AIDS and Behavior

, Volume 12, Issue 1, pp 18–26

Men Who Have Sex with Transgender Women: Challenges to Category-based HIV Prevention

Authors

    • Department of Social Policy and Social WorkUniversity of Oxford
  • Jennifer Burton
    • Department of Social Policy and Social WorkUniversity of Oxford
  • Kristen Underhill
    • Department of Social Policy and Social WorkUniversity of Oxford
  • Jae Sevelius
    • Center for AIDS Prevention StudiesUniversity of California
Original Paper

DOI: 10.1007/s10461-007-9303-y

Cite this article as:
Operario, D., Burton, J., Underhill, K. et al. AIDS Behav (2008) 12: 18. doi:10.1007/s10461-007-9303-y

Abstract

Although transgender women are acknowledged as a priority population for HIV prevention, there is little knowledge on men who have sex with transgender women (MSTGWs). MSTGWs challenge conventional sexual orientation categories in public health and HIV prevention research, and warrant increased attention from the public health community. This paper used qualitative techniques to analyze how MSTGWs describe their sexual orientation identities, and to explore the correspondence between men’s identities and sexual behaviors with transgender women. We conducted in-depth semi-structured individual interviews with 46 MSTGWs in San Francisco. We observed a diversity in the ways participants identified and explained their sexual orientation, and found no consistent patterns between how men described their sexual orientation identity versus their sexual behavior and attraction to transgender women. Findings from this qualitative study question the utility of category-based approaches to HIV prevention with MSTGWs and offer insights into developing HIV interventions for these men.

Keywords

MenTransgenderHIVSexual orientationQualitative

Introduction

HIV prevention research and intervention programs have largely focused on discrete groups identified as having high risk for HIV transmission. In the United States, priority for HIV surveillance and prevention has been given to men who have sex with men (MSM), injection drug users, and heterosexual partners of HIV-infected individuals (Lansky et al. 2007). Another group with particularly elevated HIV prevalence and incidence is transgender women (Kenagy 2002). These women are classified as male at birth, but identify as women and express a female gender identity (Israel and Tarver 1998). Studies of HIV seroprevalence in San Francisco estimate that transgender women experience the highest HIV incidence for any risk group (Clements-Nolle et al. 2001; Kellogg et al. 2001). In other metropolitan areas in the world, HIV prevalence in samples of transgender women have ranged from 11% to 78% (Elifson et al. 1993; Kellogg et al. 2001; Modan et al. 1992; Nemoto et al. 2004; Spizzichino et al. 2001; Varella et al. 1996). Local and national public health organizations accordingly have prioritized resources to improve HIV prevention research and sexual health educational programs for transgender women.

Previous studies among transgender women suggest that HIV-related risk behaviors and unprotected sex occur most frequently in the context of relationships with a primary male partner (Clements-Nolle et al. 2001; Kellogg et al. 2001; Nemoto et al. 2004). However, to date very little research has focused on men who have sex with transgender women (MSTGWs). These men challenge conventional categorical distinctions between sexual orientation and sexual behavior used in public health and HIV prevention research. These men might identify as either heterosexual, homosexual, or bisexual, or they might not identify with any of these categories. Their relationships with transgender women might range from long-term stable partnerships to episodic partnerships characterized by commercial sex. They might engage in a range of penile insertive or anal receptive sexual behaviors, and they might have sex with transgender women who are at different stages of gender transition and gender confirmation status. In addition to sexual partnerships with transgender women, they might concurrently have sexual partnerships with biological women or men. This variety in partnerships, behaviors, and identities challenges efforts to target appropriate HIV services to these men.

A review of published research on MSTGWs has discovered few relevant studies that specifically theorized men’s sexual attraction to transgender women. Identified studies generally classify and describe these men in psychiatric or psychoanalytic terms, without providing insight into their lived realities and experiences (Blanchard et al. 1993; Huxley et al. 1981; Money and Lamacz 1984). In order to improve understanding of the HIV prevention needs of MSTGWs, it is necessary to describe and analyze their subjective meanings around sexual orientation identity and sexual behaviour. Throughout this paper, sexual orientation identity refers to the descriptive terms individuals use to signify their sexual and/or erotic attraction to others. In public health and popular discourse, there currently exist a variety of previously characterized sexual orientation identity categories: these include heterosexual/straight, bisexual, homosexual/gay/lesbian, asexual, and others. However, some individuals do not identify as members of any of these categories. Notably, sexual orientation identity terminology generally reflects gendered patterns of sexual attraction, but they do not necessarily denote behavior or experience (reflected in behavior-based terms such as MSM, WSW, or MSTGW). Sexual orientation identity is also distinct from other aspects of identity such as gender identity (male, female, transgender, intersex) and gender roles (masculine, feminine). Recent debates have reminded HIV prevention researchers and health education professionals that self-ascribed sexual orientations identities do not always align with sexual behaviors (Pathela et al. 2006; Young and Meyer 2005). The validity of public health categories such as MSM and WSW has been challenged, as these terms may offer little insight into the lived experiences of individuals that fall within these behavioral parameters (Young and Meyer 2005). Prior assessments of men’s sexual behavior have often considered sex with transgender women to be a subset of MSM activity (Beyrer et al. 2005; Hernandez et al. 2006; Pisani et al. 2004) and, consequently, might not capture a potentially unique sexual and gender dynamic.

Given advances in research into transgendered populations (Boehmer 2002) and gender identity (Wylie 2004), as well as political and demographic changes due to the HIV epidemic and increased LGBT activism, an investigation into patterns of sexual orientation identity and behavior among MSTGWs is warranted. This paper uses qualitative methods to explore two questions: (1) How do MSTGWs define and describe their sexual orientation identity? and (2) Are men’s sexual orientation identities associated with the way they describe their past sexual behaviors and attraction to transgender women? By examining the men’s personal stories and narratives, we hope to assess the appropriateness of conventional sexual orientation categories for providing HIV prevention and sexual health education programs to these men.

Methods

Semi-structured individual interviews were used to explore patterns of sexual orientation, identity, and behavior in a sample of MSTGWs in San Francisco. Past research in various populations has suggested that sexual identities can be “constructed” or “narrated into existence” through the telling of “sexual stories” (Plummer 1995; Schrock 2006). Qualitative methods were therefore deemed an appropriate technique to understand how these men narrate their experiences and construct their own identities with regard to sexual experiences with transgender women.

Participants

Forty-six male participants were recruited between January and August 2004 using two techniques. First, we identified a group of transgender women in San Francisco who had currently or recently been in a relationship, and these women referred their male partners to the study. Second, we conducted outreach at venues identified through community mapping as spaces where men socialize and meet with transgender women, including bars, nightclubs, and community-based organizations. Advertisements about the study were posted at these venues, and men were individually approached and informed about the study. All men who responded to outreach or recruitment information were screened in-person or by telephone and, for those who met criteria, scheduled for an individual interview; men were asked if they had any gender preferences for interviewers (i.e., a transgender woman or a man), but none expressed preferences. Men were eligible for this study if there were a biological male, aged 18 years or older, had ever had sex with a transgender woman, and lived or worked in the San Francisco area.

Procedure

Screened participants arrived at a private research office where they were met by one of two interviewers. Both interviewers were transgender women with significant prior experience conducting HIV research using qualitative interview methods. We used transgender women interviewers based on our assumption that MSTGWs might be more comfortable and open discussing their sexual histories with transgender women, rather than with biological male or female interviewers. Conducting interviewers using biological male or female interviewers might have incurred a potential risk for self-presentational bias or censorship among participants. Interviewers received in-depth training about the aims of the study, interviewing and probing techniques, and ethical issues. Interviewers met weekly with the principal investigator to discuss the content and quality of interviews and to make adjustments to the interview protocol to improve clarity and focus.

After providing informed consent, participants completed a brief demographic questionnaire and then responded to a semi-structured interview covering four general areas: (1) relationship and dating history, (2) past sexual behavior and relationships with transgender women, (3) sexual orientation, and (4) awareness and understandings about HIV risk. There were 22 core questions overall in the interview protocol, though interviewers were trained on adding probing questions to elicit more detail or omitting certain questions if they were redundant or unnecessary. All interviews were tape-recorded, and men were encouraged to use their first names only or pseudonyms. Interviews lasted about one hour. At the completion of interviews, men received reimbursement for time and travel costs and received information on local HIV prevention and testing services. Audio tapes were immediately transcribed by a professional transcription agency; all names were omitted from transcripts. Research procedures were reviewed and approved by the Committee on Human Research at the University of California San Francisco.

Analysis

Data analysis was conducted by three separate coders, using a four-stage process. First, two coders read each transcript for general understanding of content and range of topics; a third coder read a subset of 10 transcripts. The two principal coders developed a list of thematic codes—using standard techniques of marginal remarks and memo writing—to categorize the major themes from the data. Through discussion of the identified themes, a coding book was developed. Second, to establish consistency and uniformity in coding, two coders independently coded six random interview transcripts using the codebook, with the third coder helping to resolve any discrepancies in coding. Third, after it was established that coders reached an 80% agreement, they then independently coded the remaining transcripts. Each transcript was coded by hand using the codebook, with transcript lines linked to appropriate codes. Fourth, coded data were aggregated by theme and entered into spreadsheets, which were used to address the research questions specified for this specific analysis.

The descriptive analyses here served two aims: (a) to examine how the men articulated a sexual orientation identity, noting any specific trends, and (b) to examine how sexual orientation identities related (or did not relate) to sexual behaviors and behavioral preferences with transgender partners.

Results

Sample Characteristics

Of 46 participants, 45 men were residents in the Bay Area of San Francisco and one man resided in another state but visited San Francisco regularly for business. Twenty-four men were recruited from venues, and 21 learned about the study through a personal referral. Ages ranged from 18 to 56 years. Forty-four men reported having had sex with a transgender woman within the past year, and 13 men described themselves as currently in a relationship with a transgender woman. Nineteen participants were white, 19 were African American, 6 were Latino, and 2 were of other ethnicities. Most (n = 32) of the men reported themselves as being HIV-negative, 9 as HIV-positive, and 5 reported not knowing their HIV status; we did not ask when they last were tested for HIV, so self reports on HIV status may no longer have been accurate. Half (n = 23) of the men reported having had past sexual behaviors with transgender women, biological women, and males; 22 men reported having had past sexual behaviors with transgender women and biological women; 1 man reported having had past sexual behaviors with transgender women only.

Narratives on Sexual Orientation Identity

To prompt discussion about this topic, interviewers stated, “Some guys describe themselves as heterosexual or straight, some guys describe themselves as homosexual or gay, some guys describe themselves as bisexual. Other guys have other ways of describing their sexuality. We know that these words have different meanings to different people, and that there is no right or wrong way for identifying your sexuality. We want to know how you would describe your sexual identity or sexual orientation, and what that means to you.”

Straight Identity

Nearly half (n = 20) of the men described having a straight or heterosexual sexual orientation identity, though there were noteworthy differences in response styles. Ten of these men responded without hesitation that they identified as heterosexual. These ten men offered little or no explanation about their sexual orientation identity, though when prompted they characterized being heterosexual or straight as a default identification in the absence of attraction to men. Without question, these men viewed transgender women as they would any other woman, making no distinctions or qualifications on their sexual attraction to transgender women compared with attraction to biological women.

“I like women. I’m straight. I don’t like men.”

“Straight. Gay means strictly you’re, you’re um…you’re attracted to the same sex. And straight means you’re attracted to the opposite sex.”

Other men showed more complexity in their answers. Seven men who identified as heterosexual took more time to think about their response, with lengthy pauses and deliberation about why they identify as heterosexual, for example:

“You can look at it as straight….or you can look at it [as] a man liking other females…you know what I’m saying? I mean… But still straight. ‘Cause I mean, as long as you ain’t receiving. That’s when you switch. That’s the way I look at it. You know. When you’re the receiver then you start experiencing something different now. Something else.”

“[I am] screwed up [jokingly]… No, I tell myself straight. If it doesn’t look like a girl—I’m not judging anyone—but I’m not interested.”

“[pause]…I’m pretty much straight. But if you want to say I’m bi go ahead...[pause]. No. I’m pretty much a straight guy. But I’ll have gay sex.”

“I’ve been fighting with that bisexual thing. You know what I’m saying? I’m…[pause]..But I’d describe myself as heterosexual.”

Men in this latter group expressed an internal tension between describing a heterosexual identity versus acknowledging that their sexual behavior challenged conventional categories. Some of these men considered aloud whether they should identify as bisexual, but ultimately declared a straight/heterosexual identity with the acknowledgment that their sexual orientation identity was difficult to categorize. One additional man who identified himself as heterosexual described how his sexual orientation identity had changed over time, and that his previous attraction to transgender women was linked with substance use: “Now that I’m out of my addiction [I am] probably straight, but in my addiction I was different… I only like transgenders when it’s in my addiction. That’s it. But I like women now. I mean I’m not really attracted to transgenders now”. In this example we see how sexual orientation identity can fluctuate over time, and that being heterosexual might not represent a crystallized identity for some of these men.

Bisexual Identity

Bisexual was the second most common sexual orientation identity in this sample, claimed by 14 men. Again, these men showed differences in narratives about their bisexual identity. Half of these men promptly identified themselves as bisexual with little further clarification, for example, “I’m a true bisexual, because I’ve always been attracted to real pretty girls or real pretty boys and it didn’t matter.” The other seven men deliberated on their response to the question, and eventually concluded that they identified as bisexual:

“I don’t really think of it in that term… It would be bisexual because my girl is TG [transgender] so I think of her as a girl, so in some ways I’m a straight person but in reality I guess some people would put me in the bisexual…”

“…my sexual orientation would have to be bisexual. I really couldn’t call that heterosexual because I love queens and in my mind I’m knowing that this is a man but I count them as a woman. So if you want to say heterosexual gay on that side then I would have to say heterosexual. But to me I would say bisexual.”

Here we observed men struggling with choosing a single sexual orientation identity, as they appeared aware of the disparity between how they viewed themselves versus how others might view them based on their attraction to transgender women. Consequently, these men appeared to re-examine their orientation specifically in the light of their attraction to transgender women. Two men voiced frustration and with categorical labels for sexuality, yet ultimately responded to the question by identifying as bisexual:

“[the terms]…they’re irrelevant. Like I said, I’m a man, that’s all… Well, I guess if you want to go by the labels, you know, I guess you’d have to call me bisexual, I would imagine.”

“Bisexual doesn’t mean anything…just a word. Pinning something on someone. ‘Oh, you’re gay, you like men.’ So what? I’m bisexual. I like both people. Is that label, does that label me too? So what?”

We observed that, compared with men who identified as heterosexual, narratives from men who identified as bisexual showed a stronger sense of tension and dissatisfaction with the need to declare and explain a sexual orientation identity. This tension might reflect a greater sense of awareness of the problems in category labels for sexuality, and of the potential for a disconnection between self identifications and perceptions made by others based on having sexual relationships with transgender women.

Gay Identity

Four men identified themselves as gay. Three of them showed no hesitation in describing themselves unequivocally as gay. One of these men explained his identity in greater depth:

“I guess I feel more attracted to men a lot of times, even though I’m dating this transgender and she go with the process but she’s still a man, okay? Even though she wants to be a woman. It’s just something I feel more connected with. People like that. Basically I have nothing to hide. Okay, I’m 100 percent gay. Okay, even though I may have sex with men and transgenders or females, I consider myself as a gay male.”

Another man who identified as gay showed a degree of hesitancy and ambivalence before claiming this identity, following the pattern observed among men who identified as straight or bisexual:

“I’m me. I do what I want to do. I don’t think somebody’s straight in the world. I don’t… Everybody has tendencies. Society got to label everybody so I guess I’ll go for gay.”

Choosing Not to Identify

We observed a group of men (n = 8) who responded to this question by explicitly declaring that they did not identify with any sexual orientation category. Each of these men clearly thought about the question before answering. Their responses shared two themes—(a) a disregard for labeling and (b) a focus on being sexual.

“[I am] just sexual. I don’t have any orientation. I’ve fallen in love twice with post-operative transsexuals, transgender women and so I don’t know what that says about me. I don’t really put any labels on it. I just go with what I feel.”

“I’m a try-sexual. I’ll try anything. I don’t label myself. I used to. I used to label myself as being straight but then since I find myself attracted to the transgender I’m like, ‘Well okay, so I can’t use straight. Am I gay? Am I this? Am I that?’ So I just say I don’t need to label myself.”

“…these days I tend to socially just say that I’m a sexual person and the rest is none of your business unless we’re kicking it.”

In comparison with men who labeled their sexual orientation identity (as straight, bisexual, or gay), these men’s narratives appeared more nuanced and complex, characterized by interweaving personal reflections with descriptive information on their sexuality, and noting contradictions between sexual identities and behaviors. Based on this observation, it could be hypothesized that there is an association between men choosing not to identify their sexual orientation and a greater comfort and capacity to narrate on their non-categorizable sexuality. Our data, however, did not allow us to directly examine why participants classified their sexual orientation in different ways, for example whether this might be associated with age, beyond noting substantial heterogeneity in men’s narratives and sexual identities.

Correspondence Between Sexual Orientation Identity and Sexual Behaviors

During the interview, participants were asked to describe their sexual behavior preferences and narrate on their past sexual relationships with transgender women. We examined trends in the ways men described their sexual behaviors and attractions to transgender women. We then attempted to observe patterns of association between men’s constructed identities and narratives about their stories of sexual behaviors and attractions to transgender women.

General Patterns of Attraction

We observed three general trends by which men described their personal and erotic attractions to transgender women. The first general trend was a tendency among men to describe being attracted to specific individuals who happened to be transgender women, rather than being attracted to transgender women in general. Accordingly, men described a unique interpersonal and emotional connection with a transgender partner, and this person-specific attraction appeared to prevail over preferences based on erotic or anatomically gendered tendencies. These narratives suggest that some men minimize physical preferences in favor of emotions and personality drawing them to transgender partners.

“And it’s just, I like his personal, their personality. I like this person as a person. It wasn’t just because she was transgender I was attracted to her, it was just the person… I look at her as female. I mean, well, I mean the same way I’d look at a female.

“…I guess as I’ve gotten older the personality of the person, basically that’s it. Because now I don’t distinguish between the fact that this person is or was a man and now has become a woman. I just look at this person, that this is just who they are.”

By contrast, a second observed trend was a tendency to describe categorically being attracted to transgender women because of what they represent, as a challenge to the traditional gender dichotomy. In this vein, men appeared drawn to the exoticism surrounding transgender women, who seemingly defy notions of masculinity and femininity through their personalities, appearances, and behaviors.

“It’s kind of like…more mental state. Like from my experience, they both a guy and a girl, have like the same mind, and so you’ve got the caring side of like a woman and then you have at the same time that rough side.”

“…I like the combination of the feminine and masculine characteristics because their femininity tends to be old school, the ones that I’m drawn to and interact with, but their masculine side tends to be very affirmative and aggressive, and I like the mix.”

“…there was an exoticness, a uniqueness, something that can’t be obtained elsewhere. They’re just totally unique in their sexuality in that they’re both…men and women and, at the same time, neither men or women. To me that’s my fascination.”

“I respect transgenders for being able to say ‘I’m a woman in a man’s body.’ I think that is so honorable.”

A third observed trend by which men described their attraction to transgender women was characterized by a more explicit focus on the transgendered body and appearance. Unlike the previous examples, in which men described their attraction in more holistic and esoteric terms associated with personality attributes, this third trend showed a distinct focus on physical and anatomic issues that evoked men’s personal and sexual attraction. A subgroup of these men described an attraction to a physically embodied and eroticized femininity, perhaps a hyperfemininity:

“They seem to put more into their femininity than I guess what do you call them, biological women. And they try harder.”

“I suppose erotically is the appearance. I love them dressed like a hooker, like hookers….TG girls are ultra-sexy, much more than a genetic woman.”

“…their comfort with their bodies…more feminine than real women and more…exotic. Sexually they’re more open, just like they’re a more advanced woman to me.”

Another subgroup of these men described with more explicit detail an attraction toward anatomical features of some transgender women:

“I like women with dicks. I like tits and I like dicks…. Something erotic about getting fucked by someone who is a woman.”

“I like the girls with a little something extra, you know what I mean. Pretty with a big dick.”

“There’s something about me knowing that physically they have the physical characteristics of a man. I always date pre-op. There’s something that just turns me on.”

These narratives demonstrated a continuum in the ways by which men narrated their attraction to transgender women. On the one hand, some narratives depicted an interpersonal connection with a specific transgender partner without contextualizing the attraction in physical or erotic terms. On the other, other narratives described a focused attraction toward the (mostly pre-operative) transgender body.

Weak patterns of association between sexual orientation identity and sexual behaviors. We attempted to examine whether men’s narratives on their sexual attraction and behaviors corresponded with sexual orientation identities—for example, whether heterosexual-identified men were drawn to more feminized transgender women, or whether men who chose not to identify their sexual orientation were more drawn to personality or esoteric attributes of transgender women.

In general, we found little evidence for a strong linkage based upon these qualitative data. Observed trends by which men described their attraction to transgender women (being attracted to the person, to the category, or to the body) did not appear to match consistently with the ways by which the men described their sexual orientation. Forms of attraction to transgender women were present in all sexual orientation categories.

We observed a few trends between men’s sexual preferences with transgender women and their sexual orientation categories, though these cannot be interpreted statistically. Among the 20 men who claimed a heterosexual sexual orientation identity, most (n = 18) described having a sexual history with only transgender and biological women. Two men described having ever had sex with a man, though neither had sex with a man during the past year. Over half of these heterosexual-identified men (n = 12) described preferences for being the sexually insertive partner, with the remainder describing no preference. Likewise, over half (n = 12) described having no preference regarding their transgender partner’s pre- versus post-operative status, though some clearly described favoring pre-operative transgender woman (n = 4) and other favored post-operative transgender women (n = 4).

Among the 14 men who claimed being bisexual, over half (n = 9) described having a history of sex with transgender women, biological women, and males. Four men described having only had sex with transgender and biological women, and one man reported never having had sex with a biological women though he had sex with a man several years previously. Most of these men (n = 10) described a preference for being the sexually insertive partner, whereas two had no preference and one preferred to be the anal receptive partner. Half (n = 7) expressed no preference regarding their transgener partner’s pre- versus post-operative status.

All four of the men who described themselves as gay described having a history of sex with transgender women, biological women, and males. Two of these men described recent sexual encounters with men. Only one man described a preference for being the sexually insertive partner (others described having no preference), and one man described a preference for pre-operative transgender women (others described having no preference).

Among the eight men who chose not to label their sexual orientation identity, most (n = 6) had a history of sex with transgender women, biological women, and males. Two men had never had sex with a male. Most men (n = 5) described preference for being the sexual insertive partner (others described having no preference); three men preferred post-operative transgender women and two men preferred pre-operative transgender women (others described having no preference).

We urge caution against inferring any statistical trends based upon the numbers presented here; numeric information should serve only as a way to indicate the variability in sexual behavioral preferences within the sample. Indeed, a salient finding was the lack of robust correspondence between sexual orientation categories and sexual behavior preferences.

Discussion

This qualitative research study demonstrates potential challenges in efforts to provide HIV and sexual health services to MSTGWs, due in part to limitations in existing language, cultural discourse, and terminology to understand these men. These challenges might further reflect overarching conceptual problems with using category-based approaches to HIV prevention research and outreach (e.g., targeting MSM, heterosexuals, etc.) for individuals who are not easily classifiable according to traditional public health categories.

In this study, men’s narratives about their sexual orientation identity suggested two trends. First, there was noticeable diversity in narrative explanations for how and why MSTGSWs identified their sexual orientation. Second, there appeared frustration and resistance to self-labeling and identifying with a discrete category, as many men appeared aware of the inconsistencies between conventional category distinctions versus their own sexual behaviors and preferences. Although some men chose not to identify themselves at all, most men settled on one of the conventional categories (heterosexual/straight, bisexual, or homosexual/gay), perhaps due to the perception of a forced choice. Consequently, men interviewed in this study demonstrated some of the limitations in ways by which researchers tend to reduce human sexuality into discrete categories and subsequently design programs and measure outcomes according to these categories.

This research contributes to important debates around a need for more sensitive and thoughtful conceptualization around human sexuality when measuring HIV risk and designing prevention education programs. Recently, Young and Meyer (2005) urged HIV researchers and practitioners to re-consider the usefulness of behavior-based categorical risk groups, such as MSM, for guiding public health research and interventions. According to this argument, an unquestioning overuse of behavior-based risk groups in HIV prevention can undermine personal sexual identities, obscure differences within groups, and neglect important social contextual factors that can determine HIV risk. Another compelling argument has drawn attention to potential limitations incurred by targeting HIV prevention to specific sexual identity communities, such as the gay male community (Ford et al. 2007). Insofar as people might not have formed an identity around their sexuality—such as reports of African American MSM who do not identify themselves as gay or, in the present case, MSTGWs who choose not to identify with any sexual orientation—there remains a likelihood that substantial proportions of high-risk groups might not be reached by community-targeted approaches.

Narratives from MSTGWs support both arguments—on the limitations of targeting groups based on behavioral characteristics as well as based on community identity formulations. MSTGW stories demonstrated that behavioral characteristics and community identities around sexual orientation can be temporally fluid, conceptually complex, and contested by individuals. Among this group of men, their one shared characteristic was a history of sex with transgender women. Beyond this, there were no dominant trends in sexual identities, reasons for being attracted to transgender women, or sexual behavioral preferences.

Does the heterogeneity shown here suggest that these men should not be treated as a “group” per se? This is a difficult question, and the data in this study do not allow for a clear answer. On the one hand, men did not appear to cohere within a community of choice, and their behavioral preferences appeared more idiosyncratic than consistent. But on the other hand, given repeated studies of transgender women showing high rates of HIV transmission due to sexual activity, these men might share an elevated risk for HIV. Indeed, in our sample 20% of the men reported being HIV positive and 10% did not know their HIV status. On the grounds of public health urgency, we would argue that men who engage in sex acts with transgender women warrant increased prioritization from HIV researchers, but the degree to which men are regarded as a collective unit for research and intervention should be considered more carefully.

In addition to these challenges identified, this research reveals opportunities for providing HIV prevention interventions, including targeted educational outreach and HIV testing and counseling, to this group of men. Fundamentally, this research indicates that MSTGWs can indeed be recruited into programs and research studies concerning HIV, and can be amenable to discussions of their sexual histories and their potential risk behaviors. We found that engaging members of the transgender community in outreach and recruitment, as well as employing transgender women as interviewers, facilitated a sense of trust and comfort between these men and our study. Furthermore, we respected men’s self-perceptions and definitions about identity, relationships, and patterns of erotic attraction, and withheld from judging them or placing restrictive categorical parameters on their identities and behaviors, which may have facilitated their openness to this research. It is recommended that HIV prevention and educational outreach strategies follow similar strategies, allowing MSTGWs to use their own language to express their sexual lives and characterize their personal risks for HIV. HIV prevention educators and test counsellors are reminded not to conflate sexual behavior with transgender women with any particular sexual identity.

Because these men might be difficult to identify, efforts to reach them might at first rely on establishing contact through transgender women. The data here suggest that men are more likely to identify as a transgender woman’s partner rather than view themselves as a community of men who share similar interests. By contrast, there is evidence that transgender women are more likely to view themselves as part of a cohesive community and therefore might be more easily targeted in prevention programs (Bockting et al. 2005). Making initial contact with transgender women and asking if their male partners are interested in HIV prevention services might, therefore, be an effective tactic. Indeed, couples-based approaches to HIV prevention and sexual health education might offer a promising way for reducing risk for transmission among both members of this sexual relationship dynamic. However, further research is necessary to understand how existing evidence-based approaches to HIV counselling for couples must be adapted to incorporate gender and relationship dynamics between MSTGWs and transgender women, as current approaches are specific to particular relationship dynamics (El-Bassel et al. 2005; Farquhar et al. 2004; Remien et al. 2005). Other potential strategies for reaching these men include targeting community venues and health or social service agencies where transgender women gather or socialize, and through Internet websites where men meet transgender women.

There are important study limitations to consider. First, the qualitative data preclude statistical inferences about the trends observed within the narratives. Second, the sample of men who volunteered to participate in the interviews might not be representative of other MSTGWs, therefore findings might not be generalizable. Third, because of the sensitive nature of the study topics, men might have selectively monitored or censured their discussion about particular attitudes, beliefs, or behaviors, despite attempts to heighten trust and minimize discomfort. Fourth, because there is little known academic or public discourse on MSTGWs to draw from, the depth of these interview discussions might have been limited by a lack of terminology and shared understandings about sexual behaviors and identities among MSTGWs. Fifth, we did not inquire about men’s relationships and sexual behaviors with transgender men or gender queer individuals.

In conclusion, this study offers a closer look at MSTGWs—a group that has not received significant attention in HIV research and public health literature—and provides further insights into the context of HIV risk among these men and their transgender partners. Findings underscore the need for sensitivity and deliberation when studying sexual minorities, especially those who do not fit easily within conventional categories. Findings also suggest a need to disentangle notions of sexual orientation identity from sexual behaviors and, with specific regard to measurement, to refrain from treating these as equivalent variables. Indeed, HIV risk assessment and epidemiological surveillance surveys can provide more accurate information by including items that measure specific risk behaviors MSTGWs engage in with transgender women, such as anal penetrative or receptive sex, vaginal intercourse (with post-operative transgender women), and oral sex—each of which can yield different levels of risk. Measures of risk should also be aware of the type of relationship between MSTGW and their transgender women partners, which can determine likelihood of condom use (Nemoto et al. 2004). Furthermore, this research indicates a need for further studies of MSTGWs as experiencing potentially high levels of risk for HIV transmission, as well as a need for prevention programs that situate risk within the relational dynamics between these men and their transgender partners. This research also reminds professionals who provide sexual health education information and counselling to male populations to challenge their own assumptions about sexual orientation by remaining aware that men of any orientation might be engaging in sexual activity with transgender women, biological women, and men.

Acknowledgments

We would like to thank Lally Adao and Andrea Horne, and all men who participated in this study. This research was supported by the California HIV/AIDS Research Program (grant ID03-SF-009) and the National Institute on Drug Abuse (grant R01-DA18621).

Copyright information

© Springer Science+Business Media, LLC 2007