Using an extended information-motivation-behavioral skills model to explain HPV vaccination intention among men who have sex with men only and men who have sex with men and women

ABSTRACT

Men who have sex with men and women (MSMW) have been reported to differ in psychosocial and sexual behavior patterns from men who have sex with men only (MSMO). However, results regarding the differences in HPV vaccination intention/behavior were inconclusive. We compared HPV vaccination intention between MSMO and MSMW and analyzed the differences in potentially associated factors in China. MSM participants were recruited online using a snowball sampling method. Cross-sectional data were collected via a questionnaire based on the extended information-motivation-behavioral skills model. Structural equation modeling was conducted to examine the relationship between the variables, followed by multi-group analysis to test differences between groups. Of 914 MSM, 77.68% were MSMO and 22.32% were MSMW. MSMW had a higher rate of reluctance to vaccinate than MSMO (23.53% vs. 16.20%, p = .016). Differences between the two groups were statistically significant in risky sexual behavior, behavioral skills, and promotional attitude. In both groups, promotional attitude was the most significant predictor of vaccination intention. Vaccination intention was directly influenced by motivation and indirectly by risky sexual behavior in MSMO, but not significantly in MSMW. Additionally, the direct effect of information on behavioral skills in MSMW was significantly greater than that in MSMO, but we did not find any effect of behavioral skills on vaccination intention. MSMW had lower vaccination intention than MSMO. MSMO may be influenced by risky sexual behavior and motivation, positively impacting their vaccination intention, unlike MSMW. Targeted strategies could help promote HPV vaccination, especially in MSMW.

KEYWORDS:

• Human papillomavirus vaccine
• men who have sex with men and women
• intention
• information-motivation-behavior skills model
• multi-group structural equation modeling

Introduction

Men who have sex with men (MSM) are at high risk for HPV infection and the development of anal cancer.^1,2 In China, the HPV prevalence in MSM (59.9%) is significantly higher than that in women (15.6%) and heterosexual men (14.5%).³ Since MSM are characterized by high infection rates and low clearance rates for HPV, MSM are approximately 20 times more likely to develop anal cancer than heterosexual men.⁴ The trend continuously increases at a rate of 2% per year.⁵

HPV vaccination is a cost-effective measure to prevent HPV infection and reduce HPV-related diseases,^6–9 while MSM-targeted vaccination programs have not yet been implemented in developing countries. Currently, the exploration of MSM vaccination willingness is an essential part of such vaccination programs before their implementation and promotion. However, most researchers have tended to focus on the MSM population as a whole,^10,11 which, to some extent, may have obscured the specific behavioral and psychosocial characteristics of the subgroup of men who have sex with men and women (MSMW).^12,13 In China, MSMW are common in the MSM population. Reportedly, more than 25% of MSM had heterosexual sex in the past six months,¹⁴ and about 70%-90% of MSM would eventually marry woman.¹³ Although MSMW are less likely to contract an HPV infection than men who have sex with men only (MSMO),^15,16 MSMW are more likely to have unprotected sex with a female partner than a male partner, according to their sexual behavior patterns.^17,18 Only 23.3% of MSMW in China used condoms with regular female sexual partners.¹⁴ Furthermore, MSMW manifest greater psychological vulnerability and are more secretive as they are less likely to disclose their sexual orientation.¹⁹ Conclusively, MSMW are more likely to transmit HPV to women, thus presenting a potential public health risk. Therefore, differentiating HPV vaccination intention between MSMW and MSMO is crucial for targeted vaccination efforts.

The results regarding the differences between MSMW and MSMO in HPV vaccination intention or behavior are inconclusive. Previous evidence has indicated that MSMW are less inclined to be willing to consider, initiate, or complete HPV vaccination compared to MSMO.^20–22 However, a systematic review and meta-analysis did not find a statistically significant relationship between the willingness to be vaccinated and gender identity.¹¹ Further, in other sexually transmitted disease-related studies, MSMW have demonstrated lower utilization rates of prevention and testing services relevant to the MSM population,²³ and have reported lower levels of sexual health knowledge and behavioral skills.^24,25 To our knowledge, few studies have revealed how underlying factors affect vaccination intention in these two populations and the variations among them, leading to a lack of appropriate recommendations for guidance.

To address the above issues, this study compared the differences in HPV vaccination intention and related factors between MSMO and MSMW based on the extended information-motivation-behavioral skills (IMB) model, and further explored the characteristics of the relationship between these factors in the pathway of influence, to provide a theoretical basis for improving the promotion and uptake of the HPV vaccine in the MSM population.

Methods

Study design and participants

A cross-sectional survey design was utilized in this study. In June 2021, participants were recruited in China. The questionnaire based on the extended IMB model was distributed to the MSM cohort population established in the previous study with the help of the Questionnaire Star online survey tool. Each participant could become a source of further participants, using snowball sampling to share the survey with peers to expand the study sample. The inclusion criteria were as follows: (1) MSM, (2) 15–45 y old, (3) no difficulty reading or using electronic devices, and (4) read the informed consent form and agreed to participate in the survey. The exclusion criteria was a completion time of <1.5 min or >15 min. Participants who completed the questionnaire were rewarded with 10 RMB. The study was approved by the Ethics Committee of Chongqing Medical University (2019001).

According to the published literature, the proportions of MSMO ($p_1$) and MSMW($p_2$) willing to be vaccinated are 69.3% and 56.1%, respectively.²⁶ Our previous study showed that MSMW accounted for 16.1% of MSM,²⁵ i.e., the sample size ratio ($\mathit{k}$) of MSMO to MSMW was 5.21:1. In the case of a two-sided test level $\mathit{\alpha }=0.05$ and a power of test (1$-\mathit{\beta })=80$%, using the sample size formula $n_2=\frac{{({\mathit{Z}}_{1-\mathit{\alpha }/2}\mathit{\hspace{0.17em}}+\mathit{\hspace{0.17em}}{Z}_{1-\mathit{\beta }})}^2\left[p_1\left(1-p_1\right)/\mathit{k}+p_2\left(1-p_2\right)\right]}{(p_1-p_2){ }^2}$, with $n_1=\mathit{k}\ast n_2$, we would need at least 808 cases of MSM (MSMO $n_1=$678, MSMW $n_2=$130) for our study. A total of 1477 participants were recruited for this study, of which 563 were excluded for not meeting the inclusion and exclusion criteria, duplicate fills, logical errors and missing key information, as shown in Figure 1. Ultimately, 914 participants (MSMO vs. MSMW, 710 vs. 204) were eligible for inclusion in the analysis of vaccination intention.

Figure 1. Flow chart of participants’ enrollment.

MSMO, men who have sex with men only; MSMW, men who have sex with men and women.

Measures

We designed the questionnaire based on the extended IMB model combined with the relevant literature.^27–30 MSM who reported having sex with women (both regularly and casually) in the last six months were categorized in the MSMW group, while those who did not report this behavior were classified in the MSMO group. Sociodemographic information collected included age, household registration, ethnicity, education, employment status, marital status, and monthly disposable income. Information was a 12-item self-reported questionnaire assessing participants’ knowledge of HPV and the HPV vaccine, with a score of 2 for a correct answer, 1 for a “don’t know”, and 0 for an incorrect answer (Cronbach’s α = 0.805). Motivation was assessed using a 5-point Likert scale (1 = “very low/small”, 5 = “very high/large”) containing two items for HPV infection possibility and health threat (Cronbach’s α = 0.737). Behavioral skills were mainly measured by asking about STDs and HIV testing and counseling, and the dimension consisted of four items (0 = “no”, 1 = “yes”, Cronbach’s α = 0.902). Participants were also questioned about their number of male sexual partners (regular and casual) and whether they had unprotected sex with men in the last six months, with the number of male sexual partners categorized into 3 grades: low (≤1), medium (2–4), and high (≥5), and similarly, whether they had unprotected sex with men, which was categorized into low (no partners/sex), medium (sex with a condom), and high (sex without a condom) levels (Cronbach’s α = 0.621). Finally, we surveyed their promotional attitude toward the HPV vaccine among MSM (1 = “opposed”, 2 = “neutral”, 3 = “favorable”) and asked them, “Would you be willing to be vaccinated against HPV?” (willing or unwilling). The items described here are shown in Table 1, with a measured KMO of 0.841 and a Bartlett’s test of sphericity chi-square of 6970.271 (p < .001).

Table 1. Constructs and corresponding items in the model.

Constructs	Items	Measures
Information	I1	Have you ever heard of HPV?
	I2	Have you ever heard of HPV-related diseases?
	I3	Using condom can reduce HPV infection
	I4	Both men and women can be infected with HPV
	I5	HPV can cause genital warts and malignant tumors such as cervical/penile/skin cancers
	I6	HPV is mainly transmitted through sexual contact
	I7	Most HPV infection have no visible symptoms
	I8	Have you ever heard of the HPV preventive vaccine?
	I9	HPV vaccine can effectively prevent cervical/anal cancer/genital warts.
	I10	What is the best time to be vaccinated against HPV?
	I11	Who can be vaccinated against HPV?
	I12	Does the HPV vaccine protect against all types of HPV infection?
Motivation	M1	How likely do you think you are to be infected with HPV?
	M2	How much of a threat do you think HPV is to you?
Behavioral skills	BS1	Have you ever been tested for sexually transmitted diseases?
	BS2	Have you ever initiated counseling on sexually transmitted diseases?
	BS3	Have you ever been tested for HIV?
	BS4	Have you ever initiated HIV counseling?
Risky sexual behavior	RSB1	In the last six months, number of male sexual partners (casual and regular):
	RSB2	In the last six months, Unprotected sex with men:

I, information; M, motivation; BS, Behavioral skills; RSB, Risky sexual behavior.

Statistical analysis

Continuous variables were expressed as median (M) and lower and upper quartiles (P₂₅, P₇₅), while categorical variables were described by frequencies and percentages. Differences were analyzed by χ² test and Mann-Whitney U test, and two-by-two comparisons were made via the false discovery rate (FDR) method. Correlation coefficients between variables were obtained by the Spearman correlation analysis. Full-sample structural equation modeling analysis was performed to assess the path relationships between the variables. Multi-group structural equation modeling exhibits exceptional modeling and explanatory capabilities when studying group comparisons, and in this study, it was applied to estimate the differences in path coefficients between the two groups. The mean and variance-adjusted weighted least squares with Theta parameterization were used to estimate parameters. The models were estimated with covariate adjustments for age, household registration, employment status, marital status, and monthly personal income. In this case, monthly personal income was considered a continuous variable, and the other covariates were set up with dummy variables. The model fitting evaluation indexes and criteria were as follows: chi-square degree of freedom ratio (χ²/df) < 5.0, root mean square error of approximation (RMSEA) < 0.08, comparative fit index (CFI) > 0.90, Tucker-Lewis index (TLI) > 0.90, and standardized root mean square residual (SRMR) < 0.10.^31–33 All statistical analyses were performed using R 4.3.1 and Mplus 8.3 software.

Results

Participant characteristics

This study included 914 validated participants, of which 710 (77.68%) were in the MSMO group and 204 (22.32%) were in the MSMW group. Table 2 presents the sociodemographic characteristics of the MSMO and MSMW groups. The median age of the participants was 27 (23, 32) y. More than half of MSM households were urban (59.52%) and college-educated or higher (52.52%). Also, 95.62% of MSM belonged to the Han nationality. The majority of MSM reported being unmarried (73.19%), but the MSMW group reported being married at a higher rate than the MSMO group (53.43% vs. 14.51%). Furthermore, the MSMW group was more likely to be employed (85.78% vs. 70.14%) and had higher monthly disposable income (p < .001) than the MSMO group.

Table 2. Participant characteristics of the MSMO and MSMW groups.

Variables	Total	MSMO	MSMW	p
N (%)	914(100)	710(77.68)	204(22.32)
Sociodemographic Characteristics, n (%)
Age(years)				.365
15–30	611(66.85)	480(67.61)	131(64.22)
31–45	303(33.15)	230(32.39)	73(35.78)
Household registration				.220
Urban	544(59.52)	415(58.45)	129(63.24)
Rural	370(40.48)	295(41.55)	75(36.76)
Ethnicity				.978
Han nationality	874(95.62)	679(95.63)	195(95.59)
Ethnic minorities	40(4.38)	31(4.37)	9(4.41)
Education level^a				.153
Junior high and below	64(7.00)	48(6.76)	16(7.84)
Senior high school/Vocational high school/Technical secondary school	151(16.52)	110(15.49)	41(20.10)
Junior college	219(23.96)	172(24.23)	47(23.04)
College and above	480(52.52)	380(53.52)	100(49.02)
Employment status				<.001
Employed^①	673(73.63)	498(70.14)	175(85.78)
Student^#	166(18.16)	153(21.55)	13(6.37)
Unemployed/Retired	75(8.21)	59(8.31)	16(7.84)
Marital status				<.001
Unmarried^①	669(73.19)	578(81.41)	91(44.61)
Married^#	212(23.19)	103(14.51)	109(53.43)
Divorce/widowed	33(3.61)	29(4.08)	4(1.96)
Monthly disposable income (RMB)^a				<.001
≤1000	65(7.11)	58(8.17)	7(3.43)
1001–3000	196(21.44)	176(24.79)	20(9.80)
3001–5000	233(25.49)	179(25.21)	54(26.47)
5001–10000	288(31.51)	206(29.01)	82(40.20)
>10000	132(14.44)	91(12.82)	41(20.10)
Score, M (P25, P75)
Risky sexual behavior^a	2(2,3)	2(2,4)	2(2,3)	<.001
Information^a	20(16, 22)	20(15,22)	20(17,22)	.320
Motivation^a	6(4,7)	6(4,7)	5.5(3,8)	.181
Behavioral skills^a	2(0,4)	2(0,4)	0(0,3)	<.001
Promotional attitude^a, n (%)				.008
Against	14(1.53)	10(1.41)	4(1.96)
Neutral	125(13.68)	86(12.11)	39(19.12)
Approve	775(84.79)	614(86.48)	161(78.92)
Vaccination intention, n (%)				.016
Unwilling	163(17.83)	115(16.20)	48(23.53)
Willing	751(82.17)	595(83.80)	156(76.47)

^aTested by the Mann-Whitney U test, other by χ² test.

^#Had statistical difference between two-by-two comparisons compared to ①.

MSMO, men who have sex with men only; MSMW, men who have sex with men and women.

The percentage of MSM willing to be vaccinated was 82.17% and the majority of MSM supported the promotion of the HPV vaccine (84.79%). Compared to the MSMO group, the MSMW group reported lower rates of risky sexual behavior (p < .001), lower behavioral skills (p < .001), and poorer promotional attitude (p = .008), as well as higher rates of reluctance to vaccinate (p = .016). Yet, there were no statistically significant differences in information and motivation between the two groups (p > .05) (see Table 2).

Correlation analysis

Table 3 shows the correlations between the variables within each group. In the MSMO group, the correlations between the variables were statistically significant, and the correlation coefficient between risky sexual behavior and behavioral skills was the largest (r = 0.495), showing a moderate correlation. In the MSMW group, risky sexual behavior was significantly associated with motivation, and behavioral skills were significantly associated with risky sexual behavior and motivation, while information was significantly associated with promotional attitude, and willingness to vaccinate was significantly associated with motivation and promotional attitude.

Table 3. Correlations between variables in the MSMO and MSMW groups.

	MSMO					MSMW
Constructs	1	2	3	4	5	1	2	3	4	5
1. Risky sexual behavior	1.000					1.000
2. Information	0.191***	1.000				−0.108	1.000
3. Motivation	0.301***	0.233***	1.000			0.197**	−0.050	1.000
4. Behavioral skills	0.495***	0.184***	0.270***	1.000		0.314***	0.121	0.224**	1.000
5. Promotional attitude	0.175***	0.311***	0.134***	0.160***	1.000	−0.043	0.383***	−0.043	0.096	1.000
6. Vaccination intention	0.171***	0.267***	0.211***	0.167***	0.375***	0.109	0.067	0.171*	0.120	0.280***

MSMO, men who have sex with men only; MSMW, men who have sex with men and women.

*p < .05，**p < .01, ***p < .001.

Full-sample structural equation modeling analysis

After deletion of the I12 (factor loading < 0.40) and BS3 items (factor loading > 0.95), the fit between the covariate-adjusted model and the actual data was still good (χ² /df = 3.180, RMSEA = 0.049, CFI = 0.956, TLI = 0.950, SRMR = 0.074). In this study, the extended model explained 40.9% of the vaccination intention, which was greater than the threshold of moderate explanatory power (0.33).³⁴ Risky sexual behavior could influence behavioral skills directly (β = 0.561, p < .001), as well as indirectly (via information) (β = 0.026, p = .010) with a total effect of 0.587. Motivation (β = 0.202, p = .005) and promotional attitude (β = 0.475, p < .001) had a significant positive effect on willingness to vaccinate. Information had an indirect effect on vaccination intention via promotional attitude (β = 0.262, p < .001), whereas “information → promotional attitude” (β = 0.040, p = .014) as well as motivation (β = 0.083, p = .008), functioned as mediators in the relationship between risky sexual behavior and vaccination intention, with an indirect effect of 0.123. However, the impact of behavioral skills on vaccination willingness was not significant (p > .05). In the end, the total effects of information, motivation, promotional attitude, and risky sexual behavior on vaccination willingness were 0.318, 0.182, 0.475, and 0.265, respectively (see Figure 2).

Figure 2. The adjusted standardized path estimates in the extended information-motivation-behavioral skills model.

*p < .05, **p < .01, ***p < .001.

Multi-group structural equation modeling analysis

We further constructed a multi-group model in order to compare the differences in path coefficients between the MSMO and MSMW groups. The unconstrained model showed that the covariate-adjusted model fit was generally acceptable (χ² /df = 1.869, RMSEA = 0.044, CFI = 0.963, TLI = 0.960, SRMR = 0.088). The model explained 44.6% and 43.0% of MSMO and MSMW vaccination intention, respectively.

In this study, the structural path coefficients were set to be equal between the two groups to construct the constraint model. The chi-square difference between the constrained and unconstrained models was 38.879 and the difference in the degree of freedom was 14, p < .001, suggesting that the extended IMB model differed in path coefficients between the two groups. The results showed a statistically significant difference between the MSMO and MSMW groups in the direct path of risky sexual behavior to information (t = 2.773, p = .006). Specifically, risky sexual behavior had a positive effect on information in the MSMO group (β = 0.243, p < .001), while it had no significant effect on information in the MSMW group (β=-0.235, p = .107). Meanwhile, the difference between the path coefficients of information on behavioral skills was statistically significant between the two groups (t = −2.039, p = .041). Information had a greater direct effect on behavioral skills in the MSMW group than that in the MSMO group (0.467 vs. 0.117) (Table 4). Further, on the indirect path, the effect of risky sexual behavior via information influencing promotional attitude was statistically different between the two groups (t = 2.045, p = .041). There was a positive indirect effect of risky sexual behavior by information on promotional attitude in the MSMO group (β = 0.131, p < .001), while this was not significant in the MSMW group (β = −0.122, p = .230).

Table 4. The adjusted standardized parameter estimates of the multi-group model.

Path	MSMO		MSMW		t	p
	β	p	β	p
Risky sexual behavior →Information	0.243	<.001	−0.235	.107	2.773	.006
Risky sexual behavior →Motivation	0.452	<.001	0.289	.019	0.428	.669
Risky sexual behavior →Behavioral skills	0.563	<.001	0.646	<.001	−0.429	.668
Risky sexual behavior →Promotional attitude	0.235	.137	−0.069	.826	0.781	.435
Risky sexual behavior →Vaccination intention	−0.037	.808	0.049	.890	−0.184	.854
Information →Behavioral skills	0.117	.041	0.467	<.001	−2.039	.041
Information →Promotional attitude	0.539	<.001	0.519	.013	0.015	.988
Information →Vaccination intention	0.093	.327	−0.176	.490	0.786	.432
Motivation →Behavioral skills	0.055	.402	0.121	.319	−0.363	.717
Motivation →Promotional attitude	−0.005	.960	−0.069	.591	0.337	.736
Motivation →Vaccination intention	0.208	.027	0.260	.057	0.008	.994
Behavioral skills →Promotional attitude	−0.043	.727	0.063	.818	−0.353	.724
Behavioral skills →Vaccination intention	0.069	.518	0.067	.823	0.008	.993
Promotional attitude →Vaccination intention	0.485	<.001	0.565	.001	−0.369	.712
Information <—> Motivation	0.245	<.001	0.112	.256	0.711	.477

MSMO, men who have sex with men only; MSMW, men who have sex with men and women.

Discussion

The present investigation demonstrated that although 82.17% of MSM were willing to be vaccinated, MSMW were more reluctant to be vaccinated than MSMO, which is similar to the vaccination intention and actual vaccination situations found in foreign studies.^20,21 This may be due to the fact that MSMW are less attached to the homosexual community and are more likely to benefit from herd immunity from female vaccination compared to MSMO.²² As such, MSMW may be a potential facilitator population for the HPV vaccination process. Hence, comparing HPV vaccination intention and its associated factors between MSMO and MSMW in China will provide important information for targeted HPV vaccination and promotion strategies.

The 22.32% who were MSM were categorized as MSMW, having had sex with women in the last six months, whose prevalence was higher than the 16.1% found in Western China²⁵ and the 16.87% in found in Shandong Province.³⁵ Of the participants, 23.19% of MSM reported being married, a higher rate than the pooled rate of a domestic meta-analysis (17.0%).¹⁴ Simultaneously, 53.43% of married MSMW had sex with women in the last six months, indicating that their wives might face a greater risk of HPV infection. The above results imply that MSMW might make the sexual network between MSM and heterosexuals more connected.

The correlation analysis showed that information was positively correlated with behavioral skills in MSMO but not in MSMW. Interestingly, the covariate-adjusted multi-group analysis confirmed our theoretical hypothesis that information could have a positive effect on behavioral skills in both the MSMO and MSMW groups. When comparing the path coefficients of information on behavioral skills between the two groups, we observed that the direct effect on MSMW was significantly greater than that on MSMO. This demonstrates that HPV and HPV vaccine-related knowledge might be more likely to stimulate MSMW individuals to effectively implement objective skills for reducing or avoiding HPV infection compared to MSMO.³⁶

We found that HPV vaccination intention in the two groups was primarily influenced by their promotional attitude toward vaccine. Further analysis revealed that information cannot directly change vaccination intention but can rather influence individuals’ promotional attitude, which ultimately affects their willingness to be vaccinated. Therefore, promotional attitude played a key mediating role in the relationship between information and willingness to vaccinate. This suggests that MSM individuals with positive promotional attitude are more willing to be vaccinated, and that this attitude can be stimulated by providing more information about HPV and the HPV vaccine. In other words, the knowledge and attitude of individuals regarding HPV vaccine might be key factors in improving vaccination decisions.^11,37,38

Risky sexual behavior could directly influence MSM motivation and behavioral skills, signifying that MSM might be able to perceive their HPV infection risk in terms of their own risky sexual behavior and enhance their behavioral skills to reduce their infection risk.³⁹ Additionally, in the MSMO group, risky sexual behavior positively influenced information and indirectly influenced vaccination intention via information and promotional attitude or motivation. This means that MSMO who engage in risky sexual behavior might be more willing to be vaccinated, but this process might do not occur overnight and require a series of shifts in information, motivation, attitudes, and other psychological factors. However, there was no statistically significant effect between risky sexual behavior, motivation, and the willingness to vaccinate among MSMW. Lower rates of risky sexual behavior and risk perception motivation may not contribute to the willingness to vaccinate among MSMW. Thus, fully understanding one’s own risky sexual behavior combined with one’s risk-perceived motivation will facilitate behavioral decision-making.³⁶

Surprisingly, the effect of behavioral skills on willingness to vaccinate was not significant, meaning that behavioral skills do not affect an individual’s willingness to vaccinate. This might be related to the fact that MSM have not yet been included in vaccination programs in China. When MSM engage in risky sexual behavior or have a need for HPV prophylaxis, their first thought is not to get a preventive vaccination, but it is more likely to seek the help of their healthcare providers or peers, which can be considered the most practical and effective solution for them at the moment. This can also be explained in terms of the result that risky sexual behavior directly increased behavioral skills, which further implies that the behavioral skill of communicating effectively with one’s healthcare providers and peers may facilitate MSM’s willingness to vaccinate.¹¹ Also, we identified that MSMW exhibited lower levels of behavioral skills, which is consistent with the findings of Bingham et al.²⁴ This might be related to this population being reluctant to disclose their sexual orientation.¹⁹ Therefore, the relevant institutions should enhance the proactive use of preventive services by MSM, and improve the depth of preventive services, particularly for MSMW with greater psychological vulnerability.

To summarize, the study identified a disparity in vaccination intention between MSMO and MSMW and revealed that risky sexual behavior and motivation have an effect on vaccination intention in MSMO but not in MSMW from the perspective of risky sexual behavior-driven demand. Nonetheless, some potential limitations still exist. First, due to the secretive nature of the MSM identity, participants were recruited online using snowball sampling, which could have introduced bias. Second, this study defined MSMW based on having engaged in sexual behavior with a woman in the last six months rather than sexual orientation, thereby potentially underestimating those who self-identify as bisexual.¹⁹ Meanwhile, inconsistent time definitions might have had an impact on the results.^25,40,41 Moreover, we asked for sensitive information related to sexual behavior, which might be subject to some reporting bias. Finally, this study was a cross-sectional survey, and future large-sample cohort studies are needed to further confirm the causal relationship between the variables.

Conclusions

In conclusion, MSMW had lower rates of vaccination intention than MSMO and might be a potential facilitator population for the HPV vaccination. MSMO might be influenced by risky sexual behavior and motivation, positively impacting their vaccination intention, unlike MSMW. Targeted strategies could help promote HPV vaccination, especially among MSMW.

Author contributions

GQS conceptualized and designed the paper, analyzed data and drafted the manuscript. BL, HYP, and WH conducted a survey design and collected data. XNZ critically revised and finalized the manuscript. All authors read, reviewed, and approved the final manuscript.

Ethics approval

The study was approved by the Ethics Committee of Chongqing Medical University (2019001).

Acknowledgments

We thank all participants and investigators for their help in the data collection during this research.

Disclosure statement

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

Data availability statement

Data analyzed during the current study are available from the corresponding author on reasonable request.

Additional information

Funding

This research was supported by the National Key Project for Infectious Diseases of the Ministry of Science and Technology of China, grant number 2018ZX10721102-005.