Role of universal and targeted recommendations for vaccines for sexually transmitted infections in the USA, China and Indonesia: a cross-sectional study

Abstract

Background

Vaccines for sexually transmitted infections (STIs) are in development and little is known about their future acceptance. The type of recommendations released by National Immunisation Technical Advisory Groups could influence vaccine uptake. This study aims to understand how universal and targeted wording affects theoretical acceptance to vaccines for four common STIs (chlamydia, gonorrhoea, syphilis and genital herpes) in the USA, China and Indonesia. We also aim to understand how universal and targeted wording may influence acceptance among those who engage in high-risk sexual behaviours.

Methods

A total of 1941 adults between the ages of 18 and 45 were included in the final sample. For each country, the sample was stratified by sexual behaviours, then the proportion of those who accepted each vaccine under universal or targeted wording was reported. χ² tests were used to assess differences, and the likelihood of vaccine acceptance was represented using prevalence ratios (PRs) from Poisson regression models.

Results

Overall, vaccine acceptance for each STI and in every country sampled tended to be lower among those who received targeted wording compared with universal wording. Previous knowledge of STIs significantly affected vaccine acceptance. In the USA, there was significant interaction between recommendation wording type and having heard of the STI; individuals who had previously heard of STIs were more likely to accept an STI vaccine, and that this acceptance varied significantly based on whether they received a universal (PR: 1.30, 95% CI: 1.05 to 1.62) versus targeted recommendation (PR: 2.45, 95% CI: 1.69 to 3.56).

Conclusions

Our results are important for future vaccine recommendations as vaccine hesitancy and refusal are on the rise globally. Thus, wording and education surrounding sexual health and STIs are factors that influence decision-making. Our research also highlights the importance of appealing to certain high-risk groups and providing informative language when releasing vaccine recommendations.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Sexually transmitted infection (STI) vaccines are currently under development, while vaccine hesitancy is on the rise, globally. Little research has been done to assess acceptance of STI vaccines and the effectiveness of vaccine recommendations targeted at high-risk populations.

WHAT THIS STUDY ADDS

• This study assessed attitudes towards STI vaccines in three countries; we found that targeted wording has little to no effect on acceptance.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• More research should be done to assess other methods of encouraging vaccine uptake. Targeted wording does not guarantee vaccine uptake and more needs to be done to address vaccine hesitancy for both emerging and routine vaccines.

Introduction

Annually, 1.29 billion people worldwide acquire a sexually transmitted infection (STI),¹ including 979 million cases of genital herpes, 151 million cases of chlamydia, 49.7 million cases of syphilis and 40.4 million cases of gonococcal diseases.¹ Mortality from STIs is high, including 204 000 deaths from congenital syphilis and 341 000 deaths from cervical cancer in 2016.² STI burdens vary greatly by country. In 2019, the rate of prevalent cases per 100 000 for all STIs excluding HIV among 15–49 years was 19 755 in China, 22 650 in Indonesia and 22 263 in the USA.³ The prevalence of genital herpes is highest in China (171 million prevalent cases) and far outweighs prevalence in the USA and Indonesia (56 million and 36 million, respectively).¹ Globally, new cases of STIs (excluding HIV) have increased by an estimated 2.2% each year, representing an increase from 486 million cases of STIs in 1990 to 769 million cases of STIs in 2019, indicating the need for interventions that could reduce the burden of STIs.¹

The core group theory of STIs posits that certain ‘core’ groups, including men who have sex with men (MSM), people who have multiple sexual partners, and those with inconsistent condom usage, can be at disproportionately higher risk of contracting and transmitting a given STI compared with the general population.⁴ Therefore, many countries have adopted intervention strategies that target these core groups to reduce STI rates.

National Immunisation Technical Advisory Groups (NITAGs) provide crucial guidance on immunisations, particularly for emerging vaccines.⁵ However, NITAG recommendations are different for each country. For example, in the USA, the Advisory Committee on Immunisation Practices distinguishes between ‘category A’ (vaccines broadly recommended) and ‘category B’ vaccines (vaccines that clinicians can decide to administer after a clinician–patient interaction).⁶ In China, based on the availability of vaccines and local burden of disease, the NITAG recommends vaccines to be ‘category 1’ (ie, publicly funded and mandated) or ‘category 2’ (ie, optional and requiring payment).⁷ These recommendations have the potential to influence the roll-out of vaccines as well as vaccine uptake. However, the role of NITAGs in vaccination decision-making is unclear because NITAGs have been rapidly established within the last ten years.⁵

Several vaccines against STIs are currently in clinical trials, including attenuated, subunit and DNA herpes vaccines, which are undergoing phase I or phase II clinical trials for therapy or prevention of genital herpes infections.^{8 9} A recombinant chlamydia vaccine, CTH522, has completed a phase I clinical trial.¹⁰ The existing meningitis B vaccine is being evaluated in a phase II clinical trial for its efficacy and safety in preventing gonorrhoea.¹¹ Additionally, the development of syphilis vaccines is still in the preclinical laboratory phase.^{8 12} Given that vaccines against several STIs have not been previously available, it is important to gauge the public’s interest in these emerging vaccines.

Past research on the acceptability of STI vaccines is limited,¹³ outside of research on human papillomavirus infection (HPV) and HIV vaccine acceptability.^{14 15} Acceptance of a vaccine may depend on several factors such as perceived risk of exposure to an STI, severity of infection, previous experience with STIs,^{13 16} or whether the infection can become chronic.¹⁷

Moreover, limited research exists regarding the role of NITAGs and governments in vaccine uptake. Past research has focused on the effectiveness of vaccine mandates in the global north, primarily in the USA and Europe. In different contexts and for certain groups, support for a mandatory vaccination varied substantially. For example, in Greece, 65.1% of healthcare workers thought that complete and updated vaccination should be mandated. However, in France, the proportion of general practitioners and paediatricians who felt that vaccines should be nationally mandated was only 42%.¹⁸ A 1000-person sample of parents of children under 5 in Canada found that 44% of parents felt that vaccination should ultimately be a matter of parental choice.¹⁸ In a US sample of 325 parents of children 9–17 years, a majority supported government-mandated vaccine requirements, but under half felt it was important to vaccinate their children against HPV.¹⁸

This cross-sectional study has two main objectives: (1) to determine if universal or targeted recommendations affect theoretical STI vaccine acceptance in the USA, China or Indonesia for four common STIs (chlamydia, gonorrhoea, genital herpes and syphilis) and (2) assess how universal versus targeted recommendation wording affects STI vaccine acceptance among high-risk groups: notably MSM, individuals with multiple concurrent sexual partners and individuals who use condoms infrequently.

Methods

Data collection

An online questionnaire was distributed to adults over 18 years living in the USA, China and Indonesia. Dynata, a survey research company, was responsible for recruiting participants through social media and other advertising. Responses were recorded in May through August of 2022. Those who took less than 3 min to respond to the entire questionnaire were excluded from the final sample, which was done to collect better quality data, that is, more thoughtful responses to the various questions asked. Only those who consented, spent over 3 min taking the survey and completed all survey questions were included in the final sample. Those who responded but were over the age of 45 were excluded from the final study sample, as they likely would not be a target for STI vaccine programmes. For example, in the USA, the HPV vaccine is recommended starting at 11–12 years and not recommended for those older than 26 years because the vaccine works best when administered before exposure to HPV.¹⁹ Our sample size calculation was based on a concurrent project, which was to estimate coverage of COVID-19 vaccine over different phases of vaccine roll-out. With an alpha of 0.05, a power of 80%, and an estimate of 50% (a statistically conservative assumption), we determined that a sample size of 800 would have a margin of error of 4%.

Derived variables

The questionnaire was distributed to each country and is publicly available.²⁰ Participants were given brief information about the STI, that a vaccine is currently unavailable, and then asked whether they would accept a vaccine if it were recommended to everyone in the population (universal) and asked again if they would accept the same vaccine if it were recommended for the following groups: ‘multiple sexual partners, people who have unprotected sex or MSM’ (targeted). Answering ‘yes’ to these questions was considered vaccine acceptance.

Respondents were asked about their sexual behaviours within the past 12 months. Participants were first asked how they identify (male/female/other) and later asked how their previous sexual partners identify. If one identified as male and reported having sex with those who were cis-gender men, transgender women (assigned male at birth) or non-binary persons, they were coded as MSM. Participants were then asked: “how often have you had vaginal or anal sex without using a condom?” and “how many people have you had any kind of sex with?” Participants were differentiated by their engagement in these high-risk sexual behaviours.

Participants were asked if they had previously heard of syphilis, chlamydia, gonorrhoea and genital herpes.

Vaccine hesitancy was determined by using the adult Vaccine Hesitancy Scale in which participants rated their level of agreement using a 5-point Likert scale to a series of ten statements.²¹

Statistical analysis

Survey weights were constructed to generalise the study sample to the national population distribution found in each country. Age, gender and region were variables used to create survey weights in Indonesia and China. In the USA, the above variables along with race/ethnicity and education were used to create survey weights.

For each country, we tested the effect of recommendation wording by using Rao-Scott χ² tests to detect significant differences between universal and targeted recommendations.

Subsequently, we stratified each country’s population into those who had received the universal versus the targeted vaccination recommendation. Within each of these strata, we used Poisson regression modelling using robust variance estimates²² to report the prevalence ratio (PR) of vaccine acceptance. We also report a p value of the interaction between every independent variable and recommendation type. This p value represents whether the PR significantly varied by recommendation type—in other words, if acceptance would significantly vary for certain demographic groups if given a universal versus targeted recommendation. This model included the following covariates a priori: age group, gender, MSM, multiple concurrent sexual partners, frequency of condom usage, STI vaccine and ever heard of STIs. Since we found no discernible differences across STIs, we did not further stratify by STI, and instead included this as a covariate. All analyses were conducted using SAS Studio V.9.4.

Results

Sample characteristics in the USA, China and Indonesia

The sample consisted of 1941 adults between the ages of 18 and 45: 572 respondents resided in the USA, 709 in China and 660 in Indonesia. Detailed characteristics of the respondents are presented in table 1. Almost half of the respondents (48%) in the USA were classified as vaccine-hesitant. Previous awareness of four chosen STIs (gonorrhoea, chlamydia, syphilis and genital herpes) varied greatly by country. Over 70% of respondents in the USA had heard of each STI. In China, 73% of the sample had heard of syphilis but only 58% had heard of chlamydia. In Indonesia, 85% of the sample had previously heard of syphilis but only 32% had heard of chlamydia (table 1).

Table 1. Sample characteristics in the USA, China and Indonesia (May–August, 2022).

	USA Count (weighted %±SE)	China Count (weighted %±SE)	Indonesia Count (weighted %±SE)
	N=572	N=709	N=660
Age
18–24	185 (26%±2%)	272 (23%±2%)	2–20 (28%±2%)
25–34	187 (39%±2%)	224 (42%±2%)	220 (36%±2%)
35–44	200 (36%±2%)	213 (35%±2%)	220 (36%±2%)
Gender
Male	270 (46%±2%)	387 (52%±2%)	330 (51%±2%)
Female	299 (54%±2%)	319 (48%±2%)	330 (49%±2%)
Other	3	3	0
MSM	69 (12%±1%)	87 (11%±1%)	40 (6%±1%)
Multiple concurrent sexual partners	152 (29%±2%)	96 (13%±1%)	82 (14%±2%)
Frequency of condom usage
Did not have sex (past 12 mos.)	126 (20%±2%)	155 (18%±2%)	66 (9%±1%)
Never	87 (15%±2%)	97 (14%±2%)	233 (35%±2%)
Less than half the time	66 (13%±2%)	81 (11%±1%)	63 (9%±1%)
About half of the time	70 (13%±2%)	112 (17%±2%)	71 (10%±1%)
More than half the time	52 (8%±1%)	119 (19%±2%)	65 (10%±1%)
Always	171 (29%±2%)	145 (20%±2%)	162 (26%±2%)
Vaccine hesitant	253 (48%±2%)	179 (26%±2%)	207 (32%±2%)
Previously heard of STI
Gonorrhoea	447 (78%±2%)	400 (59%±2%)	317 (47%±2%)
Chlamydia	419 (73%±2%)	407 (58%±2%)	215 (32%±2%)
Syphilis	441 (76%±2%)	513 (73%±2%)	549 (85%±2%)
Genital herpes	441 (76%±2%)	433 (61%±2%)	534 (80%±2%)

MSM, men who have sex with men; STI, sexually transmitted infection.

Vaccine acceptance

In all three countries, those who received a targeted recommendation were less willing to accept a gonorrhoea vaccine; however, this finding was only considered statistically significant for Indonesia (p=0.0005) (figure 1, online supplemental table S1). In all three countries, the same trend was observed for low-risk groups, where those who had a targeted recommendation were less accepting of a gonorrhoea vaccine. However, few low-risk groups had statistically significant differences between universal-recommendation and target-recommendation groups, such as those with no sexual partners in the past 12 months in China (p=0.0239) and those with no concurrent partners in Indonesia (p=0.0001).

Figure 1. Overall vaccine acceptance for each country by STI. The above graphs display the proportion from each country sample who accepted a theoretical STI vaccine under universal or targeted wording for each vaccine (gonorrhoea, chlamydia, syphilis, genital herpes). The 95% CIs were used to construct error bars. STI, sexually transmitted infection.

Overall, participants had lower acceptance of the chlamydia vaccine if they were in the targeted recommendation group (figure 1); the difference between universal and targeted groups was statistically significant for the U.S. and Indonesia (p=0.0233 and p<0.0001, respectively) (online supplemental table S2). Those in the USA and Indonesia who were not MSM similarly had a statistically significant difference in chlamydia-vaccine acceptance between universal and targeted groups (p=0.0168 and p=0.0001). This was also true for those who did not have multiple concurrent partners in the USA and Indonesia (p=0.0477 and p=0.0001) (online supplemental table S2).

Overall, those who received a targeted vaccine recommendation were less accepting of the syphilis vaccine (figure 1); however, the differences between the universal and targeted groups were only statistically significant for China and Indonesia (p=0.02339 and p=0.0014, respectively). There were few statistically significant differences in acceptance between universal and targeted recommendations for a syphilis vaccine (online supplemental table S3).

Overall, rates of acceptance of a vaccine for genital herpes were higher among those who received a universal recommendation compared with those who received a targeted recommendation (figure 1). Statistically significant differences for universal versus targeted groups were found in China (p=0.0481) and Indonesia (p=0.0019) (online supplemental table S4). The difference between universal and targeted vaccine acceptance was statistically significant for non-MSM in Indonesia (p=0.0016), those with no sexual partners in the past year in China (p=0.0242), and those who did not have multiple concurrent partners in China and Indonesia (p=0.0425 and p=0.0003, respectively). MSM in Indonesia had no difference in vaccine acceptance between universal and targeted groups. However, those who received a targeted recommendation and had multiple concurrent sexual partners preferred the vaccine over the universal recommendation group (74% compared with 71%). However, in China, those who had multiple concurrent sexual partners were slightly more accepting of a genital herpes vaccine, 84% for universal and 82% for targeted recommendation groups.

Poisson regression models of STI vaccine acceptance

Results from the multivariable Poisson regression models showed some consistent trends across countries, between those who received a universal recommendation and those who had a targeted recommendation, and for each of the four STI vaccines (table 2). Those vaccine-hesitant individuals had less acceptance of an STI vaccine; for example, in the USA, in the arm of those who had received a universal recommendation, acceptance was 0.73 times in those hesitant compared with those not hesitant (95% CI: 0.59 to 0.90). Those who had heard of the STI were more likely to accept the vaccine; for example, in Indonesia in the arm of those who had received a universal recommendation, acceptance was 1.34 times higher in those who had heard of the STI compared with those who had not (95% CI: 1.17 to 1.53). We did not observe large differences in vaccine acceptance by MSM status or by having concurrent sexual partners.

Table 2. Multivariable Poisson regression models of STI vaccine acceptance stratified by country (May–August, 2022).

	USA		China		Indonesia
	Universal PR (95% CI)	Targeted PR (95% CI)	Universal PR (95% CI)	Targeted PR (95% CI)	Universal PR (95% CI)	Targeted PR (95% CI)
Age
18–24 (Ref)
25–34	1.03 (0.83 to 1.29)	0.98 (0.77 to 1.26)	0.98 (0.86 to 1.12)	0.95 (0.82 to 1.09)	1.04 (0.87 to 1.24)	0.88 (0.72 to 1.08)
35–44	0.85 (0.66 to 1.10)	0.99 (0.77 to 1.27)	0.87 (0.74 to 1.02)	0.88 (0.74 to 1.04)	0.84 (0.69 to 1.02)	0.72 (0.56 to 0.91)
Gender
Not male vs male	0.85 (0.70 to 1.03)	0.80 (0.66 to 0.98)	1.16 (1.01 to 1.34)	1.06 (0.93 to 1.22)	1.12 (0.95 to 1.32)	1.11 (0.92 to 1.34)
MSM
MSM vs not MSM	0.83 (0.58 to 1.17)	0.86 (0.68 to 1.09)	1.03 (0.84 to 1.25)	1.07 (0.85 to 1.33)	0.99 (0.68 to 1.43)	1.18 (0.86 to 1.64)
Multiple concurrent sexual partners
Yes vs no	1.16 (0.95 to 1.41)	1.33 (1.08 to 1.62)	1.08 (0.92 to 1.27)	1.07 (0.92 to 1.24)	0.96 (0.77 to 1.21)	1.48 (1.24 to 1.76)
Condom use
Did not have sex (past 12 mos.) (Ref)
Condom usage≤half	1.26 (0.93 to 1.70)	1.07 (0.78 to 1.48)	1.16 (0.94 to 1.42)	1.43 (1.08 to 1.88)	1.25 (0.88 to 1.79)	1.05 (0.73 to 1.50)
Condom usage>half	1.18 (0.87 to 1.61)	1.07 (0.78 to 1.47)	1.21 (0.98 to 1.49)	1.54 (1.17 to 2.03)	1.27 (0.87 to 1.85)	1.08 (0.73 to 1.59)
Vaccine hesitant
Hesitant vs not	0.73 (0.59 to 0.90)	0.77 (0.63 to 0.94)	0.80 (0.67 to 0.96)	0.86 (0.72 to 1.03)	0.58 (0.46 to 0.73)	0.57 (0.44 to 0.74)
STI vaccine
Gonorrhoea (Ref)
Chlamydia	1.01 (1.00 to 1.03)	0.92 (0.84 to 1.00)	1.00 (0.97 to 1.03)	1.01 (0.95 to 1.08)	1.04 (1.02 to 1.07)	1.03 (0.96 to 1.10)
Syphilis	1.01 (0.99 to 1.02)	0.96 (0.87 to 1.05)	0.98 (0.95 to 1.00)	0.99 (0.93 to 1.06)	0.90 (0.86 to 0.95)	0.90 (0.83 to 0.98)
Genital Herpes	1.00 (0.99 to 1.01)	0.95 (0.85 to 1.05)	0.92 (0.89 to 0.95)	0.98 (0.91 to 1.04)	0.90 (0.86 to 0.95)	0.87 (0.80 to 0.94)
Previously heard of STI
Yes vs no	1.30 (1.05 to 1.62)	2.45 (1.69 to 3.56)	1.74 (1.48 to 2.04)	1.77 (1.49 to 2.11)	1.34 (1.17 to 1.53)	1.67 (1.38 to 2.02)

MSM, men who have sex with men; PR, prevalence ratio; STI, sexually transmitted infection.

Across all three countries—USA, China and Indonesia—vaccine acceptance generally did not vary substantially by age or MSM status. Notably, individuals with multiple concurrent sexual partners were more likely to accept a targeted STI vaccine in both the USA (PR: 1.33, 95% CI: 1.08 to 1.62) and Indonesia (PR: 1.48, 95% CI: 1.24 to 1.76), but not significantly so in China. Women and gender minorities (ie, those not identifying as male) in the USA had lower vaccine acceptance in the targeted arm (PR: 0.80, 95% CI: 0.66 to 0.98), while the pattern was reversed in China and Indonesia, where these groups had higher acceptance across both arms. Vaccine hesitancy was consistently associated with lower vaccine acceptance across all countries and recommendation types. Prior awareness of STIs was associated with substantially higher vaccine acceptance, especially in the US targeted arm (PR: 2.45, 95% CI: 1.69 to 3.56). Differences in vaccine acceptance by STI type were minimal, though acceptance was slightly lower for genital herpes and syphilis vaccines compared with gonorrhoea, particularly in Indonesia.

In the models, we placed an interaction term between what information the participant received and the variable. This interaction term was significant for two variables. In the USA, we found that individuals who had previously heard of STIs were more likely to accept an STI vaccine, and that this acceptance varied significantly based on whether they received a universal (PR: 1.30, 95% CI: 1.05 to 1.62) versus targeted recommendation (PR: 2.45, 95% CI: 1.69 to 3.56) (p value of interaction term between heard of STI and recommendation type=0.0029). And in Indonesia, acceptance of an STI vaccine did not vary between those with and without multiple concurrent sexual partners—under a universal recommendation (PR: 0.96, 95% CI: 0.77 to 1.21). But it was greater if there was a targeted recommendation (PR: 1.48, 95% CI: 1.24 to 1.76) (p value of interaction term between multiple concurrent sexual partners and recommendation type=0.0022).

We assessed the interaction between variable and recommendation type, that is, whether the PR varied significantly across recommendation type. Significant interaction terms indicate that acceptance varied based on whether the group would receive universal versus targeted recommendation. The only significant values were for multiple sexual partners in Indonesia (p=0.0022) and previously hearing of STIs in the USA (p=0.0029).

Discussion

Our study sought to assess the effect that universal wording and targeted wording had on theoretical vaccine acceptance for four common STIs, gonorrhoea, chlamydia, syphilis and genital herpes, in three different settings: the USA, China and Indonesia. We also wanted to assess how universal or targeted wording would affect those who partake in high-risk behaviours (those who do not use condoms often, those with multiple concurrent partners and those considered MSM).

Our findings differ from a previous study that assessed parents’ willingness to get their children vaccinated for HPV, in which the authors found that message and framing of vaccination did have some effect on willingness to vaccinate.²³ However, the authors were primarily interested in cancer and STI-messaging when comparing vaccine recommendations, whereas our study did not include specific language regarding prevention because it is still unclear to what degree STI vaccines will be able to prevent infection.

STI vaccines are emerging and though many studies use cross-sectional data to understand theoretical acceptance, few have used longitudinal designs to assess both willingness to be vaccinated and actual uptake of vaccines. A study assessing willingness and uptake for an H1N1 vaccine in Hong Kong had pessimistic results: only 0.8% of the 896-person sample reported receiving the H1N1 vaccine on follow-up.²⁴ While some noted concerns about receiving the vaccine due to adverse reactions, many did not comply because they perceived themselves to be at low risk for contracting H1N1. This could explain why our study found that acceptance was often lower among low-risk groups.

Targeted vaccine recommendations were issued during the 2022 outbreak of mpox that focused on LGBT+populations. In a meta-analysis that compared acceptance towards the mpox vaccine in different continents (Africa, Europe, Asia and the Americas), the authors found that mpox vaccination among all 8045 study participants was low (43%); however, 84% among the LGBT+population were vaccinated.²⁵ Thus, high perceived risk might explain the cases where targeted wording had higher acceptance, like those with multiple concurrent partners in Indonesia or MSM in the USA. This result also suggests that perception of low risk for contracting certain diseases affects vaccine uptake for the general population.

Our study elucidates an important public health problem, as vaccine hesitancy and particularly vaccine refusal are contributing to the increasing number of vaccine-preventable diseases globally.²⁶ de Figueiredo et al used model estimates to assess the level of vaccine confidence in various countries and found that confidence in vaccines fell between 2015 and 2019 within Indonesia and four other countries (Afghanistan, Pakistan, the Philippines and South Korea). The number of respondents who strongly disagreed that vaccines are safe increased within this period as well.²⁶ The authors note that in Indonesia, Muslim leaders questioned the safety of the measles-mumps-rubella vaccine and claimed that the vaccine was haram and contained ingredients derived from pigs, which was seen as unacceptable to the Muslim population. Online mobilisation against vaccines was also identified in South Korea and Malaysia. These results are concerning, especially since we found that acceptance had a wide range both between countries and between proposed STI vaccines (52%–75%).

A different study that focused on vaccine hesitancy in five study sites (Bangladesh, China, Ethiopia, Guatemala and India) assessed vaccine attitudes among mothers.²⁷ The highest vaccine-hesitancy scores were found in China and Guatemala, which also differ from our results where the USA had higher vaccine hesitancy compared with China. The construct that some vaccines are no longer needed for less common diseases was most pronounced in India.²⁷ These findings are troubling because the continued use of vaccines to achieve herd immunity in the population is still important. This is also pertinent to our research because we included a low-income, upper-middle-income and high-income country within our sample. The local context shapes beliefs about the safety and efficacy of vaccines and can also shape perceived risk of vaccination. Moreover, influence from religious institutions or the local government can also shape beliefs and attitudes towards vaccines.

We found that for all STIs, the PR of acceptance was higher for those who had previously heard of the STI in question. Zheng et al recruited a sample of MSM (homosexual and/or bisexual) and found that of the 2618 respondents, 90.2% were willing to receive an mpox vaccine, which was higher than the general population.²⁸ In this study, some of the factors that influenced vaccine acceptance were knowledge about mpox (OR 2.03), knowledge about prevention measures (OR 3.52) and concerns about susceptibility (OR 4.37).²⁸ Thus, many factors can influence vaccine behaviours, but education (especially regarding risk factors) may be beneficial for improving vaccine uptake. Thus, accurate sex education and information regarding STI vaccines is important for decision-making.

Due to the sensitive nature of STIs and the moral stigmas that may be attached to sex and sexual behaviour, another possible mechanism for vaccine hesitancy, particularly for STI vaccines, may be social stigma. A qualitative study regarding HPV vaccine acceptance among a sample of women in Hong Kong found that a reason for vaccine hesitancy regarding the HPV vaccine was the misconception that women who engage in promiscuous sexual behaviours need the vaccine, whereas women who are not as sexually active or have only one partner do not need to be vaccinated against HPV.²⁹ Others in this study expressed concern that if young women receive the HPV vaccine, then this would encourage immoral and unsafe sexual behaviours.²⁹ The perception of being at low risk due to not engaging in high-risk behaviours paired with social stigma regarding morality and sex may contribute to vaccine hesitancy, especially in contexts where sex education is lacking.

It is important to note that this study was also conducted after the COVID pandemic. During the pandemic, social restrictions and misinformation about the development of the COVID vaccine contributed to anti-vaccine sentiments on social media.³⁰ Misinformation regarding the vaccine’s ingredients and effects on health has contributed to scepticism regarding vaccine recommendations. A study of Twitter users and tweets in the USA found that indeed anti-vaccine users did increase postpandemic (20% before COVID-19, 23.28% after) and there was also a slight decrease of pro-vaccine users (61% before the pandemic, 56% after).³¹ Though these changes were small, misinformation proliferated by social media can reach the population more quickly, meaning that public health institutions may struggle to debunk this information in a timely fashion. Given the postpandemic context of this study, the vaccine hesitancy rates we found in our sample may be attributed to fears regarding the COVID-19 vaccine and the growing amount of misinformation found on social media platforms.

Limitations

Our study had several limitations. Due to the cross-sectional nature of the data and the ongoing development of STI vaccines, this study does not follow up on actual receipt of an STI vaccine. We also did not test for the change in willingness to be vaccinated based on other information about the vaccine, such as giving participants information regarding vaccine efficacy or prevention of chronic illness or cancers from an STI. This was not included because STI vaccines are still in development and information on prevention and efficacy is not yet available. Moreover, reasons for vaccine hesitancy or vaccine rejection were not recorded. We also measured a theoretical willingness to be vaccinated, which does not predict future behaviours. We are also limited because this study was an internet-based questionnaire that relied on self-reported data. It is possible that there was sampling bias based on the time period of data collection in May to August.

Portions of the questionnaire were validated,²¹ and the base of the questionnaire was developed for a COVID-19 survey in 2020, and was pretested with 16 adults in age from their early 20s to late 60s. However, other questions, including the STI vaccine acceptance questions, were not.

However, this study included three vastly different regions of the world that have different social, economic and political contexts that shape attitudes and access to vaccination. This study is also one of very few to assess the wording of vaccine recommendations among these different populations. Given that vaccine hesitancy is on the rise globally²⁶ and that many still rely on doctors, governments, or religious leaders for health information, understanding the type of information or the wording that people will most understand is key for decision-making around vaccines.

Conclusions

We wanted to assess willingness to receive an STI vaccine in three different countries (USA, China and Indonesia) and for four common STIs (gonorrhoea, chlamydia, syphilis and genital herpes). Our study assessed the wording of vaccine recommendations and aimed to understand how this wording affects theoretical vaccine acceptance. We also assessed the effect of wording among high-risk groups (MSM, infrequent condom users and those with multiple concurrent partners). Our study found that across all countries, vaccine acceptance was higher when given universal wording. However, in certain contexts, targeted recommendations resulted in higher acceptance for those who engaged in high-risk behaviours. In other contexts, universal and targeted differences made little or no difference in vaccine acceptance. The most consistent factor impacting vaccine acceptance appeared to be previous knowledge of STIs. This study highlights the need to perfect vaccine recommendations and encourage education regarding sexual risk behaviours. As these STI vaccines continue to develop, gauging interest and willingness to be vaccinated is important for vaccine roll-out strategies.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data availability statement

Data are available in a public, open access repository.