Willingness to pay for vaccines in China: A systematic review and single-arm Bayesian meta-analysis

ABSTRACT

The effective implementation of vaccination heavily depends on the society’s willingness to pay (WTP). There is currently a dearth of comprehensive evidence about WTP for vaccines in China. This systematic review aims to review studies on the WTP for vaccines, to summarize factors affect WTP in China. Base-case analysis and Sensitivity analysis of WTP for every vaccine were estimated via single-arm Bayesian meta-analysis. A total of 28 studies were included for systematic review. The point estimates and 95% Credible Interval of pooled WTP for influenza and HPV (9-valent) vaccine were $27.409 (23.230, 31.486), $464.707 (441.355, 489.456). Influencing factors to WTP were age, income, peer influence, health condition and etc. Future research should give focus to improving sample representativeness and survey tool, conducting intervention trials, identifying effective methods to promote WTP.

Introduction

Vaccination stands as one of the most successful public health interventions in human history.^1–3 The widespread application of vaccines has significantly reduced the incidence of Vaccine Preventable Disease,⁴ averting between 2 and 3 million deaths annually.⁵ As a public good, vaccine exhibits externality and is subject to the “free-rider” effect. Consequently, many countries and regions have adopted government-driven efforts to ensure the mandatory provision of vaccines, such as the inclusion of the Haemophiles influenzae type b vaccine, Rotavirus vaccine, and Human papillomavirus (HPV) vaccine in immunization programs.^6–8 China has different vaccination financing mechanisms than other countries and regions. Vaccines are classified into Class I (financed by the government) and Class II (covered by individuals) based on the source of financing.⁹ Vaccines of Class I are mainly vaccines of the national immunization program, the cost of which is fully covered by the government, such as hepatitis B vaccine, Bacille Calmette-Guérin vaccine and polio vaccine for children, and vaccine against hemorrhagic fever with renal syndrome, anthrax vaccine and human leptospirosis vaccine for high-risk populations in key regions. The vaccines mentioned above are categorized as Class II in China, resulting in higher vaccination costs for Chinese consumers compared to those in countries or regions where vaccines are included in the immunization programs. The public health system in China has an adequate capacity to deliver inoculation services of Class II vaccines, however, the coverage is inadequate.¹⁰

The successful implementation of vaccination programs often hinges on multitude of factors, with the societal Willingness to Pay (WTP) being particularly crucial. WTP refers to the monetary value an individual would forgo to maintain their current level of utility (satisfaction) when their health is enhanced by a specific health product or service.¹¹ WTP can be measured through Revealed Preference (RP) and Stated Preference (SP) methods. The former comprises Market Data approach and Experiment method, while the latter includes Contingent Valuation Method (CVM) and Discrete Choice Experiment (DCE).¹² CVM can report consumers’ WTP for vaccine through questionnaires and Iterative Bidding Games. DCE can indicate vaccine attribute levels’ preference and marginal WTP. Studying vaccine’s WTP may reveal the trade-off between its benefits of infectious disease prevention and economic costs,^13–15 which provides evidence of health economics for determining vaccine prices and catering to consumer demand. Ultimately, it contributes to advancing immunization programs, which plays an imperative role in protecting the health and lives of the population.

The outbreak of the novel coronavirus in 2020 has led to a significant focus on infectious diseases. Vaccines, being a highly efficient public good for the prevention or treatment of infectious diseases, have become the focus of societal attention. Research related to WTP has been on the rise. Currently, there is a lack of comprehensive evidence on the WTP for vaccines in China, and the overall situation of WTP for various vaccines at the national level remains unclear. Additionally, there has been no study reviewed the factors that influenced WTP for vaccines in China. Hence, this study conducted a systematic review and single-arm Bayesian meta-analysis to estimate WTP for vaccines in China, aiming to provide high-quality evidence to aid decision-making and program implementation in China and globally, thereby enhancing coverage. Therefore, the objectives of this study are to (1) quantitatively synthesize the average effect of WTP for each vaccine among the Chinese population, (2) summarize the influencing factors of WTP.

Methods

Study design, protocol, and registration

The present study is a systematic review and meta-analysis. The study protocol was registered on the Open Science Framework platform (https://osf.io/av3d2/).

Search strategy and data sources

We searched PubMed, Embase, The Cochrane Library, and Web of Science, China National Knowledge Infrastructure, Chinese Biomedical Literature, WanFang and Weipu databases from their inception to May 17, 2024. The whole process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines¹⁶ and the Cochrane Handbook for Systematic Reviews of Interventions¹⁷ (Checklist was provided in Supplemental file: Table S1).

Two researchers (YL, ZWL) and one senior researcher (LRZ) collaborated to negotiate and ultimately establish search strategies. The major concepts employed in this study encompassed vaccine, WTP. The study should be empirical study of WTP for vaccines among Chinese population. We incorporated the corresponding index terms and free text terms for “vaccine,” “WTP,” and “China.” We defined WTP as the maximum price individuals would be willing to pay for vaccines or the minimum reimbursable price they would accept. Detailed strategies for each database were listed in Supplemental file: Table S2.

Inclusion and exclusion criteria

Studies meeting the following criteria were included: (1) the research subjects were the Chinese population, (2) the study was conducted in China, included mainland China, Hongkong, Macau and Taiwan, (3) the research was empirical study on WTP for vaccines among Chinese population, including observational studies and field experiments, and (4) the language of research was Chinese or English. Seven exclusion criteria applied: (1) the subject were non-Chinese populations, (2) the study location was outside of China, (3) the vaccination targets were non-human, (4) the study were theoretical studies, literature reviews, lectures, comments, and conference abstracts related to WTP for vaccines in China, (5) the study did not report WTPs for vaccines, (6) non-Chinese and non-English language literature, (7) duplicate publications and literature for which full text were unavailable.

Study selection

Two researchers (YL, ZWL) independently screened and extracted data, with cross-verification conducted. In case of disagreement, the senior researcher (LRZ) was consulted for assistance in judgment. During the screening process, two researchers (YL, ZWL) reviewed titles and abstracts independently using the eligibility criteria. Subsequently, the full text was read by them to determine final inclusion. After studies were selected, their references were searched for subject-related literature.

Data extraction

The final literature information was filled into the data extraction form. Content extracted included: (1) basic information: title, first author, publication year, study region, (2) data information: study design, sample size, type of vaccine, method of WTP measurement, (3) results: (marginal) WTP for vaccine, attributes of vaccine preference, factors influencing the WTP, the statistical model and conclusions. When the information involved in the analysis cannot be obtained from the article, we contacted the corresponding author to try to obtain the information.

Quality assessment

Two researchers (YL, ZWL) used the checklist for prevalence studies from the Hoy risk of bias assessment tool¹⁸ and a set of criteria specific for assessing WTP studies¹⁹ in quality assessment. Each item was scored 1 (high risk) or 0 (low risk), for a total bias score of 14 points. The lower the score, the lower the risk of bias and the better the quality of the study. Disagreements in the quality scores were resolved by discussion between the two reviewers (YL, ZWL) for the final decision. For ambiguous reports, a score of 0.5 was assigned.

Data analysis

Descriptive statistical analysis

Descriptive statistical analysis was performed on the included literature. All WTP values were calculated at 2023 US dollar rates for synthesis and comparison. First, values in RMB as the currency were converted to US dollars based on Purchasing Power Parity exchange rates of the year in which the study was conducted, and then they were adjusted to 2023 US dollar values using the Consumer Price Index (unit: percent change from year ago, seasonally adjusted) of Medical Care Commodities in the United States. The data were acquired from the Organization for Economic Co-operation and Development and the U.S. Bureau of Labor Statistics, respectively (Supplementary file: Table S5-S6).^20,21 When the year in which the study was conducted was not available, the publication year were utilized.

Single-arm Bayesian meta-analysis of WTP

Single-arm Bayesian meta-analysis is applicable when fewer studies are included, all subjects are in the same group, and no other control or intervention groups are established. The data could be binary, continuous, ordinal, count, survival data, etc. Studies with same unit of WTP measurement for similar vaccines were included in the meta-analysis. When WTP was reported in terms of median, interquartile range, etc., it was converted to Mean±SD using Luo’s and Wan’s methods.^22,23

The Bayesian approach to the baseline results relied on Markov chain Monte Carlo methods (MCMC) for estimating the posterior distribution. Convergence of MCMC sampling was diagnosed based on trace plots and density plots. Inter-study heterogeneity was assessed using ${\tau }^2$. Random-effect models were employed to estimate the pooled WTP, with point estimates and 95% Credible Intervals (CI).

Sensitivity analysis was carried out based on the Divergence Restricting Conditional Tesselation (DIRECT). DIRECT aims at a covering of the conditional’s parameter space while bounding the divergence,²⁴ allowing for the efficient and reliable estimation of Maximum-likelihood (ML), Maximum-A-Posteriori, Marginal Posterior Distributions, and etc. without relying on MCMC. The prior distributions of $\mathit{\tau }$ in this section aligned with the settings of MCMC. The results were presented as a joint posterior density plot of $\mathit{\mu }$(effect parameter) and $\mathit{\tau }$ (heterogeneity parameter), where darker shades indicated higher probability densities. The red line delineated the 2-dimensional credible region, the green one represented the marginal posterior median and 95% CI, the blue one denoted the constrained posterior mean effect. The red cross (十) indicated the posterior mode, and the pink cross (×) signified the ML. The baseline results and the sensitivity analysis were presented, respectively, using R2jags, mcmcplot and bayesmeta with R (4.3.2).

Results

Study selection

The preliminary review yielded a total of 382 pertinent articles, including 243 in English and 139 in Chinese. Following a meticulous screening process, a total of 28 studies were ultimately selected in systematic review, 4 studies were selected in meta-analysis (Figure 1).

Figure 1.

PRISMA flow chart of study selection process for studies included in the systematic review.

Data extraction and synthesis

Study characteristics

This review included a total of 28 studies (Table 1). Fewer studies were conducted between 2014 and 2019 and the majority of these studies were based on CVM. In 2020 and onwards, there was an increase in the number of studies based on both methods (Figure 2). The study region was re-labeled according to the regional division of the National Bureau of Statistics in China.⁵² Five studies were multicenter studies, seven were done in the eastern region, and two were conducted in the western region. The types of vaccines included COVID-19 vaccine (9 studies) and boosters (3 studies), HPV vaccine (6 studies), influenza vaccine (4 studies), pneumococcal (conjugate) vaccine (2 studies), etc. The studies of COVID-19 vaccine (boosters) were mostly conducted among ordinary people, the studies of the HPV vaccine were mostly focus on women or students. Two studies on the pneumococcal vaccine both targeted parents as participants. In influenza vaccine studies, Lai et al.⁴³ categorized the respondents as children, patients with chronic diseases and the elderly, and one study⁴⁴ included parents of children as participants.

Table 1.

Basic information of studies included in systematic review (n = 28).

Type of vaccine	Vaccine	Region	Method	Participants	Valid sample size	Study
COVID-19 vaccine	COVID-19 vaccine	National	CVM	Ordinary people	3541	Lin Y25
	COVID-19 vaccine	National	DCE	Ordinary people	1236	Dong D26
	COVID-19 vaccine	Eastern	CVM	Internal migrants	2126	Han K27
	COVID-19 vaccine	Multicenter	CVM	Ordinary people	1188	Qin W28
	COVID-19 vaccine	National	CVM	Ordinary people	2058	Wang J29
	COVID-19 vaccine	National	DCE	Ordinary people	1066	Chen Y30
	COVID-19 vaccine	National	CVM	Ordinary people	2450	Xiao J31
	COVID-19 vaccine	National	DCE	Ordinary people	1576	Xiao J32
	COVID-19 vaccine	National	DCE	The middle-aged and elderly people	802	Li X33
	COVID-19 vaccine boosters	National	CVM	Ordinary people	1145	Lai X34
	COVID-19 vaccine boosters	National	CVM	Ordinary people	543	Zhou H35
	COVID-19 vaccine boosters	Multicenter	CVM	Hypertensive patients	453	Yan Y36
HPV vaccine	HPV vaccine	Eastern	DCE	Parents of junior high school girls	995	Zhu S37
	HPV vaccine	Eastern	DCE	Female college students	850	Wang Y38
	HPV vaccine	National	CVM	Women in health-related fields	15969	Lu X39
	HPV vaccine	Multicenter	CVM	Medical students	742	Zhou L40
	HPV vaccine	Eastern	CVM	Parents of children	2355	Zhou W41
			DCE
	HPV vaccine (9-valent)	Central	CVM	High school girls	782	Zhu L42
Influenza vaccine	Influenza vaccine	National	CVM	Children, patients with chronic diseases, the elderly	12252	Lai X43
	Influenza vaccine	National	DCE	Parents of children	1206	Jiang M44
	Influenza vaccine	Western	DCE	The elderly	144	Jiang M45
Other vaccines	Herpes zoster vaccine	Eastern	DCE	The elderly	170	Zhang H46
	Pneumococcal conjugate vaccine	Western	CVM	Parents of children	1254	Li L47
	Group B Streptococcus vaccine	Multicenter	DCE	Pregnant women	354	Du J48
	Hepatitis B vaccine	Eastern	DCE	Ordinary people	266	Guo N49
	HFMD vaccine	Eastern	CVM	Parents or grandparents of children	3626	Cheng L50
	DTaP-HBV-IPV-Hib hexavalent vaccine	National	CVM	Parents of children	581	Huang A51
Multi-vaccines	Pneumococcal conjugate vaccine (7-valent); Influenza vaccine	Multicenter	CVM	Parents of children	1924	Hou Z13

Figure 2.

Comparison of yearly number of included studies based on different methods in China. This dumbbell chart depicted the publication year and number of included studies based on CVM and DCE methods. The pink dots represented the number of studies based on CVM, and the blue dots represented the number of studies based on DCE. In 2020, the same number of studies were published based on both methods.

Study quality

All studies scored 0–7.5 with a mean of 3.66 (SD: 1.95). The majority scored 4.5 (n = 5, 17.86%). 16 (57.14%) studies had a weak sample representation. There were several studies where the sample sampling method was not indicated in the article. 12 (42.86%) studies where the sample response rate was not explicitly given. Supplementary Table S3 illustrates the scores for each item of included studies.

Levels and preferences of WTP for vaccines

Average for vaccine WTP

Among HPV vaccine-related studies, one⁴² for the nine-valent vaccine (per dose) measured an average WTP of $381.59; Zhou LR et al.⁴⁰ subdivided vaccines by producing place and valence, and mean WTP for bivalent, quadrivalent, and nine-valent vaccines respectively were $(228.03–423.15), $(466.19–555.06), and $(717.92–814.45); and one study³⁹ conducted a measurement of average WTP for vaccines, irrespective of their origin and price, which was found to be $500.82. Other reports are in Supplemental file: Table S4.

Marginal level and preference analysis of vaccine WTP

Eleven studies’ results of marginal WTP and preference for vaccine were summarized as Supplemental file: Table S7, focusing on vaccine’s own attributes and service-related features. Attributes pertaining to vaccine’s own attributes included efficacy, effectiveness, safety (including side effects and safety level), duration of protection and origin. The service-related features of the vaccine included waiting time and service time for vaccination, convenience of vaccination, vaccination site and distance to it.

Bayesian meta-analysis of pooled WTP for vaccines

As shown in Table 2, a total of 4 studies were included in the meta-analysis, including influenza vaccine (13,485 participants) and 9-valent HPV vaccine (2,071 participants). The included studies’ WTPs were assumed to follow an approximately normal distribution, as per the central limit theorem. The meta-analysis included only a limited number of research; therefore, a weakly informative prior for $\mathit{\tau }$ was specified, specifically a Half Normal (HN) with a Scale of 1. The Markov chains for each parameter converged well (Figure 3). The point estimates and 95% CIs of $\mathit{\tau }$ were shown in Supplemental file: Table S8. The point estimates and 95% CIs of pooled WTP for influenza vaccine and HPV vaccine (9-valent) were $27.409 (23.230, 31.486) and $464.707 (441.355, 489.456).

Table 2.

Basic information of studies included in meta-analysis.

Vaccine	Publication year	Year of Implementation	Sample size for modelling	Mean	SD	Participants	Unit of WTP measure	Study
Influenza vaccine	2014	2013	2019	21.46	12.88	Parents of children	Full-immunization	Hou Z13
Influenza vaccine	2020	2019	6524	33.39	10.27	Children (caregiver)	Full-immunization	Lai X43
Influenza vaccine	2020	2019	1603	25.27	14.06	Patients with chronic diseases	Full-immunization	Lai X43
Influenza vaccine	2020	2019	3339	23.07	14.88	The elderly	Full-immunization	Lai X43
HPV 9-valent vaccine	2021	2018	731	381.59	177.75	High school girl	1-dose	Zhu L42
HPV imported 9-valent vaccine	2022	2020–2021	670	814.45	517.03	Medical student	1-dose	Zhou L40
HPV domestic 9-valent vaccine	2022	2020–2021	670	717.92	446.84	Medical student	1-dose	Zhou L40

Figure 3.

Plots of trace and density of MCMC. This group of figures reported the convergence of MCMC sampling for WTP of vaccines. Taking the influenza vaccine as an example, from top to bottom, the first three graphs were trace plots of the parameters; the latter three were density plots of the parameters.

The results derived from the DIRECT were similar to those based on MCMC. The point estimates and 95% CIs of the posterior median of $\mathit{\mu }$ for influenza vaccine and HPV vaccine (9-valent) were $25.285 (22.301, 28.156), $432.435 (413.906, 450.634). The joint posteriori density plots for $\mathit{\mu }$ and $\mathit{\tau }$ are shown in Figure 4, and detailed results were in Supplemental file: Table S9.

Figure 4.

Plots of joint posterior density. This group of joint posterior density plots reported the results of sensitivity analysis for the effect parameter and heterogeneity parameter based on DIRECT.

Factors influencing WTP for vaccines

Twenty-five studies analyzed influencing factors for vaccine’s WTP (Supplemental file: Table S10). Attributes in DCE studies were considered as influencing factors for WTP. After integrating and summarizing the evidence from included studies, the influencing factors were summarized as: demographic factors, economic factors, social factors, health conditions, attitudinal factors, health literacy, vaccine’s own attributes and service-related features. The results of 16 studies indicated that demographic factors influence WTP, followed by economic factors (such as individual’s ability to pay, income, and the region’s economy). Five studies^{13,27,35,46,48} showed social factors as significant influencers of WTP. Among the attitudinal factors, three^13,25,49 concluded that individuals with higher self-perceived severity of disease had higher WTP.

Among studies of average WTP, the studies that carried out single-factor analysis used Chi-square tests and Pearson correlation coefficient as statistical methods; among the studies that carried out multifactor analysis, one²⁸ employed interval regression, three^13,39,41 used multiple linear regression and six^{29,40,42,43,47,51} applied Tobit model. Eleven studies of marginal WTP were analyzed with multifactor analysis. Mixed logit model and Latent class model were employed in five^{26,30,44,45,49} and two^32,38 studies, respectively.

Discussion

This review included 28 studies of WTP for vaccines in China, including influenza vaccine, COVID-19 vaccine (and boosters), HPV vaccine, pneumococcal (conjugate) vaccine and etc. The pooled WTP for influenza vaccine and HPV vaccine (9-valent) among Chinese population were quantitatively synthesized by single-arm Bayesian meta-analysis. The study design, influencing factors for WTP, and statistical methods of included studies were summarized. The quality of the included literature was assessed to identify the strengths and weaknesses in order to provide a reference basis for future WTP studies.

The COVID-19 vaccine (and boosters), as well as the HPV vaccine and influenza vaccine were the focus of included studies. These are all Class II vaccines in China. However, the COVID-19 vaccine (and boosters) is covered by both the basic medical insurance and the government, which might be attributed to the large epidemiologic burden of the illness that can be averted by the COVID-19 vaccine in China. Without compulsory voluntary participation, the country may face a more serious public health catastrophe. Unlike other vaccines, HPV vaccine prevents cancers, including cervical, penile, oropharyngeal cancers and etc., which impose significant disease burdens. Furthermore, it has been marketed late in China. Simultaneously, the COVID-19 pandemic has brought greater awareness and interest in vaccine among the general population. Hence there has been a surge of publications pertaining to HPV vaccine in recent years. Pregnant women, children under age of 5, the elderly, and individuals with chronic and immunosuppressive diseases are more susceptible to suffering serious illness or complications following influenza virus infection.⁵³ COVID-19 impairs the human immune system.⁵⁴ After the COVID-19 pandemic, China faced more severe influenza outbreaks. This may be associated with increased research attention.

The pooled WTPs of influenza vaccine and HPV (nine-valent) vaccine are higher than the price of winning bids in centralized procurement.⁵⁵ Consumer surplus may provide an explanation for the phenomenon. In studies of influenza and HPV vaccine, the populations were susceptible groups or their parents, those who received comprehensive medical education. These may have a higher awareness of illness susceptibility or have a higher health responsibility.⁵⁶ Another possible explanation is the flaw of SP. All the methods used in the included studies to investigate WTP were SP. It estimates the economic behavior of consumers in a hypothetical market, whereas consumers were more cautious in real market settings: first, consumers need to weigh current costs against future long-term benefits;⁵⁷ second, the costs of vaccination include not only the monetary cost of the vaccine itself, but also the cost of searching before vaccination, the discomfort experienced during the vaccination, and any potential adverse effects that may occur post-vaccination. WTP will be overestimated when consumers have insufficient perception of these two aspects. Regardless, the result suggested that the price might not be the predominant obstacle hindering the expansion of influenza and HPV vaccination coverage.

The results of this review suggest that many factors drive WTP for vaccine, including demographic factors, economic factors, social factors, and etc. Demographic (such as occupation, educational background, marital status, and place of residence), economic and social factors may reflect the portfolios of individual assets, which are related to WTP. Social circumstances of customers played an important role in the WTP for vaccination, especially when they have limited knowledge about vaccines. If the correct vaccine knowledge can be conveyed through peers, medical staff or social media, it will be beneficial to increase WTP of vaccine and expand vaccine coverage. Attitudes and health literacy were also important factors to WTP, suggesting that government may begin with raising consumer awareness of vaccine-related diseases, with the aim of boosting coverages. Evidence on the natural and service-related attributes of vaccines suggested that vaccine manufacturers should prioritize safety, effectiveness, and the duration of protection. Additionally, service providers should adapt their approaches to enhance the convenience of services and procedures. It is worth exploring that Jiang M et al.⁴⁴ assessed the population’s preference for intramuscular injection and nasal spray. Although the result suggested that there was no significantly preferred mode of vaccination, vaccine hesitant and non-vaccine hesitant populations might differ in vaccination mode preferences. A considerable proportion of vaccine hesitant population refused or delayed vaccination for fear of intramuscular injection.^58,59 At present, most vaccines are intramuscular, which require management of controlled temperature chain.^60,61 Access to vaccines is then affected by the economic and geographical conditions. Vaccine Microarray Patches provide a solution to this problem.⁶² In order to eliminate vaccine hesitancy, improve accessibility, and expand coverage, advanced technologies need to be promoted. Determining a reasonable pricing strategy is one of the effective ways to promote technologies, which requires the support of relevant evidence on the WTP of Chinese consumers.

The strengths of this study are as follows: First, we conducted a systematic review and meta-analysis according to the PRISMA guidelines and the Cochrane Handbook for Systematic Reviews of Interventions, which ensured the standardization and transparency of our study. Second, we used a series of methodological quality assessment tools to assess the risk of bias of the included studies, pointed out the methodological deficiencies of included studies, and provided some guidance for future studies on WTP. Third, to ensure that robust effect estimates were obtained under small sample conditions, we used a meta-analysis under the Bayesian framework. When conducting Meta-analysis for sparse studies, the classical frequency-based framework often leads to bias estimates, whereas more robust posterior estimation can be obtained in the Bayesian framework via the application of weakly informative or uninformative prior distribution.

This review has some limitations. First, we were unable to report publication bias. We did not estimate the publication bias of the pooled WTP due to the fact that the power of test of funnel plots would be reduced when the number of studies was too small. Second, we did not include gray literature. This may have affected the comprehension of our data sources. However, considering that gray literature has not been peer-reviewed, including such studies may reduce the credibility of our results.

Meantime, the following key points should be taken into account for conducting research on WTP in the future based on the limitations of included studies. First, some studies had poorly representative samples. The choice of sample influences the WTP of the vaccine. It is necessary to improve the representativeness of the sample to the population by making the characteristics of the sample structure and the overall structure consistent when selecting the sample. Second, most studies did not conduct pilot studies and face-to-face interviews, which could lead to inappropriate questions and options being unable to be validated and adjusted. Especially in DCE, the perception of consumers and experts is equally important. Moreover, if the appropriate attributes and levels are not selected, it will also have a greater impact on the results, such as the attributes of the survey do not influence the consumers’ WTP. Therefore, it is necessary to design questionnaires through multiple means such as pilot studies, face-to-face interviews, expert consultations, and literature reviews. Third, all the included studies used SP to estimate WTP, which may have caused the price premium. To improve the accuracy of the estimation of WTP, SP and RP can be conducted together to analyze consumers’ preferences by combining behavior data with the stated choices under hypothetical scenarios in the survey.^63,64 Fourth, the included studies were cross-sectional and did not implement interventions. The purpose of empirical studies on WTP for vaccine is to understand the public’s WTP for vaccines, attribute preferences, and to find the crucial variables that influence WTP, thus providing a basis for improving coverage. Cross-sectional studies are only a description of the current situation, and how to influence people’s WTP needs to be supported by interventional studies.

Conclusion

This study found that the pooled WTPs of influenza and HPV (nine-valent) vaccines in China were higher than price of winning bids. Demographic, economic and social factors, attitude, health literacy, natural attributes and service-related attributes and etc. were important factors to consumer’s WTP. It is recommended that the government should increase the coverage by raising consumers’ knowledge of vaccine-related diseases. Vaccine manufacturers should focus on the safety, efficacy, and duration of protection of vaccine. Service providers should modify their strategies to improve the accessibility and efficiency of services and procedures. Future research should prioritize the improvement of sample representativeness and rationality of survey tool, execution of intervention studies, and identification of effective strategies to enhance WTP. These efforts will facilitate the successful implementation of vaccination programs.

Biographies

Liangru Zhou (PhD) is the lecturer at Beijing University of Chinese Medicine. Her main research area is the health technology assessment. Recent publication include: Cost-effectiveness of pit and fissure sealing at schools for caries prevention in China: A Markov modeling analysis. (2023); Preventive health behaviors among the elderly in China: Does social capital matter? (2023); Human papillomavirus vaccination at the national and provincial levels in China: a cost-effectiveness analysis using the PRIME model (2022); On Imported and Domestic Human Papillomavirus Vaccines: Cognition, Attitude, and Willingness to Pay in Chinese Medical Students (2022); HPV Vaccine Hesitancy Among Medical Students in China: A Multicenter Survey (2022).

Ruifeng Li (PhD) is the dean of School of Management of Beijing University of Chinese Medicine. He is a distinguished researcher of the National Academy of Traditional Chinese Medicine Development and Strategy, the vice chairman and secretary general of the Humanities and Management Science Branch of the China Association of Chinese Medicine, the member of the Chinese Health Economics Association and the deputy editor-in-chief of the “Blue Book of Traditional Chinese Medicine Culture.” His main research area is health policy evaluation. Recent publication include: Empirical analysis of the impact of internet penetration on residents’ health care expenditure based on provincial panel data (2024); Research on the coupling and coordination of health resource allocation and economic development under the integration of Beijing, Tianjin and Hebei (2023); Text analysis of grassroots Chinese medicine development policy in China from the perspective of policy tools (2023).

Funding Statement

This study was funded by Fundamental Research Funds for the Central Universities [grant number 2024-JYB-XJSJJ019].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

Y.L. conceptualized and designed the study, performed data extraction and analysis, drafted the initial manuscript, and revised the manuscript. Z.W.L. performed data extraction, validation and revised the manuscript. L.R.Z. and R.F.L. contributed to the conceptualization, design and revised the manuscript critically for important intellectual content. All authors approved the final manuscript as submitted.

Data availability statement

All data presented in the current work is available in the text and supplementary materials and can also be obtained from the authors upon request.

Ethical approval and consent to participate

As this study does not involve animal and patient experiments, the ethical approval and consent to participate are not necessary.

Supplemental material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2025.2454076