Introduction

HIV infection in Peru is concentrated within the core risk group of men who have sex with men (MSM). Surveillance studies of HIV infection in Peru have noted a prevalence in MSM ranging from 11.0–18.5%, but only 1.6% in female sex workers (FSWs) and less than 1% in the general population [1]–[5]. Previous studies have identified several behavioral and biological factors associated with the higher prevalence of HIV among MSM [5]–[10]. However, current understandings of HIV epidemiology among Peruvian MSM are based on studies with venue-based convenience samples that do not provide a complete understanding of the lifetime risk factors or an accurate estimate of the prevalence of HIV and sexually transmitted infections (STIs) among all men who have engaged in sex with men.

The diversity of sexual identities associated with male homosexual behavior in Latin America has been analyzed in a number of studies [5], [11]–[19]. Researchers studying same-sex sexual contact and related risk behaviors among Latino men have described a construction of sexuality that links penetration with masculinity and receptive intercourse with femininity, through which male same-sex sexual contact does not necessarily presume a homosexual identity [19]–[27]. However, this analysis of male-male sexuality is based on studies of Latino communities in the United States, who have been partially or thoroughly influenced by U.S. gay communities and cultures. Understanding of male same-sex contact and associated risk behaviors within Latin America itself is an important area that has only begun to be explored [5], [6], [13], [18], [28], [29].

Previously published estimates of STI and HIV prevalence among MSM in Peru have found high levels of HIV and syphilis infection [5], [9], [10]. However, these estimates were based on samples of men attending MSM venues and STI clinics, or identified as MSM by peer recruiters. A recent analysis of sampling methodologies in the U.K. suggests that community or venue-based sampling not only overrepresents the highest risk segment of the population, but also excludes the potentially large number of men who have sex with men but exist outside of these community frameworks [30]. Particularly when addressing male same-sex behavior in Latin America, failure to include men who do not identify with established GBT social settings and geographic locales provides a skewed understanding of risk behavior and disease prevalence, and neglects the men most likely to serve as a “bridge” between core risk groups and the general population.

We conducted a random household survey of adults in three coastal cities in Peru to assess prevalence and risk factors for HIV and STI transmission in low-income populations. As a secondary analysis, we examined same-sex contact among men in the study population, all of whom were recruited outside of MSM venues or social circles and regardless of their identification with male homosexuality or contemporary participation in an MSM network. We sought to provide a comprehensive description of the characteristics of men who have ever had sex with men (MESM) in Peru's general population and to study the association of MESM activity with sexual risk behavior and risk for STIs.

Methods

Results

Sociodemographic Factors

A total of 2,550 men and women were targeted for recruitment, of whom 2,271 were contacted and 1,645 (72.4%) were eligible and agreed to complete the survey. Almost all participants (1,635/1,645; 99.4%) provided biological samples. There were no significant differences in refusal rates or in behavioral/biological findings between the study sites. A total of 654 men were included in the study, though 96 (14.7%) denied any previous sexual contact and were excluded from further analysis. Any lifetime same-sex sexual contact was reported by 15.2% of all sexually active men (85/558, 95% CI: 12.2%–18.2%), and by 15.9% of sexually active eighteen year olds (11/69, 95% CI: 8.2%–26.7%), with no age-related differences in MESM prevalence observed (Table 1). MESM were generally less educated, with 64.7% having graduated from high school, as opposed to 81.2% of non-MESM (p = 0.001). MESM were also more likely than exclusively heterosexual men to be married or divorced (32.9% vs. 24.3%; p = 0.040), rather than single.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participant characteristics and risk behaviors among general population males: Lima, Chiclayo, and Trujillo; Peru, 2001.

https://doi.org/10.1371/journal.pone.0000778.t001

Prevalence of STIs and STI Symptoms

HSV-2 seropositivity was dramatically higher in MESM than in men who did not report same-sex contact (21.4% vs 5.3%; p<0.001; Table 2). The prevalence of syphilis was also higher among MESM (2.4% vs. 0.7%; p = 0.126), and the prevalence of chlamydia was slightly higher (6.0% vs. 4.5%; p = 0.633), though neither difference reached statistical significance. No cases of HIV (0%, 95% CI: ≤0.002%) or urethral gonorrhea (0%, 95% CI: ≤0.002%) were diagnosed in any male study participant. MESM were more likely than non-MESM to report a previous genital ulcer (20.2% vs. 10.9%; p = 0.016), genital discharge (21.4% vs. 7.9%; p<0.001), or an STI diagnosis (26.0% vs. 6.7%; p <0.001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Prevalence and history of sexually transmitted infections (STIs) among general population males: Lima, Chiclayo, and Trujillo; Peru, 2001.

https://doi.org/10.1371/journal.pone.0000778.t002

Multivariate Analysis

Lower educational level, increased number of sex partners, provision of compensated sex, history of an STI, and HSV-2 seropositivity all remained associated with same-sex sexual contact after multivariate adjustment (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Multivariate analysis of socio-demographic and risk behavior characteristics associated with MESM behavior among general population men: Lima, Chiclayo, and Trujillo; Peru, 2001.

https://doi.org/10.1371/journal.pone.0000778.t003

Discussion

In our survey of men living in low-income neighborhoods in urban Peru, we found that men who reported engaging in same-sex sexual behavior during their lifetime were more likely than men who claimed exclusively female sexual contacts to engage in high-risk sexual behavior, to have had an STI in the past, and to be HSV-2 seropositive. Our findings are consistent with a variety of studies conducted in diverse international settings [1], [4], [5], [12], [21]–[24], [33]–[40], but should be interpreted within Peru's specific cultural context.

In our study, MESM had a greater number of lifetime sexual partners, and were less likely to report using condoms with their casual sexual partners, suggesting that they had more frequent, often unprotected, casual sexual contacts than non-MESM. Research conducted in low-income communities in Peru and other Latin American societies has suggested that the cultural segregation of sexual roles into masculine, heterosexual-identified activos and feminized, homosexual-identified pasivos enables heterosexual-identified men to engage in same-sex encounters without significantly jeopardizing their masculine social status [13], [14], [23]. However, this system of gendered sexual identities typically limits homosexual men to transient sexual encounters with heterosexual-identified men who routinely reject stable same-sex relationships [14], [18], [26].

In addition to these casual sexual contacts, more than one-third (36.5%) of men who reported previous sex with men also reported receiving money or goods in exchange for sex during their lifetime. Participation in compensated sex results in greater exposure to STIs and more frequent high-risk behavior with clients who will pay more money for unprotected intercourse [16], [18]. In the setting of compensated sex, male sexual contact is redefined as a commercial transaction where self-protective behavior is designated as a factor in a financial negotiation, rather than an intimate agreement between sexual partners.

We also observed a similar frequency of reported MESM activity between 18 year olds (15.9%) and 19–30 year olds (15.2%), suggesting that same-sex sexual contact, either as an isolated experiment or as a regular practice, is often initiated during adolescence. Initial sexual experiences with homosexual-identified men are frequently accepted in low-income Peruvian communities without necessarily identifying the adolescent as gay or bisexual [11], [23], [26]. Through this system of sexual initiation, “homosexual” men are used as sexual objects by young men in the community to gain experience before forming stable, normative heterosexual partnerships [11], [14]. In addition to marginalizing gay-identified MSM, this pattern of socialization places adolescent males in intimate contact with a high-risk group for HIV and STIs at a time in which they are least likely to possess adequate knowledge concerning prevention.

Though we did not identify any cases of HIV in our study sample, multiple elements conducive to the spread of HIV infection among MESM were present. Increased prevalences of HSV-2 and syphilis, greater rates of prior STI diagnoses and symptoms, as well as more lifetime partners and more frequent unprotected sex contacts were observed in MESM compared to non-MESM. Unfortunately, while almost one-third of the MESM in our study rated themselves as moderate or high risk for HIV infection, they were no more likely to have been tested for HIV or to have tried to obtain condoms. Though currently untouched by HIV, the MESM in our study provide a behavioral and biological context conducive to the extensive spread of infection following the introduction of HIV into their networks.

There are several limitations to our study, primarily centered on the fact that the survey was designed to evaluate risk behaviors and disease prevalence in a general population sample, without specifically assessing MSM. Participants were not asked whether they identified as heterosexual, homosexual, or bisexual, and the gender of individual sexual contacts was not specified. As a result, MESM are collapsed into a single analytic category that erases social and behavioral differences between subpopulations which may be strongly associated with risk for HIV and STIs. Participants were also not provided with a definition of the term “sex,” an omission that introduces further ambiguity into our findings by requiring participants to supply their own, inconsistent, meanings for the term. Due to stigma surrounding homosexuality and participant underreporting of same-sex activity, we have likely underestimated the prevalence of this activity. In addition, it is difficult to assess whether the higher prevalence of risk behavior reported by MESM is directly related to their sexual contact with other men or if there is a correlation between previous male same-sex sexual contact and a nonspecific propensity to engage in high-risk sexual behavior. The predominance of women in our sample reflects a potential bias in our household-based recruitment strategy since Peruvian men are traditionally expected to be engaged in work or (potentially high-risk) leisure activities outside the home and are less likely to be at home during recruitment visits. However, the household design of our study provides a more representative sample than analyses based on convenience or snowball sampling from MSM-associated venues, and provides the most accurate available measure of MESM risk behavior in Peru. Though Respondent-Driven Sampling (RDS) [41]–[44] and Time-Location Sampling (TLS) [45] may be valuable in future studies for identifying representative samples of MSM in Peru, household sampling is the only methodology known to us that can obtain a random sample that includes men whose sexual contacts with men were entirely in the past and therefore can provide an estimate of all men who have ever had sex with men (MESM). Despite their limitations, our results indicate that MESM are at higher risk for STI transmission and potentially for HIV acquisition than men who denied any lifetime sexual contact with men.

Men who reported ever having “sex” with other men in low-income urban communities in Peru were more likely to engage in high-risk sexual behavior, and had a higher prevalence of STIs than men who did not describe previous same-sex contact. HIV and STI prevention programs need to be developed that are directed to the general population and emphasize the increased risk faced by all men who have sex with men, regardless of their sexual identity, and address behavioral change within the specific contexts of male sexuality in Peruvian society. Continued study of the contexts of male sexual contact in Latin America and their relation to sexual identity and risk behavior is essential for accurate epidemiologic surveillance and improved strategies for the prevention of HIV and STIs.

Acknowledgments

We gratefully acknowledge the contributions of the study participants who shared their lives with us. We also thank the reviewers for their extensive comments on the manuscript.

These results were originally published in abstract form in the proceedings of the XVI Annual International AIDS Conference; Toronto, Canada; August 13–18, 2006 (Abstract # CDD0328).

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the U.S. Government. Dr. Andres Lescano is a civilian employee of the U.S. Government working at the U.S. Navy Medical Research Center Detachment in Lima, Peru. This work was prepared as part of his official duties. Title 17 U.S.C. § 105 provides that ‘Copyright protection under this title is not available for any work of the United States Government’. Title 17 U.S.C. § 101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person's official duties.