-PD, usualcareG1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Odds of any all-causehospitalization within 365 daysafter date of MTM enrollment (forinterventions) or randomlyassigned date in 2010 (forcomparators) (95% CI)Odds of anyCHF/COPD/diabetes-relatedhospitalization within 365 daysafter date of MTM enrollment (forinterventions) or randomlyassigned datein 2010 (for comparators) (95%CI)Congestive heart failureG1 vs. G13: 0.90 (0.86, 0.94), p0.05Chronic obstructive pulmonary diseaseG5 vs. G15: 0.90 (0.87, 0.94), p0.05DiabetesG9 vs. G17: 0.91 (0.87, 0.95), p


Table D44. Number of hospitalizations: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasYamada et al., 2012 61Cohort study/MediumBernsten et al., 2001 1,2RCT/HighPai et al., 2009 41 ; Pai et al.,2009 42RCT/HighJeong, 26 ; Jeong, 2012 27Cohort/HighG1: MTM enrolled patientsG2: Eligible MTM patients notenrolled but matched on age,gender, region and DCG riskG1: Structured communitypharmacy-based pharmaceuticalcare programG2: Usual community pharmacyservicesG1: Pharmaceutical careG2: Usual careG1: Kaiser-Permanente MTMprogram participants (2010)G2: Kaiser-Permanente patientseligible for MTM, but whodeclined enrollment ordisenrolled with a PCP visitduring first half of 2010G3: Kaiser-Permanente patientseligible for MTM, but whodeclined enrollment ordisenrolled without a PCP visitduring first half of 2010G1: 34,352G2: 138,182BaselineG1: 867G2: 7486 monthsG1: NRG2: NR12 monthsG1: NRG2: NR18 monthsG1: NRG2: NRG1: NRG2: NRG1: 23,638G2: 14,232G3: 1,810Odds of hospital admissionbetween 1 to 4 years dependingon when patient was enrolledPercent with ≥1 hospitalization inthe prior 18 months0.91 (0.88 to 0.93) p< 0.001Note: adjusted for age, sex, Charlson,CHF, and ESRD (95% CI)Pooled sampleBaseline (during 18 months prior to study)Overall: NRG1: 41.7G2: 41.3p=NS18 monthsOverall: NRG1: 35.6G2: 40.4p=NSAll-cause hospitalizations G1: 1.8 (2.4)G2: 3.1 (3.0)p: 0.02Cumulative hospitalized time(days)Percentage hospitalized within 12months of CMRG1: 9.7 (14.7)G2: 15.5 (16.3)p: 0.06G1: 30.82G2: 35.94G3: 42.38p


Table D44. Number of hospitalizations: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasRoughead et al., 2009 46Cohort/MediumWelch et al., 2009 56Cohort study/MediumG1: Collaborative home-basedmedication reviewG2: No medication reviewreceivedG1: MTM program provided tohome-based beneficiariesG2: No-MTM control group(voluntary opt-out)G1: 273G2: 5444G1: 459G2: 336Rate of hospitalization for HF atany time during studyAdjusted HR (95% CI): 0.6 (0.4 to 0.8)p: NRNOTE: Model adjusted for age, sex,comorbidity, SES, season, region ofresidence, and Ns of prescriptions,prescribers, pharmacies, changes inmedications, hospitalizations, occupationaltherapy visits, and speech therapy visitsAdjusted OR of hospitalization Adjusted OR (95% CI): 1.4 (1.1 to 2.0)from 6 month before MTMthrough 6 months after (adjusted NOTE: Model adjusted for age, sex,for age, sex, chronic disease Chronic Disease Score, specific baselinescore, specific baseline utilization) utilization(95% CI)a Perlroth et al. included MTM arms without CMR; these results were not eligible for our review because of our requirement for CMR. We have excluded groups 2, 4, 6, 8, 10, and12 from our summary tables.Abbreviations: CHF = congestive heart failure; CI = confidence interval; CMR = comprehensive medication review; DCG = diagnostic cost group (a measure of health care useand comorbidity); DRP = drug-related problems; ESRD = end-stage renal disease; G = group; HR = hazard ratio; MA-PD = Medicare Advantage Part D Plan; MTM = medicationtherapy management, N = number; NR = not reported; NS = not significant; OR = odds ratio; PCP = primary care physician; QOL = quality of life; RCT = randomized controlledtrials; RR = relative risk; SES = socioeconomic status.D-213


Table D45. Costs of hospitalization: Summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowSellors et al., 2003 47RCT/MediumMalone, 2000 29 ;Ellis, 2000 30(interventions);Malone, 2001 31 (detailedQOL outcomes);Ellis, 2000 32RCT/MediumPerlroth et al., 2013 35 aCohort/MediumG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmacist consultationprogramG2: Usual careG1: Pharmaceutical careG2: Usual careCongestive heart failureG1: enrolled in PDP receivingMTM with CMRG3: enrolled in MA-PD, receivingMTM with CMRChronic obstructive pulmonarydiseaseG5: enrolled in PDP receivingMTM with CMRG7: enrolled in MA-PD, receivingMTM with CMRDiabetesG9: enrolled in PDP receivingMTM with CMRG11: enrolled in MA-PD,receiving MTM with CMRG1: 105G2: 103G1: 379G2: 409G1: 523G2: 531G1: 12,658G3: 11,260G5: 16,372G7: 10,575G9: 16,545G11: 13,527G13: 156,441G14: 51,938G15: 184,350G16: 73,623G17: 133,925G18: 53,912Annual health care costs forinpatients 1-year closeout oradjusted to a 1-year followup andweighted by actual time forcensored patientsMean cost of all admissions tohospital $ (assumed CAD) (SE)Mean change in annualhospitalization costs in US $All hospitalization costs:Generic substitution ratio within365 days after date of MTMenrollment (for interventions) orrandomly assigned date in 2010(for comparators)Mean USD (25th–75th percentile)G1: 5751 (0–3780)G2: 3349 (0–4824)95% CI: NRp: NS at 0.05 level, specifics NRG1: 753.7 (183.1)G2: 594.9 (135.2)95% CI (calculated for standardizeddifference in means), -0.09 to 0.20p (calculated for standardized difference inmeans): 0.479G1: +542G2: +763Variance not reported95% CI, NRp: 0.21Risk adjusted costs in USD for PDP (95%CI)G1 vs. G13: -526.19 (-.919.71, -132.66),p0.05G9 vs. G17: -398.98 (-651.21, -146.75),p


Table D45. Costs of hospitalization: Summary of results (continued)StudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasPerlroth et al., 2013 35 a(continued)Comparison—Congestive heartfailureG13: enrolled in PDP, usual careG14: enrolled in MA-PD, usualcareComparison—Chronicobstructive pulmonary diseaseG15: enrolled in PDP, usual careG16: enrolled in MA-PD, usualcareAny CHF/COPD/diabetes-relatedhospitalization costs within 365days after date of MTMenrollment (for interventions) orrandomly assigned date in 2010(for comparators)Risk adjusted costs in USD for PDP (95%CI)G1 vs. G13: -222.08 (-525.99, 81.82),p>0.05G5 vs. G15: 200.21 (-55.81, 456.23),p>0.05G9 vs. G17: -363.45 (-562.00,-164.91), p


Table D46. Length of hospital stay: summary of resultsStudyStudy Arms N Analyzed Outcome and Time Period ResultsDesign/Risk of BiasHanlon et al., 1996 17,18RCT/LowPai et al., 2009 41 ;Pai et al., 2009 42RCT/HighG1: Usual care, plus clinicalpharmacist care.G2: Usual care in the GeneralMedicine ClinicG1: Pharmaceutical care,consisting of one-on-one care,with in-depth drug therapyreviews conducted by a clinicalpharmacistG2: Standard of care, consistingof brief therapy reviewsconducted by a nurseG1: 105G2: 103BaselineG1: 61G2: 44Year 1:G1: 44G2: 36Year 2:G1: 24G2: 32Abbreviations: G = group; NR = not reported; NS = not significant; RCT= randomized controlled trial.Hospitalized days (time periodNR)Cumulative hospital time (days)over 2 yearsMean (25th–75th percentile)G1: 6.7 (0–5)G2: 4.9 (0–6)95% CI: NRp: NS at 0.05 level, specifics NRCumulative hospital timeG1: 9.7 days (14.7)G2: 15.5 days (16.3)p: 0.06Pharmaceutical care reduced length of stayby 21% compared with the standard of caregroup.p=NSD-216


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of resultsStudyStudy ArmsN Outcome and Time PeriodDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/LowG1: Basic MTM services (with medication informationfrom patient interview)G2: Enhanced MTM services (pharmacist providedwith 2-page clinical summary from patient medicalrecord)G3: Usual pharmacy careG1: 211G2: 218G3: 208Percentage with ≥ 1 ADE at 0 and 3monthsResultsTime OneG1: 42.2%G2: 27.9%G3: 33.7 %G1 vs G3: 1.629(p = 0.078)G2 vs G3: 0.726(p = 0.278)G2 vs G1:0.444(p= 0.005)Time TwoG1: 36.1%G2: 31.1%G3: 34.4 %G1 vs G3: 1.107(p = 0.717)G2 vs G3: 0.889(p = 0.672)G2 vs G1: 0.803(0.432)Calculated OR: 1.294, 95% CI:0.768 to 2.180, p=0.333D-217


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of results (continued)StudyStudy ArmsN Outcome and Time PeriodResultsDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/Low(continued)Percentage with ≥ 1 emergencydepartment visit at 0 and 0 and 3monthsTime OneG1: 0.750 ± 1.113G2: 0.547 ± 1.184G3: 0.559 ± 1.202G1 vs G3:(p = 0.110)G2 vs G3:(p = 0.916)G2 vs G1:(p= 0.900)Time TwoG1: 0.814 ± 1.421G2: 0.468 ± 0.820G3: 0.530 ± 0.894G1 vs G3:(p = 0.041)G2 vs G3:(p = 0.479)G2 vs G1:(p= 0.003)Time OneG1: 20.3%G2: 20.0%G3: 20.0%G1 vs G3: 0.938(p= 0.825)G2 vs G3: 0.989(p = 0.970)G2 vs G1: 1.055(p = 0.854)D-218


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of results (continued)StudyStudy ArmsN Outcome and Time PeriodResultsDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/Low(continued)Time TwoG1: 23.6%G2: 20.8%G3: 16.8%G1 vs G3: 0.780 (p=0.380)G2 vs G3: 0.578 (p=0.064)G2 vs G1: 0.743 (p=0.317)Percentage with ≥ 1 hospitalization at 3monthsCalculated OR: 1.222, 95% CI:0.795 to 1.878, p=0.360Time OneG1: 13.9%G2: 7.9%G3: 10.4 %G1 vs G3: 1.557 (p=0.350)G2 vs G3: 0.626 (p=0.370))G2 vs G1: 0.402 (p=0.080)Time TwoG1: 17.6%G2: 12.1%G3: 9.3%G1 vs G3: 2.550 (p=0.049)G2 vs G3: 1.404 (p=0.484)G2 vs G1: 0.293 (p=0.214)Mean number of ADEs at 0 and 3monthsCalculated OR: 1.539, 95% CI:0.862 to 2.746, p=0.145Calculated mean difference: 0.346,95% CI: 0.112 to 0.580, p=0.004D-219


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of results (continued)StudyStudy ArmsN Outcome and Time PeriodResultsDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/Low(continued)Mean number of emergency departmentvisits at 0 and 3 monthsTime OneG1: 0.261 ± 0.573G2: 0.242 ± 0.558G3:0.229 ± 0.480G1 vs G3:(p = 0.735)G2 vs G3:(p = 0.963)G2 vs G1:(p= 0.769)Time TwoG1: 0.246 ± 0.512G2: 0.247 ± 0.6404G3: 0.352 ± 0.806G1 vs G3:(p = 0.077)G2 vs G3:(p = 0.057)G2 vs G1:(p= 0.900)Calculated mean difference: -0.001, 95% CI:-0.119 to 0.117, p=0.987D-220


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of results (continued)StudyStudy ArmsN Outcome and Time PeriodResultsDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/Low(continued)Mean number of hospitalizations at 0and 3 monthsTime OneG1: 0.172 ± 0.458G2: 0.111 ± 0.440G3: 0.119 ± 0.370G1 vs G3:(p = 0.265)G2 vs G3:(p = 0.619)G2 vs G1:(p= 0.109)Time TwoG1: 0.202 ± 0.477G2: 0.147 ± 0.435G3: 0.110 ± 0.362G1 vs G3:(p = 0.056)G2 vs G3:(p = 0.547)G2 vs G1:(p= 0.174)Calculated mean difference: 0.055 ,95% CI:-0.038 to 0.148, p=0.244D-221


Table D47. Access to patient records (basic MTM versus enhanced MTM): Summary of results (continued)StudyStudy ArmsN Outcome and Time PeriodResultsDesign/Risk ofAnalyzedBiasTouchette et al.,2012 53RCT/Low(continued)Mean number of physician visits at 0and 3 monthsTime OneG1: 2.57 ± 2.218G2: 2.73 ± 2.31G3: 2.57 ± 2.24G1 vs G3:(p = 0.646)G2 vs G3:(p = 0.816)G2 vs G1:(p= 0.490)Time TwoG1: 2.24 ± 2.08G2: 2.14 ± 2.08G3: 2.19 ± 2.19G1 vs G3:(p = 0.760)G2 vs G3:(p = 0.458)G2 vs G1:(p= 0.664)Abbreviations: ADE = adverse drug event; CI = confidence interval; OR = odds ratio.Calculated mean difference: 0.100 ,95% CI:-0.322 to 0.522 , p=0.643D-222


Table D48. Intensity of care coordination and followup following comprehensive medication review: Summary of resultsStudyStudy ArmsN Outcome and Time Period ResultsDesign/Risk ofAnalyzedBiasGrymonpre,2001 16Cohort/HighG1:Comprehensive drug therapy review, then issues addressed withthe client and/or the client's physician, with follow-up as required.G2: Comprehensive drug therapy review only with referral to usualpharmacist.G1: 56G2: 58Non-prescribed drugsdiscontinuedMean (SD) non-prescribeddrugsBaselineG1: 7.2(4.1)G2: 5.9(3.8)6 monthsG1:6.8 (3.4)G2: 5.8 (3.2)95% CI: NRp: 0.130Taking home remediesHoarding drugsCalculated OR: 1.08, 95%CI (0.52–2.27); p=0.830Mean home remediesBaselineG1: I 1.0(0.2)G2: 1.4(0.6)6 monthsG1: 1.1(0.3)G2: 1.1(0.4)95% CI: NRp: 0.160Calculated OR: 1.00, 95%CI (0.07–13.77); p=1.000Mean hoarded drugsBaselineG1: 3.0 (2.9)G2: 2.7(2.2),95% CI: NR6 monthsG1: 1.8 (1.2)G2: 2.0 (2.0)p: 0.018Calculated OR: 1.48, 95%CI (0.65–3.34); p=0.348D-223


Table D48. Intensity of care coordination and followup following comprehensive medication review: Summary of resultsStudyStudy ArmsN Outcome and Time Period ResultsDesign/Risk ofAnalyzedBiasGrymonpre,2001 16Cohort/High(continued)Mean symptoms reportedBaselineG1: 7.2(3.7)G2: 7.5(3.5)95% CI: NRp: NR6 monthsG1:7.9 (4.1)G2: 7.2 (3.7)95% CI: NRp: NRMean medication adherenceBetween group p-value:0.089Calculated mean: 0.70,95% CI(-0.73–2.13); p=0.34BaselineG1: 95.0 (79.4)G2: 95.5 (56.7)95% CI: NRp:6 monthsG1: 86.7(46.0)G2: 85.1(41.1)95% CI: NRp:Overall p-value: 0.895Calculated mean: 1.60,95% CI(-14.40–-17.60); p=0.84D-224


Table D48. Intensity of care coordination and followup following comprehensive medication review: Summary of resultsStudyStudy ArmsN Outcome and Time Period ResultsDesign/Risk ofAnalyzedBiasGrymonpre,2001 16Cohort/High(continued)Estimated annual prescriptioncosts in USD per clientBaselineG1: 881 (650)G2: 944 (687)95% CI: NR95% CI: NRp: NR6 monthsG1:809 (578)G2:874 (754)Abbreviations: CI = confidence interval; OR = odds ratio; USD = United States dollars.Between group P= 0.971Calculated mean: -65.00,95% CI(-305.67–175.67); p=0.60D-225


Table D49. Pharmacy intensity of adoption: Summary of resultsStudyStudy ArmsDesign/Risk ofBiasChrischilles et al., Groups 1 and 2 were combined in these analyses. Thus, all PCM-eligible2004 9people were grouped together, regardless of whether they received services.Cohort/High The subgroup analyses involved stratifying the combined group by level ofpharmacy intensity.SG1: High intensitySG2: Moderate intensityAbbreviation: NR = not reported; SG = subgroup.NAnalyzedG1: NRG2: NROutcome and Time PeriodResultsNumber of emergencyFindings NR,department claimsp=0.330Number of inpatient institutional Findings NR,claimsp=0.839Number of outpatient facility Findings NR,claimsp=0.112Number of pharmacy,Findings NR,institutional, and medical services p=0.616Emergency department claims Findings NR,p=0.652Inpatient institutional claims Findings NR,p=0.862Outpatient facility claims Findings NR,p=0.212Pharmacy, institutional, and Findings NR,medical servicesp=0.166D-226


Table D50. Community pharmacy versus call center: Summary of resultsStudyStudy ArmsNDesign/Risk of Bias AnalyzedOutcome and Time PeriodWinston et al., G1: Community pharmacy MTMG1: 21,336 Drug cost per patient per month2009 58G2: Pharmacist-staffed call center-based G2: 3,436 (USD)Cohort/High MTMG3: 49,021G3: Educational mailingsResultsPre-MTM (Jan 1 2007-April 30, 2007)G1: 669 (461)G2: 676 (463)G3: 698 (513)p: NRPost-MTM (Jan 1 2008-April 30, 2008)G1: 634 (512)G2: 661 (494)G3: 698 (556)p: NRDifference (% change)G1: -35 ± 353 (-5.2)G2: -15 ± 374 (-2.2)G3: 0 ± 406 (0)95% CI: NRp: NRCalculated mean difference: -20.0, 95% CI:-32.826 to -7.174, p=0.002Drug use per patient per month Pre-MTM (Jan 1 2007-April 30, 2007)G1: 9.79 (3.17 )G2: 9.76 (2.93)G3: 9.70 (2.94)p: NRPost-MTM (Jan 1 2008-April 30, 2008)G1: 9.29 (3.60)G2: 9.63 (3.47)G3: 9.53 (3.61)p: NRDifference (% change)G1: -0.50 ± 3.01 (-5.0)G2: -0.13 ± 2.78 (-1.3)G3: -0.18 ± 2.88 (-1.8)95% CI: NRp: NRCalculated mean difference: -0.370, 95%CI:-0.477 to -0.263, p


Table D50. Community pharmacy versus call center: Summary of results (continued)StudyStudy ArmsNDesign/Risk of Bias AnalyzedOutcome and Time Period ResultsWinston et al.,2009 58Cohort/High(continued)Weighted generic dispensing ratio Pre-MTM (Jan 1 2007-April 30, 2007)G1: 60.1 (29.8)G2: 58.6 (25.7)G3: 58.7 (27.6)p: NRPost-MTM (Jan 1 2008-April 30, 2008)G1: 65.7 (32.5)G2: 64.6 (30.5)G3: 63.5 (32.2)p: NRDifference (% change)G1: 5.6 ± 26.2 (9.4)G2: 6.0 ± 23.1 (10.2)G3: 4.8 ± 23.8 (8.1)95% CI: NRp: NRAbbreviations: CI = confidence interval; USD = United States dollars.Calculated mean difference: 9.710, 95%CI:9.583 to 9.837, p


Table D51. Type of payer: Summary of resultsStudy Study ArmsDesign/Riskof BiasWitry et al.,2011 59Cohort/HighNAnalyzedG1: Iowa Medicaid Pharmaceutical Case Management (PCM), which pays for G1: 45pharmacists to collaborate with patients and their physicians. Pharmacists G2: 469conducted CMR, looked for DRPs and untreated conditions, provided pt educationas needed, followed up with pt as needed, and coordinated care with physicians andpts.G2: PCM provided to patients with private individual-group insuranceOutcome and TimePeriodPer-patient MedicationAppropriateness Index atfollowupResultsBaselineOverall: NRG1: 9.4 (7.7)G2: 1.53 (1.64)P: NRFinalOverall: NRG1: 8.3 (7.1)G2: 1.24 (1.22)P: NRG1: 45G2: 474Proportion of patients forwhom cost was a problemat followupCalculated meandifference: 0.81, 95%CI: -1.303 to 2.923,p=0.452BaselineOverall: NRG1: 67.1G2: 60.0P: NRFinalOverall: NRG1: 65.2G2: 55.6P: 0.198G1: 91G2: 524Drug therapy problemsidentifiedCalculated OR: 1.498,95% CI: 0.807 to2.778, p=0.202.6 in both arms, p=1.0Abbreviations: CI = confidence interval; OR = odds ratio.D-229


Table D52. Confusion: Summary of resultsStudyStudy ArmsNDesign/Risk of BiasCarter et al.,1997 7 , G1: Pharmaceutical careBarnette et al., 1996 8 G2: Usual care with patients seen byCohort/High pharmacists who did not participate in theintensive skills development programAbbreviations: CI = confidence interval; OR = odds ratio.Outcome and Time PeriodAnalyzedG1: 25 Percentage that agree or strongly agree with theG2: 26 statement that they were inconvenienced by monthlyappointments with the pharmacistsResultsG1: 40 percentG2: 69 percentCalculated OR: 0.278,95% CI: 0.088 to 0.875;p=0.029D-230


Table D53. ApplicabilityAuthor, Interventions and ComparatorYear DescriptionsTrial NameBernsten etal., 2001 1 ;Sturgess etal., 2003 2Carter et al.,1997 7 ,Barnette etal., 1996 8G1: Structured communitypharmacy-based pharmaceuticalcare programG2: Normal pharmaceutical Usualcommunity pharmacy servicesG1: Pharmaceutical care providedby pharmacists within aninterdisciplinary practice model.Patient education (lifestyle, riskfactor modifications, and drugtherapy) was standardized.G2: Usual careStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Yes Yes Yes YesNoRural populationYes Yes YesD-231


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameChrischilles G1: PCM provided by pharmacistset al., 2004 9 G2: Did not receive PCM servicesChristensen G1: MTM services designed by aet al., 2007 10 health plan for its beneficiaries andprovided by either communitypharmacists or medical clinic-basedpharmacists.G2: Patients from same counties asG1 who did not receive intervention(control group 1)G3: Patients from a different countythan G1 who did not receiveintervention (control group 2)Study Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Yes Yes Yes YesYes Yes Yes YesD-232


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameClifford etal., 2002 11Fischer etal., 2000 12Fischer etal., 2002 13Study Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)G1: Pharmaceutical care providedby a clinical pharmacist , whichincluded a comprehensive reviewrelating to pharmacotherapy anddiabetes, use of proprietary andnon-proprietary medications, suchas complementary medicines, andidentification of drug therapyproblems.G2: Standard outpatient care fordiabetesYes Yes Yes YesG1: Pharmaceutical care based on Yes Yes Yes Nothe Encara Practice Systemprovided by onsite healthThe outcomes are verymaintenance organization staffintermediate (receipt ofpharmacists.information, use ofG2: Standard Community Pharmacyreminders to takePracticemedication, andG3: A set of refusers surveyed andawareness of sideincluded in some analyses amongeffects)those who were at eligible clinics butinitially declined to participate.Pharmaceutical care based on the YesUnclear or NR YesYesEncara Practice System provided bypharmacists. Communication ofPharmaciespharmacist with the patient'svolunteered tophysician about drug therapyparticipate in theproblems identified by theintervention group. Notpharmacist.clear howG2: Usual care with no additionalrepresentative they areinterventions.of communitypharmacies in general.D-233


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Fox et al., G1: Florida Health Care Plans MTM Yes Yes Yes Yes2009 14 program, consisting of a medicationtherapy review and evaluation by aclinical pharmacist that wasdocumented and sent to thepatient's physician through healthplan reviewG2: Opt-out from MTM programGattis et al., G1: Clinical pharmacy services, NoYes Yes Yes1999 15 including an assessment ofprescribed regimen, compliance, Study population limited to patientsand adverse effects, and symptoms with moderate to severe heartand response to therapy. Providing failure.patient education about the purposeof each drug and reinforcingadherence. Detailed writteninformation was also provided topatients.G2: Usual medical careHanlon etal., 1996 17G1: Pharmaceutical care providedby a clinical pharmacistG2: Usual care in the GeneralMedicine ClinicNoAll male VA patients.Yes Yes YesD-234


Table D53. Applicability (continued)Author, Year Interventions and ComparatorTrial Name DescriptionsStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)G1: MTM services provided byPharmacotherapy consultation andG1: Medication reviews led byIsetts et al.,Yes Yes Yes Yes2008 22 staff pharmacists, including theestablishment of goals of therapy,in collaboration with primary careproviders.G2: Usual medical care withoutMTMJameson,Yes Yes Yes Yes1995 23 care.VanNoord,andfollowup provided by clinicalambulatory care pharmacist.Vanderwoud, G2: Standard office-based primaryJeong et al., G1: Pharmacist-managed MTMP YesNoYesYes2007 24 provided by ambulatory carepharmacists and healthcaresupport staffG2: Eligible for Part D MTMP butdeclined enrollmentG3: Patients without Part D as theirprimary drug benefitMTM intervention itselfmay be applicable, butit was delivered withinKaiser Permanente'sintegrated healthcaresystem, which does notreflect organization ofmost healthcaresystems in the US.Krska et al.,NoYes Yes Yes2001 28 clinically-trained pharmacists.G2: Usual care involving interviews Older adultsand identification of pharmaceuticalcare issues but with nopharmaceutical care planimplemented.D-235


Table D53. Applicability (continued)Author, Year Interventions and ComparatorTrial Name DescriptionsMalone et al.,2000 29 ;Ellis et al.,2000 30 ;Malone et al.,2001 31 ;Ellis et al.,2000 32IMPROVEMcDonoughet al., 2005 36Moczygembaet al., 2011 37Moczygembaet al., 2008 38G1: Pharmaceutical care providedby clinical pharmacists practicingaccording to scope of practicewithin their respective health carefacilities.G2: Usual care withoutpharmaceutical careG1: Pharmaceutical care providedby community pharmacists. Drugtherapy monitoring focused on 5drug therapy problems:appropriateness of does, properregimen, potential interactions,nonadherence, and adverseeffects. Patient education alsoprovided.G2: Usual careG1: Opt-in telephone-based MTMprogram, in which MTM servicesprovided by clinical pharmacists ora managed care pharmacyresident based on the AmericanPharmacists Association andNational Association of Chain DrugStores Foundation MTMframework.G2: No-MTM control groupStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)YesIntervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)NoVA is a large integratedhealth system with onsitepharmacy andhighly trained clinicalpharmacists who areembedded withinambulatory care clinics.This is not typical ofmost primary carepractices.Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)YesNoYes Yes Yes YesYes Yes Yes YesUnclear how applicableVA costing methods andsystems are to the restof the healthcare system.D-236


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NamePai et al.,2009 41 ; Paiet al., 2009 42G1: Pharmaceutical care includingdrug therapy reviews conducted bya nephrology-trained clinicalpharmacist with the patient. Alsoincluded patient and health careprovider education.G2: Standard of care, consisting ofbrief therapy reviews conducted bya nursePark et al., G1: Comprehensive pharmaceutical1996 43 services including drug therapymonitoring and patient educationprovided by a community pharmacyresident.G2: Usual carePindolia etal., 2009 44G1: Telephone-based MTM servicesprovided as part of a Medicare PartD MTM program by pharmacy caremanagement clinical pharmacists.G2: Usual medical carePlanas et al., G1: MTM services provided by2009 45 community pharmacists. Alsoincluded patient education on dietand lifestyle modifications to lowerblood pressure.G2: No MTM received, but onlyinformed of blood pressure goals forpatients with diabetesRoughead etal., 2009 46G1: Home Medication Reviews(HMR), a collaborative model ofpharmaceutical care, conducted byaccredited pharmacists.G2: No medication review receivedStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)NoNarrow eligibility - Adults with ESRDwho were undergoing a stablehemodialysis.Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Yes Yes YesYes Yes Yes YesYes Yes Yes YesYes Yes Yes YesYesD-237Unclear or NRAustralia's health caresystem is different thanthe U.S. health caresystem, so it is unclearhow generalizablethese results are to theUS.YesYes


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Sellors et al., G1: Clinical pharmacistYes and No2003 47 consultations provided to familyphysicians and their patients by Yes for Canada, but may not be forcommunity pharmacists.US.G2: Usual care for family physiciansand their patients from matchedpostal codes.Sidel et al., G1: Home visits by pharmacists Unclear or NR1990 50 and, when needed, consultationswith physicians to identify and Narrow eligibility based on excludingcorrect problems associated with low and medium risk patients andmedication use.those considered to be "difficult".G2: Standard care without any visitsor information provided to G1.Staresinic et G1: MTM services provided as partal., 2007 51 of a Medicare Part D MTM programby an MTM Coordinator (non-clinicalstaff) and a pharmacistG2: Usual care provided to MTMeligibleenrollees who chose not toparticipateTaylor, Byrd, G1: Pharmaceutical care providedand Krueger, by pharmacists2003 52 G2: Standard care without advice orrecommendations given to patientsor physiciansTouchette etal., 2012 53G1: MTM basic (comprehensivemedication review and DRPassessment)G2: MTM enhanced (MTM plus 2-page clinical summary abstractedfrom patient's medical chart)G3: Usual care, consisting ofmedication counseling per clinic’snormal routine but no formal MTMfrom a study pharmacistIntervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Yes Yes YesYes Yes NoShort-term, mostsubjective and notbroadly applicable, mostsurrogate outcomes.Yes Yes Yes Unclear or NRYes Yes Yes YesYes Yes Yes YesD-238


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameVolume etal., 2001 54 ;Kassam etal., 2001 55G1: Comprehensive pharmaceuticalcare services using a nine-stepprocess as defined by Hepler andStrand provided by communitypharmacists.G2: Traditional pharmacy carePREPWelch et al., G1: MTM program provided to2009 56 home-based beneficiaries as part ofa Medicare Part D MTM programG2: No-MTM control group(voluntary opt-out)Williams etal., 2004 57G1: Medication review andoptimization of patient's medicationregimen conducted by aninterdisciplinary medicationadjustment team in addition to usualmedical care and "Bound for Health"booklet.G2: Usual medical care plusprovision of "Bound for Health"bookletStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Yes Yes Yes YesYesYesKPCO's level ofintegration notwidespread, butintervention itself isapplicableYes Yes Yes YesYesYesD-239


Table D53. Applicability (continued)Author, Interventions and ComparatorYear DescriptionsTrial NameStudy Population broadlyapplicable? (e.g., not selectedusing narrow eligibility criteria,similarity in demographicsbetween study population andcommunity patients)Intervention broadlyapplicable? (e.g.,design ofinterventionsreflected in currentpractice)Comparator(s) Outcomes broadlybroadly applicable? applicable? (e.g., not(e.g., alternative limited to short-term,therapy or usual care surrogate, orreflective of current composite outcomes)practice)Winston and G1: MTM provided in a community Yes Yes Yes YesLin, 2009 58 pharmacy (i.e., care in face-to-facemeetings or by telephone) as part ofa Medicare Part D MTM programG2: MTM provided by pharmaciststaffedcall centers as part of aMedicare Part D MTM programG3: Educational mailings (i.e.,mailed letter containing patientspecificmedication relatedinformation, personal medicationrecord, and tips to save money onprescriptions)Witry, G1: PCM provided by community Yes Yes Yes YesDoucette, pharmacists to Iowa Medicaidand Gainer, enrollees2011 59 G2: PCM provided by communitypharmacists to patients with privateindividual-group insuranceAbbreviations: BP = blood pressure; G = group; HMR = home medication review; KPCO = Kaiser Permanente Colorado; MTM = medication therapy management; MTMP =medication therapy management program; NR = not reported; PA = physician assistant; PCM = pharmaceutical case management; PREP = Pharmaceutical Care Research andEducation Project; US = United States; VA = Veterans AffairsD-240


References1. Bernsten C, Bjorkman I, Caramona M, et al.Improving the well-being of elderly patients viacommunity pharmacy-based provision ofpharmaceutical care: a multicentre study in sevenEuropean countries. Drugs Aging.2001;18(1):63-77. PMID: 11232739.2. Sturgess IK, McElnay JC, Hughes CM, et al.Community pharmacy based provision ofpharmaceutical care to older patients. PharmWorld Sci. 2003 Oct;25(5):218-26. PMID:14584229.3. Blakey SA, Hixson-Wallace JA. Clinical andeconomic effects of pharmacy services in ageriatric ambulatory clinic (Structured abstract).Pharmacotherapy. 2000(10):1198-203. PMID:NHSEED-22000001609.4. Brummel AR, Soliman AM, Carlson AM, et al.Optimal diabetes care outcomes following faceto-facemedication therapy management services.Popul Health Manag. 2013 Feb;16(1):28-34.PMID: 23113628.5. Soliman AM, Carlson AM, MacLehose RF, et al.Patient characteristics predicting the frequencyof medication therapy management visits forpatients with diabetes. Clin Ther. 2013Apr;35(4):534-40. PMID: 23541131.6. Ramalho de Oliveira D, Brummel AR, MillerDB. Medication therapy management: 10 yearsof experience in a large integrated health caresystem. J Manag Care Pharm. 2010Apr;16(3):185-95. PMID: 20331323.7. Carter BL, Barnette DJ, Chrischilles E, et al.Evaluation of hypertensive patients after careprovided by community pharmacists in a ruralsetting. Pharmacotherapy. 1997 Nov-Dec;17(6):1274-85. PMID: 9399611.8. Barnette DJ, Murphy CM, Carter BL. Clinicalskill development for community pharmacists. JAm Pharm Assoc (Wash). 1996Sep;NS36(9):573-80. PMID: 8824077.9. Chrischilles EA, Carter BL, Lund BC, et al.Evaluation of the Iowa Medicaid pharmaceuticalcase management program. J Am Pharm Assoc(2003). 2004 May-Jun;44(3):337-49. PMID:15191244.10. Christensen DB, Roth M, Trygstad T, et al.Evaluation of a pilot medication therapymanagement project within the North CarolinaState Health Plan. J Am Pharm Assoc (2003).2007 Jul-Aug;47(4):471-83. PMID: 17616493.11. Clifford RM, Batty KT, Davis TME, et al. Arandomised controlled trial of a pharmaceuticalcare programme in high-risk diabetic patients inan outpatient clinic. Int J Pharm Pract.2002;10(2):85-9.12. Fischer LR, Scott LM, Boonstra DM, et al.Pharmaceutical care for patients with chronicconditions. J Am Pharm Assoc (Wash). 2000Mar-Apr;40(2):174-80. PMID: 10730021.13. Fischer LR, Defor TA, Cooper S, et al.Pharmaceutical care and health care utilization inan HMO. Eff Clin Pract. 2002 Mar-Apr;5(2):49-57. PMID: 11990212.14. Fox D, Ried LD, Klein GE, et al. A medicationtherapy management program's impact on lowdensitylipoprotein cholesterol goal attainment inMedicare Part D patients with diabetes. J AmPharm Assoc (2003). 2009 Mar-Apr;49(2):192-9. PMID: 19289345.15. Gattis WA, Hasselblad V, Whellan DJ, et al.Reduction in heart failure events by the additionof a clinical pharmacist to the heart failuremanagement team: results of the Pharmacist inHeart Failure Assessment Recommendation andMonitoring (PHARM) Study. Arch Intern Med.1999 Sep 13;159(16):1939-45. PMID:10493325.16. Grymonpre RE, Williamson DA, MontgomeryPR. Impact of a pharmaceutical care model fornon-institutionalised elderly: Results of arandomised, controlled trial. Int J Pharm Pract.2001(4):235-41. PMID: CN-00425461.17. Hanlon JT, Weinberger M, Samsa GP, et al. Arandomized, controlled trial of a clinicalpharmacist intervention to improve inappropriateprescribing in elderly outpatients withpolypharmacy. Am J Med. 1996Apr;100(4):428-37. PMID: 8610730.18. Cowper PA, Weinberger M, Hanlon JT, et al.The cost-effectiveness of a clinical pharmacistintervention among elderly outpatients.Pharmacotherapy. 1998 Mar-Apr;18(2):327-32.PMID: 9545151.19. Hirsch JD, Gonzales M, Rosenquist A, et al.Antiretroviral therapy adherence, medicationuse, and health care costs during 3 years of acommunity pharmacy medication therapyD-241


management program for Medi-Cal beneficiarieswith HIV/AIDS. J Manag Care Pharm. 2011Apr;17(3):213-23. PMID: 21434698.20. Hirsch JD, Rosenquist A, Best BM, et al.Evaluation of the first year of a pilot program incommunity pharmacy: HIV/AIDS medicationtherapy management for Medi-Cal beneficiaries.J Manag Care Pharm. 2009 Jan-Feb;15(1):32-41.PMID: 19125548.21. Rosenquist A, Best BM, Miller TA, et al.Medication therapy management services incommunity pharmacy: a pilot programme in HIVspecialty pharmacies. J Eval Clin Pract. 2010Dec;16(6):1142-6. PMID: 21143346.22. Isetts BJ, Schondelmeyer SW, Artz MB, et al.Clinical and economic outcomes of medicationtherapy management services: the Minnesotaexperience. J Am Pharm Assoc. 2008 Mar-Apr;48(2):203-11; 3 p following 11. PMID:18359733.23. Jameson J, VanNoord G, Vanderwoud K. Theimpact of a pharmacotherapy consultation on thecost and outcome of medical therapy. J FamPract. 1995 Nov;41(5):469-72. PMID: 7595265.24. Jeong EW, Gallagher MW, Fredriks D, et al.Impact of a pharmacist-managed medicationtherapy management (MTM) program. ASHPMidyear Clinical Meeting. 2007;42(DEC)PMID:45-05344.25. Jeong E, Hui R, Gallagher M, et al. Clinical,Economical and Humanistic Outcomes from aPharmacist-managed Medication TherapyManagement Program (MTMP) in MedicarePatients from an Integrated Healthcare DeliverySystem. Kaiser Permanante; 2009. p. 1-31.26. Jeong E, Pitcher K, Fredriks D, et al. Clinical &Safety Outcomes from a Pharmacist-managedMedicare Medication Therapy ManagementProgram. Kaiser Permanente.27. Jeong E, Deguzman L, Fredriks D, et al.Comprehensive Medication Reviews ImproveHospitalization and Emergency DepartmentOutcomes. Kaiser Permanente; 2012.28. Krska J, Cromarty JA, Arris F, et al. Pharmacistledmedication review in patients over 65: arandomized, controlled trial in primary care. AgeAgeing. 2001 May;30(3):205-11. PMID:11443021.29. Malone DC, Carter BL, Billups SJ, et al. Aneconomic analysis of a randomized, controlled,multicenter study of clinical pharmacistinterventions for high-risk veterans: theIMPROVE study. Impact of ManagedPharmaceutical Care Resource Utilization andOutcomes in Veterans Affairs Medical Centers.Pharmacotherapy. 2000 Oct;20(10):1149-58.PMID: 11034037.30. Ellis SL, Billups SJ, Malone DC, et al. Types ofinterventions made by clinical pharmacists in theIMPROVE study. Impact of ManagedPharmaceutical Care on Resource Utilization andOutcomes in Veterans Affairs Medical Centers.Pharmacotherapy. 2000 Apr;20(4):429-35.PMID: 10772374.31. Malone DC, Carter BL, Billups SJ, et al. Canclinical pharmacists affect SF-36 scores inveterans at high risk for medication-relatedproblems? Med Care. 2001 Feb;39(2):113-22.PMID: 11176549.32. Ellis SL, Carter BL, Malone DC, et al. Clinicaland economic impact of ambulatory care clinicalpharmacists in management of dyslipidemia inolder adults: the IMPROVE study. Impact ofManaged Pharmaceutical Care on ResourceUtilization and Outcomes in Veterans AffairsMedical Centers. Pharmacotherapy. 2000Dec;20(12):1508-16. PMID: 11130223.33. Marques LA, Galduroz JC, Fernandes MR, et al.Assessment of the effectiveness ofpharmacotherapy follow-up in patients treatedfor depression. J Manag Care Pharm. 2013Apr;19(3):218-27. PMID: 23537456.34. Marrufo G, Dixit A, Perlroth D, et al. MedicationTherapy Management in a Chronically IllPopulation: Interim Report. (Prepared byAcumen LLC under Contract #HHSM-500-2011-00012I/TOT#0001.) Baltimore, MD:Services CfMM; 2013.35. Perlroth D, Marrufo G, Montesinos A, et al.Medication Therapy Management in ChronicallyIll Populations: Final Report. Contract #HHSM-500-2011-00012I/TOT0001. Baltimore,MD: Innovation CfMMSCCfMM; 2013.http://innovation.cms.gov/Files/reports/MTM_Final_Report.pdf.36. McDonough RP, Doucette WR, Kumbera P, etal. An evaluation of managing and educatingpatients on the risk of glucocorticoid-inducedosteoporosis. Value Health. 2005 Jan-Feb;8(1):24-31. PMID: 15841891.37. Moczygemba LR, Barner JC, Lawson KA, et al.Impact of telephone medication therapymanagement on medication and health-relatedD-242


problems, medication adherence, and MedicarePart D drug costs: a 6-month follow up. Am JGeriatr Pharmacother. 2011 Oct;9(5):328-38.PMID: 21865093.38. Moczygemba LR, Barner JC, Gabrillo ER, et al.Development and implementation of a telephonemedication therapy management program forMedicare beneficiaries. Am J Health Syst Pharm.2008 Sep 1;65(17):1655-60. PMID: 18714113.39. Moczygemba LR, Barner JC, Gabrillo ER.Outcomes of a Medicare Part D telephonemedication therapy management program.2012;52:E144.40. Moore JM, Shartle D, Faudskar L, et al. Impactof a patient-centered pharmacy program andintervention in a high-risk group. J Manag CarePharm. 2013 Apr;19(3):228-36. PMID:23537457.41. Pai AB, Boyd A, Depczynski J, et al. Reduceddrug use and hospitalization rates in patientsundergoing hemodialysis who receivedpharmaceutical care: a 2-year, randomized,controlled study. Pharmacotherapy. 2009Dec;29(12):1433-40. PMID: 19947803.42. Pai AB, Boyd A, Chavez A, et al. Health-relatedquality of life is maintained in hemodialysispatients receiving pharmaceutical care: a 2-yearrandomized, controlled study. Hemodial Int.2009 Jan;13(1):72-9. PMID: 19210281.43. Park JJ, Kelly P, Carter BL, et al.Comprehensive pharmaceutical care in the chainsetting. J Am Pharm Assoc (Wash). 1996Jul;NS36(7):443-51. PMID: 8840744.44. Pindolia VK, Stebelsky L, Romain TM, et al.Mitigation of medication mishaps via medicationtherapy management. Ann Pharmacother. 2009Apr;43(4):611-20. PMID: 19336646.45. Planas LG, Crosby KM, Mitchell KD, et al.Evaluation of a hypertension medication therapymanagement program in patients with diabetes. JAm Pharm Assoc (2003). 2009 Mar-Apr;49(2):164-70. PMID: 19289342.46. Roughead EE, Barratt JD, Ramsay E, et al. Theeffectiveness of collaborative medicine reviewsin delaying time to next hospitalization forpatients with heart failure in the practice setting:results of a cohort study. Circ Heart Fail. 2009Sep;2(5):424-8. PMID: 19808372.47. Sellors J, Kaczorowski J, Sellors C, et al. Arandomized controlled trial of a pharmacistconsultation program for family physicians andtheir elderly patients. CMAJ. 2003 Jul8;169(1):17-22. PMID: 12847034.48. Sellors C, Dalby DM, Howard M, et al. Apharmacist consultation service in communitybasedfamily practices: A randomized, controlledtrial in seniors. J Pharm Technol. 2001(6):264-9.PMID: CN-00425489.49. Shimp LA, Kucukarslan SN, Elder J, et al.Employer-based patient-centered medicationtherapy management program: Evidence andrecommendations for future programs. p. 768.50. Sidel VW, Beizer JL, Lisi-Fazio D, et al.Controlled study of the impact of educationalhome visits by pharmacists to high-risk olderpatients. J Community Health. 1990Jun;15(3):163-74. PMID: 2195066.51. Staresinic AG. Early experience with medicationmanagement - Description and preliminaryoutcomes of a Medicare Part D medicationtherapy management program. J Pharm Soc Wis.2007(SEP-OCT):8-0. PMID: 45-01183.52. Taylor CT, Byrd DC, Krueger K. Improvingprimary care in rural Alabama with a pharmacyinitiative. Am J Health Syst Pharm. 2003 Jun1;60(11):1123-9. PMID: 12816022.53. Touchette DR, Masica AL, Dolor RJ, et al.Safety-focused medication therapy management:a randomized controlled trial. J Am Pharm Assoc(2003). 2012 Sep-Oct;52(5):603-12. PMID:23023840.54. Volume CI, Farris KB, Kassam R, et al.Pharmaceutical care research and educationproject: patient outcomes. J Am Pharm Assoc(Wash). 2001 May-Jun;41(3):411-20. PMID:11372906.55. Kassam R, Farris KB, Burback L, et al.Pharmaceutical care research and educationproject: pharmacists' interventions. J Am PharmAssoc (Wash). 2001 May-Jun;41(3):401-10.PMID: 11372905.56. Welch EK, Delate T, Chester EA, et al.Assessment of the impact of medication therapymanagement delivered to home-based Medicarebeneficiaries. Ann Pharmacother. 2009Apr;43(4):603-10. PMID: 19318600.57. Williams ME, Pulliam CC, Hunter R, et al. Theshort-term effect of interdisciplinary medicationreview on function and cost in ambulatoryelderly people. J Am Geriatr Soc. 2004Jan;52(1):93-8. PMID: 14687321.D-243


58. Winston S, Lin YS. Impact on drug cost and useof Medicare part D of medication therapymanagement services delivered in 2007. J AmPharm Assoc (2003). 2009 Nov-Dec;49(6):813-20. PMID: 19858047.59. Witry MJ, Doucette WR, Gainer KL. Evaluationof the pharmaceutical case management programimplemented in a private sector health plan. JAm Pharm Assoc (2003). 2011 Sep-Oct;51(5):631-5. PMID: 21896463.60. Wittayanukorn S, Westrick SC, Hansen RA, etal. Evaluation of medication therapymanagement services for patients withcardiovascular disease in a self-insured employerhealth plan. J Manag Care Pharm. 2013Jun;19(5):385-95. PMID: 23697476.61. Yamada B. Comparison of Outcomes in theMedicare Medication Therapy ManagementProgram within Kaiser Permanente California.2012.62. Triller DM, Hamilton RA. Effect ofpharmaceutical care services on outcomes forhome care patients with heart failure. Am JHealth Syst Pharm. 2007 Nov 1;64(21):2244-9.PMID: 17959576.D-244


Appendix E. Risk of Bias Evaluations and RationaleTable E1. Risk of bias domains and ratings: Part 1Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsDid attritionresult indifferencesin groupcharacteristics?Bernsten et al., G1: Structured2001 1 ; communitySturgess, pharmacybased2003 2 pharmaceuticalcare programG2: UsualcommunitypharmacyservicesBlakey and G1: PharmacistHixson- evaluation plusWallace, 2000 3usual medicalcareG2: Usualmedical careRCT:clusterrandomizedRandomizationmethodadequate?(RCTsonly)Unclear orNRAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNo Yes Yes Unclear or NR Yes Yes YesNRCT NA NA Yes No No Unclear or NR Unclear orNRUnclear orNRUnclear orNRE-1


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignBrummel et al.,2013 4: ;Soliman et al.,2013 5 ;Ramalho deOliveira, ,Brummel, andMiller, 2010 6InterventionandComparatorDescriptionsG1: FairviewPharmacyServices' MTMprogram (opt-in)G2: controlgroup (did notopt-in)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Cohort NA NA Yes No NA No Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNADid attritionresult indifferencesin groupcharacteristics?NACarter et al., G1:1997 7 ; PharmaceuticalBarnette et al., care provided1996 8 by pharmacistswithininterdisciplinarypractice model.Standardizedpatienteducation(lifestyle, riskfactormodifications,and drugtherapy).G2: Usual careCohort NA NA Yes No Yes Unclear or NR Unclear orNRNoNAE-2


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignChrischilles etal., 2004 9Christensen etal., 2007 10InterventionandComparatorDescriptionsG1: PCMeligiblepatientswho receivedPCM servicesG2: PCMeligiblepatientswho did notreceive PCMservicesG1: MTMservicesdesigned by ahealth plan forbeneficiariesand provided byeithercommunitypharmacists ormedical clinicbasedpharmacists.G2: Patientsfrom samecounties as G1who did notreceiveintervention.G3: Patientsfrom a differentcounty than G1who did notreceiveintervention.Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Cohort NA NA Yes No Unclear orNRNRCTfor G1vs. G3,Cohortfor G1vs. G2NA NA Yes Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear or NR NA No NoDid attritionresult indifferencesin groupcharacteristics?No Yes Yes Unclear orNRE-3


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsClifford et al., G1:RCT: Yes2002 11 Pharmaceutica parallel,l care provided notby a clinical clusteredpharmacist,including acomprehensive reviewrelating topharmacotherapyanddiabetes, useof proprietaryand nonproprietarymedications,such ascomplementary medicines,andidentification ofDTPs.G2: Standardoutpatient carefor diabetesRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-4


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsFischer et al., G1:2000 12 Pharmaceuticalcare based onEncara PracticeSystemprovided byonsite healthmaintenanceorganizationstaffpharmacists(acceptors).G2: StandardCommunityPharmacyPractice.G3: A set ofthose at eligibleclinics whoinitially declinedto participate(opt-out).Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)NRCT NA NA No Yes No Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRDid attritionresult indifferencesin groupcharacteristics?YesE-5


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsFischer et al., G1:2002 13 Pharmaceuticalcare based onEncara PracticeSystemprovided bypharmacists.Communicationof pharmacistwith thepatient'sphysician aboutdrug therapyproblemsidentified by thepharmacist.G2: Usual care.Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)NRCT NA NA Yes Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?NoDid variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-6


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsFox et al., G1: Florida2009 14 Health CarePlans MTMprogram,consisting ofmedicationtherapy reviewand evaluationby a clinicalpharmacist thatwasdocumentedand sent topatient'sphysicianthrough healthplan review(acceptors)G2: Opt-outfrom MTMprogramRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionCohort NA NA Yes No NA Unclear or NR NA No NoDid attritionresult indifferencesin groupcharacteristics?E-7


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsG1: ClinicalGattis et al.,RCT: Yes1999 15 pharmacyservices,includingassessment ofprescribedregimen,compliance,and adverseeffects, andsymptoms andresponse totherapy.Providingpatienteducationabout purposeof each drugand reinforcingadherence.Detailedwritteninformationalso providedto patients.G2: Usualmedical careparallel,notclusteredRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)No Yes No No Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRDid attritionresult indifferencesin groupcharacteristics?NoE-8


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignGrymonpre, G1:Williamson,andMontgomery,2001 16Hanlon et al., G1:1996 17 ;Cowper et al.,1998 18InterventionandComparatorDescriptionsComprehensive drug therapyreview, thenissuesaddressed withthe clientand/or theclient'sphysician, withfollow-up asrequired.G2:Comprehensive drug therapyreview onlywith referral tousualpharmacist.Pharmaceutical care providedby clinicalpharmacist.G2: Usual carein the GeneralMedicine ClinicRCT: Yesparallel,notclusteredRCT: Yesparallel,notclusteredRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Unclear orNRUnclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Yes No Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRNo Yes Yes Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRUnclear orNRE-9


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?G1: MTMIsetts et al.,Cohort NA NA Yes Unclear or2008 22 servicesprovided bystaffpharmacists,including theestablishmentof goals oftherapy, incollaborationwith primarycare providers.G2: Usualmedical carewithout MTM.NRJameson et G1:al., 1995 23 Pharmacotherapyconsultationand follow-upprovided byclinicalambulatorycarepharmacist.G2: Standardoffice-basedprimary care.RCT: Yesparallel,notclusteredUnclear orNROutcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)No No Unclear orNRNo Yes No No Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRUnclear orNRE-10


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignJeong et al., G1:2007 24 ; Jeonget al., 2009 25InterventionandComparatorDescriptionsDid attritionresult indifferencesin groupcharacteristics?PharmacistmanagedMTMPprovided byambulatorycarepharmacistsand healthcaresupport staff(acceptors)G2: Eligible forPart D MTMPbut declinedenrollment(refusers)G3: Patientswithout Part Das theirprimary drugbenefitRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Cohort NA NA Yes No Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?NoDid variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear,onlyincludedsubjectswithavailabledata.Unclear orNRE-11


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsJeong et al. 26 ; G1: Kaiser-Jeong et al., Permanente2012 27MTM programparticipants(2010)G2: Kaiser-Permanentepatientseligible forMTM, but whodeclinedenrollment ordisenrolledwith a PCPvisit during firsthalf of 2010G3: Kaiser-Permanentepatientseligible forMTM, but whodeclinedenrollment ordisenrolledwithout a PCPvisit during firsthalf of 2010Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Cohort NA NA Yes No Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear,onlyincludedsubjectswithavailabledata.Did attritionresult indifferencesin groupcharacteristics?Unclear orNRE-12


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsRandomizationmethodadequate?(RCTsonly)Krska et al., G1: Medication RCT: Unclear or2001 28 reviews led by parallel, NRclinicallytrainedclusterednotpharmacists.G2: Usual careinvolvinginterviews andidentification ofpharmaceutical care issuesbut with nopharmaceutical care planimplemented.Malone et al.,2000 29 ; Ellis etal., 2000 30 ;Malone et al.,2001 31 ; Ellis etal., 2000 32IMPROVEG1:Pharmaceutical care providedby clinicalpharmacistspracticingaccording toscope ofpractice withintheirrespectivehealth carefacilitiesG2: Usual carewithoutpharmaceutical careRCT: Yesparallel,notclusteredAllocationconcealmentadequate?(RCTsonly)Unclear orNRUnclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No No Unclear orNRNo Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRYesUnclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRUnclear orNRE-13


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsRandomizationmethodadequate?(RCTsonly)Marques et al.,2013 33 G1:Interventiongroup: Dadermethodpharmacotherapyfollow-upinterventionmonthly over3-monthfollow-upperiodG2: Controlgroup: monthlypharmacistvisits withoutpharmacotherapyfollowupinterventionRCT:parallel,notclusteredUnclear orNRAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-14


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsMarrufo et al., CHF2013 34 ; G1: enrolled inPerlroth et al., Medicare PDP2013 35 receiving MTMwith a CMRG2: enrolled inPDP receivingMTM, no CMRG3: enrolled inMA-PDreceiving MTMwith CMRG4: enrolled inMA-PD,receivingMTM, no CMRCOPDG5: enrolled inMedicare PDPreceiving MTMwith a CMRG6: enrolled inPDP receivingMTM, no CMRG7: enrolled inMA-PDreceiving MTMwith CMRG8: enrolled inMA-PD,receivingMTM, no CMRRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionCohort NA NA Yes No NA Unclear or NR NA No NADid attritionresult indifferencesin groupcharacteristics?E-15


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsDid attritionresult indifferencesin groupcharacteristics?McDonough et G1:RCT:al., 2005 36 Pharmaceutica cluster-l care provided rando-mizedby communitypharmacists.Drug therapymonitoringfocused on 5DTPs:appropriateness of dose,properregimen,potentialinteractions,nonadherence,and adverseeffects. Patienteducation alsoprovided.G2: Usual careRandomizationmethodadequate?(RCTsonly)Unclear orNRAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No No Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?NoDid variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoUnclear orNRE-16


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsDid attritionresult indifferencesin groupcharacteristics?Moczygemba G1: Opt-inet al., 2011 37 ; telephonebasedMTMMoczygembaet al., 2008 38 ; program, inMoczygemba, which MTMBarner, and servicesGabrillo,2012 39provided byclinicalpharmacists ormanaged carepharmacyresident basedon AmericanPharmacistsAssociationand NationalAssociation ofChain DrugStoresFoundationMTMframework(acceptors)G2: No-MTMcontrol group(opt-out)Moore et al., G1: MTM2013 40 program (optin)G2: controlgroup(refusers)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Cohort NA NA Yes Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionCohort NA NA Yes Yes NA Unclear or NR NA NA NANoUnclear orNRE-17


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignPai et al., G1:2009 41 ; Pai etal., 2009 42InterventionandComparatorDescriptionsPharmaceutical care includingdrug therapyreviewsconducted bynephrologytrainedclinicalpharmacistwith patient.Also includedpatient andhealth careprovidereducation.G2: Standardof care,consisting ofbrief therapyreviewsconducted bynurseRCT:clusterrandomizedRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No Yes No Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionYesDid attritionresult indifferencesin groupcharacteristics?NoE-18


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsRandomizationmethodadequate?(RCTsonly)Park et al.,1996 43 G1: Comprehensivepharmaceuticalservices,including drugtherapymonitoring andpatienteducationprovided bycommunitypharmacyresident.G2: Usual careRCT:parallel,notclusteredUnclear orNRAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)No No No Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-19


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsPindolia et al., G1:2009 44 TelephonebasedMTMservicesprovided aspart ofMedicare PartD MTMprogram bypharmacy caremanagementclinicalpharmacists(acceptors)G2: Usualmedical care(opt-out)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionCohort NA NA Yes No NA Unclear or NR NA Unclear orNRDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-20


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsPlanas et al., G1: MTM2009 45 servicesprovided bycommunitypharmacists.Also includedpatienteducation ondiet andlifestylemodificationsto lower bloodpressure.G2: No MTMreceived, butonly informedof bloodpressure goalsfor patientswith diabetesRCT:parallel,notclusteredRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Yes Yes No No Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionYesDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-21


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignRoughead etal., 2009 46InterventionandComparatorDescriptionsG1: HMRsconducted byaccreditedpharmacistsG2: NomedicationreviewreceivedSellors et al., G1: Clinical2003 47 pharmacistconsultationsprovided tofamilyphysiciansand theirpatients bycommunitypharmacists.G2: Usualcare forfamilyphysiciansand theirpatients frommatchedpostal codes.Sellors et al., G1:2003 47 PharmaceuticalconsultationG2: UsualcareRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Cohort NA NA Unclear orNRRCT:clusterrandomizedGroupssimilar atbaseline,ordifferences adjustedfor inanalysis?NoOutcomeassessorsblinded?(RCTs orNRCTsonly)Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Unclear or NR NADid variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Yes Yes No Yes Yes Unclear or NR Unclear orNRRCT: Yes Yes No Unclear orparallel, notNRclusteredYesUnclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRNoNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRUnclear orNRNAE-22


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsShimp et al., G1: MTM2012 48 program forUniversity ofMichiganbeneficiaries,entitled FOMG2: UsualcareRandomizationmethodadequate?(RCTsonly)RCT: Unclear orparallel, not NRclusteredSidel et al.,1990 49 G1: Homevisits bypharmacistsand, whenneeded,consultationswith physiciansto identify andcorrectproblemsassociatedwithmedicationuse.G2: Standardcare withoutany visits orinformationprovided toG1.RCT:parallel,notclusteredUnclear orNRAllocationconcealmentadequate?(RCTsonly)Unclear orNRUnclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Unclear orNRGroupssimilar atbaseline,ordifferences adjustedfor inanalysis?Unclear orNROutcomeassessorsblinded?(RCTs orNRCTsonly)Unclear orNRNo Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRUnclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRYesDid attritionresult indifferencesin groupcharacteristics?Unclear orNRUnclear orNRE-23


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignStaresinic etal., 2007 50InterventionandComparatorDescriptionsG1: MTMservicesprovided aspart of aMedicare PartD MTMprogram byMTMCoordinator(non-clinicalstaff) andpharmacist(acceptors)G2: Usual careprovided toMTM-eligibleenrollees whochose not toparticipate(opt-out)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionDid attritionresult indifferencesin groupcharacteristics?Cohort NA NA Yes No NA Unclear or NR NA Yes Unclear orNRE-24


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignTaylor, Byrd, G1:and Krueger,2003 51InterventionandComparatorDescriptionsPharmaceutical care providedby pharmacistsG2: Standardcare withoutadvice orrecommendationsgiven topatients orphysiciansRandomizationmethodadequate?(RCTsonly)RCT: Unclear orparallel, NRnotclusteredAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-25


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignTouchette etal., 2012 52InterventionandComparatorDescriptionsG1: MTMbasic(comprehensive medicationreview andDRPassessment)G2: MTMenhanced(MTM plus 2page clinicalsummaryabstractedfrom patient'smedical chart)G3: Usualcare,consisting ofmedicationcounseling perclinic’s normalroutine but noformal MTMfrom a studypharmacistRCT:parallel,notclusteredRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionDid attritionresult indifferencesin groupcharacteristics?Yes Yes No Yes Yes Unclear or NR No No Unclear orNRE-26


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignVolume et al.,2001 53 ;Kassam et al.,2001 54PREP(Pharmaceutical CareResearch andEducationProject)InterventionandComparatorDescriptionsG1: ComprehensiveRCTpharmaceuticalcareservices usinga nine-stepprocess asdefined byHepler andStrandprovided bycommunitypharmacistsG2: Traditionalpharmacy careWelch et al., G1: MTM2009 55 programprovided tohome-basedbeneficiariesas part ofMedicare PartD MTMprogram(acceptors)G2: No-MTMcontrol group(voluntary optout)Randomizationmethodadequate?(RCTsonly)Unclear orNRAllocationconcealmentadequate?(RCTsonly)YesRecruitmentstrategy forstudydifferentacrossgroups?Unclear orNRGroupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Yes No Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionCohort NA NA Yes No NA Unclear or NR NA No NoYesDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-27


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsWilliams et al., G1:2004 56 Modification ofpatient'smedicationregimenconducted byinterdisciplinary medicationadjustmentteam inaddition tousual medicalcare and"Bound forHealth"booklet.G2: Usualmedical careplus provisionof "Bound forHealth"bookletRandomizationmethodadequate?(RCTsonly)RCT: Unclear orparallel, NRnotclusteredAllocationconcealmentadequate?(RCTsonly)Unclear orNRRecruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)No Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNoDid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-28


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignWinston andLin, 2009 57InterventionandComparatorDescriptionsG1: MTMprovided incommunitypharmacy (i.e.,care in face-tofacemeetingsor bytelephone) aspart ofMedicare PartD MTMprogramG2: MTMprovided bypharmaciststaffedcallcenters as partof MedicarePart D MTMprogramG3:Educationalmailings (i.e.,mailed lettercontainingpatient-specificmedicationrelatedinformation,personalmedicationrecord, andtips to savemoney onprescriptions)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Impact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Cohort NA NA Yes Yes NA Unclear or NR Unclear orNRE-29Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionNADid attritionresult indifferencesin groupcharacteristics?NA


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignWitry,Doucette, andGainer, 2011 58InterventionandComparatorDescriptionsG1: PCMprovided bycommunitypharmacists toIowa MedicaidenrolleesG2: PCMprovided topatients withprivateindividualgroupinsuranceWittayanukorn G1:et al., 2013 59 Interventiongroup:Pharmacistprovided faceto-faceMTMservices for30-60 minutesper encounter,not alwaysincluding afollow-up visitG2: Controlgroup: Patientswho did notreceive MTMservices(economicanalyses only)Randomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)Cohort NA NA Yes No Unclear orNRCohort NA NA Yes Yes Unclear orNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Unclear or NR NADid variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Highoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear orNRNo NA NA NADid attritionresult indifferencesin groupcharacteristics?Unclear orNRE-30


Table E1. Risk of bias domains and ratings: Part 1 (continued)Author, YearStudyTrial NameDesignInterventionandComparatorDescriptionsRandomizationmethodadequate?(RCTsonly)Allocationconcealmentadequate?(RCTsonly)Recruitmentstrategy forstudydifferentacrossgroups?Groupssimilar atbaseline,ordifferences adjustedfor inanalysis?Outcomeassessorsblinded?(RCTs orNRCTsonly)G1: Kaiser-Yamada,Cohort NA NA Yes No Unclear or2012 60 PermanenteMTM enrolledpatientsG2: Kaiserpatientsenrolled inMediare partD, but not inMTM programmatched tocontrol on age,gender, regionand DCG riskNRImpact fromconcurrentinterventionor unintendedexposureruled out byresearchers?Did variationfrom studyprotocolcompromiseconclusions? (RCTs orNRCTs only)Unclear or NR Unclear orNRHighoverall(i.e., ≥20%)ordifferential(i.e., ≥15%)attritionUnclear,onlyincludedsubjectswithavailabledata.Did attritionresult indifferencesin groupcharacteristics?Unclear orNRAbbreviations: CHF = chronic heart failure; CMR = comprehensive medication review; COPD = chronic obstructive pulmonary disease; DTP = drug therapy problem; FOM =Focus on Medicines; G = group; HMR = home medication review; IMPROVE = specific name of the MTM trial that was done in the Veterans Affairs health system; MA-PD =Medicare Advantage Part D; MTM = medication therapy management; MTMP = medication therapy management program; NA = not applicable; NR = not reported; NRCT =non-randomized controlled trial; PCM = pharmaceutical case management; PCP = primary care provider; PDP = Medicare Part D Plan; PREP = Pharmaceutical Care Research andEducation Project; RCT = randomized controlled trial; VNA = visiting nurse association.E-31


Table E2. Risk of bias domains and ratings: Part 2Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsBernsten et al., G1: Structured2001 1 ; communitySturgess, pharmacy-based2003 2 pharmaceuticalcare programG2: UsualcommunitypharmacyservicesBlakey and G1: PharmacistHixson- evaluation plusWallace, 2000 3 usual medicalcareG2: Usualmedical careBrummel et al.,2013 4: ;Soliman et al.,2013 5 ;Ramalho deOliveira, ,Brummel, andMiller, 2010 6G1: FairviewPharmacyServices' MTMprogram (opt-in)G2: controlgroup (did notopt-in)RCT:clusterrandomizedNRCTEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?No Yes Yes Yes Yes Unclear orNRUnclear orNRWere importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Partial (somevariables weretaken in toaccount)No Yes NA NA Yes No (Notaccounted for ornot identified)Cohort NA Yes Yes NA NA Yes Partial (somevariables weretakend in toaccount)Risk of BiasMedium(countryspecific)High (pooleddata)HighMediumROB formainanalysis,high ROB forsubgroupanalysisbecauseintensity ofservice iscompletelyconfoundedwith potentialoutcomesE-32


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Carter et al., G1:1997 7 ;Barnette etal., 1996 8Chrischilles etal., 2004 9Interventions/ComparatorDescriptionsPharmaceuticalcare provided bypharmacists withininterdisciplinarypractice model.Standardizedpatient education(lifestyle, riskfactormodifications, anddrug therapy).G2: Usual careG1: PCM-eligiblepatients whoreceived PCMservicesG2: PCM-eligiblepatients who didnot receive PCMservicesEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort No Yes No Yes Yes Yes No (Notaccounted for ornot identified)Cohort No Yes Yes NA Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighHighE-33


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Christensen etal., 2007 10Interventions/ComparatorDescriptionsG1: MTM NRCT Unclearservices for G1designed by a vs. G3, or NRhealth plan for observabeneficiaries and tionalprovided by (cohort)either community for G1pharmacists or vs. G2medical clinicbasedpharmacists.G2: Patientsfrom samecounties as G1who did notreceiveintervention.G3: Patientsfrom a differentcounty than G1who did notreceiveintervention.Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?Unclear orNRIntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Yes Yes Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasMediumE-34


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsClifford et al., G1:2002 11 Pharmaceuticalcare provided bya clinicalpharmacist ,including acomprehensivereview relating topharmacotherapyanddiabetes, use ofproprietary andnon-proprietarymedications,such ascomplementarymedicines, andidentification ofDTPs.G2: Standardoutpatient carefor diabetesRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Yes Yes Yes Yes Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasMediumE-35


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsFischer et al., G1:NRCT2000 12 Pharmaceuticalcare based onEncara PracticeSystem providedby onsite healthmaintenanceorganization staffpharmacists(acceptors).G2: StandardCommunityPharmacyPractice.G3: A set ofthose at eligibleclinics whoinitially declinedto participate(opt-out).Unclearor NREligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Yes No No NA Unclear orNRWere important Risk of Biasconfounding andmodifyingvariables takeninto account indesign and/oranalysis?YesMedium formostoutcomes,high foradverse drugeventsE-36


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsFischer et al., G1:2002 13 Pharmaceuticalcare based onEncara PracticeSystem providedby pharmacists.Communicationof pharmacistwith the patient'sphysician aboutdrug therapyproblemsidentified by thepharmacist.G2: Usual care.Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?NRCT No Yes NA NA Yes Yes Partial (somevariables weretakend in toaccount)Risk of BiasMediumE-37


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsFox et al., G1: Florida2009 14 Health CarePlans MTMprogram,consisting ofmedicationtherapy reviewand evaluationby a clinicalpharmacist thatwas documentedand sent topatient'sphysicianthrough healthplan review(acceptors)G2: Opt-out fromMTM programEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort NA Yes Unclear or NR NA Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighE-38


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsGattis et al., G1: Clinical1999 15 pharmacyservices,includingassessment ofprescribedregimen,compliance, andadverse effects,and symptomsand response totherapy.Providing patienteducation aboutpurpose of eachdrug andreinforcingadherence.Detailed writteninformation alsoprovided topatients.G2: Usualmedical careRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Risk of BiasYes Yes Yes No NA Yes NA MediumE-39


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Grymonpre, G1:Williamson,andMontgomery,2001 16Interventions/ComparatorDescriptionsComprehensivedrug therapyreview, thenissuesaddressed withthe client and/orthe client'sphysician, withfollow-up asrequired.G2:Comprehensivedrug therapyreview only withreferral to usualpharmacist.Hanlon et al., G1:1996 17 ; PharmaceuticalCowper et al., care provided by1998 18 clinicalpharmacist.G2: Usual carein the GeneralMedicine ClinicRCT:parallel,notclusteredRCT:parallel,notclusteredEligibilitycriteriameasuredconsistently usingvalid andreliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were important Risk of Biasconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes Yes Yes Yes Yes No (Notaccounted for ornot identified)Yes forMAI andhealthutilization,nofor SF-36Yes Yes Yes Yes Yes Yes Low for MAIoutcomesand healthutilization(ITT),medium forSF-36HighE-40


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Hirsch et al.,2011 19 ; Hirschet al., 2009 20 ;Rosenquist etal., 2010 21Interventions/ComparatorDescriptionsG1: Patientsserved atnonpilotpharmaciesG2: Patientsserved at pilotpharmaciesIsetts et al., G1: MTM2008 22 servicesprovided by staffpharmacists,including theestablishment ofgoals of therapy,in collaborationwith primary careproviders.G2: Usualmedical carewithout MTM.Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort No Yes Yes NA Yes Yes Partial (somevariables weretakend in toaccount)Cohort Unclearor NRUnclear orNRYes NA NA Unclear orNRNo (Notaccounted for ornot identified)Risk of BiasHighHighE-41


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsJameson et al., G1: Pharmacotherapy1995 23 consultation andfollow-upprovided byclinicalambulatory carepharmacist.G2: Standardoffice-basedprimary care.Jeong et al.,2007 24 ; Jeonget al., 2009 25G1: PharmacistmanagedMTMPprovided byambulatory carepharmacists andhealthcaresupport staff(acceptors)G2: Eligible forPart D MTMPbut declinedenrollment(refusers)G3: Patientswithout Part D astheir primarydrug benefitRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes Yes NA Yes Yes No (Notaccounted for ornot identified)Cohort NA Yes Yes NA Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasMedium formostoutcomes,high foradverse drugeventsMediumE-42


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsJeong et al. 26 ; G1: Kaiser-Jeong et al., Permanente2012 27MTM programparticipants(2010)G2: Kaiser-Permanentepatients eligiblefor MTM, butwho declinedenrollment ordisenrolled witha PCP visitduring first half of2010G3: Kaiser-Permanentepatients eligiblefor MTM, butwho declinedenrollment ordisenrolledwithout a PCPvisit during firsthalf of 2010Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Cohort NA Yes NA NA Unclear orNRPotentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Yes Unclear or NR HighRisk of BiasE-43


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsKrska et al., G1: Medication2001 28 reviews led byclinically-trainedpharmacists.G2: Usual careinvolvinginterviews andidentification ofpharmaceuticalcare issues butwith nopharmaceuticalcare planimplemented.Malone et al.,2000 29 ; Ellis etal., 2000 30 ;Malone et al.,2001 31 ; Ellis etal., 2000 32IMPROVERCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?No Yes Yes Yes Unclear orNRPotentialoutcomeprespecifiedandreported?G1:RCT: Yes Yes Yes Yes Yes Unclear orPharmaceutical parallel,NRcare provided by notclinical clusteredpharmacistspracticingaccording toscope of practicewithin theirrespective healthcare facilitiesG2: Usual carewithoutpharmaceuticalcareYesWere importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No (Notaccounted for ornot identified)YesRisk of BiasMediumMediumE-44


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsMarques et al., G1: Intervention2013 33 group: Dadermethodpharmacotherapy follow-upinterventionmonthly over 3-month follow-upperiodG2: Controlgroup: monthlypharmacist visitswithoutpharmacotherapy follow-upinterventionRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes NA Yes NA Yes No (Notaccounted for ornot identified)Risk of BiasMediumE-45


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Marrufo et al., CHF2013 34 ;Perlroth et al.,2013 35Interventions/ComparatorDescriptionsG1: enrolled inMedicare PDPreceiving MTMwith a CMRG2: enrolled inPDP receivingMTM, no CMRG3: enrolled inMA-PD receivingMTM with CMRG4: enrolled inMA-PD,receiving MTM,no CMRCOPDG5: enrolled inMedicare PDPreceiving MTMwith a CMRG6: enrolled inPDP receivingMTM, no CMRG7: enrolled inMA-PD receivingMTM with CMRG8: enrolled inMA-PD,receiving MTM,no CMREligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort Yes Yes Yes NA Yes Yes Partial (somevariables weretakend in toaccount)Risk of BiasMediumE-46


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsMcDonough et G1:al., 2005 36 Pharmaceuticalcare provided bycommunitypharmacists.Drug therapymonitoringfocused on 5DTPs:appropriatenessof dose, properregimen,potentialinteractions,nonadherence,and adverseeffects. Patienteducation alsoprovided.G2: Usual careRCT:clusterrandomizedEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Yes Yes Yes NA NA Unclear orNRWere importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No (Notaccounted for ornot identified)Risk of BiasMediumE-47


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsMoczygemba G1: Opt-inet al., 2011 37 ;Moczygembaet al., 2008 38 ;Moczygemba,Barner, andGabrillo,2012 39telephone-basedMTM program, inwhich MTMservicesprovided byclinicalpharmacists ormanaged carepharmacyresident basedon AmericanPharmacistsAssociation andNationalAssociation ofChain DrugStoresFoundation MTMframework(acceptors)G2: No-MTMcontrol group(opt-out)Cohort Unclearor NREligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?NA Yes NA Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasMediumE-48


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsMoore et al., G1: MTM2013 40 program (opt-in)G2: controlgroup (refusers)Pai et al., G1:2009 41 ; Pai et Pharmaceuticalal., 2009 42 care includingdrug therapyreviewsconducted bynephrologytrainedclinicalpharmacist withpatient. Alsoincluded patientand health careprovidereducation.G2: Standard ofcare, consistingof brief therapyreviewsconducted bynurseEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort NA Yes NA NA Yes Yes Partial (somevariables weretakend in toaccount)RCT:clusterrandomizedNo Yes Yes NA Yes Yes Yes HighRisk of BiasMediumE-49


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsPark et al., G1:1996 43 Comprehensivepharmaceuticalservices,including drugtherapymonitoring andpatient educationprovided bycommunitypharmacyresident.G2: Usual carePindolia et al., G1: TelephonebasedMTM2009 44 servicesprovided as partof Medicare PartD MTM programby pharmacycaremanagementclinicalpharmacists(acceptors)G2: Usualmedical care(opt-out)RCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Yes Yes Yes Yes NA Unclear orNRWere importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No (Notaccounted for ornot identified)Cohort Yes Yes Unclear or NR Unclear or NR Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighHighE-50


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsPlanas et al., G1: MTM2009 45 servicesprovided bycommunitypharmacists.Also includedpatient educationon diet andlifestylemodifications tolower bloodpressure.G2: No MTMreceived, butonly informed ofblood pressuregoals for patientswith diabetesRoughead etal., 2009 46G1: HMRsconducted byaccreditedpharmacistsG2: Nomedicationreview receivedRCT:parallel,notclusteredCohort Unclearor NREligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes Unclear or NR NA NA Yes No (Notaccounted for ornot identified)Yes NA NA Yes Yes Partial (somevariables weretakend in toaccount)Risk of BiasHighMediumE-51


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsSellors et al., G1: Clinical RCT: Unclear2003 47 pharmacist clusterrandomior NRconsultationsprovided to zedfamily physiciansand their patientsby communitypharmacists.G2: Usual carefor familyphysicians andtheir patientsfrom matchedpostal codes.Sellors et al., G1:2003 47 PharmaceuticalconsultationG2: Usual careShimp et al., G1: MTM2012 48 program forUniversity ofMichiganbeneficiaries,entitled FOMG2: Usual careRCT:parallel,notclusteredRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Risk of BiasYes NA No Yes Yes Yes Medium riskof bias forotheroutcomes;high risk ofbias forquality of lifemeasures.No Yes NA NA Unclear orNRNoUnclear orNRNA NA Unclear orNRUnclear orNRUnclear orNRNo (Notaccounted for ornot identified)No (Notaccounted for ornot identified)MediumMediumE-52


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsSidel et al., G1: Home visits1990 49 by pharmacistsand, whenneeded,consultationswith physiciansto identify andcorrect problemsassociated withmedication use.G2: Standardcare without anyvisits orinformationprovided to G1.RCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No No Yes NA Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighE-53


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Staresinic etal., 2007 50Taylor, Byrd, G1:and Krueger,2003 51Interventions/ComparatorDescriptionsG1: MTMservicesprovided as partof a MedicarePart D MTMprogram by MTMCoordinator(non-clinicalstaff) andpharmacist(acceptors)G2: Usual careprovided toMTM-eligibleenrollees whochose not toparticipate (optout)Pharmaceuticalcare provided bypharmacistsG2: Standardcare withoutadvice or recommendationsgiven to patientsor physiciansEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Cohort Yes Yes NA NA Yes Unclear orNRWere importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No (Notaccounted for ornot identified)RCT:parallel,notclusteredNo Yes Yes Yes Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighMedium formostoutcomes,high foradverse drugeventsE-54


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Touchette etal., 2012 52Interventions/ComparatorDescriptionsG1: MTM basic RCT:(comprehensive parallel,medication notreview and DRP clusteredassessment)G2: MTMenhanced (MTMplus 2 pageclinical summaryabstracted frompatient's medicalchart)G3: Usual care,consisting ofmedicationcounseling perclinic’s normalroutine but noformal MTM froma studypharmacistEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Yes Yes No Yes Yes Yes Yes LowRisk of BiasE-55


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Volume et al.,2001 53 ;Kassam et al.,2001 54PREP(Pharmaceutical CareResearch andEducationProject)Interventions/ComparatorDescriptionsG1:RCT:Comprehensive cluster-pharmaceutical rando-mizedcare servicesusing a nine-stepprocess asdefined byHepler andStrand providedby communitypharmacistsG2: Traditionalpharmacy careWelch et al., G1: MTM2009 55 programprovided tohome-basedbeneficiaries aspart of MedicarePart D MTMprogram(acceptors)G2: No-MTMcontrol group(voluntary optout)Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes Yes Yes NA Yes No (Notaccounted for ornot identified)Cohort NA Yes Unclear or NR Yes Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasMediumMedium foroutcomesreported asadjustedORs, highfor outcomesreportedwithoutadjustmentsforconfoundingE-56


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsWilliams et al., G1: Modification2004 56 of patient'smedicationregimenconducted byinterdisciplinarymedicationadjustment teamin addition tousual medicalcare and "Boundfor Health"booklet.G2: Usualmedical careplus provision of"Bound forHealth" bookletRCT:parallel,notclusteredEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?No Yes Yes Yes No Yes No (Notaccounted for ornot identified)Risk of BiasMediumE-57


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Winston andLin, 2009 57Interventions/ComparatorDescriptionsG1: MTMprovided incommunitypharmacy (i.e.,care in face-tofacemeetings orby telephone) aspart of MedicarePart D MTMprogramG2: MTMprovided bypharmaciststaffedcallcenters as partof Medicare PartD MTM programG3: Educationalmailings (i.e.,mailed lettercontainingpatient-specificmedicationrelatedinformation,personalmedicationrecord, and tipsto save moneyon prescriptions)Eligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort NA Yes NA NA Yes Yes No (Notaccounted for ornot identified)Risk of BiasHighE-58


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Witry,Doucette, andGainer, 2011 58Wittayanukornet al., 2013 59Interventions/ComparatorDescriptionsG1: PCMprovided bycommunitypharmacists toIowa MedicaidenrolleesG2: PCMprovided topatients withprivateindividual-groupinsuranceG1: Interventiongroup:Pharmacistprovided face-tofaceMTMservices for 30-60 minutes perencounter, notalways includinga follow-up visitG2: Controlgroup: Patientswho did notreceive MTMservices(economicanalyses only)Cohort Unclearor NREligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Yes Unclear or NR NA NA Yes No (Notaccounted for ornot identified)Cohort Yes Yes NA NA Yes Yes Partial (somevariables weretakend in toaccount)Risk of BiasHighMediumE-59


Table E2. Risk of bias domains and ratings: Part 2 (continued)Author, YearTrial NameStudyDesignITT?Interventions/ComparatorDescriptionsYamada, G1: Kaiser-2012 60 PermanenteMTM enrolledpatientsG2: Kaiserpatients enrolledin Medicare partD, but not inMTM programmatched tocontrol on age,gender, regionand DCG riskEligibilitycriteriameasuredconsistentlyusing validand reliablemeasures?IntermediateOutcomes:assessedconsistentlyusing valid andreliablemeasures?Patient-CenteredOutcomes:assessedconsistentlyusing valid andreliablemeasures?UtilizationOutcomes:assessedconsistentlyusing validand reliablemeasures?Potentialoutcomeprespecifiedandreported?Were importantconfounding andmodifyingvariables takeninto account indesign and/oranalysis?Cohort NA Yes NA Yes Yes Yes Partial (somevariables weretaken in toaccount)Risk of BiasAbbreviations: BP = blood pressure; CHF = chronic heart failure; CMR = comprehensive medication review; COPD = chronic obstructive pulmonary disease; DTP