Abstract
Pre-exposure prophylaxis (PrEP) is an effective tool to prevent HIV and has recently been approved in Taiwan. However, more research regarding PrEP acceptability and effective PrEP implementation is needed in Taiwan. Little is known about the intention to use PrEP and salient psychosocial factors among men who have sex with men (MSM). The purpose of this study was to examine health beliefs linked to PrEP use intention among young MSM (YMSM) in Taiwan. Using the Health Belief Model (HBM) as a guiding framework, an anonymous online survey was used to collect data from participants who were recruited using multiple channels to ensure diversity. Multivariate logistic regression was performed to examine the health beliefs linked to high PrEP use intention. Data from 1329 HIV-negative YMSM were analyzed; 488 (36.7%) were classified as having high PrEP use intention, which was found to be associated with having high perceived susceptibility to HIV infection [adjusted odds ratio (AOR) = 2.11], high perceived benefits of PrEP use (AOR = 2.14), high self-efficacy in PrEP use (AOR = 4.19), and many cues to action concerning PrEP use (AOR = 1.49). YMSM with high perceived barriers to PrEP use (AOR = 0.64) had lower PrEP use intention. In addition, the effects of HBM constructs and specific health beliefs on PrEP use intention varied by preferred penile–anal sexual role. The study findings may inform efforts aimed at improving PrEP acceptability and optimizing PrEP implementation programs tailored for YMSM of different sexual roles.
Introduction
The total number of HIV-infected people in Taiwan reached over 30,000 in 2020. 1 Despite the government's preventive efforts, an average of around 1700 people per year over the past 3 years were infected by HIV in Taiwan. 1 In the past 3 years, over 75% of new HIV infections in Taiwan occurred among men who have sex with men (MSM), and more than 65% of newly infected persons were between 15 and 34 years of age. 1
Antiretroviral therapy (ART) and the concept of “Undetectable = Untransmittable (U = U)” are important in prevention because of the resulting reduction in HIV transmission and destigmatization of HIV infection. 2 In Taiwan, the importance of ART use and the concept of U = U have been recognized and advertised by Taiwan CDC and promoted through many LGBT health centers and nongovernmental organizations. Therefore, despite limited existing research, many MSM in Taiwan have likely been exposed to related information about this concept. However, to combat the HIV/AIDS epidemic more effectively, methods other than ART and condoms are also needed. Pre-exposure prophylaxis (PrEP) is an effective tool that can prevent HIV transmission and has recently been approved for use in Taiwan.
Studies have demonstrated that PrEP can reduce the risk of HIV infection by up to 86–92%, 3 and PrEP has been recommended as a prevention strategy by the World Health Organization (WHO). 4 Prior research has also examined PrEP access, use, concerns, and adherence among MSM. For example, studies examining PrEP access have suggested that PrEP protocols be tailored for specific populations 5 and that intervention messages be more diverse and inclusive. 6 A study examining PrEP use and concerns among MSM found that perceiving daily PrEP regimens to be burdensome, worrying about side effects, and perceiving PrEP-related stigma were associated with not using PrEP. 7 Another study examining PrEP adherence found that MSM who reported more risky behaviors and were in a partnership had higher PrEP adherence. 8
PrEP is relatively new in Taiwan, and PrEP-related studies are lacking. Further, HIV stigma and mistrust of health care providers can negatively impact HIV prevention efforts 9 ; in Taiwan, MSM reported frequently encountering less LGBT-friendly medical staff and having fewer resources, 10 and MSM perceiving that physicians were friendly to people living with HIV were more likely to seek health care. 11 Therefore, to better understand PrEP acceptability in Taiwan and to facilitate development of localized PrEP implementation programs, it is crucial to analyze PrEP use intention among MSM.
The Health Belief Model (HBM) posits that one's beliefs about health-related problems, perceived benefits and barriers to taking action, and self-efficacy determine whether one would engage in a certain health-promoting behavior. 12 Cues to action can also prompt the occurrence of the behavior. These HBM constructs are modifiable and may be important drivers of PrEP acceptability and may contribute to optimal PrEP implementation. The HBM has been commonly used in sexual behavior research on abstinence, condom use, having steady sexual partners, and being tested for HIV infection. 13 –15 PrEP use is an HIV preventive behavior, but little research has employed the HBM in examining PrEP use intention.
The risk of HIV infection differs by penile–anal sexual role among MSM. Receptive MSM had more than a 10-fold higher per-act risk of HIV infection than insertive MSM. 16 In addition, MSM of different sexual roles appear to engage in risky 17,18 and preventive behaviors differently, which might be influenced by MSM's different HIV risk perceptions depending on sexual role.
A study found that insertive MSM were more likely to use condoms than versatile MSM; 19 however, another study found that insertive MSM more commonly engaged in condomless anal intercourse (CAI). 20 Therefore, because of different risk perceptions, MSM of different sexual roles might have differential intentions to use PrEP as a preventive measure against HIV infection. This may have important implications regarding PrEP implementation, as a tailored approach by sexual role may be more effective as population-specific protocols can contribute to more effective care related to PrEP. 5
The current study employed a data-driven and theory-based approach to systematically examine PrEP use intention and its associated health beliefs using the HBM among young MSM (YMSM) in Taiwan recruited using multiple channels to capture a diverse sample. In addition, the analyses were stratified by preferred penile–anal sexual role to compare the differences across receptive, versatile, and insertive MSM in their health beliefs associated with high PrEP use intention. Identification of factors associated with PrEP use intention will shine some light on facilitators and barriers to PrEP acceptability and implementation. Further, the findings may be used to inform future implementation research.
Methods
Participants
Data were collected using an anonymous online survey. As recommended by the European MSM Internet Survey (EMIS) study, 21 we recruited participants through multiple channels. We advertised the survey to participants through MSM community health centers and anonymous HIV testing sites located in different regions of Taiwan and social media. The inclusion criteria for participants were (1) self-identified MSM, (2) aged 15–39 years, and (3) HIV-negative status. Participants were informed that voluntary completion of the survey would constitute informed consent to participate in this study. The study protocol was reviewed and approved by the Research Ethics Committee of the National Taiwan University Hospital.
Measures
A structured questionnaire was designed with the HBM as a guiding framework. The intention to use PrEP in the next year was measured by a semantic differential scale, scored from 1 (very unlikely) to 5 (very likely). Demographic characteristics included age, education level, employment status, monthly income, and current relationship status. Sex/gender-related characteristics included HIV status, sexual orientation, gender characteristic (scored from 1, relatively feminine, to 10, relatively masculine), preferred penile–anal sexual role (during penile–anal intercourse: exclusively insertive, mostly insertive, exclusively receptive, mostly receptive, or both receptive and insertive), and anal sex experience (yes/no).
HBM constructs were measured by semantic differential scales, scored from 1 (strongly disagree, very unlikely, or not at all confident) to 5 (strongly agree, very likely, or very confident). Perceived susceptibility (five items, Cronbach's α = 0.878) evaluated the perceived likelihood of getting infected with HIV under various circumstances. Perceived severity (eight items, Cronbach's α = 0.813) assessed the perceived impact of multi-dimensional consequences of HIV infection. Perceived benefits (four items, Cronbach's α = 0.785) measured the perceived levels of various beneficial effects of PrEP use. Perceived barriers (10 items, Cronbach's α = 0.800) assessed perceived levels of difficulty in taking PrEP. Self-efficacy (six items, Cronbach's α = 0.864) evaluated the confidence in taking PrEP in different situations. Cues to action inquired about specific experiences that might prompt MSM to take PrEP, scored 1 (“Yes, I had the experience”) or 0 (“No, I did not have the experience”). For each of the above HBM constructs, a construct score was created by summing the scores of all the items composing the construct.
Procedure
The intention to use PrEP and gender characteristics were dichotomized into high versus low and masculine versus feminine, respectively, using the median. The preferred penile–anal sexual role was recoded into insertive (exclusively insertive and mostly insertive), receptive (exclusively receptive and mostly receptive), and versatile (both receptive and insertive). The HBM construct scores were each trichotomized into high, moderate, and low, using tertiles as cutoffs, except for cues to action, which were classified as many, some, and few cues.
We compared the high and low PrEP use intention groups in relation to all variables, using bivariate tests as appropriate. Multivariate logistic regression analysis, adjusting for demographic and sex/gender-related characteristics, was performed to examine the HBM constructs associated with high PrEP use intention among MSM. Finally, to identify specific health beliefs significantly associated with high PrEP use intention, a stepwise logistic regression analysis was conducted separately by preferred penile–anal sexual role.
Results
A total of 1329 HIV-negative MSM completed the survey. Table 1 shows that 488 (36.7%) and 841 (63.3%) MSM were classified as having high and low PrEP use intention, respectively. A large proportion of MSM were single (59.1%) and between the ages of 15 and 34 years (87%). Most of the MSM (45%) were receptive, followed by versatile (28.5%) and insertive (26.5%). Table 2 shows that the proportions of MSM with high PrEP use intention increased as perceived susceptibility, perceived benefits, self-efficacy, and cues to action increased. By contrast, as the perceived barriers increased, the proportions of MSM with high PrEP use intention decreased.
Sample Characteristics of Young Men Who Have Sex with Men in Taiwan (n = 1329)
NT$20,000 at the time of the survey was approximately US$630.
As detailed in the Methods section: receptive = exclusively or mostly receptive; insertive = exclusively or mostly insertive; and versatile = both receptive and insertive.
NT, New Taiwan; PrEP, pre-exposure prophylaxis.
Health Belief Model Constructs in Relation to Low Versus High Pre-Exposure Prophylaxis Use Intention
HBM, Health Belief Model.
As shown in Table 3, the multivariate logistic regression model for all MSM found that MSM with high perceived susceptibility [adjusted odds ratio (AOR) = 2.11; 95% confidence interval (CI): 1.53–2.90] and those with high perceived benefits (AOR = 2.14; 95% CI: 1.58–2.88) had greater odds of having high PrEP use intention. In addition, MSM with high self-efficacy (AOR = 4.19; 95% CI: 3.03–5.79) had more than four times the odds of having high PrEP use intention. Having many cues to action (AOR = 1.49; 95% CI: 1.06–2.10) was also found to be associated with high PrEP use intention. Last, MSM with high perceived barriers (AOR = 0.64; 95% CI: 0.46–0.90) had lower odds of having high PrEP use intention.
Multivariate Logistic Regression Estimating the Associations Between Health Belief Model Constructs and High Pre-Exposure Prophylaxis Use Intention Among Young Men Who Have Sex with Men in Taiwan, by Preferred Penile–Anal Sexual Role
As detailed in the Methods section: receptive = exclusively or mostly receptive; insertive = exclusively or mostly insertive; and versatile = both receptive and insertive. All the above models controlled for the following variables: age, education level, employment status, monthly income, relationship status, sexual orientation, gender characteristic, and anal sex experience.
p < 0.05.
AOR, adjusted odds ratio; CI, confidence interval; HBM, Health Belief Model.
Further, when examined separately by preferred penile–anal sexual role, receptive and versatile MSM with high and moderate perceived susceptibility had greater odds of having high PrEP use intention (AOR = 1.78–2.86; 95% CI: 1.09–5.37). High perceived severity was associated with high PrEP use intention only among versatile MSM (AOR = 2.43; 95% CI: 1.20–4.93). High and moderate perceived benefits were significantly associated with high PrEP use intention among MSM of various sexual roles (AOR = 1.94–2.86; 95% CI: 1.24–5.94), except among versatile MSM with moderate perceived benefits. Finally, high perceived barriers were associated with lower PrEP use intention only among insertive MSM (AOR = 0.46; 95% CI: 0.22–0.94).
Table 4 presents results of the multivariate stepwise logistic regression analysis that examined specific health beliefs significantly associated with high PrEP use intention. Perceiving high HIV risk through having multiple sexual partners was associated with increased PrEP use intention only among receptive MSM (AOR = 1.74; 95% CI: 1.16–2.60). Only insertive MSM who considered HIV transmission to partners a serious consequence would have significantly reduced PrEP use intention (AOR = 0.57; 95% CI: 0.34–0.98). Elevated PrEP use intention was found among receptive MSM who felt they would not need to worry about getting HIV if taking PrEP (AOR = 1.92; 95% CI: 1.29–2.85), versatile MSM who felt they could try having sex without condoms (AOR = 1.84; 95% CI: 1.11–3.04), and insertive MSM who felt PrEP use could prevent HIV infection (AOR = 2.28; 95% CI: 1.29–4.02).
Multivariate Logistic Regression Estimating the Associations Between Specific Health Beliefs and High Pre-Exposure Prophylaxis Use Intention Among Young Men Who Have Sex with Men in Taiwan, by Preferred Penile–Anal Sexual Role a
As detailed in the Methods section: receptive = exclusively or mostly receptive; insertive = exclusively or mostly insertive; and versatile = both receptive and insertive. All the above models controlled for the following variables: age, education level, employment status, monthly income, relationship status, sexual orientation, gender characteristic, and anal sex experience. All the above specific health beliefs were dichotomized into high versus low, using the median as the cutoff. Each cue to action was originally dichotomous (yes/no).
p < 0.05; “—” indicates this variable was not included in the model.
AOR, adjusted odds ratio; CI, confidence interval.
Receptive and insertive MSM had four and three barriers, respectively, but versatile MSM had no specific barrier associated with high PrEP use intention. Versatile MSM who were confident about taking PrEP despite side effects (AOR = 2.01; 95% CI: 1.18–3.42) and insertive MSM who were self-efficacious about taking PrEP despite having partners unhappy about their PrEP use (AOR = 2.72; 95% CI: 1.43–5.15) tended to have high PrEP use intention. Having HIV-infected friends (AOR = 1.74; 95% CI: 1.15–2.65) among receptive MSM and having obtained HIV information from apps among versatile MSM (AOR = 1.75; 95% CI: 1.07–2.87) were associated with high PrEP use intention.
Discussion
Overall, higher self-efficacy had the strongest effect on increased PrEP use intention, followed by higher perceived susceptibility and perceived benefits. Consistent with prior research, this study also found self-efficacy to be a significant predictor of willingness to use PrEP. 22 A prior study found that MSM with more risky behaviors had higher adherence to PrEP. 8 In this study, MSM perceiving low risk of HIV infection were less inclined to take PrEP, which is similar to the findings of other previous studies. 3,23 Further, similar to earlier reports, the current study also confirmed that MSM who were aware of the benefits of PrEP were prone to use it. 22,24 –26 In addition, the better informed MSM are about the benefits of PrEP use, the higher their PrEP use intention. These findings underscore the importance of reemphasizing all possible routes of HIV infection and associated risks.
Higher perceived severity was not associated with higher PrEP use intention. One possible explanation is that MSM generally regarded HIV infection as a serious condition. Hence, because of a potential ceiling effect, increased levels of perceived severity could not significantly further elevate PrEP use intention among MSM. In view of such findings, perhaps future PrEP implementation should avoid overplaying the severity of HIV infection, and HIV stigma can also negatively impact the effectiveness of interventions. 9
As anticipated, high perceived barriers were associated with lower PrEP use intention. Interestingly, moderate levels of perceived barriers were not significant, suggesting that there might be a threshold beyond which MSM are hindered from taking PrEP. To our knowledge, this threshold phenomenon has not been reported in the current literature. Hence, PrEP implementation programs should try to address as many barriers as revealed in the current and previous research, which make MSM less willing to use PrEP. 3,5 –7,9,24,27,28 For multi-dimensional cues to action, again, there appeared to be a threshold for this construct to have a significant effect on PrEP use intention. Therefore, PrEP implementation using channels for targeted health communication should be broadened and diversified to optimize message dissemination. In addition to diversification, the communication should also aim to be more inclusive. 6
The effects of HBM constructs and specific health beliefs on PrEP use intention were found to vary by preferred penile–anal sexual role. Among receptive MSM, those who felt at risk for HIV infection, had related cues, and perceived benefits associated with PrEP use were more inclined to use PrEP. This suggests that concerns about HIV risk were strongly felt by receptive MSM to the degree that they would want to take precautions such as PrEP to reduce their worries about HIV infection. Therefore, it may be an effective approach to increase PrEP use intention by reinforcing such risk perception and protective effects of PrEP among receptive MSM, especially those with multiple sexual partners, who have also been found to have greater adherence to PrEP in prior research. 8
Several specific barrier beliefs among receptive MSM were significantly related to PrEP use intention. These findings suggest that PrEP implementation may need to focus more on provider–client communication and public health information to demystify and clarify the prophylactic purpose of PrEP use and to emphasize the importance of expressed partner support of PrEP use. In contrast to a study that found not using PrEP was related to stigma associated with being a PrEP user, 7 receptive MSM who thought they would be misperceived as promiscuous by their doctors turned out to have higher PrEP use intention. Given the cross-sectional nature of this survey, this could be explained reversely such that those receptive MSM intending to use PrEP were more likely to have such concerns of misperception. Despite this finding, providers ought not to regard the request for PrEP as a sign of promiscuity and label their clients as such. Notably, this finding is particularly important in light of prior research in Taiwan, revealing that MSM reported frequently encountering less LGBT-friendly medical staff 10 and that MSM perceiving less HIV-related stigma from physicians were more likely to seek health care. 11
Versatile MSM had some unique health beliefs. Those more concerned about being able to try sex without condoms, feeling self-efficacious despite having side effects of PrEP use, and ever obtaining HIV information from apps were more willing to use PrEP. These empirical results suggest that it is critical to inform versatile MSM who have engaged in CAI about the protective effects of PrEP and the infrequent occurrence of associated side effects. Finally, our finding reveals apps as an important channel to communicate HIV health messages to versatile MSM and to increase their PrEP use intention.
Unlike receptive and versatile MSM, insertive MSM's decision to use PrEP was not associated with their overall perceived HIV risk. However, if insertive MSM felt that CAI with steady partners could put them at risk of HIV, they would be, just like their receptive and versatile counterparts, inclined to take PrEP, suggesting that this key message should be incorporated into future PrEP implementation for all MSM. Among insertive MSM, those deeply concerned about HIV transmission to their partners were found to have lower odds of having high PrEP use intention. A possible explanation is that if they regard transmitting their HIV to their partners as a serious consequence, they might prefer methods perceived to be safer, such as condom use rather than PrEP.
If insertive MSM felt lack of support by their partners, their PrEP use intention decreased; however, if they felt efficacious about PrEP use, even in the face of partners unhappy about their PrEP use, their intention would increase. These findings underscore the importance of partner support and suggest that perhaps through partner counseling, PrEP implementation may focus on enhancing their mutual understanding to help them reach a consensus about PrEP use. Last, insertive MSM expressing concerns about possible decreased condom use had greater odds of having high PrEP use intention; reversely, it could be interpreted as insertive MSM with high PrEP use intention anticipating possible increased CAI as a result of PrEP use.
This study had some limitations. First, the cross-sectional nature of this survey constrained our ability to confirm the causal relationships of the results. However, the associations found in this study provide insights into possible mechanisms through which MSM's intention to use PrEP was determined. Another limitation of this study pertains to its generalizability to the entire YMSM population in Taiwan. However, we advertised the online survey through multiple channels to ensure the diversity of our sample.
In conclusion, this study was among the first to examine PrEP use intention among YMSM of different preferred penile–anal sexual roles in Taiwan using the HBM as a theoretical framework. Less than 40% of the YMSM were classified as having high PrEP use intention. This study also identified several key health beliefs that could enable or hinder PrEP use. These empirical findings of facilitators and barriers associated with PrEP use intention could inform future PrEP implementation. Further, this study revealed several specific health concerns that were uniquely and independently associated with receptive, versatile, and insertive MSM. These important differential associations may provide the foundation upon which tailored PrEP implementation could be devised.
Footnotes
Authors' Contributions
S.-T.H. and J.-H.H. contributed equally to this work, including conception of ideas, design of the study, development of measures, and analysis and interpretation of data, as well as writing and revision of the article. S.-T.H. contributed to data collection. J.-H.H. conceptualized the research framework, supervised the study, and also edited the earlier versions and final draft of the article. J.-H.C. contributed to the preparation, writing, and revision of the article.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
Preparation of this article was supported, in part, by grants, MOST 105-2410-H-002-143-MY2, MOST 103-2410-H-002-225-MY2, NSC 102-2410-H-002-220-, and NSC 100-2410-H-002-174-MY2, from the Ministry of Science and Technology, Taiwan (formerly, National Science Council), awarded to J.-H.H.