Breakfast-with-the-Chiefs

gobi.k

Breakfast-with-the-Chiefs

Breakfast with the ChiefsRe-focusing from Efficiency to Appropriateness:The Value of Understanding Variation in HealthcareJune 10, 2014


Hospitals and healthsystem funding needto shift focusfrom improvingefficiency toreducing variation inclinical practice


Hospitals and healthsystem funding needto shift focusfrom improvingefficiency toreducing variation inclinical practice


Re-focusing from Efficiency to Appropriateness:The Value of Understanding Variation in Healthcare Ontario hospitals have focused on becoming the mostefficient hospitals in Canada Shortest length of stay in acute care Lowest worked hours per RIW weighted case Need (and demand) for health care will increase fasterthan our ability (or willingness) to pay for it Ongoing search for more cost-effective ways to respond to needs Diminishing returns from efficiency, and greaterawareness of evidence-based health care, will forceattention to appropriateness E.g. Choosing Wisely Canada To both: Reduce costs Improve quality© 2012 Hay Group. All rights reserved4


Ontario hospitalsare running out ofopportunities tofurther improveefficiency toreduce costs


Health Spending per CapitaTotal Health Spending per Capita, by Province, 2013f$8,000$5,531 $5,775 $5,835$6,354 $6,474 $6,514 $6,626$6,633 $6,787$7,132PrivatePublic$4,000$0Que. B.C. Ont. P.E.I. N.B. N.S. Sask. Man. Alta. N.L.Canada $5,988© 2012 Hay Group. All rights reservedNote: f: forecast.Sources: National Health Expenditure Database, CIHI; Statistics Canada.6


Ontario Has Lowest Public HealthSpending per Capita of all the provincesTotal Health Spending per Capita, by Province, 2013f$8,000$5,531 $5,775 $5,835$6,354 $6,474 $6,514 $6,626$6,633 $6,787$7,132PrivatePublic$4,000$0Que. B.C. Ont. P.E.I. N.B. N.S. Sask. Man. Alta. N.L.Canada $5,988© 2012 Hay Group. All rights reservedNote: f: forecast.Sources: National Health Expenditure Database, CIHI; Statistics Canada.7


Proportion of health spending on hospitalsin Canada has declined in the last decade from 35% to 29%19932013 f11%15%35%10%19%29%11%13%15%11%16%15%Hospitals Physicians Drugs Other Professionals Other Institutions Other ExpendituresNote: f: forecast. Source: National Health Expenditure Database, CIHI.© 2012 Hay Group. All rights reserved8


Ontario has lowest number of hospital bedsper 1,000 population of all the provinces Ontario hasalso beenreducing itshospitalcapacity Now haslowestnumber ofhospitalbeds per1,000populationof all theprovinces2012 Hospital Beds (All Types) per 1,000 Population© 2012 Hay Group. All rights reserved10


Enormous variationin hospital utilizationrates in Canada andacross Ontario; apotential problemand a potentialopportunity


Enormous variationin hospital utilizationrates in Canada andacross Ontario; apotential problemand a potentialopportunity


Variations in Health Care in Canada –Hysterectomy Rates Within Ontario, age/gender standardizedhysterectomy ratesfor residents of theNE LHIN are 3 timesthe rates for Torontoresidents Hysterectomy rate forSW LHIN more thantwice the rate forToronto residentsPer 100,000 Age/GenderStandardized Population© 2012 Hay Group. All rights reserved13


Variations in Health Care in Canada –Cardiac Revascularization Rates Residents of NWLHIN have double therevascularization rateof residents of theWaterloo WellingtonLHIN Residents of SE LHIN(& CW LHIN) haverevascularizationsrates 50% higher thanWW LHINPer 100,000 Age/GenderStandardized Population© 2012 Hay Group. All rights reserved14


Variation in Hospital Care in OntarioCOPDRate ofhospitalization forCOPD is 3.6 timeshigher for NW LHINresidents than forCentral LHINresidentsRate ofhospitalization forCOPD is 2.5 timeshigher for SE LHINresidents than forCentral LHINresidentsHospitalizations per10,000 Age/GenderStandardizedPopulation© 2012 Hay Group. All rights reserved15


Variation in Hospital Care in OntarioHip Replacement Rate ofhospitalization forhip replacement forSouth Westresidents is almostdouble the rate forCentral Westresidents Similar wait times:SW wait time is 220days, CW wait timeis 234 days Hospitalizations per10,000 Age/GenderStandardizedPopulation© 2012 Hay Group. All rights reserved16


Admission Through ED: % of ED ChestPain Patients Admitted to IP Acute Care In highest volumeOntario EDs, % of EDpatients with ChestPain admitted to IPcare ranges from 6%to almost 30% An ED patient withchest pain is 5 timesmore likely to beadmitted to IP care atHHS than at SMGHEDs with > 400 Chest PainVisits/Yr in 2012/13© 2012 Hay Group. All rights reserved17


Admissions Through ED: % of CTAS 1, 2 & 3COPD ED Patients Admitted to IP Acute Care % of ED CTAS 31-3(urgent & emergent )patients with COPDadmitted to IP careranges from 32% toalmost 63% in highestvolume Ontario EDs A COPD patient istwice as likely to beadmitted to IP care atWOHC than at QHCEDs with > 600 COPD Visits/Yr in 2012/13© 2012 Hay Group. All rights reserved18


Admissions Through ED: % of EDPaed. Asthma Patients Admitted to IP Acute Care % of Paediatric EDpatients with Asthmaadmitted to IP careranges from 7% to22% in highestvolume Ontario EDs A paediatric EDpatient with Asthma is3 times more likely tobe admitted to IPcare at William Oslerthan at HaltonHealthcareEDs with > 200 PaediatricAsthma Visits in 2012/13© 2012 Hay Group. All rights reserved19


And the issue willbecome a biggerproblem as populationgrows and ages


Projected % Change in Population byAge/Gender Cohort 2014 to 2029 Projectedincrease inOntariopopulationfrom 2014 to2029 is 18% Growth ofjust over 1%per year But growthis not evenlydistributedacross agegroups69% increase inpopulation 65+120% increasein malepopulation 85+© 2012 Hay Group. All rights reserved21


Average Annual Per Capita Acute CareHospital Cost by Population Age and Gender Increase innumber of peopleover 65 will putgrowing pressureon health system Move fromrelatively lowcost use as‘Near Elderly’ To increasinglyhigher cost useof hospital carewhen age past65 years. Growth in ‘Old –Old even moredramaticincrease indemand andcosts© 2012 Hay Group. All rights reservedNear ElderlyYoung andMedium OldOld - Old22


Other Perspectives on Aging ofPopulationAging of babyboomers in15 yearsmeans:Shift offastestgrowing agecohort fromlow cost tohigh costLoss of mostexperiencedsenior staffSwitch ofbaby boomerfrom sourceof taxationrevenue toconsumers oftax fundedhealth careBaby BoomersNowBaby Boomersin 15 Years© 2012 Hay Group. All rights reserved23


Need new models ofpayment that rewardproviders forimproving quality,and reducingunnecessary care


Variations in Health Care, PatientPreferences & High-Quality Decision Making “Variations in practice should disturb physicians notmerely because they may indicate wasteful practicesbut because of the possibility that such variations donot optimally serve the best interests of patients. The health care system should allow variation inpractice, provided that variation is based on patientclinical differences and preferences rather than onother factors such as payment method, geography, orsystem proclivities.”Harlan M. Krumholz, MD, SM.JAMA. 2013;310(2):151© 2012 Hay Group. All rights reserved25


Reflections on Geographic Variationsin U.S. Health Care “The key take-home message is that we believethat there is enormous scope for improving theefficiency and quality of US health care. TheDartmouth research suggests that improvementsin both cost and quality can be achieved bysupporting new models of payment that rewardproviders for improving quality, managing capacitywisely, and reducing unnecessary care.”Jonathan Skinner and Elliott Fisher (5/12/2010)http://www.dartmouthatlas.org/downloads/press/Skinner_Fisher_DA_05_10.pdf© 2012 Hay Group. All rights reserved26


Can HSFR, HBAMand QBP Fundingprovide incentives tobest address theissue ofappropriateness ofcare?


How Does the Ontario Health CareFunding System Address Variation?MOHLTC HSFR Educational Slides: “Prior to developing HBAM, the Ministry andstakeholders established a set of guiding principles.The principles stated that HBAM was to: Provide an evidence-based distribution of funding to LHINswithin the government’s budget limits Recognize provider characteristics that are accepted to affectthe cost of providing care Maintain patient freedom to choose their health serviceproviders Ensure stability in funding from year to year Facilitate health sector integration Encourage quality improvement in health outcomes Be simple to understand and communicate”© 2012 Hay Group. All rights reserved28


HBAM Service Models HBAM service models compare actual weightedactivity for a community with the “expected”weighted activity MOHLTC HBAM Educational Slides: “What does “Expected” mean in HBAM?• The term “Expected” is used in a statistical sense, referring to an“Average”• “Expected” service is an average level of service based on casemix, age, SES and rurality specific to the patients to which anorganization provides service• In this context, “Expected” does not mean optimal”© 2012 Hay Group. All rights reserved29


2008 MOHLTC Presentation of Strengthof HBAM“The model’s strength comes from its focus on individualhealth care needs and it’s about the entire population ofOntario. We associate the appropriate costs with thoseservices and then we get a good picture about what is goingon with the entire population.”“Some of the highlights of the model, so that you can beginto understand how we do it: We create a profile of everylast Ontarian; all 12.5 million Ontarians have a profile that isbuilt through the information sources we have that we canidentify, where individuals get their care.”“We also have people in this profile who have blanks. Thereare lots of us in Ontario who don’t use the health caresystem. It’s important to understand where the needsare and where they’re not. We also look at that.”© 2012 Hay Group. All rights reserved30


2012/13 HBAM Acute Service ModelResults – ED and Acute CareDespite significant variation in utilization rates, HBAM Service Model Resultsshow service provided in almost all LHINs is within 1% of expected© 2012 Hay Group. All rights reserved31


Rehab Beds per 1,000 PopulationAged 65 and Older Large variationin availabilityof rehab bedsper populationacross LHINs Efficiency inacute careverydependent onaccess topost-acuteservices© 2012 Hay Group. All rights reserved32


HBAM 2012/13 Inpatient RehabilitationService Model Results Most LHINs actualrehab activity waswithin 1% of HBAMexpected Toronto Central providedslightly less activity thanexpected Central East (with one fifththe beds per populationas Toronto Central)provided 6% more activitythan expected© 2012 Hay Group. All rights reserved33


HBAM Service Model – Person ProfileOriginal Description: “A person profile is created for each Ontarioresident” Necessary for a true population-based model Requires mechanism to assign every person to acategory reflecting their need for care “Population Risk Adjustment Grouper” (e.g. JohnsHopkins Aggregates Diagnosis Groups, used by ICES)© 2012 Hay Group. All rights reserved34


HBAM Service Model – Person ProfileRevised Description: “A person profile is created for each Ontario residentthat received care funded by the Ontario HealthInsurance Plan (OHIP) in an Ontario hospital.” No longer looks at entire population, only hospitalpatients Compares intensity of service per patient If admit patients who would not be admitted in other hospitals,can reduce average intensity of service and thus will becharacterized as providing less ‘service’ than expected. More likely to be characterized as under servicing the‘population’© 2012 Hay Group. All rights reserved35


% of ED TIA Patients Admitted to IP AcuteCare – 2012/13 Hospitals > 200 Visits Admission of TIApatients presentingat ED varies from8% to 38% inhighest volumehospitals Other hospitalsadmit 80% of TIAED patients© 2012 Hay Group. All rights reserved36


Also an issue ofinadvertentlyperverseincentives withinQBP funding


HQO COPD Expert Panel – Incentivesfor Inappropriate Utilization “The Expert Panel recognized that in defining COPDpatient groups largely based on utilization and dispositionthere is the potential for perverse incentives to becreated when these groups are assigned “prices” in afunding methodology. The cost of an average admittedCOPD patient is often more than 10 times the cost oftreating and discharging a COPD patient in the ED. Ifprices for the QBP funding system reflect these costs,care must be taken to ensure hospitals are notincentivized to admit greater proportions of patientsfor a higher payment. The QBP funding system can potentially mitigate theserisks through bundling payment across the ED andinpatient settings, and setting policies aroundappropriateness.”© 2012 Hay Group. All rights reserved38


Variation in Admissions via ED forSelected QBPs What drives variation in propensity to admit EDpatients? Severity of illness, comorbidities, availability of ambulatoryand community resources, home support, physicianpractice, ...or funding system incentives?© 2012 Hay Group. All rights reserved39


Need for HBAMand QBP fundingto more effectivelyaddress issues ofvariation in clinicalpractice


Getting funding models toaddress variation and incent best practicesNeed to return to first principlesA true population-based service funding methodology:would be based on population need andwould incent appropriate system responses to needA true quality based procedure funding methodology:would be based on best practices related to both:• Admission decisions and• processes of carewould be based on best practices for the full episode of carewould incent adoption of these evidence-based best practicesIncentives in funding methodologies should be:aligned with best practice evidencetransparent and easily understoodMasking (or ignoring) variation in populations’ use of health services is:keeping us from addressing significant, potentially inappropriate variations in clinical practicesLeaves us focusing on cost reduction through increasingly elusive improvements in efficiencyIt is time to review the current Ontario approaches to health system fundingreform with consideration of a return to and/or a more comprehensive adoption of:the original, population need-based concept of the HBAM service modelthe original evidence based best practice model of QBP funding model© 2012 Hay Group. All rights reserved41


Questions?Comments

Similar magazines