Side Effects of Vaccination - Anthrax Vaccine Immunization Program

CONTINUING EDUCATION SERIESONCOLOGY/IMMUNOLOGYThis CE is published through an unrestricted educational grant fromBristol-Myers Squibb.Side Effects of Vaccination:A Consumer’s Guide to Cause and EffectJohn D. Grabenstein, RPh, PhD, FASHPIgot vaccinated four weeks ago. Igot sick two weeks ago. Are thetwo events related? The twoevents are related in time, certainly.But do the vaccination and theillness have a cause-and-effect relationship?Investigating relationshipsbetween vaccines and adverse eventsis important,to keep people healthy.Ifa vaccine does cause an adverse reaction,we need to find out as soon aspossible and respond appropriately.But jumping to the conclusionthat two events are causally related isscientifically improper. Jumping to aconclusion can also delay finding thetrue cause of an illness. For example,assuming that egg allergy was linkedto anaphylaxis after measles–mumps–rubella (MMR) vaccinationmay have delayed recognition of therole of gelatin allergy. 1–2Vaccination programs providegreat public benefit, both to individualsand to communities. By severalmeasures, vaccines provide morebenefits than any other medical intervention.3–4 Vaccination programs are*Directorate of Health Care Operations, U.S.Army Medical Command, 5111 Leesburg Pike,Falls Church,VA 22041.The opinions or assertions contained hereinare the private views of the author and arenot to be construed as official or reflecting theviews of the US Department of the Army or theDepartment of Defense.Credit — This lesson is good for 0.2 CE units, with a passing grade of70%.Objectives1. To describe historical examples of adverse events truly and spuriouslyassociated with vaccination.2. To describe a rational basis for deciding whether adverse events arecausally or coincidentally linked to a medication.3. To demonstrate how to apply this rational basis to a variety ofexposure—outcome associations.Key Words — adverse events, causality, immunization, side effects,vaccinescomplicated, intricate social endeavorsthat require vast amounts of energy(eg, time, labor, money) to set inmotion and sustain.Making an ill-founded claimabout vaccine safety is like pullingthe stop cord on a moving train. It iscertainly possible to bring a vaccinationtrain to a complete halt, if needed.Butsuch a drastic measure shouldfollow objective analysis, not unsubstantiatedcriticism. The energyrequired to get the train rolling againleads us to want solid evidence,rather than false alarms.SIDE-EFFECT DETECTIVESAs we investigate the relationshipbetween a vaccine and anadverse event, we collect clues, siftevidence, and consider conflictinglines of argument. What caused theadverse effect? Did the vaccine precipitatethe event, or was it an innocentbystander? To render a verdict,what evidence does the court of scienceneed?A good example of the objective,analytical approach is the promptassessment in 1999 of rotavirus vaccineas a cause of intussusception. 5–6Clinical trials detected intussusceptionin both the vaccine and placebogroups, enough to warrant mentionin the package insert, but not a risksufficient to withhold licensing. Thesentinel surveillance program withinthe Vaccine Adverse Event ReportingSystem (VAERS) identified more casesof intussusception than expected,1308 Volume 35, December 2000

Oncology/Immunology CE SeriesSteps in Assessing Cause-and-Effect RelationshipsStep 0. Gather Information.If considering an individual case, gather objective information about the personand the condition of interest, to use in later steps. Confirm the diagnosis. Confirmexposure to the medication of interest.Step 1. Already Acknowledged?Find out whether recognized experts already concede that the vaccine (medication)causes the effect (see also Table 2).Step 2. Novel Event?Is the condition novel, unique, never reported before? New syndromes are rare.Step 3. Determine Baseline.If not novel, determine how often the symptom or condition or disease affectsthe general population (preferably a corresponding population of the same age).Step 4. Compare Vaccinated to Unvaccinated.Leveling the playing field as much as possible, compare the occurrence of thesymptom or condition among vaccinated people to its occurrence among unvaccinatedpeople.Step 5. Full Analysis.If the baseline rate of occurrence is not known, then use criteria for cause-andeffectrelationship proposed by Robert Koch and further refined by Sir AustinBradford Hill and others 38 (see also Table 3).TABLE 1enough to warrant a public warningand to commission controlled studies.Thosecontrolled studies revealedan attributable risk of one case ofintussusception per 5,000 to 12,000vaccine recipients.An objective approach is not theonly one available. Holding rigidly toa “vaccines can do no harm” philosophyis one approach, but an inappropriateone. Similarly, assuming that avaccine causes harm before the factssupport this conclusion is also inappropriate.Examples of recent, wellpublicized,unsubstantiated claimsappear below.This article focuses specificallyon vaccines, but this approach to reasoningapplies to all medications andmany environmental exposures.Doesfluoxetine lead to suicide? Does sildenafilcause heart attacks? Do cellulartelephones cause brain cancer? Dopower lines cause leukemia? Theseand analogous questions of this sortcan be addressed in a similar manner.So, how should we evaluate concernsabout adverse events? This articlesuggests a five-step methodapplicable to vaccines and othermedications (Table 1).STEP 0: GATHER INFORMATIONBegin with Step 0 if you areinvestigating the health of a particularperson. If not, and you are consideringadverse events in a broader perspective,proceed directly to Step 1.Ultimately, that broader perspectiverelies on the validity and veracity ofthe individual clinical cases that populateit. So, the reliability of the individualcases is key.For Step 0,gather objective informationabout this individual whointerests you. Confirm the diagnosis.Distinguish objective signs or valuesfrom subjective symptoms.Ask aboutpre-existing conditions. Collectappropriate laboratory data. See ifsigns or symptoms appeared beforevaccination. Ask what specialists orsubspecialists had been consulted,and so on.I know several people whoreached erroneous conclusionsabout their health after being toldtheir blood work yielded a positiveantinuclear antibody (ANA) test. AnANA value can be used to assess systemiclupus erythematosus (SLE), anautoimmune disease. These peopleassumed that they had an autoimmunedisease solely on the basis ofone ANA test. It takes less time toreach that conclusion than to learnthe intricacies of the ANA test,including its propensity for false-negativeand false-positive results.Indeed, the main use of ANA testsmay be to exclude SLE as a diagnosis.With the disease, the test is positive95% of the time.Without the disease,the test is positive 50% of the time.In other words, a negative test hasmore meaning than a positive test. 7Similarly, one should confirmexposure to the vaccine by checkingdates of vaccination and lot numbers,for example. I have talked with severalmilitary personnel and their familiesconcerned about the adverseeffects of anthrax vaccine theyreceived during military basic training.The irony here is that anthraxvaccine is not administered duringbasic training, so any adverse eventsin that setting certainly would be misattributedif blamed on anthrax vaccine.After individual exposure andclinical data are collected and considered,proceed to Step 1.STEP 1: ALREADYACKNOWLEDGED?Step 1 applies information thatscience has already accumulated, andfor which there is consensus. Do recognizedexperts already concede thatthe vaccine causes the adverse effectin question?Vaccines have been administeredsince Edward Jenner gave smallpoxvaccination in 1796. With over 200years of experience, medicine hasHospital Pharmacy 1309

Oncology/Immunology CE SeriesTABLE 2Summarized Version of the Vaccine Injury TableVaccine contents Illness, disability, injury, Time period for first symptomor condition coveredor manifestation of onset or of(acute complications refer tosignificant aggravation aftercomplications of conditions listed vaccine administrationin this column)Limit forLimit forCompensation aReporting bTetanus toxoid (eg, Anaphylaxis and anaphylactic shock 4 hours 7 daysDTaP, DTP, DT; Td, TT) Brachial neuritis 2–28 days 28 daysAny acute complication (including death) Within period specifiedWhole-cell pertussis Anaphylaxis and anaphylactic shock 4 hours 7 daysbacteria or specific Encephalopathy (or encephalitis) 72 hours 7 dayspertussis antigen(s) Any acute complication (including death) Within period specified(eg, DTaP, DTP, P, DTP-Hib)Measles, mumps, or Anaphylaxis and anaphylactic shock. 4 hours 7 daysrubella viruses (eg, MMR, Encephalopathy (or encephalitis) 5 to 15 days 7 daysMR, M, R) Any acute complication (including death) Within period specifiedMeasles virus (eg, MMR, MR, M) Thrombocytopenic purpura 7–30 days 30 daysVaccine-strain measles viral infection 6 months 6 monthsin an immunodeficient recipientAny acute complication (including death) Within period specifiedRubella virus (eg, MMR, MR, R) Chronic Arthritis 7-42 days 42 daysAny acute complication (including death) Within period specifiedLive poliovirus (OPV)Paralytic Polio—in a non-immunodeficient recipient 30 days 30 days—in an immunodeficient recipient 6 months 6 months—in a vaccine-associated community case No limit No limitVaccine-strain polio viral infection—in a non-immunodeficient recipient 30 days 6 months—in an immunodeficient recipient 6 months No limit—in a vaccine-associated community case No limit No limitAny acute complication (including death) Within period specifiedInactivated poliovirus (eg, IPV) Anaphylaxis and anaphylactic shock 4 hours 7 daysAny acute complication (including death) Within period specifiedHepatitis B vaccines Anaphylaxis and anaphylactic shock 4 hours 7 daysAny acute complication (including death) Within period specifiedHaemophilus influenzae Early-onset Hib disease 7 days 7 daystype b polysaccharides Any acute complication (including death) Within period specified(unconjugated, PRP vaccines)Haemophilus influenzae No condition specified Not applicabletype b polysaccharideconjugatesVaricella virus No condition specified Not applicableRotavirus No condition specified Not applicableaClaims may also be filed for a condition with onset outside the time intervals or a condition not included in this table. Information on filinga claim may be obtained by calling 800-338-2382 or at www.hrsa.gov/bhpr/vicp.bFrom the Reportable Events Table, 42 USC 300aa-25, which lists conditions reportable by law. Events described in the manufacturers’package insert as contraindications to additional doses of a vaccine must also be reported. Individuals are encouraged to report any clinicallysignificant or unexpected events (even if not certain the vaccine caused the event) for any vaccine, whether or not listed here. Toreport call 800-822-7967 or go to www.fda.gov/cber/vaers/report.htm.Source:Vaccine Injury Compensation Program, version of October 1998, www.hrsa.dhhs.gov/bhpr/vicp/table.htm1310 Volume 35, December 2000

Oncology/Immunology CE SeriesIntricate Criteria for Assessing Cause-and-Effect Relationships1. How strong is the association between the exposure and the outcome?2. What is the quality of the evidence for an association?3. Is there a dose-response relationship?4. Is there consistency among several studies?5. Is there a specific cause for the effect observed?6. Did the cause exist before the effect occurred?7. Is the outcome plausible, given what we know about biology?Source: Adapted from Rothman & Greenland 38 TABLE 3assembled considerable understandingof adverse events to considerafter vaccination.We understand quite well thatinjection of any vaccine can result inswelling, redness, and soreness at theinjection site. 8 We should acknowledgethat people also reportswelling, redness, and soreness afterinjection of isotonic sodium chloride.This suggests that at least some injection-sitesymptoms are related to themechanical process of depositing avolume of fluid within a muscle orsubcutaneous space.More rarely, vaccine recipientscan have a severe allergic reaction tovaccination, such as anaphylaxis oranaphylactic shock. 9 Anaphylaxis canalso be caused by peanuts, penicillin,and other agents, of course. But existingdata, plus the rapid time coursebetween exposure and reaction,leave little doubt that anaphylaxisshortly after vaccination is associatedin a causal way. When anaphylaxishappens within minutes of a vaccinationthere are few alternate explanations.Some other adverse events arepresumed to be caused by a vaccineafter a vaccination. The more seriousof these events are included in theVaccine Injury Table, the list used inthe Vaccine Injury CompensationProgram. 10 Adverse events currentlylisted on the Vaccine Injury Tableappear in Table 2.In some cases, patient advocacygroups may make a point of disavowinga purported link between vaccinationand an adverse event. Forexample, the National Multiple SclerosisSociety and the World HealthOrganization agree that there is noreliable evidence for a link betweenhepatitis B vaccination and multiplesclerosis. 11–13An adverse-event controversyrarely involves a case where vaccineadvocates acknowledge causality.Accepted warnings form the basisfor the indications and contraindicationsof vaccine product inserts. Toconsider the matter further, proceedto Step 2.STEP 2: NOVEL EVENT?Is the adverse event in questionnovel? Has it ever been reportedbefore? Step 3 asks if the symptom ordisease is unique, one-of-a-kind.New adverse events are rare.None of the recent allegations ofharm after vaccination is a novelevent.The allegations regard the linkagebetween the adverse event andthe vaccination, not the novelty ofthe adverse event itself.Anaphylaxis was “new” in 1902when Portier and Richet coined theterm to describe a syndrome theyobserved in dogs given repeatedinjections of a toxin derived from thesea anemone. 9 The death of an Egyptianpharaoh from a wasp sting,recorded in hieroglyphics around2640 BC, is the oldest known reportof anaphylaxis. Anaphylaxis, alsocalled “serum shock,”was increasinglyrecognized in the early 1900s asequine antitoxins and sera becamecommon modes of treating infectiousdiseases. 14“Serum sickness” (Serumkrankheit)was a novel diagnosis when itwas associated with injections of animalsera in 1905 by von Pirquet andSchick. 14 The symptoms that compriseserum sickness (eg, edema,fever, urticaria, maculopapular rash,myalgia) had certainly been recognizedin the 1890s with repeatedinjections, but the convergence ofthese individual symptoms into aconsistent syndrome, even with thefirst injection, became widelyacknowledged in the early 1900s. 2,14Scientists recognized in the1960s that 20% to 30% of the womenwho took the sedative thalidomideduring pregnancy delivered childrenwith limb and organ deformities.Aside from the increased rate of birthdefects, it was the pattern of birthdefects of the limbs (phocomely) thatwas unusual, almost unprecedented.15 When veterans of the PersianGulf War returned home, somereported symptoms and illnesses thatraised questions about the healthconsequences of their service. Afterconsiderable inquiry, illnesses amongGulf War veterans are generallyacknowledged by scientists as notbeing qualitatively unique.No uniquesyndrome or set of syndromes hasbeen found.In other words,these veteransdevelop the same symptomsand diseases as people who did notdeploy to the Persian Gulf. It alsodoes not appear that these veteransdie or develop disease more oftenHospital Pharmacy 1311

Oncology/Immunology CE Seriesthan people who did not serve in thePersian Gulf, 16–19 but that is a questionmore properly considered by themethods described at Step 3 and Step4.If the condition or disease wasunrecognized before people startedbeing immunized with a particularvaccine, then the likelihood of acause-and-effect relationship rises.This situation is very rare. In such arare case, we would consider othercriteria, as discussed in Step 5.The more common situation,however, is that unvaccinated peopleget a given illness too.Which leads usto Step 3.STEP 3: DETERMINE BASELINEThe more common situation willbe that the adverse event in questionis not novel, not unique. In that case,Step 3 asks how often the symptomor the disease affects people unexposedto the medication of interest.For vaccines, Step 3 asks how oftenunvaccinated people contract ordevelop the symptom or disease.To illustrate, we will considerGuillain-Barré syndrome (GBS). GBSis a demyelinating disease that damagesnerves,causing temporary weakness.From 80% to 85% recover fullyfrom GBS.About two-thirds of all GBScases are provoked by an acute infection,at a rate of about once per62,000 people per year. GBS clearlyoccurs in the absence of vaccination.Indeed, the Centers for Disease Controland Prevention (CDC) reportsthat more than 99% of people whocontract GBS were not vaccinatedduring the weeks before vaccination.20If the background rate of GBS isone among 62,000 unvaccinated peopleper year, then we begin byexpecting GBS to occur once per62,000 vaccinated people per yearalso. If the rate was much less commonthan once per 62,000, then wemight wonder whether the vaccineoffered some protection against GBS.Conversely, if the GBS rate amongvaccinated people was much morecommon than once per 62,000, weshould wonder whether vaccinationincreases the risk of contracting GBS.The same logic applies to otheradverse events. Between 6% to 10%of unvaccinated pregnant womenwill spontaneously miscarry theirfetuses. 21 With this rate as a startingpoint, we then expect that 6% to 10%of vaccinated pregnant womenwould miscarry. Similarly, we need toknow how often heart attacks,leukemia, thyroid disease, diabetes, orany other event happens amongunvaccinated people.The ideal situation is to knowhow often the adverse event happensamong people who share everyother risk factor except vaccination.In other words, the comparisongroup should have the same age distribution,gender mix, health status,and other personal characteristics aspeople who get the vaccine. Returningto the GBS example,it might helpto know the specific incidence ratesof GBS among adolescents, 20- to 45-year-old adults, or people 65 andolder,depending on the specific comparisonbeing made. This helps compareapples to apples.If reliable data do not exist, thenscientists and health officials have aresponsibility to determine the baseline.Only from the baseline can elevationsabove baseline be recognized.Whenever the symptom or conditionhappens in people who havenot been vaccinated,then the burdenof proof in our “court”is to show thatthe condition happens more often invaccinated people than unvaccinatedpeople. Step 3 establishes the essentialbaseline that permits the criticalcomparison step, Step 4.STEP 4: COMPAREVACCINATED TOUNVACCINATEDFor Step 4, leveling the playingfield as much as possible, comparethe rates of adverse events amongvaccinated people and unvaccinatedpeople. For example, we would findevidence for an association betweenvaccination and Guillain-Barré syndromeonly if the rate among vaccinatedpeople was substantially higherthan once per 62,000 people peryear.A slightly elevated rate of GBShas been associated with influenzavaccine in some years (eg, 1976,1992, 1993), but not in most years.The risk of GBS was 2 to 8 cases permillion influenza vaccine recipientshigher in those three years than thebackground incidence rate. In otheryears,the GBS rate was essentially thesame among people vaccinated orunvaccinated against influenza. GBSdoes occur in people recently immunizedwith other vaccines. In isolatedindividuals, GBS recurred with revaccination.22 But other vaccines havenot been found to induce a rate ofGBS higher than baseline.Step 4 is the critical step wheremany purported links betweenadverse event and vaccine fail tomeet objective, scientific standards.For example,brain damage (ie,chronicencephalopathy) after whole-cellpertussis vaccination does not substantiallydiffer among vaccinatedand unvaccinated children. 22–25 Thisalleged side effect, which launchedmuch of the current antivaccinemovement,has no substantiated basisin scientific fact.Similarly, sudden infant deathsyndrome (SIDS) occurs no moreoften in the interval shortly afterinfant vaccination than in comparableintervals before or after vaccination.Indeed, the rate of SIDS fell inthe US as infant vaccination rates1312 Volume 35, December 2000

Oncology/Immunology CE Seriesincreased, but without a cause-andeffectrelationship. The fall in SIDScases is more properly attributed tothe recommendations to put infants“back to sleep,” rather than on theirstomachs. 2,24On the other hand, peopleexposed to oral poliovirus vaccine(OPV) are certainly subject to a higherrisk of paralytic poliomyelitis thanpeople not exposed to OPV.The associationis very rare, about once per2.4 million doses, but it is a truecause-and-effect relationship. 22,26Leveling the playing field toassess side effects is easier said thandone. A proper comparison involvestwo groups of similar age, gendermix,state of health,employment,likelinessto report symptoms, and manyother factors. Information about selfselectedgroups of people is substantiallyless reliable and generalizablethan information derived from a randomsample of a large population.Epidemiologists invest whole careersin the pursuit of valid comparisons.An apples-to-apples comparisonrequires rigorous planning and oftenrelies on hard-to-find or expensive-toassembledata sets.Some people used a set of inappropriatecomparisons to concludethat childhood vaccination increasesthe risk of diabetes. The issue startswith some ecological comparisons,analyses that look at broad populationsin which multiple factors canexert effects simultaneously.Epidemiologistsinterpret ecological datawith caution,because this design providesonly weak evidence uponwhich to gain cause-and-effect understanding.These diabetes claims are basedon analyzing nations with differentrates of diabetes and different policiesfor timing of childhood vaccinations.27–28 No adjustments were madefor varying degrees of scrutiny indiagnosing diabetes, varying case definitionsof diabetes,or other biases orconfounding effects that imperil simplecomparisons.Workshops at JohnsHopkins University and the NationalInstitute for Allergy and InfectiousDiseases found no basis for a causeand-effectlink between diabetes andvaccination. 29–30 In addition, the investigatorsof a study upon which theassertions are based recently publishedresults showing there is noassociation between vaccination anddiabetes. 31 Similarly,other researchershave not found an increased risk ofdiabetes associated with vaccination.32–36If there is no increased rate of illnessamong vaccinated people, relativeto a proper comparison withunvaccinated people, there is no evidenceof a cause-and-effect relationship.It cannot and should not beruled out entirely, however. The studiesmay not have been large enoughto rule out rare effects or effects ofsmall magnitude. Scientists mustkeep an open mind, of course, strivingto avoid the words “never” or“always.”And what if an increased rate ofillness in a proper comparison isfound? Is the vaccine doomed ordamned? Not necessarily. The stepsabove are the first steps in epidemiologicanalysis,but there is much moreto consider.Thus, we next apply Step5.STEP 5: FULL ANALYSISStep 5 involves a broader set ofcriteria than those considered above.Criteria for cause-and-effect relationshipsstarted with the infectious-diseasework of Jacob Henle in 1840 andRobert Koch in 1882. 37Henle’s and Koch’s postulatesconsist of three criteria for causation:(1) the parasite occurs in every caseof the disease and under circumstancesthat account for pathologicchanges and clinical course; (2) itoccurs in no other disease as nonpathogenicparasite; and (3) afterbeing fully isolated from the bodyand repeatedly grown in pure culture,it can induce the disease anew. 37Henle and Koch did not considerthe world beyond microbiology, sotheir list was short. Their postulateshave been augmented and refinedover the years, perhaps most notablyby Sir Austin Bradford Hill, leader ofthe team that advanced our understandingof the link between tobaccoand lung cancer and cardiovasculardisease.Hill suggested nine aspects ofan exposure-outcome association tohelp distinguish causal from noncausalassociations: (1) strength ofassociation, (2) consistency of studies,(3) specificity of effect, (4) temporality,(5) biologic gradient, (6)plausibility, (7) coherence, (8) experimentalevidence,and (9) analogy.Themodern variants of these criteria aredescribed by Rothman and Greenland(Table 3). 38The CDC publication,Epidemiologyand Prevention of Vaccine-PreventableDisease discusses the mostreliable and conclusive ways to establishcausal relationships for vaccineadverse events, 39 and they are relativelyfew.Causal links between a vaccineand an adverse event may beestablished if they produce a uniquelaboratory result, a unique clinicalsyndrome, or an epidemiologic studyshows vaccinated persons are morelikely than unvaccinated persons toexperience the adverse event.Beyond the narrow confines ofmedicine, other scientists proposeways of distinguishing true sciencefrom pseudoscience. James Lett suggestssix rules for evidential reasoning:falsifiability, logic, comprehensiveness,honesty, replicability, andsufficiency (Table 4). 40USING THE STEPSWakefield and colleagues raisedconcerns about MMR vaccine leadingto inflammatory bowel disease (IBD,eg, ulcerative colitis, Crohn’s dis-Hospital Pharmacy 1313

Oncology/Immunology CE Seriesease). 41 Wakefield’s theory is basedon 12 people referred to a specialtyservice, even though no vaccineviruses were isolated from thesepatients. Other laboratories usingmore sensitive and specific testsfailed to detect any findings to supportan association. 42–45 Epidemiologicstudies failed to confirm an associationbetween measles virus andIBD. 46–51 A purported associationbetween MMR vaccine and IBD failsat Steps 0 and Step 4: The individualclinical information is unpersuasiveand there is no evidence for anincreased risk, compared with unvaccinatedpeople. 52–54Several alleged associations sharethe common feature that the diseasesare diagnosed at the typical age ofvaccination.Two current naïve examplesof this confusion are MMR vaccinewith autism and hepatitis B vaccinewith multiple sclerosis.Each willbe discussed in turn.Autism is typically diagnosedafter the first birthday, as a child’svocal skills begin to manifest. This isthe time when MMR vaccine isadministered. The other “evidence”offered for an association betweenMMR and autism is Wakefield’s inferencethat bowel problems could leadLett’s Rules for Evidential ReasoningFalsifiability: It must be possible to conceive of evidence that would prove theclaim false.Logic: Any argument offered as evidence in support of a claim must be sound.Comprehensiveness: All of the available evidence must be considered.Honesty: Evidence must be evaluated without self-deception.Replicability: Evidence from experimental results or evidence that might becoincidental must be repeated.Sufficiency: Evidence must be adequate to establish the truth, where(a) the burden of proof rests on the claimant,(b) extraordinary claims demand extraordinary evidence, and(c) evidence based upon authority or testimony is always inadequate for a biologicallyimplausible claim.Source: Adapted from Lett 40 TABLE 4to decreased absorption of nutrients,permitting developmental disorderslike autism. 41 This case against MMRis based on speculation and eventscoinciding in time, no more. 51,55–56Again, this alleged association fails onthe basis of Steps 0 and 4. Moreover,scientifically valid evidence runs contraryto the speculation. 57–59Multiple sclerosis (MS) is typicallydiagnosed in the third and fourthdecade of life, coincidentally a timewhen health care providers are vaccinatedagainst hepatitis B. Isolatedcase reports of MS in adults vaccinatedagainst hepatitis B received widespreadattention in France.At least sixcontrolled studies are currentlyunderway, which will objectivelyaddress Step 4. But the hundreds ofmillions of people around the worldvaccinated against hepatitis B whodid not contract MS already suggestthat an association is at worst veryrare. 11–12 Other evidence shows thatinfluenza vaccination of people whoalready have MS does not affectattack rate or disease progression. 60Can vaccination cause cancer?Soon after simian virus 40 (SV40) wasdiscovered in 1960, SV40 was foundto be a contaminant of monkey kidneycells used to manufacture inactivated(and to a lesser extent oral)poliovirus vaccines. In 1961, the governmentrequired all polio vaccine tobe free of SV40, but more than a millionpeople had already received vaccinescontaining this virus. 61 SV40can cause some cancers in rodents.Recently, researchers found SV40 inpeople with rare cancers (eg,ependymomas, osteosarcomas,mesotheliomas), but many of thesepeople were too young to havereceived poliovirus vaccine containingSV40. So if unvaccinated peoplehave SV40 virus in their tumors, Step4 calls on us to ask if vaccinated peoplewith these rare cancers are morelikely to involve SV40 than unvaccinatedpeople. The evidence showsthat the risk is the same,not elevated.There is no indication that anyincreased risk due to SV40 exists. 62Space does not allow an exhaustiveconsideration of every side effectever blamed on a vaccine.But the discussionabove provides a uniformframework within which to addressthe role of any vaccine (indeed, anymedication) on the incidence of allergicdiseases, arthritis, thyroid disease,or any other medical condition.IMPLICATIONS FORPHARMACISTSAfter the diagnosis of a seriousdisease, it is rational to search for thedisease’s precipitating cause.Vaccinations,however, are memorable,painful, well-documented events thathave become convenient foci forblame. 11,55,63–65Vaccinations are intentional stimuliof the immune system. But thehuman immune system gets unintentionalstimuli all the time. Every tripto a church, school, shopping center,or other place where humans congregateleads to viral and bacterialexchange. We ingest microbes dailyin our food and water. Breaches ofthe skin or mucosa introduce moremicrobes. Families with children in1314 Volume 35, December 2000

Oncology/Immunology CE Seriesdaycare know well that they getmore infections than other people,because their kids bring other kids’germs home with them.Despite humanity’s daily encounterswith microbial antigens,vaccinesstill seem unnatural and frighteningto some people. We can expect additionalalarms over vaccine sideeffects in coming years. How will weknow if it is a true alarm (like intussusception)or a false alarm (like diabetes,IBD,autism,multiplesclerosis)?Good science — with intellectuallyhonest inquiry — is the answer.An oft-quoted book on adverseevents after vaccination is Sir GrahamWilson’s The Hazards of Immunization.66 Wilson concludes his book bysaying: “Vaccines, of one sort oranother, have conferred immensebenefit on mankind but, like aeroplanesand motor-cars, they havetheir dangers. . . .” It is for us, and forthose who come after us, to see thatthe sword which vaccines and antiserahave put into our hands is neverallowed to tarnish through over-confidence,negligence, carelessness, orwant of foresight on our part.”Reliable information about vaccinesafety is available from a varietyof sources, including the CDC atwww.cdc.gov/nip/vacsafe/concerns/default.htm.REFERENCES1. Khakoo GA, Lack G. Recommendationsfor using MMR vaccine in children allergicto eggs. BMJ. 2000;320:929–32.2. Pickering LK,ed.2000 Red Book:Reportof the Committee on Infectious Diseases,25th ed. Elk Grove Village, IL: AmericanAcademy of Pediatrics; 2000.3. Centers for Disease Control and Prevention.Achievements in public health,1900–1999: Impact of vaccines universallyrecommended for children—United States,1990–1998. MMWR. 1999;48:243–8.4. Grabenstein JD. Immunization from theperspective of a millennium. Hosp Pharm.2000;35:619–30.5. Centers for Disease Control and Prevention.Intussusception among recipients ofrotavirus vaccine—United States,1998–1999. MMWR. 1999;48:577–81.6. Advisory Committee on ImmunizationPractices. Withdrawal of rotavirus vaccinerecommendation. MMWR. 1999;48:1007.7. Wallach J. Interpretation of DiagnosticTests. 6th ed. Boston: Little, Brown & Co.;1996.8. Grabenstein JD. ImmunoFacts:Vaccinesand Immunologic Drugs. St. Louis, MO:Facts and Comparisons, Inc.; November2000.9. Grabenstein JD. Anaphylaxis: Epinephrineand emergency responses. HospPharm. 1997;32:1377–8,1382–9.10. Grabenstein JD. Compensation for vaccineinjury: Balancing society’s need andpersonal risk.Hosp Pharm. 1995;30:831–6.11. World Health Organization. Lack of evidencethat hepatitis B vaccine causes multiplesclerosis. Weekly Epidemiol Rec.1997;72:149–52.12. Dorozynski A. Suspension of hepatitis Bvaccination condemned. BMJ. 1998;317:1034.13. National Multiple Sclerosis Society. Nofoundation for hepatitis B alarm. Inside MS.1999;17(1):3.14. Parish HJ. A History of Immunization.Edinburgh: E & S Livingstone, Ltd; 1965.15. Sherman M, Strauss S. Thalidomide: Atwenty-five year perspective. Food DrugCosmetic Law J. 1986;41:458–66.16. Sartin JS. Gulf war illnesses: Causes andcontroversies. Mayo Clin Proc. 2000;75:811–9.17. Riddle JR, Hyams KC, Murphy FM, et al.In the borderland between health and diseasefollowing the Gulf War. Mayo ClinProc. 2000;75:777–9.18. Ismail K, Everitt B, Blatchley N, et al. Isthere a Gulf War syndrome? Lancet.1999;353:179–82.19. Straus SE. Bridging the gulf in war syndromes.Lancet. 1999;353:162–3.20. Centers for Disease Control and Prevention.Guillain-Barré syndrome and influenzavaccine. www.cdc.gov/nip/vacsafe/concerns/GBS/default.htm.Accessed September9, 2000.21. Grabenstein JD.Vaccines and antibodiesin relation to pregnancy and lactation.HospPharm. 1999;34:949–56,959–60.22. Fenichel GM. Assessment: Neurologicrisk of immunization. Neurology. 1999;52:1546–52.23. Advisory Committee on ImmunizationPractices. Pertussis vaccination: Use of acellularpertussis vaccines among infants andyoung children. MMWR. 1997;46(RR-7):1–25.24. Institute of Medicine. Vaccine SafetyForum: Summaries of Two Workshops.Washington, DC: National Academy Press;1997.25. Gangarosa EJ, Galazka AM, Wolfe CR, etal. Impact of anti-vaccine movements onpertussis control: The untold story. Lancet.1998;351:356–61.26. Advisory Committee on ImmunizationPractices. Poliomyelitis prevention in theUnited States: Updated recommendations.MMWR. 2000;49(RR-5):1–22.27. Classen JB. The timing of immunizationaffects the development of diabetes inrodents. Autoimmunity. 1996;24:137–45.28. Classen DC, Classen JB. The timing ofpediatric immunization and the risk ofinsulin-dependent diabetes mellitus. InfectDis Clin Pract. 1997;6:449–54.29. Institute for Vaccine Safety WorkshopPanel. Childhood immunizations and type 1diabetes: Summary of an Institute for VaccineSafety Workshop. Pediatr Infect Dis J.1999;18:217–22.30. Vaccination and its adverse events: Realor perceived. BMJ. 1999;317:159–60.31. Karvonen M, Cepaitis Z, Tuomilehto J.Association between type 1 diabetes andHaemophilus influenzae type b vaccination:Birth cohort study. BMJ. 1999;318:1169–72.32. Blom L, Nystrom L, Dahlquist G. TheSwedish childhood diabetes study:Vaccinationsand infections as risk determinants fordiabetes in childhood. Diabetologia.1991;34:176–81.33. Dahlquist G, Gothefors L. The cumulativeincidence of childhood diabetes mellitusin Sweden unaffected by BCG-vaccination.Diabetologia. 1995;38:87374.34. Heijbel H, Chen RT, Dahlquist G. Cumulativeincidence of childhood-onset IDDMis unaffected by pertussis immunization.Diabetes Care. 1997;20:173–5.35. Hyoty H, Hiltunen M, Reunanen A, et al.Decline of mumps antibodies in Type 1(insulin-dependent) diabetic children and aplateau in the rising incidence of type 1 diabetesafter introduction of the mumpsmeasles-rubellavaccine in Finland. Diabetologia.1993;36:1303–8.36. Parent M, Fritschi L, Siemiatycki J, et al.Bacille Calmette-Guérin vaccination andincidence of IDDM in Montreal, Canada.Diabetes Care. 1997;20:767–72.37. Kelsey JL, Thompson WD, Evans AS.Methods in Observational Epidemiology.New York: Oxford University Press; 1986.38. Rothman KJ, Greenland S. Modern Epidemiology.2nded.Philadelphia:Lippincott-Raven; 1998.39. Atkinson W, Wolfe C, Humiston S, et al.,Hospital Pharmacy 1315

Oncology/Immunology CE Seriesed. Epidemiology and Prevention of Vaccine-PreventableDiseases. 6th ed. Atlanta:Centers for Disease Control and Prevention;2000.40. Lett J. Field guide to critical thinking.Skeptical Inquirer. 1990;(Winter):1–7.41. Wakefield AJ, Murch SH, Anthony A, etal. Ileal-lymphoid-nodular hyperplasia, nonspecificcolitis, and pervasive developmentaldisorder in children. Lancet.1998;351:637–41.42. Liu Y,Van Kruiningen HJ,West AB, et al.Immunocytochemical evidence of Listeria,Escherichia coli and Streptococcus antigensin Crohn's disease. Gastroenterology.1995;108:1396–404.43. Iizuka M, Nakagomi O, Chiba M, et al.Absence of measles virus in Crohn’s disease.Lancet. 1995;345:199.44. Haga Y, Funakoshi O, Kuroe K, et al.Absence of measles viral genomic sequencein intestinal tissues from Crohn’s disease bynested polymerase chain reaction. Gut.1996;38:211–5.45. Afzal MA, Minor PD, Begley J, et al.Absence of measles virus genome in inflammatorybowel disease. Lancet. 1998;351:646–7.46. Nielson LLW, Nielsen NM, Melbye M, etal. Exposure to measles in utero andCrohn's disease: Danish register study. BMJ.1998;316:196–7.47. Hermon-Taylor J, Ford J, Sumar N, et al.Lancet. 1995;345:922–3.48. Jones P,Fine P,Piracha S.Crohn's diseaseand measles. Lancet. 1997;349:473.49. Feeney M, Clegg A, Winwood P, et al. Acase-control study of measles vaccinationand inflammatory bowel disease. Lancet.1997;350:764–6.50. Miller E,Waight P. Measles, measles vaccination,and Crohn’s disease: Secondimmunisation has not affected incidence inEngland [letter]. BMJ. 1998;316:1745.51. Peltola H,Patja A,Leinikki P,et al.No evidencefor measles, mumps, and rubella vaccine-associatedinflammatory bowel diseaseor autism in a 14-year prospectivestudy. Lancet. 1998;351:1327–8.52. Van Damme W, Lynen L, Kegels G, et al.Measles vaccination and inflammatorybowel disease. Lancet. 1997;350:1174–5.53. Miller E, Goldblatt D, Cutts F. Measlesvaccination and inflammatory bowel disease.Lancet. 1998;351:755–6.54. Metcalf J.Is measles infection associatedwith Crohn's disease? The current evidencedoes not prove a causal link [editorial].BMJ. 1998;316:166.55. Nicoll A, Elliman D, Ross E. MMR vaccinationand autism 1998: Déjà vu-pertussisand brain damage 1974 [editorial]? BMJ.1998;316:715–6.56. Chen RT, DeStefano F. Vaccine adverseevents: Causal or coincidental? Lancet.1998;351:611–3.57. Taylor B, Miller E, Farrington CP, et al.Autism and measles, mumps, and rubellavaccine: No epidemiological evidence for acausal association. Lancet. 1999;353:2026–9.58. Gillberg C, Heijbel H. MMR and autism.Autism. 1998;2:423–4.59. Miller D, Wadsworth J, Diamond J, RossE. Measles vaccination and neurologicalevents. Lancet. 1997;349:730–1.60. Miller AE, Morgante LA, Buchwald LY, etal. A multicenter, randomized, double-blind,placebo-controlled trial of influenza immunizationin multiple sclerosis. Neurology.1997;48:312–4.61. Brown F, Lewis AM, eds. Simian virus 40(SV40): A possible human polyomavirus;symposium proceedings. Dev Biol Stand.1998;94:1–406.62. Strickler HD, Rosenberg PS, Devesa SS,et al. Contamination of poliovirus vaccineswith simian virus 40 (1955-1963) and subsequentcancer rates. JAMA. 1998;279:292–5.63. Grabenstein JD. Vaccine side effects:Separating mirage from reality.J Am PharmAssoc. 1999;39:417–9.64. Grabenstein JD. The natural history of avaccine and its disease. Hosp Pharm.1996;31:559,563–8,571–2.65. Grabenstein JD, Wilson JP. Are vaccinessafe? Risk communication applied to vaccines.Hosp Pharm. 1999;34:713–8,721–9.66. Wilson GS. The Hazards of Immunization.London: Athlone Press; 1967.1316 Volume 35, December 2000

CONTINUING EDUCATION SERIESONCOLOGY/IMMUNOLOGYThis CE is published through an unrestrictededucational grant from Bristol-Myers Squibb.Continuing Education QuizTo take this CE quiz online, go to www.drugfacts.com1. Which of the following is anexample of a vaccine-adverseevent association later foundto be unrelated to vaccination?A. anthrax vaccine and injectionsitesymptomsB. diphtheria-tetanus-pertussisvaccine and sudden infantdeath syndromeC. influenza vaccine and eggallergyD. tetanus toxoid and anaphylaxis2. An example of an adverseevent that led to the cessationof a national vaccination programwas:A. rotavirus vaccine and intussusceptionB. influenza vaccine and brachialneuritisC. tetanus toxoid and Guillain-Barré syndromeD. measles vaccine and chronicarthritis3. An objective rationale forassessing causal linksbetween vaccines and adverseevents can also be applied toother medications, as well asenvironmental exposures.A. TrueB. False4. The first signal of an associationbetween rotavirus vaccineand intussusceptioncame from:A. the VICP programB. a toll-free hotlineC. the VAERS programD. clinical trials5. Step 0 in the scheme proposedin this article involves:A. verifying an individual’s exposureto the vaccineB. verifying that an adverse eventactually occurredC. both A and BD. neither A nor B6. Which of the following is anadverse event acknowledgedto be caused by vaccination?A. Systemic lupus erythematosusafter diphtheria vaccinationB. Heart attacks after hepatitis AvaccinationC. Serum sickness after yellowfevervaccinationD. Anaphylaxis after any vaccination7. Name a relatively novel medication-adverseevent combination:A. diphtheria antitoxin and anaphylaxisin 1902B. influenza vaccine and Guillain-Barrésyndrome in 1976C. varicella vaccine andheadache in 1995D. hepatitis B vaccine and multiplesclerosis in 19988. If an adverse event after vaccinationis not novel, it meansthat the event also occursamong unvaccinated people.A. TrueB. False9. If an adverse event occursamong unvaccinated people,the key question in Step 3 is:A. Does it really occur amongvaccinated people?B. Does it really occur amongunvaccinated people?C. How often does it occuramong vaccinated people?D. How often does it occuramong unvaccinated people?10. Roughly what fraction ofcases of Guillain-Barré syndromeoccur in the absenceof vaccination?A. from 6% to 10%B. 50%C. Two-thirdsM. More than 99%11. If 6% to 10% of unvaccinatedwomen who are pregnantspontaneously miscarry theirfetuses, what proportion ofvaccinated women areexpected to do so?A. 3% to 5%B. 6% to 10%C. 12% to 20%D. 24% to 40%12. In some years, but not others,people vaccinated againstinfluenza have a substantiallyhigher rate of Guillain-Barrésyndrome than unvaccinatedpeople.A. TrueB. False13. Infants vaccinated againstdiphtheria, tetanus, and pertussishave a substantiallyhigher rate of chronicencephalopathy than unvaccinatedinfants.A. TrueB. False14. Which of the following factorsweakens the hypothesisthat vaccination leads to diabetes?A. reliance on ecological comparisonsB. failure to adjust for diagnosticscrutinyC. Failure to use common casedefinitionsD. all of the above15. Which of the following alternativeexplanations couldexplain a temporal relationshipbetween hepatitis B vaccineand multiple sclerosis?Hospital Pharmacy 1317

Oncology/Immunology CE SeriesA. hepatitis A vaccine is knownto cause multiple sclerosisB. multiple sclerosis is known tofollow hepatitis B viral infectionC. adults vaccinated againsthepatitis B virus because ofoccupational risks are often ofthe same age as people newlydiagnosed with multiple sclerosisD. not enough reports of multiplesclerosis in hepatitis B vaccinerecipient have beenmailed in16. The Vaccine Injury Tableincludes which of the followingevents after tetanus toxoidadministration?A. inflammatory bowel diseaseB. Guillain-Barré syndromeC. brachial neuritisD. intussusception17. The Vaccine Injury Tableincludes which of the followingevents after rubella vaccination?A. chronic arthritisB. acute arthritisC. thrombocytopenic purpuraD. paralytic poliomyelitis18. Reports to the VaccineAdverse Event Reporting System(VAERS) are required foradverse events listed in theVaccine Injury Table.A. TrueB. False19. Which of the following is oneof the intricate criteria forassessing cause-and-effectrelationships?A. Any biological hypothesis issufficientB. Effect must occur beforecauseC. Strength of associationbetween exposure and outcomeD. Any one of several studiesshows a statistically significantfinding20. Reports to the VaccineAdverse Event Reporting System(VAERS) are required foradverse events listed as contraindicationsto additionaldoses of a vaccine.A. TrueB. FalseOncology/Immunology CE SeriesSide Effects of VaccinationACPE #071-999-00-031-H01 (0.2 CEU)Program Expires: December 2002To receive continuing education credit, complete thisform and mail with your $7 processing fee (made payable toWSU College of Pharmacy) to:Continuing Education DepartmentCollege of PharmacyWashington State University at Spokane601 West First AvenueSpokane, WA 99201-3899Tel: (509) 358-7660Print clearly or type. Please allow 4 weeks for processing.Name: ____________________________________________Address:___________________________________________City: __________________ State: ________ Zip:_______Social Security #: ___________________________________Note: Your answer sheet will be graded confidentially andyou will receive prompt notification of your score. In orderto receive continuing education credit for this program, youneed a minimum correct response rate of 70%.PROGRAM EVALUATIONPlease rate our continuing education offering by respondingto the following questions:1. Were the educational objectives met?■ completely ■ fairly well ■ not at all2. The overall quality of the program was:■ excellent ■ good ■ fair ■ poor3. Was the content of this article relevant to the practiceof pharmacy?■ excellent ■ good ■ fair ■ poor4. How long did it take you to complete this continuingeducation program? _______ hours5. What other continuing education programs or topicswould you like to see?________________________________________________1. A B C D E2. A B C D E3. A B C D E4. A B C D E5. A B C D EAnswer Sheet6. A B C D E 11. A B C D E7. A B C D E 12. A B C D E8. A B C D E 13. A B C D E9. A B C D E 14. A B C D E10. A B C D E 15. A B C D E16. A B. C D E17. A B C D E18. A B C D E19. A B C D E20. A B C D EThe Washington State UniversityCollege of Pharmacyis approved by theAmerican Council onPharmaceutical Education(ACPE) as a provider ofcontinuing pharmaceuticaleducation.1318 Volume 35, December 2000