EL

SBCa

Bb

c

d

a

ARRAA

KEVHLR

1

fo

Nf

h0

Vaccine 32 (2014) 4089–4096

Contents lists available at ScienceDirect

Vaccine

j our na l ho me page: www.elsev ier .com/ locate /vacc ine

vidence-to-policy gap on hepatitis A vaccine adoption in 6 countries:iterature vs. policymakers’ beliefs

achiko Ozawaa,∗, Lois A. Privor-Dumma, Angeline Nannib, Emily Durdenc,rett A. Maiesec, Chizoba U. Nwankwod, Kimberly G. Brodoviczd,amilo J. Acostad, Kathleen A. Foleyc

International Vaccine Access Center, Johns Hopkins Bloomberg School of Public Health, Department of International Health, 615 North Wolfe Street,altimore, MD 21205, USAAeras Foundation, 1405 Research Boulevard, Rockville, MD 20850, USATruven Health Analytics, 150 Cambridge Park Drive, Cambridge, MA 02140, USAMerck & Co, Inc., 770 Sumneytown Pike, West Point, PA 19486, USA

r t i c l e i n f o

rticle history:eceived 14 June 2013eceived in revised form 23 April 2014ccepted 1 May 2014vailable online 15 May 2014

eywords:vidence-to-policyaccineepatitis Aiterature reviewesearch methods

a b s t r a c t

Background: National vaccine adoption decisions may be better understood by linking multiple datasources. When examining countries’ decisions to adopt the hepatitis A vaccine, applying multiple researchmethods can facilitate assessments of gaps between evidence and policy. We conducted a literaturereview on hepatitis A and stakeholder interviews about decisions to adopt the vaccine in six countries(Chile, India, South Korea, Mexico, Russia, and Taiwan).Methods: A systematic literature review was conducted across five literature databases. The review identi-fied and abstracted 340 articles, supplemented by internet search. In addition, we interviewed 62 expertsand opinion leaders on hepatitis A and/or vaccines. Data from the two sources were analyzed to identifygaps around epidemiologic data, economic data, and barriers/facilitators of hepatitis A vaccine adoption.Results: Epidemiologic data gaps were found in Chile and Russia, where stakeholders believed data to bemore solid than the literature documented. Economic data on hepatitis A was found to be weak across allcountries despite stakeholders’ agreement on its importance. Barriers and facilitators of vaccine adoptionsuch as political will, prioritization among vaccines, and global or local recommendations were discussedmore by stakeholders than the literature. Stakeholders in India and Mexico were not concerned with thelack of data, despite growing recognition in the literature of the epidemiological transition and threat ofoutbreaks.Conclusions: Triangulation of results from two methods captured a richer story behind vaccine adoption

decisions for hepatitis A. The discrepancy between policymakers’ beliefs and existing data suggest adecline in priority of hepatitis A or weak investment in data collection. Filling the confirmed data gaps inseroprevalence or economic data is important to help guide policy decisions. Greater communication ofthe risk of hepatitis A and the benefits of the vaccine may help countries undergoing the epidemiologictransition.

© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

. Introduction

Using one research methodology is often not enough to tell aull story especially for national vaccine adoption decisions, whichften require diverse viewpoints to understand the complete

∗ Corresponding author at: Johns Hopkins Bloomberg School of Public Health, 615orth Wolfe Street, Suite E8003, Baltimore, MD 21205, USA. Tel.: +1 410 955 3928;

ax: +1 410 614 1419.E-mail address: sozawa@jhsph.edu (S. Ozawa).

ttp://dx.doi.org/10.1016/j.vaccine.2014.05.026264-410X/© 2014 The Authors. Published by Elsevier Ltd. This is an open access article un

license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

picture. Applying multiple research methods and triangulatingresults may capture elements of the story that might be over-looked by one method or the other. In this paper, we apply tworesearch methods in examining decisions to adopt a new vaccinewhere notable gaps may exist between evidence and policy. Thesegaps may be particularly important for considerations to add thehepatitis A vaccine into national immunization schedules given

the unique characteristics of the epidemiological transition thatmoves countries from high to low endemicity of hepatitis A.

Hepatitis A is an acute liver disease caused by the hepatitis Avirus, which is preventable by available safe and highly efficacious

der the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

4090 S. Ozawa et al. / Vaccine 32 (2014) 4089–4096

Table 1Background characteristics of six countries.

Key indicators Chile India South Korea Mexico Russia Taiwan

Health Endemicity of Hep A, WHO Classificationb Intermediate High Low Intermediate Low Low% DTP3 official coverage [2010]c 92% 72% 94% 95% 97% 96%Total population, in thousands [2010]d 17,114 1,224,614 48,184 113,423 142,958 23,162Birth cohort, in thousands [2010]d 245 27,165 478 2217 1682 167<5 mortality rate [2010]d 9 63 5 17 12 6

Economic GAVI eligibilitye No Yes No No No NoWorld Bank Income Groupf Upper middle Lower middle High Upper middle Upper middle HighGNI per capita, atlas method, US$ [2010]g $9940 $1340 $19,890 $9330 $9910 $18,573Health exp as % of total govt exp [2009]a,h 16% 3.7% 12.3% 11.9% 8.50% 6.5

a Health expenditure as a percentage of total government expenditure.b World Health Organization. WHO Position Paper on Hepatitis A Vaccines. Weekly Epidemiologic Record. 2012. 87: 261–276. http://www.who.int/wer/

2012/wer8728 29/en/index.htmlc World Health Organization. WHO vaccine-preventable disease monitoring system, 2012 global immunization profile (WHO/UNICEF estimates). Accessed May 1, 2014 at

http://www.who.int/immunization monitoring/data/data subject/en/index.htmld UNICEF. The State of the World’s Children. 2012. Accessed May 1, 2014 at: http://www.unicef.org/sowc/e GAVI Alliance. GAVI Eligible Countries. 2009. Accessed May 1, 2014 at http://www.gavialliance.org/support/who/eligible/index.phpf World Bank. World Bank Country and Lending Groups. Accessed May 1, 2013 at http://data.worldbank.org/about/country-classifications/country-and-lending-groups

at httditure

c

vflivcottrptw

hvmMttsAi(g

2

2

aeoieriidplir

g World Bank. Development Data and Statistics. 2011. Last accessed May 1, 2013h World Health Organization. National Health Accounts: Global Health Expen

ountry/en/index.html

accines [1]. Since hepatitis A virus is transmitted through theecal-oral route, the incidence of hepatitis A vary according toevel of socio-economic development. As countries develop andmprove sanitation and water supply, childhood exposure to theirus decreases and countries begin an epidemiologic transition,haracterized by later age at first infection and increasing incidencef symptomatic hepatitis A. The disease may represent a substan-ial economic burden in countries transitioning from developingo developed economies with intermediate and high incidenceates, where slow recovery and rare serious complications result inroductivity loss, caregiver burden and medical resource utiliza-ion. Despite its high efficacy, the hepatitis A vaccine has not beenidely adopted into national immunization programs to date [2,3].

This study simultaneously carried out a literature review onepatitis A, supplemented by an internet search and policy inter-iews about the adoption process for hepatitis A vaccine in sixiddle- and high-income countries (Chile, India, South Korea,exico, Russia, and Taiwan). The literature review focused on cap-

uring epidemiologic, economic or policy articles on hepatitis A inhese countries, while key informant interviews set out to under-tand local stakeholder perceptions about the evidence on hepatitis

infections and its vaccines. The study focused on countries in var-ous stages of epidemiologic transition and economic developmentsee Table 1 for key health and economic indicators) to assess if anyaps exist between existing evidence and policies.

. Materials and methods

.1. Literature review

We conducted a systematic literature search in October 2011cross five electronic databases: PubMed®, ISI Web of Knowl-dge, EMBASE, Scopus, and EconLit. The search used variationsf two search terms: “hepatitis A” and [one of six countries]. Wencluded articles primarily focused on hepatitis A epidemiology,conomics and/or policy. Epidemiologic articles included thoseeporting seroprevalence, incidence, prevalence, endemicity, clin-cal manifestations or risk factors of hepatitis A. Policy articlesncluded government reports, editorials or reports without primaryata, which were focused on issues related to vaccine adoption,

revention or control efforts for hepatitis A. We excluded articles

ess relevant to this analysis, such as papers focusing on biolog-cal mechanisms of hepatitis A, non-human studies, vaccine trialesults, and case studies. Given that hepatitis A was not high on

p://data.worldbank.org/indicator Database. April, 2011. Last accessed May 1, 2013 at http://www.who.int/nha/

the global agenda prior to 1990, our search was limited to articlespublished since then. For most countries, pre-1990 seroprevalencedata was reported in articles published after 1990 providing his-torical data with trends in seroprevalence over several decades. Insome instances, however, it was necessary to search pre-1990 lit-erature to fill in data gaps on seroprevalence prior to 1990. Articlesin each of the local languages (Chinese, Korean, Russian, Spanish)were included in the search. Reference lists of primary studies andsystematic reviews were also scanned to identify additional studiesmissed by the initial search. Articles were first reviewed for inclu-sion based on title. Abstracts and full articles were reviewed nextto determine study inclusion.

A supplementary internet search was conducted to fill in gapsin country-specific epidemiological data or vaccine policy informa-tion. Direct scan of ministry of health, pediatric society, infectiousdisease society, immunization technical advisory councils, medicaljournal databases or other relevant websites was also conductedfor each country to identify relevant articles or reports, find currentrecommendations or fill specific data gaps.

For articles meeting the inclusion criteria, we abstracted dataon background information (authors, title, year of publication anddata collection, journal, country/region, type of article), as well asstudy design, study subject characteristics, results, policy recom-mendations and perceived barriers and facilitators to hepatitis Avaccine adoption. We summarized results separately for epidemi-ologic and policy-focused articles. Articles in Russian, Spanish, andChinese were abstracted by native language speakers and writersof those languages, with a background in healthcare analysis.

2.2. Policy interviews

In-depth interviews were conducted with experts and opin-ion leaders on hepatitis A in each country, including current andformer government officials, authors of articles on hepatitis A,academics, clinicians, hepatologists, individuals working for inter-national organizations, vaccine manufacturers, and those in civilsociety. Individuals were identified through the literature searchand personal contacts using snowball sampling. The contact list wasreviewed by country experts to identify the most relevant contactsand facilitate interviews in some cases. All interviews were carried

out face-to-face by two interviewers, where one individual tookdetailed notes. Interviews were held in the capital cities, lasted onehour, and not digitally recorded. Questions were asked mostly inEnglish with professional translators used in Taiwan and Russia. In

S. Ozawa et al. / Vaccine 32 (2014) 4089–4096 4091

Fig. 1. Flow chart for study selection.a Other reasons included: methods-focused article, abstract or conference abstract, or unable to obtain full article.b s by ca

CS

tspnaa

uqrafil

2

mtpcs

Some articles reported data on multiple countries. We identified 352 publicationuthor.

hile and Mexico, some respondents explained some answers inpanish in response to questions in English.

An interview guide was developed and pretested where ques-ions focused on perceptions of disease burden and the evidenceupporting hepatitis A vaccination, as well as the decision-makingrocesses for adoption of a hepatitis A vaccine into national immu-ization programs. Interviews also assessed respondent beliefsbout general policymaker agreement with a series of statementsbout hepatitis A severity and its vaccine.

Detailed interview notes were analyzed by line-by-line codingsing ATLAS.ti software. A codebook including a priori researchuestions was developed and applied. We present numbers ofesponses among those who answered specific questions. Resultsre presented in aggregate across respondents to protect the con-dentiality of individuals. Analyses were conducted at the country

evel and by themes across countries.

.3. Combining the literature review and policy interviews

Data from the literature review, internet search and key infor-ant interviews were analyzed together to identify gaps between

he two sources around epidemiological data, economic data andolicies around hepatitis A vaccine adoption. For each topic, weompared what was said or reported in the literature with whattakeholders reported.

ountry. A list of included papers can be obtained by contacting the corresponding

3. Results

The literature and internet search yielded 797 articles. The ini-tial screening removed 343 articles based on titles and abstracts.Another 114 articles were excluded upon reading of full-lengtharticles. This resulted in 340 articles, or 352 by country, as somearticles covered multiple countries (see Fig. 1 for a flow diagram).The majority of included articles were identified through PubMed.India, South Korea and Taiwan (88, 77 and 72 articles) had twiceas many publications as Russia, Chile and Mexico (43, 40 and 32articles). 312 articles discussed the epidemiology of hepatitis A, 36articles were on policy and 4 articles on economic analyses. Whileall the articles on India were in English, many of the articles inthe other countries were in local languages (Russia 83%, Chile 75%,Mexico 63%, South Korea 47% and Taiwan 13%).

For policy interviews, a total of 143 individuals were contactedand 62 interviews were completed in 2012. In Mexico, Russia andChile, current and former government employees represented 67%,50% and 42% of respondents, respectively, compared to 20–30% ofrespondents in other countries. Other respondents included clini-cians (29%), academics (23%), members of civil society (6%), vaccinemanufacturers (2%), and international organization representatives(2%). Among those not interviewed, 72% did not respond to inter-view invitations, 15% were unable to participate due to travel, 11%

stated they were not experts on hepatitis A, and 2% could not beconducted without permission in Russia.

4092 S. Ozawa et al. / Vaccine 32 (2014) 4089–4096

Table 2Epidemiologic data on hepatitis A disease.

Chile India South Korea Mexico Russia Taiwan

Literature Endemicity of Hepatitis A,WHO Classification

Intermediate High Low Intermediate Low Low

Most Recent Year ofSeroprevalence Data

Santiago: 2001;Los Lagos: 1999

Western: 2008;North: 2002;South: 2002;Central: 1998;East: 1998

Seoul: 2010;Central: 2009;Northern:None;Southern:2009

Mexico City:1995; Other:2006

Moscow/StPetersburg:2009; Other:1999

Northern(Taipei): 2007;Central: 2008;Southern:2005; Eastern:2010

Support for theEpidemiological Transition

Santiago: Late1970s; LosLagos:Mid-1980s

Western: Yes;North: No;South: No;Central: No;East: No

Seoul: Late1980–90s;Central:variable data;Northern:N/A;Southern:variable data

Mexico City:unclear;Other: Late1990s

Data arevariable;trends unclear

Northern:Early 1990s;Central:variable data;Southern:variable data;Eastern: fullyimmunized

No. of Epidemiologypublications

33 80 75 25 38 65

Most Recent Year ofIncidence Data

2011 Not identified Not identified 2010 2000 2010

Most Recent Year ofOutbreak Data

2003 2005 2008 None 2002 2007

Perceptions of hepatitis Adisease

Mostdecision-makerswould:

Chile India South Korea Mexico Russia Taiwan

Interviews “More infections are nowoccurring in youngchildren and teenagersthan in infants”

Agree 92% 64% 60% 67% 100% 50%

Disagree 0% 18% 30% 17% 0% 40%Don’t know 8% 18% 10% 8% 0% 10%

“Because seroprevalenceamong children isdeclining, we can be lessconcerned about HepatitisA disease”

Agree 42% 36% 50% 33% 0% 50%

Disagree 50% 64% 40% 58% 80% 40%Don’t know 8% 0% 10% 8% 20% 10%

“With better water,sanitation and nutritionchildren are not gettinginfected early in life,therefore we don’t need tobe concerned aboutHepatitis A vaccines”

Agree 58% 18% 10% 17% 0% 30%

Disagree 42% 82% 90% 75% 100% 60%Don’t know 0% 0% 0% 8% 0% 10%

“The consequences ofHepatitis A infectionacquired later in life arevery severe and should be aconcern for policy makers”

Agree 58% 80% 80% 100% 40% 100%

Disagree 42% 20% 10% 0% 60% 0%Don’t know 0% 0% 10% 0% 0% 0%

“Government needs to beconcerned about the severeillness, and not just deaths,that can be caused byHepatitis A infections”

Agree 92% 89% 70% 100% 100% 90%

Disagree 8% 11% 10% 0% 0% 10%Don’t know 0% 0% 20% 0% 0% 0%

“How important a factor ispatients with fulminanthepatitis A (requiringtransplant or dying) indecision-making foradoption of Hepatitis A

Very important 50% 45% 78% 50% 0% 80%

Moderate 25% 18% 0% 17% 40% 20%Not important 25% 27% 22% 25% 40% 0%Don’t know 0% 9% 0% 8% 20% 0%

3

ih

vaccine?”

.1. Epidemiologic data on hepatitis A disease

Epidemiologic data from the literature were compared withnterviewees’ general perceptions of data availability and risk ofepatitis A disease (Table 2).

There was strong agreement between the literature and inter-

viewees’ perceptions of the ample epidemiologic evidence onhepatitis A in South Korea (75 articles) and Taiwan (65 articles).Many Korean interviewees mentioned epidemiologic data includ-ing disease burden and infection source of hepatitis A. In Taiwan,

S. Ozawa et al. / Vaccine 32 (2014) 4089–4096 4093

Table 3Economic data on hepatitis A vaccine.

Key economic data Chile India South Korea Mexico Russia Taiwan

Interviews “How important areeconomic studies inthe decision-makingprocess for a HepatitisA vaccine in[country]?”

Very important 73% 64% 100% 83% 33% 89%Moderate 18% 36% 0% 17% 33% 11%Not imp./Don’t know 9% 0% 0% 8% 34% 0%

“Were any types ofeconomic studies forHepatitis A done in[country]?”

Yes 90% 0% 89% 17% 60% 10%No 0% 64% 0% 42% 20% 80%Don’t know 10% 36% 11% 42% 20% 10%

“[If Yes above] Are theeconomic studiessufficient to support anadoption of a HepatitisA vaccine?”

Yes 67% N/A 71% N/A 25% N/A

No 33% N/A 29% N/A 75% N/A

Literature Type of economic study Cost-effectiveness

None Economicanalysis

Financialburden

None None

Main conclusions Chile Valenzuela, 2005 From the health system perspective, vaccination would cost $460 US per life-year saved, or$281 US per QALY gained. Savings from healthcare costs would exceed the cost of vaccinationin 7–13 years ($3.9 million US saved within 5- years).

Quezada, 2008 $4,984 per life year gained (base case scenario: 95% vaccine coverage, 100-year time horizon,1% annual decrease in force of infection); Cost savings were reached after 6 years; Break-evenprice for one dose of the vaccine: $48.

SouthKorea

Korea CDC, 2010 Most cost beneficial vaccination strategy: vaccinate 90% of 1-year olds and 50% of those 13–18years; Most effective strategy: vaccinate 50% of those 19–39 years along with 90% of 1-yearolds.

l cost case:(US$7

asvTtobTT

toe

Mexico Rivas-Oropeza, 2009 Annuaminorpesos

number of interviewees expressed confidence in the country’surveillance system: “We have disease burden and reported cases,ery excellent surveillance.” Published data in South Korea andaiwan show a downward shift in population seroprevalence overime and trends toward infection at older ages [4–7]. A numberf Korean studies showed most people aged 10–29 have no anti-odies against hepatitis A virus [6,8–11], a trend also mentioned inaiwan. Recent outbreaks were reported in both countries (2007 inaiwan, 2008–9 in South Korea) [12–15].

In Chile and Russia, the majority of interviewees suggestedhat routine surveillance provided reasonable epidemiological datan hepatitis A, but recent data were not verified from the lit-rature review. Many Chilean respondents were positive about

0% 1

Local vaccine manufacturing capacityVaccine safety & effectiveness data

Cost-effectiveness; economic burden dataAvailability of combination vaccine

Availaibility of data on Hep A diseaseAnti-vaccine movement; distrust of manufacturers

Vaccine procurement, delivery & regulatory challengesPublic awareness; media coverage

Emphasis on sanitationGlobal or local recommendations on vaccine

Priority vis-à-vis other vaccinesPolitical support from government leaders; elections

Disease burden & outbreaksBudget & price of the vaccine

Both Interviews onlyDiscussed by:

Fig. 2. Barriers and facilitators to h

of hepatitis A and its complications: $75.9 million pesos (US$5.9 million); Cost per $1,186 pesos (US$92), severe case: $5,942 pesos (US$463), cholestatic case: $9,96777), fulminant case: $55,798 pesos (US$4,349).

the surveillance data, and our review found sufficient litera-ture through the 1990s documenting the transition to lowerendemicity [16–22]. The most recent hepatitis A specific data,however, were from 2001, with only two studies [23,24] exam-ining the changing epidemiology of hepatitis A and the potentialthreat it poses. Although the Chile Ministry of Health reportsincidence data from 1975 to 2011, all hepatitis cases are com-bined, leaving doubts as to the specific role of hepatitis A: “Wedon’t have routine hepatitis A tested. Typing is for B only, and

if not B, then “non-B.” Overall, respondents in Chile reported ahigh level of confidence that water and sanitation improvementshad largely addressed disease, except for a small number ofareas.

0% 30% 50% 70% 90%20% 40% 60% 80% 100%

Literature only Neither

epatitis A vaccine adoption.

4 ine 32

ieToOfdeaRfb

tdtttmclMn“ftttosasA

gidddod

3

emnoeacbKtaa

OnhstEb

094 S. Ozawa et al. / Vacc

In Russia, several respondents reported that disease burden datas available and cited numbers of cases by region and year; how-ver, we could not identify such data through the literature review.he review identified 38 epidemiological studies in Russia, withnly two studies reporting seroprevalence data after 2000 [25,26].ne Russian government respondent noted: “seroprevalence data

or some regions show high antibodies; however, we do not have exactata for most regions in different age groups.” Overall, the publishedpidemiological data in Russia were quite variable, suggesting vari-tions in measurement, reporting, or interpretation [27–29]. Inussia, the literature reported several outbreaks in cities [30] and

ollowing natural disasters [31–33], some of which were mentionedy respondents.

In India and Mexico, respondents and the literature agreed thathe hepatitis A epidemiological evidence is weak, but some respon-ents did not find this alarming. In India, two respondents saidhere were no epidemiologic data available: “[We have] no mor-ality, no morbidity, no estimates of economic loss for the poor. Buthe technical advisory groups need to have these data to review toake decisions.” A few respondents noted recent studies not yet

ompleted and published. The literature review confirmed theack of recent seroprevalence data in most areas of India [34–39].

eanwhile, several respondents believed hepatitis A disease isot in India and that seroprevalence in India has not changed:We don’t have [data] and we really don’t need it.” Policy articlesrom 1995 through 2011, however, indicate a growing recogni-ion of the epidemiological transition in India and the growinghreat of outbreaks [40–47]: “The epidemiological transition needso be documented as well as the potential for outbreak; Kerala wasne state with a recent outbreak.” A 2005 outbreak in Hyderabaduggested a change in adult seroprevalence, warranting furtherssessment for vaccination [48]. Currently, there is no nationalurveillance system to track outbreaks and the burden of hepatitis

in India.In Mexico, respondents noted there is no data by age group,

eography, or socioeconomic status, or data capturing privatemmunizations, disease severity and the extent of fulminantisease. The overall body of Mexican literature on hepatitis A epi-emiology was relatively small, with old (1996) seroprevalenceata for Mexico City [49] and more recent data through 2006 forther areas [50–52]. Older data suggest the initiation of the epi-emiological transition in Mexico [53].

.2. Economic data on hepatitis A vaccine

The majority of stakeholders in 5 out of 6 countries reported thatconomic and financial data were very important in the decisionaking process (Table 3). A government implementer in Mexico

oted the Ministry of Health is “quite willing to have a discussionn hepatitis A; that is why we need cost-effectiveness [data].” How-ver, the literature and internet search identified only 4 economicnalyses on hepatitis A in the six countries. These included twoost-effectiveness studies in Chile [54,55] which were referred toy the interviewees in Chile, one economic analysis model in Southorea [56] also known to Korean respondents, and one study of

he costs of hepatitis A in Mexico [57] which is only published asn abstract and was not known to the respondents. No economicnalyses were found in India, Russia or Taiwan.

Even among the published economic studies, data gaps remain.f the two cost-effectiveness studies in Chile [54,55] respondentsoted the studies are missing the cost of illness for a patient withepatitis A, and that they were suspicious of economic studies

ponsored by pharmaceutical companies. We also found that nei-her models used Chilean cost data, and instead relied on US anduropean costs of hepatitis A. The 2010 economic model publishedy the South Korean Centers for Disease Control did not include

(2014) 4089–4096

detailed data on incidence by severity of hepatitis A cases and onlyreported per unit costs for different services, leaving gaps in costsof hepatitis A in South Korea [56].

While economic data are important, respondents cautioned thatit is not the sole decision maker. A vaccine manufacturer in Indianoted that economic data are “not the only issue as India looks atseveral other impact factors such as infant and maternal mortality.” InMexico, a government official noted: “The introduction of the vaccinecould be more costly than the disease itself. For example, pneumococcalvaccine was controversial at one time because of the cost. One studyshowed that it wasn’t cost-effective, but it was still introduced becauseof the number of deaths and cases reported.”

3.3. Barriers and facilitators to hepatitis A vaccine adoption

We identified 14 barriers and facilitators to adopting the hep-atitis A vaccine by comparing those discussed in the literature withthose described in interviews by country. Fig. 2 presents thesebarriers/facilitators and whether each was discussed in the liter-ature and/or interviews.

In general, we found a large gap between barriers and facili-tators for adoption perceived by stakeholders compared to thosediscussed in policy papers. The importance of political sup-port from government leaders and the role of elections werebrought up as a barrier or facilitator in interviews in every coun-try (e.g. “this is an election year and it is not good to introduceanything that costs money.”), but were not mentioned in theliterature. The interviews also discussed the priority for this vaccinevis-à-vis other vaccines and mentioned global or local recommen-dations on vaccine adoption, which were rarely discussed in theliterature. A Mexican government official noted, “There are manyother needs for the country and the [Ministry of Health] spends largesums of money on immunization. It is the money that is the prob-lem, it is not available.” Other areas less frequently discussed inthe literature but described by stakeholders as barriers were: astrong emphasis on sanitation, weak public awareness and lowmedia coverage, challenges in vaccine procurement, delivery andregulations, anti-vaccine movements and distrust of manufactur-ers.

On the other hand, barriers more commonly discussed in theliterature were: the lack of data on hepatitis A disease, cost-effectiveness and other economic data, combination vaccines forhepatitis A, and the potential for safety and effectiveness data ofthe vaccine to facilitate decision making. Immunization budget orprice of the vaccine, and outbreaks of hepatitis A were the onlyfactors consistently discussed by both sources.

4. Discussion

Our analysis identified gaps between the published literatureand what key stakeholders believe about epidemiologic data, eco-nomic data and barriers and facilitators of vaccine adoption forhepatitis A in six countries. The results of this study highlight sev-eral areas in which having data from both the literature review andstakeholder interviews provided additional insights into the fac-tors driving policy decisions for the hepatitis A vaccine. Regardingthe evidence in support of an epidemiologic transition for hepatitisA seroprevalence, we found that most often the stakeholders wereaware of the existing data or that very little data existed. How-ever, in Chile and Russia, stakeholders believed the data to be moresupportive of their positions or more solid than the literature could

document. This discrepancy between the belief in existing data andwhat was found suggest a decline in investment in data collectionor priority of hepatitis A, perhaps due to a reliance on improve-ments in hygiene and sanitation. The lack of solid data on current

ine 32

slwsc

tfmistFalzTh

hbftgmmaeawDtic

5

rdacnciirgutco

A

tMa

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

S. Ozawa et al. / Vacc

eroprevalence rates underscores the potential for outbreaks and aingering threat of hepatitis A. In India and Mexico, although there

as recognition that data were lacking, there were a surprisinglymall number of seroprevalence studies despite the size of theseountries.

Our findings of limited economic data were consistent betweenhe literature and the interviews. However, investigation into theour economic models identified areas in which current economic

odeling falls short in meeting the needs of policy makers andn utilizing the best and most relevant data to support countrypecific decision making. Our review suggests the need for addi-ional investment in economic analyses using country specific data.inally, comparison of the barriers and drivers of hepatitis A vaccinedoption noted several differences in factors emphasized by theiterature and stakeholders. For example, political will and prioriti-ation of vaccines were barriers rarely mentioned in the literature.hese data clearly demonstrate that neither source alone wouldave provided the complete picture of relevant factors.

Despite the benefits of using two separate methods for assessingepatitis A vaccine policy decision making, our results are limitedy the search strategies for the literature review and the samplingrame for interviews. Our literature search focused primarily onhe published literature supplemented by an internet search of theray literature which varied significantly in quality. For key infor-ant interviews, our study resulted in a relatively small sample sizeainly due to the study’s very specific topic (hepatitis A vaccine

doption) and focus on the viewpoints of government officials, sci-ntists, clinicians and other administrators who know somethingbout the topic. People with program and private sector experienceere contacted, but many did not respond to interview requests.espite these limitations, we believe we have identified and syn-

hesized articles in a systematic manner and provide a glimpsento the understandings of key stakeholders of Hepatitis A in eachountry.

. Conclusions

This study concurrently carried out a systematic literatureeview and key stakeholder interviews to assess gaps betweenocumentation and policy makers’ perceptions in six countries. Tri-ngulation of results allowed us to identify countries where betterommunication of existing evidence or greater sharing of existingon-published evidence would be fruitful. It also highlighted andonfirmed data gaps in seroprevalence or cost-effectiveness stud-es where both the literature and stakeholders agree that evidences missing and would be important to gather. Applying multipleesearch methods resulted in a more focused attention on the dataaps and evidence-to-policy gaps than if only one method had beensed. This study also highlights the dearth of seroprevalence datahat exist in India and Mexico. Further research is needed in theseountries to highlight the potential health and economic impactsf hepatitis A disease to help guide vaccination decisions.

cknowledgements

We thank Kyung Min Song, Amanda Debes and Lauren Oldija forheir support with interviews and analysis. We also thank Leslie

ontejano, Nianwen Shi, and Elnara Eynullayeva for translationssistance and Orin Levine for his guidance on the project.

eferences

[1] World Health Organization. WHO position paper on hepatitis A vaccines –June 2012. Weekly epidemiological record. Geneva, Switzerland: World HealthOrganization; 2012. p. 261–76.

[2] Wasley A, Fiore A, Bell BP. Hepatitis A in the era of vaccination. Epidemiol Rev2006;28:101–11.

[

(2014) 4089–4096 4095

[3] World Health Organization. In: WHO Strategic Advisory Group of Expertson Immunization HAwg, editor. Evidence based recommendations for use ofhepatitis A vaccines in immunization services: background paper for SAGEdiscussions. Geneva, Switzerland: World Health Organization; 2011.

[4] Choi HK, Song YG, Kim CO, Shin SY, Chin BS, Han SH, et al. Clinical features ofre-emerging hepatitis A: an analysis of patients hospitalized during an urbanepidemic in Korea. Yonsei Med J 2011;52:686–91.

[5] Wang C, Lin M, Wang A, Tschen S. Incidence and seroprevalence of hepatitisA virus infections among Taiwanese living in Taipei metropolis. In: A globalhepatitis A meeting: has the time come to control hepatitis A globally? Miami,FL: Meeting organized by Center for Disease Control (CDC), Center for the Eval-uation of Vaccination at the University of Antwerp, World Health Organization(WHO) and the Pan American Health Organization (PAHO); 2007. p. 106.

[6] Lee H, Cho HK, Kim JH, Kim KH. Seroepidemiology of hepatitis A in Korea:changes over the past 30 years. J Korean Med Sci 2011;26:791–6.

[7] Chen JY, Chiang JC, Lu SN, Hung SF, Kao JT, Yen YH, et al. Changing prevalenceof anti-hepatitis A virus in adolescents in a rural township in Taiwan. ChangGung Med J 2010;33:321–6.

[8] Lee D, Ki M, Lee A, Lee KR, Park HB, Kim CS, et al. A nationwide seroprevalenceof total antibody to hepatitis A virus from 2005 to 2009: age and area-adjustedprevalence rates. Korean J Hepatol 2011;17:44–50.

[9] Yun SW, Lee WK, Cho SY, Moon SH, Shin HD, Yun SY, et al. The seroprevalencerate, vaccination rate and seroconversion rate of hepatitis A in central regionof Korea. Korean J Gastroenterol 2011;57:166–72.

10] Noh du Y, Cho YC, Jun WJ, Kim SK, Yun KW, Park SY, et al. Seroprevalence ofIgG anti-HAV in hospital employees below 40 years old. Korean J Gastroenterol2010;55:183–8.

11] Lee CS, Kwon KS, Koh DH, Youm JH, Gwack J, Lee JH. Declining hepatitis Aantibody seroprevalence in the Korean military personnel. Jpn J Infect Dis2010;63:192–4.

12] Yoon YK, Chun BC, Lee HK, Seo YS, Shin JH, Hong YS, et al. Epidemiologicaland genetic analysis of a sustained community-wide outbreak of hepatitis A inthe Republic of Korea, 2008: a hospital-based case-control study. J Clin Virol2009;46:184–8.

13] Jiang DD-S, Yu C-J, Yeh H-C, Pzeng M-Y, Lai M-H. An outbreak of hepatitis Ain the Taipei metropolitan area. Taiwan Epidemiol Bull (Center for DiseaseControl) 2007:310–1.

14] Korean Centers for Disease Control. Outbreak of hepatitis A infection at a highschool in Dobong-gu, Seoul. In: Public Health Weekly Report. Seoul, Korea:Korean Centers for Disease Control; 2009.

15] Korean Centers for Disease Control. Hepatitis A outbreak in Daegu, May 2008.In: Public Health Weekly Report. Seoul, Korea: Korean Centers for Disease Con-trol; 2008.

16] Fix AD, Martin OS, Gallicchio L, Vial PA, Lagos R. Age-specific prevalence of anti-bodies to hepatitis A in Santiago, Chile: risk factors and shift in age of infectionamong children and young adults. Am J Trop Med Hyg 2002;66:628–32.

17] Ibarra H, Riedemann S, Prado V, Reinhardt G, Vega I, Potin M, et al. Currentstatus of immunity to hepatitis A virus in various adult groups. Rev Med Chile1999;127:1165–8.

18] Zacarias J, Rakela J, Riveros C, Brinck P. Anti-hepatitis A antibod-ies in healthy children and in patients with hepatitis. Rev Méd Chile1981;109(September):833–6.

19] Tapia-Conyer R, Santos JI, Cavalcanti AM, Urdaneta E, Rivera L, Manterola A,et al. Hepatitis A in Latin America: a changing epidemiologic pattern. Am J TropMed Hyg 1999;61:825–9.

20] Riedemann S, Hochstein-Mintzel V, Reinhardt G. Prevalence of hepatitis A andB in the population of Valdivia: a seroepidemiological study. Rev Méd Chile1984;112(July):672–4.

21] Riedemann S, Ibarra H, Hochstein-Mintzel V, Reinhardt G, Niedda M, FroesnerG. Viral hepatitis in rural area of Chile. Rev Méd Chile 1987;115(January):16–8.

22] Ibarra H, Riedemann S, Reinhardt G, Hochstein-Mintzel V, Froesner G. Hepatitisantibody prevalence in children from two rural areas in Chile. Rev Méd Chile1988;116(November):1115–8.

23] Ibarra H, Riedemann S, Toledo C. Hepatitis A and E virus antibodies in Chileanchildren of low socioeconomic status: a one year follow-up study]. Rev MedChile 2006;134:139–44.

24] Ibarra H. The changing epidemiology of viral hepatitis in Chile. Rev Med Chile2007;135:229–39.

25] Mukomolov SL, Stalevsklaia AV, Zheleznova NV, Sinayskaya EV, Levakova IA,Vasilyeva VA. Seroepidemiology of hepatitis a in Saint-Petersburg in 2009. ZhMikrobiol Epidemiol Immunobiol 2010:15–20.

26] Popova OEK, Ilchenko KK, Yu L, Isaeva OV, Kozhanova TV, Dmitriev PN, et al.Epidemiological and molecular biological analysis of causes of rise of hepati-tis a incidence in Republic of Tuva in. Zh Mikrobiol Epidemiol Immunobiol2008;2010:23–6.

27] Reizis AR, Drondina AK, Nikitina TS, Zamiatina NA, Mikhailov MI, Asratian AA,et al. An unusual course of viral hepatitis A in children. Zh Mikrobiol EpidemiolImmunobiol 1997:84–6.

28] Mindlina AY. Infectious intestinal morbidity in Russia. Vestn Ross Akad MedNauk 2010:30–3.

29] Zverev V. Vaccinoprophylaxis of hepatitis A: successes and challenges [Selected

data on the incidence of hepatitis A: table]. Vaccination 2001;16(June–August).

30] Solodovnikov Iu P, Tibekin AT, Cherkasova LV, Berglezova LN, Zaitsev BE,Lytkina IN. Epidemics of enteric infections in Moscow in the last 10 years(1993–2002) institutions amd contingents of epidemic risk. Zh Mikrobiol Epi-demiol Immunobiol 2004:115–7.

4 ine 32

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

096 S. Ozawa et al. / Vacc

31] Jung YM, Park SJ, Kim JS, Jang J-H, Lee SH, Kim J-W, et al. Atypical manifestationsof hepatitis A infection: a prospective, multicenter study in Korea. J Med Virol2010;82:1318–26.

32] Onishchenko GGA, Kh A, Doleva LA, Konchenko AV. Prevention of acute entericinfections and viral hepatitis A in the Republic of Adygeia in connection with anatural disaster. Zh Mikrobiol Epidemiol Immunobiol 2003:97–8.

33] Kovalev NG, Balaban OA, Koval’chuk IV, Romanova TI, Kashirina IB, PugachevaON. Prevention of acute enteric infections and viral hepatitis A in the StavropolTerritory in connection with a natural disaster. Zh Mikrobiol EpidemiolImmunobiol 2003:99–101.

34] Acharya SK, Batra Y, Bhatkal B, Ojha B, Kaur K, Hazari S, et al. Seroepidemi-ology of hepatitis A virus infection among school children in Delhi and northIndian patients with chronic liver disease: implications for HAV vaccination. JGastroenterol Hepatol 2003;18:822–7.

35] World Health Organization. In: The Department of Immunization VaB, edi-tor. The global prevalence of hepatitis A virus infection and susceptibility: asystematic review. Geneva: World Health Organization; 2009.

36] Anand AC, Nagpal AK, Seth AK, Dhot PS. Should one vaccinate patients withchronic liver disease for hepatitis A virus in India? J Assoc Physicians India2004;52:785–7.

37] Gadgil PS, Fadnis RS, Joshi MS, Rao PS, Chitambar SD. Seroepidemiology ofhepatitis A in voluntary blood donors from Pune, western India (2002 and2004–2005). Epidemiol Infect 2008;136:406–9.

38] Rath CP, Akki A, Patil SV, Kalyanshettar SS. Seroprevalence of hepatitis A virusantibody in Bijapur, Karnataka. Indian Pediatr 2011;48:71–3.

39] Mohanavalli B, Dhevahi E, Menon T, Malathi S, Thyagarajan SP. Prevalenceof antibodies to hepatitis A and hepatitis E virus in urban school children inChennai. Indian Pediatr 2003;40:328–31.

40] John TJ, Chandy GM. What priority for prevention of hepatitis A in India? IndianJ Gastroenterol 1998;17:2–3.

41] Indian Academy of Pediatrics. IAP guide book on immunization: IAP committeeon immunization (COI) 2009–2011. New Delhi: Indian Academy of Pediatrics;2011.

42] John TJ. Immunization dialogue: hepatitis A vaccine. Indian Pediatr

1995;1995:1249–51.

43] Mathur P, Arora NK. Considerations of HAV vaccine in India. Indian J Pediatr1999;66:111–20.

44] Arankalle VA, Chadha MS. Who should receive hepatitis A vaccine? J Viral Hepat2003;10:157–8.

[

[

(2014) 4089–4096

45] Arankalle VA. Hepatitis A vaccine strategies and relevance in the present sce-nario. Indian J Med Res 2004;119:iii–vi.

46] Singh S. Update on immunization policies, guidelines and recommendations.Indian Pediatr 2004;41:239–44.

47] Karande S. Update on available vaccines in India: report of the APPA VU 2010:I. Indian J Pediatr 2011;78:845–53.

48] Sarguna P, Rao A, Sudha Ramana KN. Outbreak of acute viral hepatitisdue to hepatitis E virus in Hyderabad. Indian J Med Microbiol 2007;25:378–82.

49] Ortiz-Ibarra FJ, Figueroa-Damian R, Lara-Sanchez J, Arredondo-Garcia JL,Ahued-Ahued JR. Prevalence of serologic markers of hepatitis A, B, C, and Dviruses in pregnant women. Salud Publica Mex 1996;38:317–22.

50] Garcia-Juarez I, Solorzano Santos F, Alvarez-y-Munoz MT, Vazquez-Rosales JG.Is there a shift in the epidemiology of hepatitis A in Mexican children? RevInvest Clin 2008;60:292–6.

51] Escobedo-Melendez, Fierro NA, Roman S, Maldonado-Gonzalez M, Zepeda-Carrillo E, Panduro A. Prevalence of hepatitis A, B and C serological markersin children from western Mexico. Ann Hepatol 2012;11:194–201.

52] Sotelo N, de los A, ngeles Durazo M, Gonzalez A, Dhanakotti N. Early treatmentwith N-acetylcysteine in children with acute liver failure secondary to hepatitisA. Ann Hepatol 2009;8:353–8.

53] Ruiz-Gomez J, Castaeda-Desales D, Huerta-Nez L, Pama-Cola O, Mascareas-delos Santos C, Kuri-Morales P, et al. Hepatitis A virus infection among Mexicanyoung population: National health and nutrition survey 2006. In: A global hep-atitis A meeting: has the time come to control hepatitis A globally? Miami, FL:Meeting organized by Center for Disease Control (CDC), Center for the Evalu-ation of Vaccination at the University of Antwerp, World Health Organization(WHO) and the Pan American Health Organization (PAHO); 2007.

54] Quezada A, Baron-Papillon F, Coudeville L, Maggi L. Universal vaccination ofchildren against hepatitis A in Chile: a cost-effectiveness study. Rev PanamSalud Publica 2008;23:303–12.

55] Valenzuela MT, Jacobs RJ, Arteaga O, Navarrete MS, Meyerhoff AS, Innis BL. Cost-effectiveness of universal childhood hepatitis A vaccination in Chile. Vaccine2005;23:4110–9.

56] Korean Centers for Disease Control. Risk factors, mathematical modeling andeconomic analysis for hepatitis A in Korea. In: Public Health Weekly Report.Seoul, Korea: Korean Centers for Disease Control; 2010.

57] Rivas-Oropeza I, Valencia-Mendoza A, Sanchez-Gonzalez G. Financial burdenfor hepatitis A in Mexico. Value Health 2009;12:A516.