2011 JAMA - Tseng - Herpes Zoster Vaccine in Older Adults and the Risk of Subsequent Herpes Zoster Disease

ORIGINAL CONTRIBUTION

Herpes Zoster Vaccine in Older AdultsandtheRiskofSubsequentHerpesZosterDiseaseHung Fu Tseng, PhD, MPHNing Smith, PhDRafael Harpaz, MD, MPHStephanie R. Bialek, MD, MPHLina S. Sy, MPHSteven J. Jacobsen, MD, PhD

HERPES ZOSTER, COMMONLYknown as shingles, is a pain-ful vesicular rash caused byreactivation of varicella zos-ter virus, persisting latently in dorsalroot ganglia.1,2 The pain of herpes zos-ter is often disabling and can last formonths or even years, a complicationtermed postherpetic neuralgia. Approxi-mately 1million episodes of herpes zos-ter occur in the United States annually,but aside from age and immunosup-pression, risk factors for this condi-tion are not known.3,4

The Shingles Prevention Study (SPS)is a clinical trial of a live-attenuated vac-cine prepared from the Oka/Merckstrain of varicella zoster virus.5 The SPStrial included 38 546 participants aged60 years or older who had no historyof herpes zoster, immunosuppres-sion, or conditions that could inter-fere with participation. In SPS, herpeszoster vaccine reduced herpes zosterand postherpetic neuralgia incidence by51% (P .001) and 67% (P .001), re-spectively; laboratorymarkers indicat-ing protection were not identified.6

Consequently, Zostavax (Merck&CoInc, Whitehouse Station, New Jersey)was licensed by the US Food and DrugAdministration in 2006 and subse-quently recommended by the Advi-

soryCommittee for ImmunizationPrac-tices for healthy individuals aged 60years or older.7

Although the SPS provided evi-dence that herpes zoster vaccine worksunder idealized conditions (ie, vac-cine efficacy), confirmation of these re-sults is needed in field conditions (ie,vaccine effectiveness) to showwhetherbenefits of the vaccine can be general-ized to conditions of clinical prac-tice.8-10 This is particularly importantfor herpes zoster vaccine, given themedical and physiological diversity in

the elderly population for whom thevaccine is indicated1 and since there arestringent storage and handling require-ments for this live-attenuated vac-

See also Patient Page.

AuthorAffiliations:Department of Research and Evalu-ation, Southern California Kaiser Permanente, Pasa-dena (Drs Tseng, Smith, and Jacobsen andMs Sy); andDivision of Viral Diseases, National Center for Immu-nization and Respiratory Diseases, Centers for Dis-easeControl andPrevention, Atlanta,Georgia (DrsHar-paz and Bialek).Corresponding Author: Hung Fu Tseng, PhD, MPH,Department of Research and Evaluation, KaiserPermanente, Southern CaliforniaMedical Group, 100S Los Robles Ave, Second Floor, Pasadena, CA 91101([email protected]).

Context Approximately 1million episodes of herpes zoster occur annually in theUnitedStates. Although prelicensure data provided evidence that herpes zoster vaccine worksin a select study population under idealized circumstances, the vaccine needs to beevaluated in field conditions.

Objective To evaluate risk of herpes zoster after receipt of herpes zoster vaccine amongindividuals in general practice settings.

Design, Setting, and Participants A retrospective cohort study from January 1,2007, through December 31, 2009, of individuals enrolled in the Kaiser PermanenteSouthern California health plan. Participants were immunocompetent community-dwelling adults aged 60 years or older. The 75 761 members in the vaccinated cohortwere age matched (1:3) to 227 283 unvaccinated members.

Main Outcome Measure Incidence of herpes zoster.

Results Herpes zoster vaccine recipients were more likely to be white, women, withmore outpatient visits, and fewer chronic diseases. The number of herpes zoster casesamong vaccinated individuals was 828 in 130 415 person-years (6.4 per 1000 person-years; 95% confidence interval [CI], 5.9-6.8), and for unvaccinated individuals it was4606 in 355 659 person-years (13.0 per 1000 person-years; 95% CI, 12.6-13.3). Inadjusted analysis, vaccination was associated with a reduced risk of herpes zoster (haz-ard ratio [HR], 0.45; 95% CI, 0.42-0.48); this reduction occurred in all age strata andamong individuals with chronic diseases. Risk of herpes zoster differed by vaccinationstatus to a greater magnitude than the risk of unrelated acute medical conditions, sug-gesting results for herpes zoster were not due to bias. Ophthalmic herpes zoster (HR,0.37; 95%CI, 0.23-0.61) and hospitalizations coded as herpes zoster (HR, 0.35; 95%CI, 0.24-0.51) were less likely among vaccine recipients.

Conclusions Among immunocompetent community-dwelling adults aged 60 yearsor older, receipt of the herpes zoster vaccine was associated with a lower incidence ofherpes zoster. The risk was reduced among all age strata and among individuals withchronic diseases.JAMA. 2011;305(2):160-166 www.jama.com

160 JAMA, January 12, 2011Vol 305, No. 2 (Reprinted) 2011 American Medical Association. All rights reserved.

at CDC-Information Center on January 14, 2011jama.ama-assn.orgDownloaded from

cine.7 Moreover, a large observationalstudy also allows for exploration of vac-cine benefits among important patientsubgroups that would be unfeasible inmost randomized controlled trials.Wereport the results of a large observa-tional study among community-dwelling adults aged 60 years or olderevaluating the risk of herpes zoster af-ter herpes zoster vaccination in aman-aged care organization.

METHODSSetting

The studywas conducted amongmem-bers of Kaiser Permanente, SouthernCalifornia (KPSC), and the protocolwasreviewed and approved by theKPSC in-stitutional review board, which waivedrequirement for informedconsent.KPSCis an integrated health care system thatprovides comprehensive prepaid healthservices for its 3.2 million members.Members are socioeconomically di-verse and more than 99% are commu-nity dwelling. Patient data regarding de-mographics, services, and diagnoses aretracked in the KPSC electronic healthrecord databases from the outpatient,emergencydepartment, andhospital set-tings. The databases provide a primarydiagnosis for hospitalizations; second-ary hospital diagnoses, as well as diag-noses from outpatient and emergencydepartment visits, are not rank or-dered. Data regarding care from non-KPSC providers are likely to be cap-tured in KPSC databases becausedocumentation is required for reim-bursement of such services. Vaccina-tions received by members are trackedin the Kaiser Immunization TrackingSystem,11 regardless of whether admin-istered in or outside of KPSC. Herpeszoster vaccine isprovided toKPSCmem-bers at little or no charge.

Participants

This retrospective cohort study in-cluded data from January 1, 2007,through December 31, 2009. The vac-cinated cohort consisted of KPSCmem-bers who received herpes zoster vac-cine at age 60 years or older, fromJanuary 1, 2007, through June 30, 2009;

the date of vaccination was termed theindex date. The unvaccinated cohortconsisted of randomly sampled mem-bers who were matched 3:1 to the vac-cinated cohort based on birth date (1year). Unvaccinated individuals wereassigned the same index dates as thematching vaccinated persons. Partici-pants in both cohorts were limited toindividuals with continuous member-ship and drug benefits for the 1 year be-fore their index date, and were ob-served for first occurrence of herpeszoster, termination of KPSC member-ship, or study completion. Individualswith herpes zoster codes during theprior 6 months were excluded fromanalysis.Herpes zoster vaccine is not recom-

mended for immunocompromised pa-tients.7 The study therefore excludedimmunocompromised patients so thatthe vaccinated and unvaccinated co-horts would be comparable in terms ofunderlying herpes zoster risk and sothat the reduced risk associated withherpes zoster vaccine could be inter-preted in the context of national rec-ommendations and the SPS.5,7 The studydefined immunocompromised indi-viduals as those with human immuno-deficiency virus, leukemia, or lym-phoma diagnoses within 1 year beforethe index date until the end of follow-up, or having immunosuppressingagents dispensed within 1 year beforethe index date (eBox, available at http://www.jama.com).

Outcomes, Subgroups, andCovariates

Incident herpes zoster and ophthal-mic herpes zoster were defined by In-ternational Classification of Diseases,Ninth Revision (ICD-9) codes (053.XXand 053.2X, respectively; in any posi-tion) from hospital, outpatient, andemergency department settings dur-ing the study period.We categorized in-dividuals to assess factors associatedwith vaccination and risk of herpes zos-ter including sex, health care utiliza-tion, and comorbid chronic diseases.Race (self-reported) was also includedin the analysis because herpes zoster

vaccine coverage and risk of herpes zos-ter both likely differ by race.7,12 Healthcare utilizationwas defined as the num-ber of hospitalizations or outpatient oremergency department visits within 1year before the index date, and chronicdiseases were defined as 1 or more di-agnoses for diabetes or for heart, lung,kidney, or liver disease within 1 yearbefore the index date. Patients could beassigned multiple conditions.

Statistical Analysis

Incidencewas calculated bydividing thenumber of herpes zoster cases by thetotal number of person-years. The 95%confidence intervals (CIs) were esti-mated assuming occurrence of herpeszoster follows a Poisson distribution.Hazard ratios (HRs) and 95%CIs wereestimated for age, sex, race, and chroniccomorbid diseases using Cox propor-tional hazards regressionmodels. Thesemodels were also used to assess the as-sociation between vaccination and sex,race, health care utilization, and chroniccomorbid diseases.Our sample sizewassufficient to detect a relative risk of her-pes zoster incidence between the vac-cinated and unvaccinated cohorts of0.85 during the study interval, withpower of greater than 80% and type Ierror rate of .05. Significance was setat .05 based on a 2-sided test. We usedSAS Enterprise Guide 4.2 (SAS Insti-tute Inc, Cary, North Carolina) for allanalyses.

Assessment of Bias

Individuals who received herpes zostervaccine in the study population mayhave differed from thosewho did not intheir underlying risk for herpes zosteror in their ability anddesire to access carefor herpes zoster. To assesswhether ourresults might have been distorted bysuchconfounders, rate ratios for13acutesymptomatic conditions in the vacci-nated vs age-matched unvaccinated co-horts were estimated, as we did for her-pes zoster (eTable 1). Because herpeszoster vaccine shouldnot protect againstthese conditions,we reasoned that thesevalues shouldcluster around1.0and thatany deviation from 1.0 would alert us

EFFECTIVENESS OF HERPES ZOSTER VACCINE IN OLDER ADULTS

2011 American Medical Association. All rights reserved. (Reprinted) JAMA, January 12, 2011Vol 305, No. 2 161


to biases thatmight be confounding re-sults regarding herpes zoster. This strat-egy is analogous to approaches used toassess validity of estimates of vaccine ef-fectiveness and other outcomes.13-18

To conduct this analysis, conditionswere selected thatwere likely to be com-parable to herpes zoster in terms of ac-cess, health care seeking, and documen-

tation, butwith no plausible associationwith herpes zoster itself (eg, Bell palsy).We identified conditions thatwere acuteand self-limited, with disturbing symp-toms that would typically trigger effortsto seek care in outpatient primary caresettings.Wesought conditionswithhighincidence among elderly individuals butreasonablybalancedacross sex, race, and

socioeconomic status; they were condi-tions that are easily diagnosed and spe-cifically coded with 1 or few categoriesof ICD-9 codes. The conditionswere se-lected without first estimating their in-cidence in the vaccinated and unvacci-nated cohorts.The incidence of the 13 indicator

conditions was ascertained using therelevant ICD-9 codes (eTable 1) in anydiagnostic position of the outpatientand emergency department files of theKPSC databases, from the index datethrough follow-up.Caseswith codes forindicator conditions during the prior 6months were excluded from analysis.

RESULTSCharacteristics of Participants inVaccinatedandUnvaccinatedCohorts

There were 75 761 vaccinated and227 283 unvaccinated individuals in-cluded in the study. Comparedwith theunvaccinated cohort, individuals in thevaccinated cohort were more likely tobe white, women, and to have had alarger number of outpatient visits, butfewer emergency department visits andhospitalizations in the 12months priorto the index date (TABLE 1). The vac-cinated cohort had a lower prevalenceof chronic diseases.

Herpes Zoster Cases in Vaccinatedand Unvaccinated Cohorts

There were 5434 herpes zoster casesidentified in the study: 4593 (84.5%) inoutpatient settings, 574 (10.6%) inemer-gencydepartments, and267 (4.9%)hos-pitalized. Among the unvaccinated andvaccinated cohorts, follow-up averaged1.56 and1.72 years, yieldingherpes zos-ter incidence of 13.0 (95% CI, 12.6-13.3) and6.4 (95%CI, 5.9-6.8) per 1000person-years, respectively, in univari-ate analysis (TABLE 2).Amongunvaccinatedindividuals,her-

peszoster incidencevariedbyseveral fac-tors. It increasedwith age (80years vs60-64years:HR,1.45;95%CI,1.30-1.63),waslowerinmen(HR,0.75;95%CI,0.70-0.79), andwas also lower in black indi-viduals (HR, 0.69; 95% CI, 0.62-0.76).Itvariedbychronicdisease,beinghigherinindividualswithlungdisease(HR,1.34;

Table 1. Baseline Characteristics of Study Cohorts by Herpes Zoster Vaccination Statusa

Characteristic

No. (%)b

Vaccinated (n = 75761) Unvaccinated (n = 227283)

Age of participants, y 23195 (30.6) 69585 (30.6)

60-64 20166 (26.6) 60498 (26.6)

65-69 15426 (20.4) 46278 (20.4)

70-74 10978 (14.5) 32934 (14.5)

75-79 5966 (7.9) 17988 (7.9)

80 0 0

Mean (SD) 69.6 (6.8) 69.6 (6.8)

SexWomen 42822 (56.5) 118506 (52.1)

Men 32939 (43.5) 108771 (47.9)

RaceWhite 45435 (60.0) 122050 (53.7)

Black 4729 (6.2) 27905 (12.3)

Asian 6599 (8.7) 15496 (6.8)

Native American/multiple 87 (0.1) 362 (0.2)

Other 2125 (2.8) 9176 (4.0)

Unknown 16786 (22.2) 52294 (23.0)

No. of outpatient visitsc0 1970 (2.60) 20199 (8.89)

1-4 16897 (22.3) 54938 (24.2)

5-10 25546 (33.7) 68488 (30.1)

11 31348 (41.4) 83658 (36.8)

Mean (SD) 11.8 (11.3) 10.8 (11.8)

Median (25th-75th percentile) 9 (5-15) 7 (3-14)

No. of emergency visitsc0 61139 (80.7) 177578 (78.1)

1 10064 (13.3) 30057 (13.2)

2 4558 (6.0) 19648 (8.6)

Mean (SD) 0.3 (0.8) 0.4 (1.1)

No. of hospitalizationsc0 68117 (89.9) 199608 (87.8)

1 5684 (7.5) 18435 (8.1)

2 1960 (2.6) 9240 (4.1)

Mean (SD) 0.1 (0.5) 0.2 (0.7)

Chronic conditions/diseasesDiabetes 14720 (19.4) 50212 (22.1)

Kidney 9449 (12.5) 31873 (14.0)

Heart 9266 (12.2) 31238 (13.8)

Lung 1282 (1.7) 4940 (2.2)

Liver 1030 (1.4) 3417 (1.5)aP .001 for all variables except age, which was the matching factor.bValues are presented as No. (%) unless otherwise indicated.cData indicate 1 year prior to index date. Index date is the date of vaccination among vaccine recipients and their matchedunvaccinated group.




95%CI,1.13-1.59)comparedwiththosewithout.Anincrease in incidenceamongindividualswithkidneydisease(HR,1.04;95%CI,0.95-1.13)andheartdisease(HR,1.06;95%CI,0.97-1.16)wasnot signifi-cantly significant.Differences inherpeszoster incidence remainedwhenwe ad-justed forsex, race,chronicdiseases,andhealth care utilization. In addition, theriskofherpeszosterwasassociatedwithnumber of outpatient visits during theyearbefore the indexdates, an indicatorofhealthcareseekingbehavior.Thead-justedHRswere1.94(95%CI,1.62-2.31)for1 to4outpatientvisits, 2.44(95%CI,2.05-2.91)for5to10visits,and3.03(95%CI,2.55-3.61) for11ormoreoutpatientvisits.Therewasnoassociationbetweenhospitalizationsoremergencydepartmentvisits and herpes zoster.The Kaplan-Meier plot showing cu-

mulative risk of herpes zoster in the vac-cinated and unvaccinated cohorts isshown in FIGURE 1. In the fully ad-justed analysis, vaccination was asso-

ciated with reduced risk of herpes zos-ter (HR, 0.45; 95% CI, 0.42-0.48); thereduction in risk did not vary by age atvaccination (TABLE 3), sex, race, orwithpresence of chronic diseases. Herpeszoster vaccine recipients had reduced

risks of ophthalmic herpes zoster(HR, 0.37; 95% CI, 0.23-0.61) andhospitalizations coded as herpeszoster (HR, 0.35; 95% CI, 0.24-0.51).In a secondary analysis, definitionof immunocompetent was further

Figure 1. Kaplan-Meier Estimates of the Cumulative Risk of Herpes Zoster by Herpes ZosterVaccination Status

4

3

2

1

0

No. at riskUnvaccinatedVaccinated

227 28375 761

0.5

208 37473 722

1.0

158 88758 425

1.5

128 92048 658

2.0

77 36730 574

2.5

29 22612 527

3.0Follow-up, y

Her

pes

Zost

er In

cide

nce,

%

Unvaccinated

Vaccinated

Table 2. Comparison of Herpes Zoster Incidence in Study Cohorts by Herpes Zoster Vaccination Status

Characteristic

Vaccinated (n = 75 761) Unvaccinated (n = 227 283)

Rate Ratio(95% CI)

PValue

No. ofParticipants

No. ofCases

Total No. ofPerson-years

Incidence/1000Person-years

(95% CI)No. of

ParticipantsNo. ofCases

Total No. ofPerson-years

Incidence/1000Person-years

(95% CI)

Age of participants, y60-64 23 195 204 38 405 5.3 (4.6-6.1) 69 585 1027 105 700 9.7 (9.1-10.3) 0.55 (0.47-0.64) .001

65-69 20 166 197 34 975 5.6 (4.9-6.5) 60 498 1206 94 835 12.7 (12.0-13.5) 0.44 (0.38-0.52) .001

70-74 15 426 202 27 635 7.3 (6.3-8.4) 46 278 1090 74 532 14.6 (13.8-15.5) 0.50 (0.43-0.58) .001

75-79 10 978 146 19 894 7.3 (6.2-8.6) 32 934 824 54 074 15.2 (14.2-16.3) 0.48 (0.40-0.58) .001

80 5996 79 9506 8.3 (6.6-10.4) 17 988 459 26 518 17.3 (15.8-19.0) 0.48 (0.38-0.61) .001

SexWomen 42 822 518 73 713 7.0 (6.4-7.7) 118 506 2770 186 096 14.9 (14.3-15.5) 0.47 (0.43-0.52) .001

Men 32 939 310 56 703 5.5 (4.9-6.1) 108 771 1836 169 552 10.8 (10.3-11.3) 0.50 (0.45-0.57) .001

RaceaWhite 45 435 597 80 358 7.4 (6.9-8.1) 122 050 2887 193 202 14.9 (14.4-15.5) 0.50 (0.46-0.54) .001

Black 4729 25 7200 3.5 (2.3-5.1) 27 905 452 44 983 10.1 (9.1-11.0) 0.34 (0.23-0.51) .001

Asian 6599 61 11 093 5.5 (4.2-7.1) 15 496 351 24 812 14.2 (12.7-15.7) 0.39 (0.30-0.51) .001

Native American/multiple

87 1 146 6.9 (0.2-38.3) 362 5 585 8.5 (2.8-19.9) 0.79 (0.09-6.78) .84

Other 2125 19 3571 5.3 (3.2-8.3) 9176 198 14 199 13.9 (12.1-16.0) 0.38 (0.24-0.61) .001

Unknown 16 786 125 28 047 4.5 (3.7-5.3) 52 294 713 77 879 9.2 (8.5-9.9) 0.49 (0.40-0.59) .001

Chronic conditions/diseases

Diabetes 14 720 174 25 490 6.8 (5.9-7.9) 50 212 1052 77 523 13.6 (12.8-14.4) 0.50 (0.43-0.59) .001

Kidney 9449 117 16 060 7.3 (6.0-8.7) 31 873 758 48 177 15.7 (14.6-16.9) 0.46 (0.38-0.56) .001

Heart 9266 116 16 143 7.2 (5.9-8.6) 31 283 754 47 323 15.9 (14.8-17.1) 0.45 (0.37-0.55) .001

Lung 1282 20 2205 9.1 (5.5-14.0) 4940 145 6808 21.3 (18.0-25.1) 0.43 (0.27-0.69) .001

Liver 1030 12 1634 7.3 (3.8-12.8) 3417 59 4673 12.6 (9.6-16.3) 0.58 (0.31-1.08) .09

Overall 75 761 828 130 415 6.4 (5.9-6.8) 227 283 4606 355 659 13.0 (12.6-13.3) 0.49 (0.46-0.53) .001

Abbreviation: CI, confidence interval.a Individuals were self-identified.




restricted by excluding individualshaving documentation of at least 1 cor-ticosteroid dispensed within 1 year be-fore the index date, including those

taken bymouth, injection, nasal spray,or inhalation. The HR in this fullyadjusted analysis was 0.46 (95% CI,0.43-0.49).

The incidence of 13 acute sympto-matic indicator conditions was com-pared in the vaccinated and age-matchedunvaccinatedcohorts (FIGURE2and eTable 2). The adjusted rate ra-tios for the 13 conditions ranged from0.76 to 1.38 (mean, 1.05; SD, 0.19),being 1.0 or greater for 7 of the 13 con-ditions. None of the conditions had ad-justed rate ratios as low as 0.45, the rateratio for herpes zoster.

COMMENTOur data complement the results of theoriginal clinical trial of herpes zoster vac-cine,5 indicating that the vaccinewas as-sociated with a reduced risk of herpeszoster in a community setting with itsmixed population and routine clinicalpractices. In addition, our data indicatethat the vaccine was associated with re-duced risks of ophthalmic herpes zos-ter and hospitalizations potentially at-tributable to herpes zoster. The resultsalso suggest that these potential ben-efits extended to individuals of all agesfor whom herpes zoster vaccine is rec-ommended, and to individuals withchronic diseases.Overall, we found that herpes zoster

vaccine was associated with a 55% re-duction in incidence of herpes zoster,

Table 3. Hazard Ratio of Herpes Zoster in Study Cohorts

Subgroups

Hazard Ratio (95% Confidence Interval)

Unadjusted Adjusteda

All 0.49 (0.46-0.53) 0.45 (0.42-0.48)

Age at vaccination, y60-64 0.55 (0.47-0.64) 0.50 (0.43-0.58)

65-69 0.44 (0.38-0.52) 0.40 (0.34-0.47)

70-74 0.50 (0.43-0.58) 0.46 (0.39-0.53)

75-79 0.48 (0.40-0.58) 0.45 (0.38-0.54)

80 0.48 (0.38-0.61) 0.44 (0.35-0.56)

SexMen 0.50 (0.45-0.57) 0.47 (0.42-0.53)

Women 0.47 (0.43-0.52) 0.44 (0.40-0.48)

RacebWhite 0.50 (0.46-0.54) 0.48 (0.44-0.52)

Black 0.34 (0.23-0.51) 0.33 (0.22-0.49)

Asian 0.39 (0.30-0.51) 0.37 (0.29-0.49)

Native American/multiple 0.79 (0.09-6.78) 0.61 (0.05-6.96)

Other 0.38 (0.24-0.61) 0.36 (0.23-0.58)

Unknown 0.49 (0.40-0.59) 0.42 (0.35-0.51)

Chronic conditions/diseasesDiabetes 0.50 (0.43-0.59) 0.49 (0.42-0.57)

Kidney 0.46 (0.38-0.56) 0.45 (0.37-0.54)

Heart 0.45 (0.37-0.55) 0.44 (0.36-0.53)

Lung 0.43 (0.27-0.69) 0.39 (0.25-0.63)

Liver 0.58 (0.31-1.08) 0.56 (0.30-1.05)aAdjusted for age, sex, race, health care utilization, and chronic disease in the model.b Individuals were self-identified.

Figure 2. Comparison of Adjusted Hazard Ratios of 13 Indicator Conditions and Herpes Zoster in Study Cohorts by Herpes Zoster VaccinationStatus

Adjusted Hazard Ratio (95% CI)

Indicator ConditionsAdjusted

Hazard Ratio

Unvaccinated Group(n = 227 283)

No. ofCases

Total No. ofPerson-Years

Vaccinated Group(n = 75 761)

No. ofCases

Total No. ofPerson-Years

Thrombosis 0.80623 359 851171 131 015

Hip fracture 0.76811 359 676237 130 938

Herpes zoster 0.454606 355 659828 130 415

Gout 0.826886 352 5651947 128 919

Cholelithiasis and cholecystitis 0.951114 359 268401 130 731

Wrist fracture 0.971172 359 223471 130 667

Epistaxis 1.082513 357 8891025 130 155

Lipomas 1.142457 357 9251066 130 105

Eyelid disorders 1.167895 352 4543744 127 224

Cataracts 1.1746 609 309 96419 730 108 180

Ingrown nail 1.224716 355 5792133 128 849

Wound of hand or finger 1.231759 358 635806 130 360

13 184 347 2996790 123 826Hemorrhoids 1.38

Renal stone 3349 356 8471213 129 863 0.98

0.40 0.55 0.75 1.00 1.25 1.60

CI indicates confidence interval.




which is consistent with the 51% vac-cineefficacyreportedfromtheSPS.Theseresultswould be consistentwith an ab-solute reduction inherpes zoster riskof1.4% at 30 months of follow-up; alter-natively,1episodeofherpeszosterwouldbe averted for every 71 individuals re-ceivingthevaccine.IncontrasttotheSPS,however, we found that the lower her-pes zoster risk associated with the vac-cine was retained across all age strata(P=.62).5,19Thus,ourresultssupportrec-ommendationstoofferherpeszostervac-cine toeligiblepatientsofall ages includ-ingtheoldestpopulation.Notonlymightthesepatients experience a reduction intheir relative risk of herpes zoster, butfor theoldest group, this could translateinto a very large absolute reduction indisease because they bear the greatestburdenofherpeszosterandpostherpeticneuralgia andarealsoespeciallyvulner-able to these disabling conditions.1,20,21

The efficacy of herpes zoster vaccineat preventing ophthalmic herpes zosterwas not assessed in the SPS. Our find-ing that the vaccine recipients had a re-duced risk of these episodes was there-fore particularly important.Ophthalmicinvolvement is a common manifesta-tion of herpes zoster and it can lead toseriousvision-threateningsequelae.2,4,22-24

Our study provides new informationregarding the riskofherpeszoster inkeyvaccinatedandunvaccinatedsubgroups.Herpeszostervaccinewasassociatedwitha similar reduction across sexes and ra-cialgroups.Althoughweconfirmresultsof other studies showing that black per-sonsareat lowerriskofherpeszosterthanwhitepersons,3,25,26incidenceamongblackindividualsremainsconsiderableandpre-vention in this population is importantbecauseracialdisparities inaccess tocaremayhinderthecomplicatedmanagementofherpeszosterandpostherpeticneural-gia. Our data demonstrating disparitiesin herpes zoster vaccine uptake, consis-tentwithothernationaldata,12,27,28 high-light the need to offer this vaccine to allracial and ethnic groups.We also foundthat individualswithcertainchronicdis-easeswereat increasedunderlyingriskofherpeszoster,but thatherpeszostervac-cine was associated with a reduced risk

in these individuals too. This is reassur-ing informationbecause itwasplausiblethat thesediseasesmighthave interferedwithfunctional immunityandvaccineef-fectiveness.29 Control of pain from her-pes zoster and postherpetic neuralgia iscomplicated in thesepatientsbecauseoftheirunderlyingconditionsandthemedi-cationstheymusttake.Herpeszostervac-cine provides an opportunity to preventherpes zoster and the clinical challengeof managing its symptoms.Observational studies can be subject

to biases because patients who seek thevaccinemaydifferintheirunderlyingriskofdiseaseor in their ability anddesire toaccesscare for thecondition.Suchbiaseshave, for instance, confounded interpre-tationof results fromstudiesof influenzavaccine in elderly individuals becausethose at greatest risk of hospitalizationordeathfrominfluenzamaybeleastlikelyto be offered the vaccine.8-10 The poten-tial for bias in studies of herpes zostervaccine is particularly large because theriskfactorsthatdistinguishimmunocom-petent individuals who develop herpeszoster and who receive herpes zostervaccine are not known.3 The markedheterogeneity inphysiology, fitness, andhealth care seeking that exists in elder-ly individuals targeted for herpes zostervaccine couldpotentiallyhide these im-portantyetunrecognizedriskfactors.Weassessed the association between vacci-nationand13indicatorconditions in thevaccinated and unvaccinated study co-horts, justaswehadforherpeszoster.Theadjusted rate ratios for all 13 conditionswere considerably greater than that forherpes zoster; in fact, 7 adjusted rate ra-tios were significantly greater than 1.0.Apossibleexplanationforthisexcessriskis that individuals receiving herpes zos-ter vaccine are generally more inclinedor better able to obtainmedical care foracute conditions. This interpretation isconsistentwithourobservationthather-pes zoster vaccine recipients had moreoutpatientencounters. If correct,herpeszoster vaccine recipientsmight bemorelikely to seekcare for episodesofherpeszoster as well, making our estimates ofreducedriskassociatedwithherpeszos-ter vaccine conservative.

Ourstudyhad limitations. It includeda fully insured population in 1 region ofthecountry; therefore, theresultsneedtobe generalized carefully. It was not de-signed to assess severity or duration ofsymptomsamongherpes zoster casesorto assess the effectiveness of herpes zos-tervaccineatpreventingpostherpeticneu-ralgia. In the SPS, herpes zoster vaccinewasassociatedwithlesspainanddiscom-fort among individuals in whomherpeszoster developed and reduced the inci-denceofpostherpeticneuralgiaby66.5%.5

Average lengthof follow-up inourstudywasshort and itwasnotdesigned tocap-tureanydeclineinprotectionthatis likelytooccurwithtime.30Misclassificationwaspossiblebecauseweascertainedstudyout-comesandvariablesusingpassivefollow-up through health care encounters andusingelectronichealthrecords.Suchdatahave proven fairly reliable for detectingherpes zoster, although their role in de-tectingophthalmic involvementhasnotbeenadequatelyexplored.4,31-33 Itwould,however, seem that misclassification ofvaccinationorherpeszosterstatusismostlikely toresult inanunderestimateof thereduction in risk associatedwith herpeszostervaccine.Theprevalenceofchronicdiseases andnumber of hospitalizationsprior to the indexdatewerehigher in theunvaccinatedcohort;weadjustedforthesefactors and the magnitude of the differ-encewas small, so they shouldnot affectour conclusions substantially.Regarding chronic disease status, our

results were unchanged when we ex-panded our definition to include indi-viduals with at least 1 relevant diagno-sis at any timeduring follow-up(datanotshown). Although our conclusion thatherpes zoster vaccinewasassociatedwithfewer herpes zostercoded hospitaliza-tions is important, it should be inter-preted cautiously because herpes zosteris often incidental to herpes zostercoded hospitalization rather than itscause.7,34 Furthermore, althoughwecon-trolled for age and excluded immuno-suppressed individuals from our study,the risk of hospitalization due to herpeszoster is so strongly associatedwith thesefactors that residual confounding is pos-sible.4,7,20 A sourceofpotential confound-




ing we could not address in this studywas family history of herpes zoster,which could predispose family mem-bers to herpes zoster, encourage them toseek vaccination, or both.35 Such a biaswould result in an underestimate of re-duced risk associated with the vaccine.In conclusion, we found that indi-

viduals aged 60 years or older who re-ceived herpes zoster vaccine had a re-duced risk of herpes zoster regardless ofage, race, and presence of chronic dis-eases. Herpes zoster vaccine was li-censed recently, whichmeans the dura-bility of its protection needs to beassessed in future studies. Meanwhile,however, this vaccine has the potential

to annually prevent tens of thousands ofcases of herpes zoster and postherpeticneuralgia nationally.7 To date, herpeszoster vaccine uptake has beenpoor dueto weaknesses in the adult vaccine in-frastructure and also due to serious bar-riers to the vaccine among clinicians andpatients.12,36 Solutions to these chal-lenges need to be found so that individu-als seeking to receive herpes zoster vac-cine will be able to reduce their risk ofexperiencing this serious condition.

Author Contributions: Dr Tseng had full access to allof the data in the study and takes responsibility for theintegrity of thedata and theaccuracyof thedata analysis.Study concept and design: Tseng, Smith, Harpaz,Bialek, Jacobsen.Acquisition of data: Tseng, Smith.

Analysis and interpretation of data: Smith, Harpaz,Bialek, Sy, Jacobsen.Drafting of the manuscript: Tseng, Harpaz, Bialek.Critical revision of the manuscript for important in-tellectual content: Smith, Harpaz, Bialek, Sy, Jacobsen.Statistical analysis: Tseng, Smith, Harpaz.Obtained funding: Jacobsen.Administrative, technical, or material support: Tseng,Bialek, Jacobsen.Conflict of Interest Disclosures:All authors have com-pleted and submitted the ICMJE Form for Disclosureof Potential Conflicts of Interest. Drs Tseng, Smith, andJacobsen and Ms Sy report having received researchfunding from Merck for other vaccine studies. Dr Ja-cobsen reports having served as an unpaid consul-tant forMerck Research Laboratories. Drs Harpaz andBialek report no disclosures.Disclaimer: The findings and conclusions in this re-port are those of the authors and do not necessarilyrepresent the views of the Centers for Disease Con-trol and Prevention, USDepartment of Health andHu-man Services.Online-Only Material: An eBox and eTables 1 and 2are available at http://www.jama.com.

REFERENCES

1. Dworkin RH, Johnson RW, Breuer J, et al. Recom-mendations for themanagement of herpes zoster.ClinInfect Dis. 2007;44(suppl 1):S1-S26.2. Gnann JW Jr, Whitley RJ. Clinical practice: herpeszoster. N Engl J Med. 2002;347(5):340-346.3. Thomas SL, Hall AJ. What does epidemiology tellus about risk factors for herpes zoster? Lancet InfectDis. 2004;4(1):26-33.4. YawnBP, Saddier P,Wollan PC, St Sauver JL, KurlandMJ, Sy LSA. A population-based study of the inci-dence and complication rates of herpes zoster beforezoster vaccine introduction. Mayo Clin Proc. 2007;82(11):1341-1349.5. OxmanMN, LevinMJ, Johnson GR, et al; ShinglesPrevention Study Group. A vaccine to prevent her-pes zoster and postherpetic neuralgia in older adults.N Engl J Med. 2005;352(22):2271-2284.6. LevinMJ,OxmanMN, Zhang JH, et al; Veterans Af-fairs Cooperative Studies Program Shingles PreventionStudy Investigators. Varicella-zoster virus-specific im-mune responses in elderly recipients of a herpes zostervaccine. J Infect Dis. 2008;197(6):825-835.7. Harpaz R, Ortega-Sanchez IR, Seward JF; Advi-soryCommittee on Immunization Practices (ACIP)Cen-ters for Disease Control and Prevention (CDC). Pre-vention of herpes zoster: recommendations of theAdvisoryCommittee on Immunization Practices (ACIP).MMWRRecommRep. 2008;57(rr-5):1-30, quiz CE2-CE4.8. Clemens J, Brenner R, Rao M, Tafari N, Lowe C.Evaluating new vaccines for developing countries: ef-ficacy or effectiveness? JAMA. 1996;275(5):390-397.9. Orenstein WA, Bernier RH, Hinman AR. Assess-ing vaccine efficacy in the field: further observations.Epidemiol Rev. 1988;10:212-241.10. Weinberg GA, Szilagyi PG. Vaccine epidemiol-ogy: efficacy, effectiveness, and the translational re-search roadmap. J Infect Dis. 2010;201(11):1607-1610.11. Sy LS, Liu IL, Solano Z, et al. Accuracy of influenzavaccination status in a computer-based immunizationtracking systemofamanagedcareorganization.Vaccine.2010;28(32):5254-5259.12. Schiller JS, Euler GL. Vaccination coverage esti-mates from the National Health Interview Survey:United States, 2008. http://www.cdc.gov/nchs/data/hestat/vaccine_coverage/vaccine_coverage.pdf. Ac-cessed July 28, 2010.13. Broome CV, Facklam RR, Fraser DW. Pneumo-

coccal disease after pneumococcal vaccination: an al-ternative method to estimate the efficacy of pneu-mococcal vaccine. N Engl J Med. 1980;303(10):549-552.14. Eurich DT, Marrie TJ, Johnstone J, Majumdar SR.Mortality reduction with influenza vaccine in pa-tients with pneumonia outside flu season: pleiotro-pic benefits or residual confounding?Am J Respir CritCare Med. 2008;178(5):527-533.15. Jackson LA, Jackson ML, Nelson JC, Neuzil KM,Weiss NS. Evidence of bias in estimates of influenzavaccine effectiveness in seniors. Int J Epidemiol. 2006;35(2):337-344.16. Leung J, Harpaz R, Molinari NM, Jumaan A. Her-pes zoster incidence among insured persons in theUnited States, 1993-2006. Ann InternMed. In Press.17. Nelson JC, JacksonML,Weiss NS, Jackson LA.Newstrategies are needed to improve the accuracy of in-fluenza vaccine effectiveness estimates among seniors.J Clin Epidemiol. 2009;62(7):687-694.18. Tate JE, Curns AT, Cortese MM, et al. Burden ofacute gastroenteritis hospitalizations and emergencydepartment visits in US children that is potentially pre-ventable by rotavirus vaccination: a probe study usingthe now-withdrawn rotashield vaccine. Pediatrics.2009;123(3):744-749.19. Merck&Co. ZOSTAVAX (Zoster vaccine live): im-portant information regarding storage and handling[package insert]. http://www.merck.com/product/usa/pi_circulars/z/zostavax/zostavax_pi.pdf. Ac-cessed June 15, 2010.20. Lin F, Hadler JL. Epidemiology of primary vari-cella and herpes zoster hospitalizations: the pre-varicella vaccine era. J Infect Dis. 2000;181(6):1897-1905.21. Schmader K. Herpes zoster in the elderly: issuesrelated to geriatrics. Clin Infect Dis. 1999;28(4):736-739.22. Liesegang TJ. Herpes zoster ophthalmicus natu-ral history, risk factors, clinical presentation, andmorbidity. Ophthalmology. 2008;115(2)(suppl):S3-S12.23. Severson EA, Baratz KH,HodgeDO, Burke JP. Her-pes zoster ophthalmicus in Olmsted county, Minne-sota: have systemic antivirals made a difference?ArchOphthalmol. 2003;121(3):386-390.24. Shaikh S, Ta CN. Evaluation and management ofherpes zoster ophthalmicus.Am Fam Physician. 2002;66(9):1723-1730.25. Schmader K, George LK, Burchett BM, Pieper CF.

Racial and psychosocial risk factors for herpes zosterin the elderly. J Infect Dis. 1998;178(suppl 1):S67-S70.26. Tseng HF, Smith N, Marcy SM, Sy LS, Chao CR,Jacobsen SJ. Risk factors of herpes zoster among chil-dren immunized with varicella vaccine: results from anested case-control study. Pediatr Infect Dis J. 2010;29(3):205-208.27. Lindley M. Zoster vaccine for adults 60 or older:do they know about it and do theywant it? Paper pre-sented at: 44th National Immunization Conference;April 19-22, 2010; Atlanta, GA.28. Lu PJ, Euler GL, Jumaan AO, Harpaz R. Herpeszoster vaccination among adults aged 60 years or olderin the United States, 2007: uptake of the first new vac-cine to target seniors. Vaccine. 2009;27(6):882-887.29. Okamoto S, Hata A, Sadaoka K, Yamanishi K,MoriY. Comparison of varicella-zoster virus-specific im-munity of patients with diabetes mellitus and healthyindividuals. J Infect Dis. 2009;200(10):1606-1610.30. MerckResearch Laboratories. ZOSTAVAXZostervaccine live (Oka/Merck): presented by Merck Re-search Laboratories at the Vaccines and Releated Bio-logical Products Advisory Committee meeting of theFood and Drug Administration, December 15, 2005.http://www.fda.gov/ohrms/dockets/ac/05/slides/5-4198S2_2.ppt. Accessed July 28, 2010.31. Donahue JG, Choo PW, Manson JE, Platt R. Theincidence of herpes zoster. Arch Intern Med. 1995;155(15):1605-1609.32. Jumaan AO, Yu O, Jackson LA, Bohlke K, GalilK, Seward JF. Incidence of herpes zoster, before andafter varicella-vaccination-associated decreases in theincidence of varicella, 1992-2002. J Infect Dis. 2005;191(12):2002-2007.33. Mullooly JP, Riedlinger K, Chun C,Weinmann S,Houston H. Incidence of herpes zoster, 1997-2002.Epidemiol Infect. 2005;133(2):245-253.34. Jackson LA, Reynolds MA, Harpaz R. Hospital-izations to treat herpes zoster in older adults: causesand validated rates. Clin Infect Dis. 2008;47(6):754-759.35. Hicks LD, Cook-Norris RH,Mendoza N,MadkanV, Arora A, Tyring SK. Family history as a risk factorfor herpes zoster: a case-control study.ArchDermatol.2008;144(5):603-608.36. Hurley LP, Lindley MC, Harpaz R, et al. Barriersto the use of herpes zoster vaccine. Ann Intern Med.2010;152(9):555-560.




Documents

2011 JAMA - Tseng - Herpes Zoster Vaccine in Older Adults and the Risk of Subsequent Herpes Zoster Disease