Objective

8406899

7945

Vaccine

0264-410X1873-2518

23588084

4667720

10.1016/j.vaccine.2013.03.047

HHSPA739881

Article

Cost-utility analysis of 10- and 13-valent pneumococcal conjugate vaccines: Protection at what price in the Thai context?

Kulpeng

Wantanee

ab*Leelahavarong

Pattara

aRattanavipapong

Waranya

aSornsrivichai

Vorasith

bBaggett

Henry C.

cMeeyai

Aronrag

adPunpanich

Warunee

eTeerawattananon

Yot

a Health Intervention and Technology Assessment Program (HITAP), 6th Floor, 6th Building, Department of Health, Ministry of Public Health, Tiwanon Rd., Muang, Nonthaburi 11000, Thailandb Epidemiology Unit, Faculty of Medicine, Prince of Songkla University, P.O. Box 5, Khohong, Hat Yai, Songkhla 90112, Thailandc International Emerging Infections Program (IEIP), Global Disease Detection Regional Center, Thailand MOPH-US Centers for Disease Control and Prevention Collaboration, Department of Disease Control, Ministry of Public Health, Tiwanon Rd., Muang, Nonthaburi 11000, Thailandd Department of Epidemiology, Faculty of Public Health, Mahidol University, 420/1 Ratchawithi Rd., Ratchathewi District, Bangkok 10400, Thailande Department of Pediatrics, Queen Sirikit National Institute of Child Health, 420/8 Rajvithi Rd., Rajthevi, Bangkok 10400, Thailand

* Corresponding author at: Health Intervention and Technology Assessment Program (HITAP), 6th Floor, 6th Building, Department of Health, Ministry of Public Health, Tiwanon Road, Muang, Nonthaburi, 11000 Thailand. Tel.: +66 2 590 4374 5; fax: +66 2 590 4369. wantanee.k@hitap.net, wantaney@hotmail.com

pattara.l@hitap.net (P. Leelahavarong), waranya.r@hitap.net (W. Rattanavipapong), vorasith@msn.com (V. Sornsrivichai), Kipb@th.cdc.gov (H.C. Baggett), phamy@mahidol.ac.th, aronrag.c@hitap.net (A. Meeyai), waruneep@gmail.com (W. Punpanich), yot.t@hitap.net (Y. Teerawattananon).

24112015

1242013

1062013

02122015

312628392847Objective

This study aims to evaluate the costs and outcomes of offering the 10-valent pneumococcal conjugate vaccine (PCV10) and 13-valent pneumococcal conjugate vaccine (PCV13) in Thailand compared to the current situation of no PCV vaccination.

Methods

Two vaccination schedules were considered: two-dose primary series plus a booster dose (2 + 1) and three-dose primary series plus a booster dose (3 + 1). A cost-utility analysis was conducted using a societal perspective. A Markov simulation model was used to estimate the relevant costs and health outcomes for a lifetime horizon. Costs were collected and values were calculated for the year 2010. The results were reported as incremental cost-effectiveness ratios (ICERs) in Thai Baht (THB) per quality adjusted life year (QALY) gained, with future costs and outcomes being discounted at 3% per annum. One-way sensitivity analysis and probabilistic sensitivity analysis using a Monte Carlo simulation were performed to assess parameter uncertainty.

Results

Under the base case-scenario of 2 + 1 dose schedule and a five-year protection, without indirect vaccine effects, the ICER for PCV10 and PCV13 were THB 1,368,072 and THB 1,490,305 per QALY gained, respectively. With indirect vaccine effects, the ICER of PCV10 was THB 519,399, and for PCV13 was THB 527,378. The model was sensitive to discount rate, the change in duration of vaccine protection and the incidence of pneumonia for all age groups.

Conclusions

At current prices, PCV10 and PCV13 are not cost-effective in Thailand. Inclusion of indirect vaccine effects substantially reduced the ICERs for both vaccines, but did not result in cost effectiveness.

Pneumococcal conjugate vaccineIndirect effectCost-effectiveCost-utility analysisThailand

1. Introduction

Bacterial meningitis, pneumonia and otitis media caused by Streptococcus pneumoniae (S. pneumoniae) are serious but preventable health problems in young children. Pneumococcal conjugate vaccines (PCVs) have been proven safe and effective in children less than 5 years old to prevent both invasive (e.g., meningitis, bacteremia) and non-invasive (e.g., pneumonia, otitis media) pneumococcal diseases [1–3]. Moreover, clinical studies in the United States and Europe have demonstrated that vaccinating young children with PCV can lead to a significant decline in the incidence of pneumococcal disease among unvaccinated populations, notably older children, adults and the elderly [4–6]. Although PCV has been available for more than a decade, its use has been limited in many areas due to high cost.

The cost-effectiveness of PCV has been documented in many high-income countries, and the governments in these settings have adopted the vaccine as part of their national immunization programs [7–13]. However, few economic evaluations have been conducted in low- or middle-income settings, where the burden of pneumococcal disease is at least as high [14–16]. In recent years, many low-income countries, especially in Africa, have introduced PCV programs with substantial support from the GAVI Alliance, a broad partnership that works to improve access to immunization [17]. Most middle income countries such as Thailand, which are not eligible for GAVI support and therefore face potentially substantial financial barriers to PCV implementation, have not yet implemented PCV programs. Cost-effectiveness studies are especially important to inform decision-making in these settings.

This study was conducted at the request of policy makers in Thailand to inform decisions about the adoption of PCV as part of this country’s Expanded Program on Immunization (EPI). It was believed that if the vaccine is included in the EPI, its coverage would be almost 100%. Given that Thailand achieves 99% coverage with DTaP 3 dose vaccine [18], such an assumption is not unrealistic. This economic evaluation considered costs and impact of offering 10-valent PCV (PCV10), which covers 10 of approximately 90 S. pneumoniae serotypes, or recently licensed 13-valent PCV (PCV13), which covers 3 additional serotypes, compared to the current situation without a PCV program.

2. Methods

A model-based economic evaluation was performed to estimate costs as well as outcomes of vaccination with PCV10 and PCV13 compared to ‘no vaccination’. Because there are different options for vaccination schedules [19], this study considered two commonly recommended regimens: two-dose primary series at 2 and 4 months of age plus a booster dose at age 13 months (2 + 1) and three-dose primary series at 2, 4 and 6 months of age plus a booster dose at age between 12 to 15 months (3 + 1). The study adopted a societal viewpoint using a life-time horizon with 3% discounting for both costs and outcomes beyond one year, as recommended in the by the Thai Health Technology Assessment guideline [20].

2.1. Model structure and assumptions

A Markov model was constructed based on the natural history of disease related to S. pneumoniae infection (Fig. 1). The model consisted of three major health states: healthy, S. pneumoniae infection and death. For S. pneumoniae infection, the model accounts for four health conditions based on their association with high case fatality or permanent disability (e.g., epilepsy, neurodevelopmental impairment or chronic lung disease): pneumococcal meningitis, pneumococcal bacteremia, all-cause pneumonia and all-cause acute otitis media (AOM). A one-year cycle was deployed in the model, and it was assumed that more than one infection is possible during a lifetime but each Markov cycle allows for only one infection.

2.2. Model input parameters2.2.1. Epidemiological data

Estimated age-specific incidences of pneumococcal diseases in Thailand are presented (Supplementary Table 1). Pneumococcal bacteremia incidence was based on findings from active surveillance for bacteremia requiring hospitalization in two rural Thailand provinces [21] and does not include outpatient cases. All-cause meningitis and pneumonia incidence were derived from national surveillance conducted by the Bureau of Epidemiology, Ministry of Public Health (MoPH) [22]. For this model, all hospitalized meningitis cases reported to the national surveillance system were assumed to be caused by bacteria. The proportion of pneumococcal meningitis cases among all bacterial meningitis (mean = 14.27%, standard error (SE) = 3) was derived from hospital databases [23,24]. AOM incidence was obtained from the Thailand Burden of Disease Project [25].

Table 1 illustrates probabilities of hospitalization and developing complications from pneumococcal disease. Mortality rate and case fatality data were acquired from the Burden of Disease Project and literature review, utilizing data from Thailand or the East Asia region whenever available [23–28].

2.2.2. Direct effects (vaccine efficacy)

For a 3 + 1 dosing schedule, vaccine efficacy (VE) against vaccinetype invasive pneumococcal disease (IPD) was considered 89% based on a 2009 meta-analysis of randomized controlled trials (RCTs) [29]. This figure was used to estimate the efficacy of PCV10 and PCV13 against vaccine-type IPD (Table 1) by assuming the same overall efficacy against vaccine-type IPD, accounting for the additional serotype coverage [30–33]. Because sufficient data on serotype coverage were not available for pneumonia and AOM, VE against all-cause pneumonia and AOM for PCV10 and PCV13 were extrapolated from the efficacy of PCV7 against all-cause pneumonia (6%) [3] and AOM (6%) [29]. It was assumed that the efficacy of PCV10 and PCV13 against pneumonia and AOM increased proportionally with the increase in serotype coverage.

VE for a 2 + 1 schedule was modified to account for reduced immunogenicity for serotypes 6B and 23F [34] compared to the 3 + 1 schedule; a 20% reduction in efficacy against these serotypes was assumed. Serotypes 6B and 23F accounted for approximately 40% of PCV7 serotypes in Thai children [30–32]. As a result, an overall reduction of 8% in VE for the 2 + 1 schedule was estimated using the following formula: VE2−1=VE3−1x(1−0.08)

2.2.3. Indirect effects (herd protection)

This model accounted for the indirect effect of the vaccine to prevent disease in unvaccinated populations (Supplementary Table 2). The percentage reduction in IPD incidence among unvaccinated populations was based on survey data after mass vaccination in the United States [4] with the adjustment for differences in serotype distribution between Thailand and the United States [35]. The indirect effect for IPD was based using the following formula: %IPD fall in Thailand=%IPD fall in the United States×Serotype coverage in Thailand/Serotype coverage in the United States Because the indirect effects can occur in every population cohort ranging from aged 16–99 years, we manually calculated the indirect effects in each age group using the static model.

The indirect effect for pneumonia was estimated for unvaccinated populations, assuming that the protective effect would be equivalent to the decrease in IPD incidence among the same groups and adjusted for proportion of hospitalized pneumonia caused by S. pneumoniae. To estimate the proportion of hospitalized pneumonia cases caused by S. pneumoniae, we used data from Prapasiri et al. [26], who found that 11.76% (SE = 2.35) of bacteremic pneumonia cases in two Thai provinces were S. pneumoniae. The calculation of indirect effect of vaccine was base on the following formula: %Hospitalized pneumonia fall in Thailand=Proportion of pneumococcal pnemonia×%IPD fall in Thailand

2.2.4. Costs and outcomes

The cost analysis was performed based on a societal perspective, and included both direct medical and direct non-medical costs (Table 1). Direct medical costs for outpatient and inpatient care were obtained from the Thailand’s Centre for Health Equity Monitoring [36] and the Central Office for Healthcare Information [24], respectively. The cost of the vaccination program included the vaccine cost and delivery cost [37]. Direct non-medical costs, such as costs for transportation, meals, accommodation, facilities, productivity loss [38] by parents or caregivers for hospital visits or providing informal care, were derived from face-to-face interviews with caregivers of 192 ill children aged 5–14 years in seven public hospitals in five provinces throughout Thailand. All cost parameters are presented in 2010 Thai Baht (THB) (THB 31 = US$ 1).

Outcomes were measured in quality-adjusted life years (QALYs) using the Health Utilities Index Mark 3 [39] (Table 1). Utility measures were derived from interviews with the aforementioned 192 caregivers and the results previously described [40].

2.3. Uncertainty analyses2.3.1. One-way sensitivity analysis

One-way sensitivity analysis was performed to examine the uncertainty surrounding each parameter individually (e.g., discounting rate at 0% and 6% per annum, disease incidence, vaccine efficacy, vaccine serotype coverage, percentage incidence reduction among unvaccinated groups, utility and cost). The impact of serotype replacement and indirect vaccine effects were also examined. The former was done by adjusting the serotype coverage parameter whereas the latter was explored by varying the disease incidence reduction among unvaccinated groups in the United States [4]. For pneumonia incidence, there were two data sources in Thailand. We used data from Thailand’s national surveillance (Bureau of Epidemiology, MoPH) [22] as the base-case and data from an active, population-based surveillance system operated collaboratively by MoPH and the International Emerging Infections Program (IEIP, US Centers for Disease Control and Prevention) [41] in the sensitivity analysis. We also assessed the effect of two different durations of vaccine protection: 5 and 10 years.

This analysis used the cost-effectiveness ceiling threshold of one per-capita gross domestic product or THB 100,000 (US$ 3226) per QALY gained as recommended by the Subcommittee for Development of the National List of Essential Drugs 2007 [42]. The Subcommittee sets the threshold for considering new medicines and vaccines for public reimbursement. For PCV vaccination scenarios determined to be not cost-effective at the current price, we examined the maximum cost of the vaccine that would make it cost-effective as well as cost-saving in the Thai setting. Cost-saving implies that no additional budget would be required for vaccination, because resources saved from averted pneumococcal disease could be used to cover vaccination costs.

2.3.2. Probabilistic sensitivity analysis

Probabilistic sensitivity analysis was conducted to examine the effect of all parameter uncertainty simultaneously using a Monte Carlo simulation using Microsoft Office Excel 2007. The simulation was run for 1000 iterations to yield a range of possible values for total costs, health outcomes, and incremental cost-effectiveness ratios (ICERs) in THB per QALY gained. The probability distributions were determined according to the range of each input parameter value. The normal distribution was used as a default. The beta distribution was used when parameter values ranged between zero and one, such as in probability and utility parameters. The gamma distribution was used when parameter values ranged between zero and positive infinity, such as costs parameters.

3. Results

Compared to ‘no vaccination’, the 3 + 1 dose schedule of PCV10 and PCV13 would prevent an estimated 4262 and 5241 episodes of pneumococcal disease in the vaccinated population, respectively (Fig. 2). In addition, 4510 and 6211 episodes of pneumococcal disease would be averted in unvaccinated populations due to indirect effects. It was estimated that 369 and 495 pneumococcal deaths would be avoided by introducing PCV10 and PCV13, respectively.

Table 2 shows the ICERs of different PCV vaccination schedules with and without inclusion of indirect vaccine effects. Without the indirect effects of vaccine, the 2 + 1 dose schedule produced ICERs of THB 1,368,072 and THB 1,490,305 per QALY gained for PCV10 and PCV13, respectively. The 3 + 1 dose schedule without accounting for indirect effects produced ICERs of THB 1,677,379 for PCV10 and THB 1,830,716 for PCV13. When the indirect effects of vaccination were included in the analysis, ICERs of PCV vaccination decreased by more than half. In one-way sensitivity analysis, the important determinants were discount rate, the change in duration of vaccine protection (5 vs. 10 years) and the incidence of pneumonia for all age groups. A 10-year protection duration including indirect effects, ICERs of PCV10 decreased to THB 287,353 and THB 363,248 for the 2 + 1 and 3 + 1 dose schedules, respectively; for PCV13, the corresponding ICERs were THB 290,420 and THB 367,339. When we used pneumonia incidence from active, population-based surveillance [41] and included indirect effects, the ICERs were reduced by almost 50% for the 3 + 1 schedule to THB 360,891 (PCV10) and THB 371,723 (PCV13) as well as by approximately 50% for the 2 + 1 schedule to THB 287,353 (PCV10) and THB 290,420 (PCV13). The model was less sensitive to variations in direct medical and non-medical costs and serotype replacement.

At current pricing, neither PCV10 nor PCV13 would be costeffective compared to ‘no vaccination’ at a ceiling threshold of THB 100,000 per QALY gained, with or without inclusion of indirect vaccine effects (Fig. 3). Including the indirect vaccine effects, PCV13 had a higher probability of being cost-effective compared to ‘no vaccination’ at a ceiling threshold between THB 600,000 and THB 750,000, depending on dosing schedule (Fig. 3A and 3B). Compared to PCV10, PCV13 had a higher probability of being cost-effective at a ceiling threshold between THB 550,000 and THB 600,000.

Without indirect vaccine effects, PCV10 yielded a higher probability of being cost-effective compared to ‘no vaccination’ at a ceiling threshold between THB 1,450,000 and THB 1,750,000, and PCV13 had a higher probability of being cost-effective compared to PCV10 at a ceiling threshold between THB 2,050,000 to THB 2,550,000 (Fig. 3C and D).

Threshold analysis demonstrated that using the 2 + 1 dosing schedule and considering indirect vaccine effects, PCV10 and PCV13 costs would have to be 75% lower (to THB 373 and THB 494), to be cost-effective; 92% cost reduction for both PCV10 and PCV13 (to THB 121 and THB 165) would be needed for either vaccine to be cost-saving (Fig. 4). Using a 3 + 1 dosing schedule, PCV10 and PCV13 costs would have to be 79% lower (to THB 304 and THB 403), to be cost-effective, and 93% lower (to THB 99 and THB 134), respectively, to be cost-saving.

When indirect vaccine effects were excluded, the maximum vaccine costs for both PCV10 and PCV13 to achieve cost-effective ranged from THB 107 to THB 162, and to be cost-saving, maximum costs ranged from THB 14 to THB 21.

4. Discussion

This study indicates that, at current pricing, neither PCV10 nor PCV13 would be considered cost-effective in Thailand at either dosing schedule examined, using Thailand’s standard ceiling threshold to assess health interventions. This finding results largely from the relatively high cost of the vaccine (per dose), which is equivalent to 5–6 times Thailand’s daily minimum wage. Our findings also reveal that the vaccine can become cost-effective or even cost-saving if vaccine costs were reduced by around 70–90% of current market prices.

Our findings stand in contrast to previous studies conducted in Argentina and Singapore which found PCV to be cost-effective [43,44]. The differences may be explained by differences of model structure and input parameters, especially epidemiological and economic data that vary across settings. In addition, the VE estimate used in our model was lower than that used in other studies. In this study, VE against vaccine-type IPD (89%) was derived from a systematic review and meta-analysis of RCTs [29], while other studies used 97% as reported from a single RCT conducted in the United States [1]. Difference in country specific serotype coverage may also have influenced the results. PCV10 serotype coverage for IPD among children aged less than 5 years is 75%, 81%, and 71% in Argentina, Singapore and Thailand, respectively [30–32,43,45]. This study also assumed a vaccine protection duration of 5 years, which is in line with several other economic evaluations of PCV studies [9,46], whereas some studies assumed protection lasted 10 years [7,47,48]. Our decision to use a 5-year protection duration was based on an immunogenicity study of PCV9 in South Africa [49], although this study did not follow participants beyond 5–6 years. Recognizing the limited data available, we applied a conservative assumption for the duration of vaccine protection. Furthermore, lower treatment costs in Thailand compared to other settings [12,13,43,44], contributed to the different conclusions about vaccine cost effectiveness in this study.

The model was very sensitive to pneumonia incidence. The ICERs decreased significantly when the pneumonia incidence was based on active, population-based surveillance compared to Thailand’s national surveillance system. However, even using the higher pneumonia incidence rate, PCV was not considered cost-effective for Thailand in our model.

4.1. Strengths and limitations

Parameters used in this model were obtained from high quality studies, including systematic reviews and metaanalyses. All parameters were contextualized for Thailand; thus, applying results of this study to other settings should be performed with caution. Our study examined two PCV formulations (10- and 13-valent) and two dosing schedules (2 + 1 and 3 + 1). Although this study adopted a static modeling rather than dynamic one, it included indirect effect of vaccination that protects infection in population who are not vaccinated. The use of static model also facilitates transparency of this study because many Thai decision makers and academics are more familiar with Markov, and the use of dynamic model in this study will require a number of assumptions given that this study considers four health conditions.

Nonetheless, this study has some limitations. First, due to the lack of local data on indirect vaccine effects, the model made assumptions based on findings from the United States [4]. Data from the United States showed a significant decline in IPD incidence among unvaccinated populations aged 20 years and above only. This ignored herd protection among young children (1–4 years) and teenagers, which could not be assessed in the United States, because children in this age group (1–4 years) were vaccinated as part of catch-up vaccination efforts. Second, IPD incidence rates used in this model were likely underestimates, because the available studies were conducted in public health facilities (i.e. government hospitals and health centers); thus, patients without access to public hospitals or at private hospitals were not included. Additionally, it has been shown that antibiotic use before blood culture collection in Thailand leads to underestimation of IPD incidence in hospital-based studies [21]. Perhaps more importantly, IPD rates cited for this analysis did not include outpatients because most of them were suspected and not confirmed cases. Including outpatient IPD cases in the model inputs would have resulted in lower ICERs. Lastly, the ceiling threshold used in this analysis is based on the preference of decision maker in Thailand. Decision makers in different settings may have their own preference regarding health investment, we encourage readers to compare the results to any threshold they consider it appropriate.

4.2. Implications

In summary, based on a societal perspective with a ceiling threshold of THB 100,000 per QALY, PCV10 and PCV13 would not be considered cost-effective, whether or not indirect vaccine effects were included in the model. Therefore, it cannot be recommended that PCV be included in Thailand’s EPI until prices decline to recommended values. Reduction in vaccine cost, which seems possible given the widespread introduction of PCV in many countries, could improve the feasibility of introduction in Thailand, which could result in substantial public health impact. Based on analyses that include indirect vaccine effects, PCV would become cost-effective at a price per-dose between THB 304 (PCV10, 3 + 1 schedule) and THB 494 (PCV13, 2 + 1 schedule) and cost-saving at a per-dose price between THB 99 and THB 165.

Supplementary Material

Supplementary Tables - Cost-utility analysis of 10- and 13-valent pneumococcal conjugate vaccines - protection at what price in the Thai context

Acknowledgements

We would like to thank the Thai Health Promotion Foundation and the Thailand Research Fund under the Senior Research Scholar Program on the Development of Health Technology Assessment Capacity in Thailand (RTA5580010) for funding support through the Health Intervention and Technology Assessment Program (HITAP). We are grateful for the support from Health Utilities Inc. We wish to acknowledge the valuable information provided by Pasakorn Akarasewi from the Bureau of Epidemiology, Department of Disease Control, Julia Rhodes and Susan Maloney from the International Emerging Infections Program (IEIP). Finally, we would like to thank Virasakdi Chongsuvivatwong, Anchalee Aramtiantamrong, Supalert Netsuwan, Manit Kongpan and Anek Mungaomklang, for helpful comments.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine.2013.03.047.

Abbreviations

PCV

pneumococcal conjugate vaccine

EPI

Expanded Program on Immunization

AOM

acute otitis media

MoPH

Ministry of Public Health

standard error

vaccine efficacy

IPD

invasive pneumococcal disease

RCT

randomized controlled trial

THB

Thai Baht

QALY

quality-adjusted life year

ICER

incremental cost-effectiveness ratio

References[1]

Black

Shinefield

Fireman

Lewis

Ray

Hansen

Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children, Northern California Kaiser Permanente Vaccine Study Center Group

Pediatr Infect Dis J3200019318795

10749457

[2]

Eskola

Kilpi

Palmu

Jokinen

Haapakoski

Herva

Efficacy of a pneumococcal conjugate vaccine against acute otitis media

N Engl J Med2200134464039

11172176

[3]

Black

Shinefield

Ling

Hansen

Fireman

Spring

Effectiveness of heptavalent pneumococcal conjugate vaccine in children younger than five years of age for prevention of pneumonia

Pediatr Infect Dis J920022198105

12352800

[4]

Whitney

Farley

Hadler

Harrison

Bennett

Lynfield

Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine

N Engl J Med5200334818173746

12724479

[5]

Grijalva

Nuorti

Arbogast

Martin

Edwards

Griffin

Decline in pneumonia admissions after routine childhood immunisation with pneumococcal conjugate vaccine in the USA: a time-series analysis

Lancet420073699568117986

17416262

[6]

Miller

Andrews

Waight

Slack

George

Herd immunity and serotype replacement 4 years after seven-valent pneumococcal conjugate vaccination in England and Wales: an observational cohort study

Lancet Infect Dis10201111107608

21621466

[7]

Asensi

De Jose

Lorente

Moraga

Ciuryla

Arikian

A pharmacoeconomic evaluation of seven-valent pneumococcal conjugate vaccine in Spain

Value HealthJan-Feb2004713651

14720129

[8]

Bergman

Hjelmgren

Ortqvist

Wisloff

Kristiansen

Hogberg

Cost-effectiveness analysis of a universal vaccination programme with the 7-valent pneumococcal conjugate vaccine (PCV-7) in Sweden

Scand J Infect Dis20084097219

18712627

[9]

Salo

Sintonen

Nuorti

Linna

Nohynek

Verho

Economic evaluation of pneumococcal conjugate vaccination in Finland

Scand J Infect Dis20053711–1282132

16308215

[10]

Tilson

Usher

Butler

Fitzsimons

O’Hare

Cotter

Economic evaluation of a universal childhood pneumococcal conjugate vaccination strategy in Ireland

Value HealthSep-Oct2008115898903

18489504

[11]

Hubben

Bos

Glynn

van der Ende

van Alphen

Postma

Enhanced decision support for policy makers using a web interface to health-economic models—illustrated with a cost-effectiveness analysis of nation-wide infant vaccination with the 7-valent pneumococcal conjugate vaccine in the Netherlands

Vaccine520072518366978

17360082

[12]

Chuck

Jacobs

Tyrrell

Kellner

Pharmacoeconomic evaluation of 10- and 13-valent pneumococcal conjugate vaccines

Vaccine720102833548590

20554066

[13]

Rozenbaum

Sanders

van Hoek

Jansen

van der Ende

van den Dobbelsteen

Cost effectiveness of pneumococcal vaccination among Dutch infants: economic analysis of the seven valent pneumococcal conjugated vaccine and forecast for the 10 valent and 13 valent vaccines

BMJ2010340c2509

20519267

[14]

Giglio

Micone

Gentile

The pharmacoeconomics of pneumococcal conjugate vaccines in Latin America

Vaccine9201129Suppl. 3C3542

21896351

[15]

Castaneda-Orjuela

Alvis-Guzman

Velandia-Gonzalez

De la Hoz-Restrepo

Cost-effectiveness of pneumococcal conjugate vaccines of 7, 10, and 13 valences in Colombian children

Vaccine320123011193643

22266291

[16]

Aljunid

Abuduxike

Ahmed

Sulong

Nur

Goh

Impact of routine PCV7 (Prevenar) vaccination of infants on the clinical and economic burden of pneumococcal disease in Malaysia

BMC Infect Dis201111248

21936928

[17]

Global Alliance for Vaccines and Immunization

Pneumococcal vaccine support; 2011Available from: http://www.gavialliance.org/support/nvs/pneumococcal/ [cited 2011 October 10]

[18]

World Health Organization. WHO-UNICEF estimates of dtp3 coverage; 2012

Available from: http://apps.who.int/immunization monitoring/en/globalsummary/timeseries/tswucoveragedtp3.htm [updated 14 July 2012; cited 2012 December 20]

[19]

World Health Organization. Target Product Profile for the Advance Market Commitment for Pneumococcal Conjugate Vaccines; 2008

Available from: http://www.who.int/immunization/sage/target product profile.pdf [cited 2011 December 2]

[20]

Guidelines Development Working Group

Thai Health Technology Assessment Guideline2008

Bangkok

[21]

Rhodes

Hyder

Peruski

Fisher

Jorakate

Kaewpan

Antibiotic use in Thailand: quantifying impact on blood culture yield and estimates of pneumococcal bacteremia incidence

Am J Trop Med Hyg820108323016

20682872

[22]

2009 National Disease Surveillance (Report 506) Nonthaburi: Bureau of Epidemiology, Department of Disease Control2011

Ministry of Public Health

Thailand

[23]

Chomchai

Kim

Kriengsoonthornkij

Sutchritpongsa

Manaboriboon

Ungcharoen

Active hospital-based epidemiologic surveillance for invasive pneumococcal disease and pneumonia burden in children in Bangkok, Thailand

5th Asian congress of pediatric infectious diseases2010

[24]

Central office for Health Information

Data Structure of Inpatient 2006–20102011

Thailand. Bangkok

[25]

Working group of Burden of Disease Project

Burden of disease and injuries in Thailand 2004

Nonthaburi: International Health Policy Program2007

Ministry of Public Health

Thailand

[26]

Prapasiri

Jareinpituk

Keawpan

Chuxnum

Baggett

Thamathitiwat

Epidemiology of radiographically-confirmed and bacteremic pneumonia in rural Thailand

Southeast Asian J Trop Med Public Health7200839470618

19058610

[27]

Hsieh

Hsueh

Lee

Huang

Clinical manifestations and molecular epidemiology of necrotizing pneumonia and empyema caused by Streptococcus pneumoniae in children in Taiwan

Clin Infect Dis320043868305

14999627

[28]

Netsawang

Punpanich

Treeratweeraphong

Chotpitayasunondh

Invasive pneumococcal infection in urban Thai children: a 10-year review

J Med Assoc Thai11201093Suppl. 5S612

21298830

[29]

Pavia

Bianco

Nobile

Marinelli

Angelillo

Efficacy of pneumococcal vaccination in children younger than 24 months: a meta-analysis

Pediatrics620091236e110310

19482744

[30]

Srifeungfung

Tribuddharat

Comerungsee

Chatsuwan

Treer-authanaweeraphong

Rungnobhakhun

Serotype coverage of pneumococcal conjugate vaccine and drug susceptibility of Streptococcus pneumoniae isolated from invasive or non-invasive diseases in central Thailand, 2006–2009

Vaccine42010281934404

20199759

[31]

Baggett

Peruski

Olsen

Thamthitiwat

Rhodes

Dejsirilert

Incidence of pneumococcal bacteremia requiring hospitalization in rural Thailand

Clin Infect Dis3200948Suppl. 2S6574

19191621

[32]

Phongsamart

Srifeungfung

Dejsirilert

Chatsuwan

Nunthapisud

Treerauthaweeraphong

Serotype distribution and antimicrobial susceptibility of S. pneumoniae causing invasive disease in Thai children younger than 5 years old, 2000–2005

Vaccine12007257127580

17092618

[33]

Levine

Dejsirilert

Sangsuk

Chantra

Feikin

Dowell

Serotypes and antimicrobial resistance of streptococcus pneumoniae in Thailand 2002–2004

Pediatr Infect Dis J220062521768

16462300

[34]

Ruckinger

Dagan

Albers

Schonberger

von Kries

Immunogenicity of pneumococcal conjugate vaccines in infants after two or three primary vaccinations: a systematic review and meta-analysis

Vaccine122011295296006

21933696

[35]

Lloyd

Patel

Scott

Runge

Claes

Rose

Cost-effectiveness of heptavalent conjugate pneumococcal vaccine (Prevenar) in Germany: considering a high-risk population and herd immunity effects

Eur J Health Econ2200891715

17333089

[36]

Centre for Health Equity Monitoring

Outpatient database 2005–20062011

Thailand Phitsnulok: Faculty of Medicine, Naraesuan University

Thailand

[37]

Ut-Ang

Reiewpaiboon

Cost of logistics of vaccines in the expanded programme on immunization in Thailand2009

Mahidol University

Bangkok

[38]

Statistical Forecasting Bureau

2009 Household Socio - Economic Survey2010

National Statistical Office of Thailand

Bangkok

[39]

Health Utilities Inc. Questionnaire Development, Translations and Support; 2008

Available from: http://www.healthutilities.com/ [cited 2010 March 2]

[40]

Kulpeng

Sornsrivichai

Chongsuvivatwong

Rattanavipapong

Leela-havarong

Cairns

Variation of health-related quality of life assessed by caregivers and patients affected by childhood infections

Nonthaburi: Health Intervention and Technology Assessment Program (HITAP)2012

Ministry of Public Health

Thailand

[41]

International Emerging Infections Program

An active, population-based surveillance for pneumonia requiring hospitalization2009

Nonthaburi Global Disease Detection Regional Center, Thailand MOPH-US Centers for Disease Control and Prevention Collaboration

[42]

Wibulpolprasert

The need for guidelines and the use of economic evidence in decision-making in Thailand: lessons learnt from the development of the national list of essential drugs

J Med Assoc Thai6200891Suppl. 2S13

19255984

[43]

Uruena

Pippo

Betelu

Virgilio

Giglio

Gentile

Cost-effectiveness analysis of the 10- and 13-valent pneumococcal conjugate vaccines in Argentina

Vaccine720112931496372

21621575

[44]

Tyo

Rosen

Zeng

Yap

Pwee

Ang

Cost-effectiveness of conjugate pneumococcal vaccination in Singapore: comparing estimates for 7-valent, 10-valent, and 13-valent vaccines

Vaccine920112938668694

21745516

[45]

Siti Zulaina

Jureen

Tee

WSN

Lin

RVTP

Laboratory-based surveillance system of invasive pneumococcal diseases in Singapore: a nationwide monitoring of vaccine serotype coverage

Epidemiol News BullApr-Jun2010362405

[46]

Butler

McIntyre

MacIntyre

Gilmour

Howarth

Sander

The cost-effectiveness of pneumococcal conjugate vaccination in Australia

Vaccine32004229–10113849

15003641

[47]

McIntosh

Conway

Willingham

Lloyd

The cost-burden of paediatric pneumococcal disease in the UK and the potential cost-effectiveness of prevention using 7-valent pneumococcal conjugate vaccine

Vaccine62003211920256472

[48]

Bos

Rumke

Welte

Postma

Epidemiologic impact and cost-effectiveness of universal infant vaccination with a 7-valent conjugated pneumococcal vaccine in the Netherlands

Clin Ther1020032510261430

14667962

[49]

Madhi

Adrian

Kuwanda

Jassat

Jones

Little

Long-term immunogenicity and efficacy of a 9-valent conjugate pneumococcal vaccine in human immunodeficient virus infected and non-infected children in the absence of a booster dose of vaccine

Vaccine32007251324517

17023095

Fig. 1

Markov model used for assessing costs and outcomes of pneumococcal conjugate vaccine (PCV) vaccination compared to ‘no vaccination’. The structure of the ‘PCV’ node is identical to the ‘no vaccination’ node and is thus omitted.

Fig. 2

Predicted numbers of life-time pneumococcal disease cases and deaths averted due to vaccination with 10- and 13-valent pneumococcal conjugate vaccines (PCV10 and PCV13) by clinical syndrome and age at entry to the cohort. (A) pneumococcal meningitis; (B) pneumococcal bacteremia; (C) all-cause pneumonia; (D) all-cause acute otitis media.

Fig. 3

Cost-effectiveness acceptability curves for 10- and 13-valent pneumococcal conjugate vaccines (PCV10 and PCV13), and ‘no vaccination’. (A) 3 + 1 schedule with indirect vaccine effects; (B) 2 + 1 schedule with indirect vaccine effects; (C) 3 + 1 schedule without indirect vaccine effects; (D) 2 + 1 schedule without indirect vaccine effects.

Fig. 4

Threshold analysis for maximum per-dose price for 10- and 13-valent pneumococcal conjugate vaccines (PCV10 and PCV13) to achieve cost-effective (incremental cost-effectiveness ratio (ICER) = THB 100,000) or cost-saving (ICER = THB 0). Current price per dose: THB 1440 for PCV10; THB 1930 for PCV13.

Table 1

Input parameters used in the model.

Parameter description	Distribution	Mean	SE	References
Epidemiological parameters
Proportion of bacterial meningitis due to S. pneumoniae	Beta	0.14	0.03	Meta analysis [23,24]
Epilepsy after pneumococcal (Pnc.) meningitis	Beta	0.10	0.06	[28]
Hearing loss after Pnc. meningitis	Beta	0.03	0.03	[28]
Neurodevelopmental impairment after Pnc. meningitis	Beta	0.34	0.09	[28]
Death after Pnc. meningitis	Beta	0.03	0.03	[28]
Death after Pnc. bacteremia	Beta	0.08	0.04	[28]
Necrotizing pneumonia after Pnc. pneumonia^a	Beta	0.18	0.05	[27]
Death after hospitalized pneumonia	Beta	0.01	0.00	[22]
Hearing loss after AOM	Beta	0.05	0.00	[25]
Risk ratio of mortality compared to general population				[25]
Epilepsy		1.01–1.14^b
Hearing loss		1.00–1.01^b
Neurodevelopmental impairment		5.16–7.17^b
Chronic lung		1^b
Baseline vaccine parameters
Vaccine efficacy (PCV7; 3 + 1 schedule)
IPD caused by vaccine serotype	Normal	89.00%	3.57%	[29]
Clinical pneumonia	Beta	6.00%	2.30%	[3]
AOM	Normal	6.00%	1.53%	[29]
Vaccine serotype coverage in Thais
PCV7 serotype coverage in aged <5	Normal	67.60%	5.36%	Meta analysis [30–33]
PCV10 serotypes coverage in aged <5	Normal	70.60%	5.66%	Meta analysis [30–33]
PCV13 serotypes coverage in aged <5	Normal	86.80%	4.03%	Meta analysis [30–33]
PCV7 serotypes coverage in aged ≥5	Normal	38.09%	2.29%	Meta analysis [30–33]
PCV10 serotypes coverage in aged ≥5	Normal	43.71%	3.00%	Meta analysis [30,31,33]
PCV13 serotypes coverage in aged ≥5	Beta	60.19%	4.69%	[30]
Serotypes coverage US				[35]
PCV7 serotypes coverage in aged 10 to 39	Not varied	71.30%
PCV7 serotypes coverage in aged 40 to 64	Not varied	65.40%
PCV7 serotypes coverage in aged ≥65	Not varied	69.70%
% IPD fall among unvaccinated group in US				[4]
% fall among who aged 20 to 39	Beta	40.00%	4.59%
% fall among who aged 40 to 64	Beta	14.00%	4.59%
% fall among who aged ≥65	Beta	29.00%	3.57%
Cost parameters (THB)
Vaccine costs
PCV10 cost per dose	Not varied	1440		GlaxoSmithKline (Thailand)
PCV13 cost per dose	Not varied	1930		Pfizer (Thailand) Limited
Delivery cost per dose	Not varied	5% of vaccine price		[37]
Direct medical costs
Cost per episode
Meningitis aged ≥14	Gamma	63,775	20,830	[24]
Meningitis aged 15 to 59	Gamma	59,210	15,570	[24]
Meningitis aged ≥60	Gamma	31,980	15,260	[24]
Bacteremia aged ≥14	Gamma	14,120	4587	[24]
Bacteremia aged 15 to 59	Normal	22,120	743	[24]
Bacteremia aged ≥60	Gamma	22,440	5372	[24]
Hospitalized pneumonia aged ≤14	Normal	9099	46	[24]
Hospitalized pneumonia aged 15 to 59	Normal	23,952	122	[24]
Hospitalized pneumonia aged ≥60	Normal	31,948	278	[24]
Non-hospitalized pneumonia aged ≤14	Normal	39	2	[36]
Non-hospitalized pneumonia aged 15 to 59	Normal	103	5	[36]
Non-hospitalized pneumonia aged ≥60	Normal	98	5	[36]
AOM aged ≤14	Normal	350	7	[36]
AOM aged 15 to 59	Normal	520	7	[36]
AOM aged ≥60	Normal	764	17	[36]
Cost per year
Epilepsy aged <14	Gamma	3962	475	[36]
Epilepsy aged 15 to 59	Normal	1600	21	[36]
Epilepsy aged ≥60	Gamma	1672	85	[36]
Hearing loss aged ≤14	Gamma	896	385	[36]
Hearing loss aged 15 to 59	Gamma	838	48	[36]
Hearing loss aged ≥60	Gamma	1312	123	[36]
Neurodevelopmental impairment aged ≤14	Gamma	3582	2333	[36]
Neurodevelopmental impairment aged 15 to 59	Gamma	936	72	[36]
Neurodevelopmental impairment aged ≥60	Gamma	5811	2892	[36]
Chronic lung aged ≤14	Gamma	1404	1404	[36]
Chronic lung aged 15 to 59	Normal	3306	62	[36]
Chronic lung aged ≥60	Normal	3636	31	[36]
Direct non-medical costs^c				Primary data collection
Meningitis (per episode)		15,485
Bacteremia (per episode)		9987
Hospitalized pneumonia (per episode)		5674
Non-hospitalized pneumonia (per episode)		527
AOM (per episode)		527
Epilepsy (per year)		4489
Hearing loss (per year)		868
Neurodevelopmental impairment (per year)		17,548
Chronic lung (per year)		7133
Age-specific productivity loss (per day)				[38]
15–29	Not varied	196
30–39	Not varied	409
40–59	Not varied	571
60–69	Not varied	246
70–79	Not varied	98
Utility parameters (using HUI3)				Primary data collection
Meningitis	Beta	0.96	0.00
Bacteremia	Beta	0.99	0.00
Pneumonia	Beta	0.99	0.00
AOM	Beta	1.00	0.00
Epilepsy	Beta	0.64	0.07
Hearing loss	Beta	0.55	0.06
Neurodevelopmental impairment
Mild mental retardation	Beta	0.69	0.07
Severe mental retardation	Beta	0.10	0.11
Mental retardation + epilepsy	Normal	0.00	0.09
Chronic lung disease	Beta	0.59	0.06

Assuming all necrotizing pneumonia cases would develop chronic lung disease.

Risk ratio of mortality varied by age

Including travel costs, foods, accommodation, informal care and special education, each component is gamma distributed.

Table 2

Incremental cost effectiveness ratios (ICER, in THB/QALY) classified by vaccination schedules and inclusion of indirect vaccine effects.

	PCV10 vs. No vaccine	PCV13 vs. No vaccine
2 + 1 schedule with indirect effects
Incremental cost (THB)	4178	5593
Incremental LYs	0.00674	0.00898
Incremental QALYs	0.00804	0.01061
Episode averted	0.01867	0.02501
Death averted	0.00200	0.00275
ICER per QALY gained (THB/QALY)	519,399	527,378
3 + 1 schedule with indirect effects
Incremental cost (THB)	5658	7576
Incremental LYs	0.00726	0.00967
Incremental QALYs	0.00870	0.01147
Episode averted	0.02030	0.02723
Death averted	0.00217	0.00299
ICER per QALY gained (THB/QALY)	650,087	660,662
2 + 1 schedule without indirect effects
Incremental cost (THB)	4492	6026
Incremental LYs	0.00212	0.00261
Incremental QALYs	0.00328	0.00404
Episode averted	0.00469	0.00577
Death averted	0.00007	0.00009
ICER per QALY gained (THB/QALY)	1,368,072	1,490,305
3 + 1 schedule without indirect effects
Incremental cost (THB)	6001	8048
Incremental LYs	0.00229	0.00282
Incremental QALYs	0.00358	0.00440
Episode averted	0.00508	0.00625
Death averted	0.00008	0.00010
ICER per QALY gained (THB/QALY)	1,677,379	1,830,716

LY = life year, QALY = quality-adjusted life year.