High-Risk Sexual Behaviors and Health Services Needs Among Bisexual Men in Peru: A Cross-Sectional Study

Abstract

Objectives

This study aims to describe sexual behaviors and access to health care services among self-identified bisexual men and to examine differences between those who reported recent sexual intercourse with a woman (MSMW) and those who did not (MSM).

Methods

We conducted a cross-sectional study with 560 self-reported bisexual men who responded to a semi-structured survey. Those who reported having intercourse with a woman in the year prior to the survey were classified as MSMW.

Results

MSMW behavior (48.9%) was associated with being 36 years or older and engaging in sexual intercourse under the influence of alcohol (OR: 2.90, 95% CI: 1.65 − 5.10, p < 0.01; OR: 3.33, 95% CI: 2.13 − 5.21, p < 0.01; respectively). Conversely, MSM behavior was significantly associated with self-reporting HIV positive status and engaging in unprotected anal intercourse with any male sexual partner (OR: 0.43, 95% CI: 0.19 − 0.95, p: 0.037; OR: 0.69, 95% CI: 0.46-1.02, p: 0.04; respectively).

Conclusion

Bisexual self-identification does not necessarily correlate with intercourse with both men and women. Instead, it relates to various psychosocial factors that result in distinct behavioral patterns and health needs. It is crucial to develop prevention strategies targeting bisexual men, considering their limited access to health services, sociodemographic characteristics, and the prevalence of high-risk sexual behaviors.

Introduction

HIV remains a critical public health issue in Peru, where approximately 110,000 people are living with HIV/AIDS (PLWHA) (Joint United Nations Programme on HIV/AIDS (UNAIDS), 2023). In 2022, the Ministry of Health reported 9,598 new HIV and 1,424 AIDS cases (Ministerio de Salud del Perú (MINSA), 2023). HIV/AIDS cases in Peru in recent years have increased, especially among people between 25 and 29 years old (Ministerio de Salud del Perú (MINSA), 2023). Men who have sex with men (MSM) are disproportionately impacted by the disease. Official data indicate that 12.4% of MSM in Peru are HIV positive, compared to 0.5% of the general population (Cárcamo et al., 2012). Furthermore, over 50% of new infections in Peru occur among gay men, bisexual men and transgender women (Joint United Nations Programme on HIV/AIDS (UNAIDS), 2018). Moreover, it is estimated that less than 50% of Peruvian gay and bisexual men have been tested for HIV, and approximately 50,000 are unaware of their HIV positive status (Menacho et al., 2013), indicating a need for improved, targeted HIV testing and prevention efforts.

Bisexual men could act as a bridging population for the transmission of HIV and other sexually transmitted infections (STIs) between other MSM and their female partners (Konda et al., 2008; Mercer et al., 2009). This subgroup often faces social stigma, which leads them to conceal their sexual orientation and behavior, and they encounter barriers when accessing health services, such as HIV testing (Bostwick & Dodge, 2019; Dodge et al., 2016; Konda et al., 2008). Despite their unique sexual behaviors and cognitions, bisexual men are often grouped with gay men who have sex with men (MSM) when designing and implementing prevention interventions, which may result in excluding them from important services (Institute of Medicine (US) Committee on Lesbian, Gay, Bisexual, and Transgender Health Issues and Research Gaps and Opportunities, 2011; Zule et al., 2009).

Not all men who self-identify as bisexual engage in sexual intercourse with both men and women regularly. For instance, studies have found that only between 45% and 54% of men who self-identify as bisexual have had sexual intercourse with women throughout their lives (Everett, 2013; Mustanski et al., 2014). Therefore, a significant proportion of bisexuals may not be represented if only men who have had sexual intercourse with both men and women are considered. Moreover, the context in which the bisexual behavior occurs, such as the sociocultural characteristics and values of the community, must be considered to better understand their differences (Bauer & Brennan, 2013).

Classifying men as bisexual based solely on sexual behavior does not take self-determination into account and provides little information about specific sexual behaviors such as sexual role. Furthermore, bisexual behavior can change over time (Bauer & Brennan, 2013; Young & Meyer, 2005). A deeper understanding of this population group is crucial, given its unique cognitive, psychosexual, and sociocultural characteristics, as well as specific healthcare needs (Feinstein & Dodge, 2020). However, in the Peruvian national context, there is limited information about the distinct sexual behaviors and cognitions of bisexual men, resulting in a lack of tailored preventive health interventions. Therefore, it is essential to conduct studies that describe and explain various psychosocial aspects, as differentiated care is necessary for self-identified bisexual men (Feinstein & Dodge, 2020). This study aims to describe sexual behaviors and access to health care services among self-identified bisexual men and to examine differences between those who reported recent sexual intercourse with a woman and those who did not.

Methods

Design and participants

This cross-sectional study comprised a convenience sample of 560 individuals who: a) self-identified as bisexual; b) self-reported engaging in anal intercourse with male partners; c) were at least 18 years of age, signifying they were adults capable of providing informed consent; d) and were residents of Lima, ensuring the study’s relevance to the local context.

Recruitment and data collection

The data collection occurred from July to October 2019 through a semi-structured online survey. The recruitment strategy employed targeted convenience sampling via the Internet, which was designed to: engage a wide-reaching network of this relatively hidden population and prevent under-sampling and ensure the enrollment of participants from diverse sociodemographic backgrounds across the various urban zones of Lima (Allen et al., 2019).

Participants were recruited through targeted advertisements on two popular websites used by member of the bisexual male community and leveraged Facebook’s extensive reach, given its status as one of Peru’s most visited social media sites. The ads featured thematic banners that succinctly communicated the study’s objectives and guided potential participants to the online informed consent form. After consenting to participate, the system automatically requested an email to validate that each survey corresponded to a person and not a bot or computer virus. This process confirmed the uniqueness of each response, mitigating the risk of duplicate entries and safeguarding the integrity of the survey data (Menacho et al., 2018).

Measures

The survey was developed on the SurveyMonkey platform. It consisted of 21 questions divided into three sections: demographics, access to health services, and high-risk sexual behaviors (Table 1). The tool was reviewed by experts working in the field with this population who determined it to be a valid measure on the constructs. It was then piloted with 50 members of the target population, from whom we received qualitative feedback in order to improve the survey before we initiated full online participant recruitment.

Table 1.

Demographic data, access to health services and sexual behaviors in bisexual men in Lima.

Demographics	n	%
Age (m ± SD: 29.74 ± 10.58)
18-24	230	41.07
25-35	185	33.04
36 or older	145	25.89
Education
≤ High school	96	17.14
University/Technical non-graduate	240	42.86
University/Technical graduate	224	40
Smartphone use
Yes	348	73.11
No	128	26.89
Access to health services
Receiving ART
Yes	28	73.68
No	10	26.32
Attended at a center for gays and bisexuals
Yes	21	23.6
No	68	76.4
Last HIV test
Never	229	40.89
In the last year	182	32.5
More than one year ago	149	26.61
HIV self-report
Positive	38	12.26
Negative	272	87.74
Ever treated for a sexual health problem
Yes	210	38.25
No	339	61.75
Preferred center for receiving care
Center for Bisexuals	122	24.85
Center for Gay and Bisexuals	102	20.77
Male Sexual Health Center	267	54.38
Search for sexual health information on the Internet in the last year
Yes	306	60.96
No	196	39.04
Search for sexual health information on the Internet at least once a month
Yes	134	43.93
No	171	56.07
Sexual behavior
Sexual intercourse with a woman during the last year*
Yes (MSMW)	274	48.93
No (MSM)	286	51.07
Sexual role
Top	131	23.86
Versatile	365	66.48
Bottom	53	9.65
Symptoms of STIs in the last year
Yes	147	26.25
No	413	73.75
Being attended for STI symptoms
Yes	90	61.22
No	57	38.78
Told their sexual partner
Yes	42	46.15
No	49	53.85
Alcohol use
Sexual intercourse under the influence of alcohol in the last year
Yes	149	27.14
No	400	72.86
Sexual intercourse under the influence of alcohol in the last year
With men only	72	48.32
With women only	8	5.37
With both	69	46.31
Unprotected vaginal intercourse
With stable female partner
Yes	143	81.25
No	33	18.75
With an acquaintance woman
Yes	80	51.61
No	75	48.39
With sex worker
Yes	8	12.31
No	57	87.69
With any type of female partner
Yes	181	32.32
No	379	67.68
Unprotected anal intercourse with men
With stable male partner
Yes	128	69.95
No	55	30.05
With an acquaintance man
Yes	198	51.43
No	187	48.57
With sex worker
Yes	24	31.17
No	53	68.83
With any type of male partner
Yes	254	45.36
No	306	54.64

*Effective bisexuality: self-identified bisexual men who engaged in sexual intercourse with a woman within the last year; ART: Antiretroviral treatment; STIs: sexually transmitted infections.

The primary outcome was sex with a woman (dichotomized as yes/no), which refers in our study to bisexual individuals who reported engaging in sexual intercourse with at least one woman during the year prior to the survey (MSMW). Those who did not are referred to as MSM throughout this paper. Age was categorized to align with demographic results from similar studies conducted in Peru (Blas et al., 2010, 2011; Menacho et al., 2018). In terms of access to health services, the variable ‘attended a center for gays and bisexuals’ indicates whether participants have ever sought treatment for a sexual health problem at centers working with this population. Sexual role was categorized as top (insertive anal sex), bottom (receptive anal sex) and versatile (insertive and receptive anal sex).

Data analysis

Our analytic sample comprised 560 participants who fulfilled the eligibility criteria from a total of 639 responses. The focus was on discerning the differences between two distinct groups: those who reported sex with women in the past year and those who did not. Utilizing Stata^® (version 15.1, Stata Corporation, College Station, TX, US), we computed frequencies and percentages for categorical variables, alongside the mean and standard deviation for age. Bivariate analysis employed the Chi-square test and Fisher’s exact test for categorical variables. A logistic regression model was constructed for multivariate analysis. This model included variables that were statistically significant in the bivariate analysis (p < 0.05) and those substantiated by the literature. To address multicollinearity, we estimated the variance inflation factor (VIF) and removed any variable with a VIF exceeding 5, enhancing the model’s fit. We assessed the area under the receiver operating characteristic curve (AUC) and evaluated model fit using the Akaike information criterion (AIC). All confidence intervals were established at a 95% level (95% CI).

Ethical considerations

Participants were not provided with any incentives for completing the online survey. The study received approval from the Institutional Ethics Committee of the Universidad Peruana Cayetano Heredia in Lima, Peru.

Results

Univariate analysis

Of the 560 participants, the average age was 29.74 years (M ± SD: 10.58). Over 40% were between 18 and 24 years old, over 80% reported having some level of university/technical education, and 73% reported having a smartphone (Table 2).

Table 2.

Factors associated with having had sexual intercourse with a woman among self-identified bisexual men in Lima, Peru.

	Engaging in sexual intercourse with a woman in the last year
	No (MSM)		Yes (MSMW)
	n	%	n	%	P-value
Demographics
Age					< 0.01
18-24	142	49.65	88	32.12
25-35	95	33.22	90	32.85
36 or older	49	17.13	96	35.04
Education					< 0.01
≤ High school	49	17.13	47	17.15
University/Technical non-graduate	145	50.7	95	34.67
University/Technical graduate	92	32.17	132	48.18
Access to health services
Preferred center for receiving care					0.003
Center for Bisexuals	69	26.74	53	22.75
Center for Gay and Bisexuals	66	25.58	36	15.45
Male Sexual Health Center	123	47.67	144	61.8
Symptoms of STIs in the last year					0.004
Yes	90	31.47	57	20.8
No	196	68.53	217	79.2
Told their sexual partner					0.82
Yes	23	45.1	19	47.5
No	28	54.9	21	52.5
Went to the doctor for symptoms of STIs					0.314
Yes	58	64.44	32	56.14
No	32	35.56	25	43.86
Last HIV test					0.041
Never	126	44.06	103	37.59
In the last year	97	33.92	85	31.02
More than one year ago	63	22.03	86	31.39
HIV self-report					0.009
Positive	26	17.22	12	7.55
Negative	125	82.78	147	92.45
Intention to be tested for HIV in the next 6 months					0.012
Yes	91	72.22	58	56.31
No	35	27.78	45	43.69
Sexual role					0.296
Top	66	23.49	65	24.25
Versatile	193	68.68	172	64.18
Bottom	22	7.83	31	11.57
Sexual intercourse under the influence of alcohol					< 0.01
Yes	58	20.64	91	33.96
No	223	79.36	177	66.04
Sexual behavior
Unprotected anal intercourse with stable male partner					0.989
Yes	72	69.9	56	70
No	31	30.1	24	30
Unprotected anal intercourse with a acquaintance man					0.206
Yes	107	54.59	91	48.15
No	89	45.41	98	51.85
Unprotected anal intercourse with sex worker					0.783
Yes	8	33.33	16	30.19
No	16	66.67	37	69.81
Unprotected anal intercourse with any type of male partner					0.037
Yes	142	49.65	112	40.88
No	144	50.35	162	59.12

STIs: sexually transmitted infections.

Concerning HIV testing, 40.1% reported never having been tested, and 26.6% had undergone testing over a year ago. Of those who had never been tested, 65% reported an intention to do so within the next six months. Among those tested for HIV, 12.3% self-reported being HIV positive. Additionally, 26.3% reported experiencing some STI symptoms in the last year (such as pus, blisters, wounds, warts on the penis and/or anus), with 53.9% failing to notify their sexual partners. Only 38.3% reported ever being treated for a sexual health problem, and 54% preferred treatment at a sexual health center exclusively for men. Regarding information-seeking behavior on sexual health, 61% searched for information online, with 44% doing so at least once a month (Table 2).

In terms of access to health services, over 70% (28/38) of those self-reporting as HIV positive disclosed using antiretroviral treatment (ART). Among those with reported STI symptoms in the last year, 61.2% (90/147) sought medical attention for their symptoms. Of these, 23.6% visited a healthcare center for gays and bisexuals, while 76.4% sought care at a center for the general population.

Regarding sexual behavior, 48.9% reported engaging in vaginal sex with women in the prior year (MSMW). Approximately 68% self-identified as "versatile" in their sexual role with another man. When asked about high-risk sexual behaviors, 27% reported engaging in sexual intercourse under the influence of alcohol in the last year. During vaginal intercourse with stable female partners, 81.3% did not use a condom, while 70% did not use one during anal intercourse with stable male partners. Over 50% reported having unprotected intercourse with an acquaintance sexual partner of either sex. Moreover, 12.3% did not use protection during vaginal intercourse with sex workers, and 31.2% reported unprotected anal intercourse with male sex workers. Overall, 32.3% of participants reported condomless sex with women (vaginal intercourse) and 45.4% with men (anal intercourse) regardless of partner type (e.g., stable, acquaintance, or sex worker) (Table 2).

Bivariate analysis

Bisexual individuals who reported engaging in vaginal sex with women in the last year (MSMW) were older (p < 0.01), had a higher level of education (p < 0.01), had a longer time since their last HIV test (p = 0.041), and preferred treatment at a sexual health center for men (p = 0.003). Conversely, those who did not report engaging in vaginal sex with women in the last (MSM) year were younger (p < 0.01), had a lower level of education (p < 0.01), had never been tested for HIV (p = 0.041), and preferred treatment at a sexual health center for gays and/or bisexuals (p = 0.003). Furthermore, compared to bisexual MSM, MSMW reported fewer STI symptoms in the last year (p = 0.04), more frequently reported being HIV negative (p = 0.009), expressed a weaker intention to get tested for HIV (p = 0.012), and reported more sexual intercourse under the influence of alcohol (p < 0.01) (Table 2).

Multivariate analysis

After adjusting for confounding variables, MSMW behavior in the last year was associated with being 36 or older (compared to those aged 18 to 24) and having sexual intercourse under the influence of alcohol (OR: 2.90, 95% CI: 1.65 − 5.10, p < 0.01; OR: 3.33, 95% CI: 2.13 − 5.21, p < 0.01, respectively) (Table 3). On the other hand, MSM behavior was significantly associated with being HIV positive and having unprotected anal intercourse with any type of male sexual partner (OR: 0.43, 95% CI: 0.19 − 0.95, p: 0.037; OR: 0.69, 95% CI: 0.46 − 1.02, p: 0.04, respectively). The final model retained variables such as education level, care for STI symptoms, recent HIV testing, future HIV testing intentions, online sexual health information seeking in the past year, and preferred healthcare center, due to their potential confounding effects.

Table 3.

Multivariate logistic regression analysis of the impact of sociodemographic, high-risk sexual behaviors, and access to health services factors on MSMW behavior among self-identified bisexual men in Lima, Peru.

	Crude OR (95% CI)	P-value	Adjusted OR (95% CI)	P-value
Demographics
Age
18-24	–	–	–	–
25-35	1.53 (1.03 − 2.26)	0.034	1.32 (0.82 − 2.14)	0.254
36 or older	3.16 (2.05 − 4.88)	≈ 0	2.90 (1.65 − 5.10)	< 0.01
Education
≤ High school	–	–	–	–
University/Technical non-graduate	0.68 (0.42 − 1.10)	0.117	0.76 (0.45 − 1.30)	0.325
University/Technical graduate	1.50 (0.93 − 2.42)	0.101	1.11 (0.63 − 1.94)	0.719
Sexual behavior
Intercourse under the influence of alcohol
No	–	–	–	–
Yes	1.98 (1.35 − 2.90)	0.001	3.33 (2.13 − 5.21)	< 0.01
Did not answer	1.51 (0.45 − 5.04)	0.501	1.00 (0.24 − 4.24)	1
Unprotected anal intercourse with any type of male partner
No	–	–	–	–
Yes	0.70 (0.50 − 0.98)	0.037	0.69 (0.46 − 1.02)	0.04
Access to health services
Preferred center for receiving care
Center for Bisexuals	–	–	–	–
Gay and Bisexual Center	0.71 (0.41 − 1.22)	0.215	0.77 (0.43 − 1.39)	0.391
Male Sexual Health Center	1.52 (0.99 − 2.35)	0.055	1.41 (0.88 − 2.26)	0.157
Did not answer	1.91 (1.05 − 3.48)	0.035	0.72 (0.28 − 1.84)	0.488
Symptoms of STIs in the last year
No	0.57 (0.39 − 0.84)	0.004
Yes	–	–	–	–
Last HIV test
Never	–	–	–	–
In the last year	1.07 (0.73 − 1.58)	0.727	0.82 (0.27 − 2.54)	0.734
More than one year ago	1.67 (1.10 − 2.53)	0.016	1.04 (0.33 − 3.23)	0.952
Being attended for STI symptoms
No	–	–	–	–
Yes	0.71 (0.36 − 1.39)	0.315	0.55 (0.26 − 1.18)	0.123
No symptoms reported	1.42 (0.81 − 2.48)	0.22	1.03 (0.55 − 1.94)	0.928
HIV self-report
Negative	–	–	–	–
Positive	0.39 (1.19 − 0.81)	0.011	0.43 (0.19 − 0.95)	0.037
Never been tested	0.72 (0.51 − 1.02)	0.067	1.14 (0.42 − 3.07)	0.795
Intention to be tested for HIV in the next 6 months
No	–	–	–	–
Yes	0.50 (0.29 − 0.86)	0.013	0.60 (0.33 − 1.09)	0.096
Did not answer or had it done recently	0.83 (0.51 − 1.36)	0.461	–	–
Search for sexual health information on the Internet in the last year
No	–	–	–	–
Yes	0.73 (0.51 − 1.04)	0.083	0.82 (0.55 − 1.23)	0.332
Did not answer	1.62 (0.89 − 1.44)	0.116	2.95 (1.08 − 8.08)	0.035

OR: Odds ratio; 95% CI: 95% Confidence Intervals; STIs: sexually transmitted infections.

Discussion

The results of this study describe the sexual behaviors, sociodemographic aspects, and access to health services of a sample of self-identified bisexual men from Lima, Peru. Additionally, we observed some differences between bisexual individuals who reported engaging in intercourse with a woman in the past year (MSMW) and those who did not (MSM).

The mean age and education of the participants were similar to those reported in other local studies using the same methodology (Blas et al., 2011; Menacho et al., 2018). Self-reported HIV prevalence is the same as the one reported nationally, which is notably higher compared to the general population (Ministerio de Salud del Perú (MINSA), 2023; Garcia-Fernandez et al., 2018). The majority of participants had either never been tested for HIV or had not been tested in the last year, with a higher self-report of HIV among those who did not recently have sex with women. As other studies in different settings have highlighted, it is critical to strengthen strategies to promote HIV testing and linkage to care among bisexual men (Feinstein & Dodge, 2020).

Bisexual MSMW reported less frequent HIV testing and lower intention to get an HIV test. This aligns with findings from other researchers who reported that bisexual men get tested for HIV less frequently than gay men (Feinstein et al., 2019; Jeffries, 2010; Jin et al., 2002). Extant research also demonstrates that MSMW are more likely to be unaware of their HIV-positive status, to avoid HIV care, to receive a late diagnosis, and to be diagnosed with AIDS one to three years after the initial HIV diagnosis (Friedman et al., 2014, 2018; Singh et al., 2014). Given the decreased engagement in HIV testing observed among MSMW in the current study, Peruvian MSMW with HIV may also experience late diagnosis and be less likely to engage in care. This reflects the societal stigma surrounding bisexuality and the intersection of multiple identities, which can affect individuals’ willingness to seek healthcare and sexual health services. Additional research is needed to further understand gaps in the HIV care cascade for bisexual MSMW in Peru. Moreover, lower rates of HIV testing engagement suggest a critical need to design and implement tailored HIV prevention interventions for this population (Feinstein & Dodge, 2020).

Half of the sample did not report engaging in vaginal sex with a woman in the last year even though they self-identified as bisexual. History of intercourse with both men and women seems to be insufficient to categorize a man as bisexual; indeed, bisexuality is a dynamic concept encompassing affective, circumstantial, and temporal aspects toward both sexes (Federación estatal de lesbianas, gays, trans y bisexuales \(FELGTB\), 2012). For example, men self-identified as bisexual differ from men who have sex with men and women who are not self-identified as bisexual. For instance, self-identified bisexual men marry and have children less frequently, engage in more anal intercourse with men, and are more open about their sexual behavior and orientation (Lever et al., 1992; Ortiz Hernández & García Torres, 2005; Schrimshaw et al., 2013). These individuals may participate in social networks and communities with which MSMW might not engage in that could limit their access to health services.

We found a high rate of condomless sex, consistent with other studies among men who have sex with men (Everett, 2013; Konda et al., 2008; Ortiz Hernández & García Torres, 2005; Zule et al., 2009). Our study supports the findings of other authors regarding the higher rates of unprotected sex with stable female and male partners (Konda et al., 2008; Stokes et al., 1993). It is important to emphasize, as reported by Stokes et al., that the rate of condom use was lower among steady female partners (Stokes et al., 1993). Condom use appears to be associated with the type of partner. They are used more frequently with unknown or less familiar women, while they are less likely to be used with stable female partners (Chirinos et al., 2006; Konda et al., 2008; Stokes et al., 1993). Similar to gay men, bisexual men face a higher risk of acquiring HIV/STIs compared to heterosexual men (Jeffries, 2010). One study reported that men who had sex with both men and women were five times more likely to have HIV than men who exclusively had sex with women, but half as much as men who exclusively had sex with men (Hubach et al., 2014). Moreover, self-identified bisexual men also experience a higher incidence of other STIs than men who exclusively have sex with women (Jeffries, 2010). Bisexual men with high-risk sexual behaviors and who do not receive regular HIV testing potentially put their long-term female partners at risk. These findings suggest that bisexual men may play a role in the transmission of STIs and HIV to the general population, highlighting the importance of designing tailored interventions for them (Konda et al., 2008; Zule et al., 2009).

Participants reported more unprotected sex with male partners than with female partners when considering any type of partner (e.g., stable or casual). Other studies have also reported condom use patterns based on the partner’s gender (Feinstein & Dodge, 2020). For instance, some bisexual men reported not using condoms if the partner could not get pregnant (Liao et al., 2011). In a Peruvian study, men used condoms with stable female partners to prevent pregnancy but not to prevent STI-HIV transmission (Chirinos et al., 2006). Cognitions regarding non-condom use often start during adolescence and are linked to reduced sexual pleasure and the perception that condoms are used more by adults, and that they are not completely effective. These cognitions may persist in subsequent stages with stable female partners. Bisexual men consistently report unprotected sex and prevention strategies should not only consider the promotion of condom but also the delivery of preexposure prophylaxis (PrEP). Efforts to explore and understand implementation gaps, as well as opportunities to improve access to biomedical interventions like PrEP are needed for bisexual men in Peru (Murphy et al., 2023).

Alcohol use is a risk factor for HIV/STIs for bisexual men and their sexual partners. In our study, MSMW were more likely to engage in intercourse with women under the influence of alcohol. Other authors have found that bisexual men used more alcohol and marijuana with women, which might contribute to condomless sex (Feinstein & Dodge, 2020). Additionally, it has been reported that alcohol consumption before or during sex was more prevalent among those worried about their sexual orientation being disclosed (Feinstein & Dodge, 2020). In fact, internalized stigma is a proximal stressor that is related to substance use and this association is stronger among youth (Watson et al., 2019). One study found that sexual and gender minority with higher levels of internalized stigma, reported more episodes of binge and heavy drinking (Baiocco et al., 2010). In addition, bisexual men can face stigma against their identity from other subgroups of MSM and it is important to explore if internalized stigma is stronger associated with alcohol use among them (Watson et al., 2019).

Bisexual men are often not reached by traditional methods, such as the delivery of prevention interventions through peer educators (Blas et al., 2007; Orellana et al., 2006). It is critical to create targeted services for self-identified bisexual men, both with MSM and MSMW behavior, based on preferences and risk factors identified in this study; highlighting that those who report more recent sexual intercourse with women have different high-risk sexual behaviors. Such targeted services must include the delivery of PrEP, among those who meet the criteria, which is provided by the Department of Health since 2023. Effective implementation of PrEP for this populations may require tailored approaches. As such, it is important to study barriers and facilitators to PrEP uptake and retention in these populations, and design strategies to overcome these barriers and capitalize on facilitators (Menacho et al., 2024). Moreover, we need to promote access to HIV and STI prevention and care services with targeted campaigns, similar to what we did to enroll more bisexual men in the current study. It is also important to consider the stigma and discrimination faced by bisexual men, as well as the influence of machismo in Latin America, which hinders their access to care (Bostwick & Dodge, 2019; Dodge et al., 2016; Konda et al., 2008; Manji et al., 2007). Understanding these cultural dynamics is essential for designing effective health interventions that are both culturally relevant and empowering for the individuals they are intended to serve. This is directly related to the training and sensitivity of healthcare workers in using neutral, non-stigmatizing language and communication, which is essential to providing care for bisexual men without discrimination (Menacho et al., 2015; Menacho & Blas, 2015).

Limitations and strengths

Findings from our study must be interpreted in the context of its limitations. The data collection method, which involved using an online survey, may yield a sample that is not representative of bisexual men in Peru. On the one hand, it is crucial to consider self-selection bias in online surveys, where those who are most interested are more likely to participate. Although no incentives for participation were provided, some participant responses indicated likely misrepresentation and duplicity. These respondents were identified and excluded from the analysis. Participation bias may also exist because not all bisexual men have access to the websites where the advertisement was posted online. In addition, our study population is limited to bisexual men residing in Lima, the capital city. This geographical focus inherently limits the diversity of experiences represented in our research, as the social, cultural, and economic conditions in Lima may differ significantly from those in rural areas or other regions of Peru. On the other hand, recruiting participants online is an effective way to access large samples from otherwise inaccessible groups. Indeed, recruiting a significant number of bisexual men is a strength of our study, given the hidden nature of this population (Menacho et al., 2018).

Few studies have examined the specific sexual risk factor of Peruvian self-identified bisexual MSM and MSMW. We found that among young bisexual men with an average educational level in Lima, there is limited access to health services, including HIV testing. This situation may be associated with low awareness of the importance of prevention and various barriers to health service access. Bisexual self-identification extends beyond the frequency of intercourse with both men and women, incorporating psychosocial factors that shape behavioral patterns and preferences within this group. The association between the absence of recent intercourse with women and unprotected intercourse with male partners, coupled with the high prevalence of unprotected intercourse with stable female partners among MSMW, could exacerbate the transmission of HIV/STIs.

Disclosure statement

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