Masculinity, body image, and sexual behavior in HIV-seropositive gay men: a two-phase formative behavioral investigation using the internet.

The purpose of this study was to understand the synergistic relationships between conceptions of masculinity, body image, and sexual behavior in HIV-positive gay men. The data were drawn from a two-phase formative behavioral mixed methodologies investigation with the use of the Internet. Findings demonstrated that conceptions of masculinity were intimately linked to body image and sexual adventurism, such that men defined their masculinity by their physical appearance and sexual behavior. Further, the data support high levels of risk taking, including steroid use and intentional unprotected anal intercourse (barebacking), which correlated with the participants' conception of physical masculinity. Clinical implications suggest that it is critical to examine the interaction that occurs between the individual's gay, masculine, and HIV-positive identities as these appear to be overlapping realities that have an impact on the decisions and behaviors in which these men engage.

Keywords: masculinity, body image, gay men, sexual behavior, HIV, barebacking, steroid use, Internet, sexual adventurism

**********

The developmental trajectory of gay culture to construct a conception of masculinity distinct from heterosexual males may have been thwarted in the 1980s with the onset of the HIV epidemic (Halkitis, 1999; 2001). After the emergence of HIV, the emphasis on physicality among some gay men in large urban centers was directed to preserve and increase the health of men infected with the virus who were experiencing weight loss, muscular wasting and deterioration, decreases in libido, and eventual death (Shilts, 1987). As many gay men began to physically deteriorate due to the disease, heterosexual perceptions of gay men as non-masculine and fragile were realized.

In order to combat the physical weakness brought on by the virus, many HIV-positive gay men undertook complementary therapy measures such as steroid replacement as well as weight training and nutritional supplements to increase strength and maintain health. Today, while these therapies are still important elements in the lives of HIV-positive gay men, they have moved beyond the goals of health and survival and have become associated with a physically based conception of masculinity espoused by many gay men, regardless of HIV serostatus (Signorile, 1997).

Central to this conception of physical masculinity is body image and muscularity (Halkitis, 1999; 2001). To this end, some HIV-positive gay men describe lives driven by a desire to achieve an ideal physical appearance of muscularity, characterized by what has been labeled as the "buff agenda" (Halkitis, 1999). While therapeutic approaches to body enhancement serve to decrease HIV-related morbidity and mortality (Charlin, Carrasco, Sepulveda, Tortes, & Kehr, 2002)) and the associated body dissatisfaction experienced by many (Tate & George, 2001), some HIV-positive men have taken this to an extreme. Even given the advances in HIV treatment in the last decade, which have drastically reduced AIDS-related deaths (Mocroft et al., 2003), many HIV-positive men still implement exercise, steroid use, and nutritional supplements, perhaps not as a way of avoiding their wasting and mortality, but rather as a way to achieve psychical perfection.

A second tenet of this conception of physical masculinity among HIV-positive gay men is that of sexual prowess. Sexual behaviors and associated risk-taking have been linked to masculine identity (Chapple, Kippax, & Smith, 1998; Halkitis, 1999; 2001 Kippax & Smith, 2001). HIV-positive men describe sexual partnering as a means of enhancing conceptions of attractiveness; thus some seek out sex as an affirmation of self and as a means of eradicating feelings of undesirability (Halkitis & Wilton, in press). Recent increases in sexually transmitted infections (Centers for Disease Control & Prevention, CDC, 2003a) and HIV (CDC, 2003b) indicate that certainly some HIV-positive men-who-have-sex-with-men are engaging in unprotected sexual behaviors. These unprotected behaviors, and the recent emergence of the phenomenon of barebacking, or intentional unprotected anal sexual acts (Halkitis & Parsons, 2003; Halkitis, Parsons, & Wilton, 2003; Mansergh et al. 2002; Suarez & Miller, 2001) are described by some HIV-positive men as "sexy" and "masculine" and provide a direct route to physical and emotional connection with others (Halkitis, 1999). A physically attractive body often provides the route by which many of these men attract partners. Thus, these sexual acts provide further reinforcement to achieve body perfection, perpetuating the "buff agenda."

Body image and sexual prowess are central to the conceptions of masculinity espoused by some HIV-positive gay men. Yet these interconnections between masculinity, body image, and sexual prowess are complex and not always apparent among individuals who are attempting to balance their identities as male, gay, and HIV-positive. Thus, we undertook an investigation using mixed methodologies to further elaborate these constructs as well as to disentangle the relationships that exist between body image, sexual prowess, and conceptions of masculinity as physicality among gay men living with HIV/AIDS.

METHOD

STUDY DESIGN

The ideas presented here are based on two sources of data that represent an integration of an investigation incorporating both qualitative/ethnographic and quantitative methodologies with the use of the Internet as a source of data collection. The qualitative study (Phase I) was undertaken to explore and more fully understand the meanings that HIV-positive gay men assigned to their conceptions of physical masculinity. Based on the data from the qualitative study, measures were developed as part of a quantitative investigation (Phase II) to assess these emerging relationships among a larger sample of participants. It should be noted that the qualitative data presented here expand upon our previously published ethnographic work (Halkitis, 1999; 2001).

PARTICIPANTS AND PROCEDURE

The sample for Phase I consisted of 15 men ranging in age from 29 to 56 with a mean age of 36 years (SD = 8.16) and eight years (SD = 3.08) since they had been diagnosed with HIV. Approximately 77% of the participants were currently taking medications to treat HIV. Racial and ethnic data were not collected from this sample.

A total of 114 men participated in Phase II. The mean age for the sample was 42 (SD = 6.82) and yielded a similar age range (27 to 57) to Phase I participants. Participants had first been diagnosed with HIV 11 (SD = 5.27) years prior to the study, slightly earlier than those who participated in the qualitative interviews. Comparable to Phase I, 86.0% (n = 98) of the participants in Phase II indicated that they were currently on some form of HIV antiretroviral therapy. The sample for Phase H was predominantly White, 77.2% (n = 88); 9.6% (n = 11) Black; and 6.1% (n = 7) Latino; 7.0% (n = 8) provided no racial/ethnic identification.

PHASE I: QUALITATIVE/ETHNOGRAPHIC INVESTIGATION

In Phase I, we sought to consider the behaviors and meanings associated with physical notions of masculinity among HIV-positive gay men residing in New York City. Men were identified through use of the Internet using passive recruitment (postings on listserv accounts) as well as active methodologies (engaging in conversations in Information Relay Chats [IRCs] and chat rooms). Participants were screened on the following criteria: a self-reported HIV-positive serostatus test result, self-identification as a gay man, and a self-report of regular (three or more clays per week) exercise at a physical fitness center. Those exercising regularly were intentionally selected in order to examine the interconnections of masculinity, physical appearance, and sexual behaviors. Men who met the eligibility criteria and were willing to participate were scheduled for an interview.

All communications between the researcher and the participants were conducted through an electronic medium (e-mail). To assure confidentiality, no identifying information was obtained, and all participants were identified through the use of pseudonyms. Participants were sent the interview via e-mail and were asked to return the responses to the researcher within 10 days. Upon obtaining the responses, answers were reviewed for clarity. When answers were ambiguous or unclear, follow-up questions were posed to the participants regarding their responses and communications continued until such point that participant responses were clarified. At the conclusion of the data collection period, 15 electronic interviews had been conducted.

Interviews consisted of 10 questions focused on definitions of masculinity, the impact of HIV on masculinity, and the interplay of physicality and masculinity among HIV-positive gay men. Questions, which sought to ascertain how the participants defined masculinity for themselves, focused on personal definitions of masculinity rather than cultural norms and expectations. The 10 questions were as follows: (1) How do you define masculinity? (2) Give an example of an act that you label as masculine or an example of a masculine person. Explain. (3) To what extent is masculinity related to physical appearance? (4) How does masculinity relate to sex or sexual behavior? (5) How much do others' perceptions of your masculinity matter to you? Explain. (6) Have you made changes in your lifestyle since you became HIV-positive to enhance your masculinity? Describe these. (7) How has being HIV-positive affected the way you perceive your masculinity? (8) How do you think that HIV affects or threatens one's masculinity? (9) Do you think HIV-positive men overcompensate on the physical aspects of their masculinity? Explain. (10) To what extent do matters of masculinity impact on your daily life, and how much time per day would you say you spend on these matters?

After the data were collected, a thematic codebook was developed based on the structure proposed by Miles and Huberman (1984) such that independent readers noted themes and sub-themes that emerged in the words of the participants. Each set of responses was coded independently by the two researchers and yielded an inter-rater agreement of 89.6%.

PHASE II: QUANTITATIVE INVESTIGATION

The development of the quantitative survey was based on the knowledge generated from Phase I. We sought to quantify definitions and perceptions of masculinity as well as the behaviors that were clearly delineated in the qualitative interviews: body building, physical appearance, and sexual adventurism including sexual risk-taking in the form of intentional unprotected anal intercourse (barebacking). In this phase, 114 participants throughout the United States were administered surveys via the Internet. Electronic postings tailored to HIV-positive gay men seeking sexual partners were placed on listserv accounts describing the study and asking interested participants to contact the researcher via e-mail. Men were also directly recruited through IRCs and chat rooms.

Participants were required to meet the following eligibility criteria: 18 years of age or older, self-reported as HIV-positive, self-identified as a gay man, and ability to read/comprehend English. Participants who completed the survey were encouraged to return their responses via e-mail yet were given the option of sending the responses through traditional mail or fax. All e-mail correspondences were deleted from the investigator's account upon downloading responses. To prevent duplication of participants, the entered data were checked for matches on age, postal code, and years since testing HIV-positive. This analysis yielded one set of duplicate data, which were eliminated from the final data set.

MEASURES

Participants were assessed on several measures developed for the purposes of this investigation.

Sociodemographics. Participants were asked to self-report their age, race/ethnicity, HIV serostatus, length of time since HIV diagnosis, and whether they were taking HIV meds.

Sexual behavior (barebacking). Participants first were asked to indicate whether or not they had engaged in bareback sex with HIV-positive, HIV-negative, and status-unknown partners in the three months prior to assessment. In cases when an affirmative answer was given, the participant was then asked to indicate the number of times in which the bareback sex was anal insertive ("How many times did you top") and anal receptive ("How many times did you bottom?") for each type of partner.

Sexual Adventurism Scale. This construct was assessed through an initial set of 12 items and factor analytic methods yielded a final set of 10 items loading on one factor with an explained variance of 59.91% and loadings ranging from .61 to .88. (alpha = .92). Item responses were based on a five-point, Likert-type scale (1 = not at all like me to 5 = completely like me). Sample items included "'I try to have bareback sex," "I go to bathhouses and sex clubs to look for sex," and "I love taking the cum [semen] of other men."

Conceptions of Masculinity Scale. An original set of 34 items were used to assess the men's conceptions of masculinity based on a five-point, Likert-type Scale (1 = completely disagree to 5 = completely agree), and factor-analytic methods yielded three subscales: conceptions of masculinity as physical appearance (n = 9, alpha = .81, explained variance of 41.05% with factor loadings ranging from .30 to .90); conceptions of masculinity as sexual behavior (n = 4, alpha = .83, explained variance of 66.41% with factor loadings from .72 to .87); conceptions of masculinity as social behavior (n = 5, alpha = .67, explained variance of 43.58% with factor loadings of .51 to .75). Based on factor analyses, the remaining items were omitted from the scale. Sample items for the three subscales were as follows: (1) Conceptions of Masculinity as Physical Appearance. Items included the following: "Physical appearance is an important dement of masculinity in the gay community," "Masculinity is more about how one looks than how one acts," "Physical appearance does define masculinity," and "Well-built men give the impression of masculinity at first sight"; (2) Conceptions of Masculinity as Sexual Behavior. The four items for this subscale are as follows: "Sexual performance is an important part of masculinity," "Sex is a celebration of masculinity," "A masculine man has lots of sex"; and "Masculinity celebrates male form and virility"; (3) Conceptions of Masculinity as Social Behavior. The five items for this subscale are as follows: "I am not comfortable around unmasculine gay men," "I would not have sex with a masculine looking man who acted in any way feminine," "A masculine man is both butch in behavior and appearance," "I watch my behavior to make sure that I act masculine around other gay men," and "Drag queens undermine the idea of masculinity in the gay community."

Furthermore, three items that were dropped from the original scale were maintained as stand-alone items in our analyses because they provided interesting information about the participants' view of societal norms regarding masculinity. These three stand-alone items are as follows: "Our society reserves the idea of masculinity for straight men," "A man can be gay and be masculine," and "For most people in the U.S., masculinity makes no room for same-sex desires."

Emphasis on Body Scale. A set of five items based on a five-point, Likert-type scale (1 = not at all like me to 5 = completely like me) were used to assess emphasis that the participant placed on his physical appearance, specifically with regard to muscularity and bodybuilding (alpha = .82). Factor-analytic methods yielded one factor for this scale with an explained variance of 57.90% and loadings ranging from .64 to .83. Items included "I work out regularly each week," "I use testosterone or deca durabolin to help pump up," "I use other steroids to help build up my body," "I use creatine, vitamins, or other nutritional supplements for bodybuilding," and "I work hard to look muscular."

RESULTS

CONCEPTIONS OF MASCULINITY

Based on the qualitative interviews, many participants believed that heterosexuals oftentimes did not view them as masculine or "manly." This notion was elaborated in the words of a Phase I qualitative interview participant:

   I know that I am masculine, mostly because of the way I look. Women
   look at me on the street all the time. But when I tell people I'm
   gay, that seems to erase what they think of me as a man. They just
   think of me as gay and not a real man.

Similarly, as demonstrated in the Phase II quantitative data, 83.3% (n = 87) of the

participants agreed that society reserved the idea of masculinity for heterosexual men only; 76.3% (n = 73) agreed that most people think of gay men as non-masculine; and 78.4% (n = 78) believed that for most people in the United States, masculinity could not be associated with same-sex desires. Regardless of what the participants believed about society's heterosexist views, the majority of the men in both phases of the study maintained that gay men have both a gay and a masculine self. In fact, 99.1% (n = 113) of the participants in Phase II agreed that an individual could be concurrently gay and masculine.

The majority of the participants in the qualitative interviews clearly indicated their definition of masculinity as physicality, both in appearance and behavior: "Society makes masculinity and physical appearance go hand in hand," according to one of the qualitative interview participants. An association between conceptions of masculinity and physicality emerged when participants described the idealized masculine male, their sexual partners, their exercise routines, and their sexual behaviors. These are reflected in the words of a 40-year-old HIV-positive gay man, who suggested: "I do a significant amount to make myself more masculine looking. My workout, diet, and grooming are all pointed toward maintaining attractiveness and sexual opportunity."

BODY IMAGE IN RELATION TO MASCULINITY

In the qualitative interviews, almost all of the participants described the emphasis they placed on their physical appearances. The statements reflected the importance the men placed on "looking good" and "feeling strong" and how these elements interacted with their sense of masculinity. As stated by one interview participant, "Physical appearance does define masculinity on the outer shell."

For some of the interview participants, the emphasis on physical appearance was related to the impressions they emitted when they were seen in social situations (e.g., bars, dance clubs, pride events). A few of the participants expressed an ongoing concern that potential sexual partners perceived them as masculine, because such a perception would increase their probabilities of meeting these men. This concern, on the part of the interview participants, was related to what they believed was revered by the gay community and what was sanctioned as appropriate masculine appearance. One of the participants in the qualitative interview, a 46-year-old gay man who has been HIV-positive for 11 years, summarized this reality:

   Initial perceptions of masculinity are strongly tied to physical
   appearance. Well-built men usually give an initial feeling of
   masculinity at first sight. Since I like this perception, since it
   matters a lot to me to be masculine, I work out and watch my diet.

Some of the men articulated the social pressures within the gay community to conform to this ideal standard of masculinity associated with the leather man, construction worker, and sports jock, which are iconic in this masculine view of the world (Halkitis, 2001; Signorile, 1997). An example of this ideal was provided by a 38-year-old interview participant actively involved in the "leather scene" and a former contestant in the International Male Leather Contest, who unconditionally supported this concept of masculinity suggesting that it is the core of what he is and what he desires: "I am a man and want to be with men. Men are strong, muscular, and tough. They roll around with each other." Similarly, in describing this prescribed concept of masculinity, a 37-year-old man who has been HIV-positive since 1986 suggested the following:

   Of course, you are told what to look like. When I first came out, I
   had long hair and didn't really work out. One trip to the Eagle [a
   bar in New York City], and I knew that had to change. I cut my
   hair real short and haven't missed the gym for more than three
   days in a row in years.

The importance of physical exercise and bodybuilding was revealed by the fact that 59.7% (n = 68) of the HIV-positive gay men assessed in the quantitative survey indicated they had a regular workout plan.

In the qualitative interviews, several participants clearly discussed the steps they take to ensure the manifestation of their physical masculinity, including a regimented exercise schedule, the use of nutritional supplements, and/or the utilization of anabolic and androgenic steroids, such as testosterone and deca durabolin, to enhance muscularity and growth. In one interview, a participant commented:

   I use a collection of weight-training enhancements such as creatine,
   androstene, and HMB for the personal choice to build muscle,
   not necessarily for my HIV. As a secondary consideration, by
   keeping my body as healthy as possible, I can help stave off
   infections and maintain my t-cells and low viral load naturally.

In the quantitative survey, 33.3% (n = 38) reported the use of testosterone, and 17.5% (n = 20) reported the use of deca durabolin while 15.8 % (n = 18) used both substances in combination. Furthermore, 25.4% (n = 29) revealed they used testosterone or deca durabolin specifically to "pump up," while 36% (n = 41) reported the use of other substances, such as creatine and vitamins for bodybuilding.

The relationship between the use of steroids and the emphasis on muscularity among HIV-positive men is further demonstrated by the data of the quantitative survey as shown in Table 1. Specifically, with regard to the Emphasis on Body Scale, those reporting the use of testosterone and/or deca durabolin indicated that they assigned greater importance to looking muscular and strong (M = 17.73, SD = 5.38) than those who did not report use of either steroid (M = 10.58, SD = 4.42) (t (110) = 7.58, p < .001). Similarly, there was a trend toward significance between the use of steroids and the relative importance placed on the "Conceptions of Masculinity as Physical Appearance" subscale (t (104) = 1.85, p = .07), with those using steroids tending to place greater emphasis on masculinity as defined by physical appearance. This is further confirmed by the relationship between emphasis on body and conceptions of masculinity. As shown in Table 2, those who placed greater emphasis on the appearance of their bodies also scored higher on the "Conceptions of Masculinity as Physical Appearance" subscale (r = .371, p < .001). In addition, this body image construct was related to conceptions of masculinity as a social behavior and masculinity as appearance but was unrelated to age or any HIV-related characteristics (e.g., years living with HIV, AIDS diagnosis, treatment characteristics, or opportunistic infections). These data suggest that the emphasis HIV-positive gay men place on their overall physical appearance and muscularity supersedes developmental and HIV progressive factors.

SEXUAL BEHAVIOR IN RELATION TO MASCULINITY

For many of the interview participants, sexuality was intimately tied to the construct of masculinity as physicality. When describing their sexual behaviors, these men spoke of sex in terms of the frequency of the sexual behaviors as well as the adventurism associated with sexual encounters, including HIV related sexual risk-taking. In the words of one of the interview participants, "I try to have sex as much as possible with as many men as possible and as anonymously as possible."

For a few of the participants, the need for frequent and adventurous sex was heightened by their HIV-positive serostatus. The impact of an HIV-positive status on sexual behavior was clearly discussed by some of the men as an affirmation of their attractiveness and, in effect, their masculinity. In this regard, a 28-year-old interviewee stated:

   Sex is so important to me, maybe more important than before,
   when I was negative. By being with men, I feel desired, I feel
   wanted. When a man wants me, it means that all my efforts to look
   hot, to have a great body have paid off. Yeah, that makes me feel
   surer of my masculinity.

Sexual adventurism as an affirmation of masculinity was a central theme in many of the interviews we conducted. One of the participants described the role of the masculine man as the "hunter-gatherer who seeks his conquest in other men; the more he succeeds, the more he is able to prove his manhood, his desirability, his masculinity." Similarly, reflecting upon the importance and meaning of sex as an affirmation of his masculinity in tight of his HIV-positive status, a 36-year-old interview participant who had been HIV-positive for five years commented as follows:

   HIV-positive men take advantage of sexual opportunities more
   that HIV-negative men, perhaps because they feel, due to their
   status, they may get less opportunities in total. For example,
   in a bar you might strike up a conversation, which seems to be
   headed to the bedroom. But it quickly dissipates when your
   positive status is revealed. When your partner is willing, you
   move fast.

The importance of sexual adventurism as a part of the masculine self was supported by the quantitative data. Men who were more sexually adventurous, as measured by the Sexual Adventurism Scale, also scored higher in terms of the importance that they placed on sex as a means of defining their own masculinity ("Conceptions of Masculinity as Sexual Behavior" subscale of the Conceptions of Masculinity Scale) (r = .359, p < .001). As shown in Table 2, this construct was also related to masculinity as social behavior as well as masculinity as appearance.

A few of the interview participants suggested that the HIV epidemic has forced gay men, collectively, to re-examine their sexuality. One of the participants suggested, "Sex affirms strong physical needs as well as social needs." With regard to safer sexual practices, about half the men in the qualitative study revealed their dislike of condoms and spoke of intentional anal sexual intercourse without condoms (i.e., barebacking) with other HIV-positive men (and in some cases HIV-negative or status-unknown men) as a means of overcoming this emotional barrier associated with protected sexual behaviors. Several men suggested that sex was an affirmation of life, and by barebacking they could "share their manhood with others." To this point, one of the participants noted: "An HIV-positive man is able to perform in the way a man is prescribed to perform, to overcome the impact of the virus and prove to oneself that one is still strong, sexual, virile, and masculine." And one other stated:

   I was dating this other really hot poz guy for about two months,
   and I thought for sure when we had sex we would bareback. But
   he kept insisting that when I fuck him I use a condom, and so I
   did. But I didn't really like it. I did it but never got really
   hard. And then one day we agreed to do it raw, and I came inside
   him, and we ended up in bed for like three days. Now, even when I
   think about doing it raw, it excites me. It's the most amazing
   feeling. It's what two men are supposed to do. It's what real men
   do. It feels right.

This link between masculinity and sex was corroborated by the survey data that indicated those who scored higher on the "Conceptions of Masculinity as Sexual Behavior" subscale of the Conceptions of Masculinity Scale also reported more frequent bareback anal insertive intercourse with HIV-positive partners (r = .246, p = .01), bareback anal receptive sex with HIV-positive partners (r = .257, p < .01), and bareback anal receptive sex with HIV-negative/status-unknown partners (r = .273, p < .01).

Furthermore, this link between sex and masculinity corresponds with the use of steroids and Viagra, suggesting the synergistic relationship between physical appearances/functioning, sexual performance, and conceptions of masculinity. Specifically, those who reported the use of Viagra in the three months prior to assessment reported more frequent bareback anal insertive sex (t (103) = 1.99, p = .05) and bareback anal receptive sex with other HIV-positive partners (t (42) = 3.70, p = .001) as well as more frequent bareback anal receptive sex with HIV-negative/status-unknown partners (t (42) = 3.92, p < .001) than those who did not report Viagra use. Similar patterns emerged when we compared bareback sexual behaviors to steroid use. These data are fully shown in Table 3.

DISCUSSION

The data collected from HIV-positive gay men in this two-phase investigation certainly suggest a conception of masculinity based on physical and sexual ideals that is embraced by certain segments of the gay community. Levine (1998) described men who adhere to this ideal as the "manliest of men," embracing a type of "hyper-masculinity." And while certainly it may be argued that masculinity is not a monolithic concept and that multiple types of masculinities may function within gay culture (Nardi, 2000; Plummer, 1999), the masculinity described by the participants in this study represents a type of masculinity that is revered and respected by many gay men. In a pluralistic society, where diversity is seemingly celebrated and individual expression is encouraged, the idea that multiple types of masculinities exist certainly is politically correct. Unfortunately, the workings of our society seem to transgress these types of notions. In the day-to-day workings of gay life, it is the buff men--the muscular, masculine men--who are the objects of much attention and desire.

The development of masculine identities separate from heterosexual ideals has been an ongoing struggle for gay men. Homophobic attitudes as well as stereotyping of gay men have served as obstacles to such efforts. The HIV epidemic may have added to this struggle by perpetuating homophobic attitudes of gay men as physically weak and thus thwarted efforts to clearly define a separate gay masculinity (Halkitis, 1999). Gay men themselves who, according to our data, have embraced the idea that the ideal masculine identity is based on physical appearance and sexual performance may perpetuate this reality. Perhaps in part because these efforts at developing a gay masculinity have been so difficult, many gay men have adopted the heterosexual standard of the tough, physically strong male (Kleinberg, 1992).

For HIV-positive gay men who attempt to maintain their own health while at the same time remain desirable, the emphasis on the physical definitions of masculinity seems perfectly justifiable (Halkitis, 1999). For these men, physical appearance and sexual expression are the cornerstones of their masculinity. Having a strong muscular body is essential to HIV-positive gay men embracing this ideal and has become increasingly associated with other physical attributes that enhance a masculine appearance. Therefore, these men often associate with others who celebrate this definition of masculinity in a variety of environments where sexual adventurism and multiple sexual partners are encouraged (Halkitis, 1999; Halkitis, Parsons, & Wilton, 2003; Halkitis & Wilton, in press).

In order to be attractive and to be attracted to others, this ideal conception of masculinity described in the words and behaviors of our study participants requires conformation to an abstract model (Gough, 1989). Clark (1997) contends that some gay men emphasize the values associated with "hypermasculinity," perhaps at the expense of learning the "important lessons of life," such as how to form a loving and caring same-gender relationship. It is unlikely, however, that HIV-positive gay men who hold this ideological stance have forgotten these important lessons, and in fact recent data indicate that many of these men seek intimate connections with others (Halkitis & Wilton, in press). More likely, in their struggle to remain alive and their desire to be loved and accepted given their HIV-positive status, they have learned that their "hypermasculinity" provides a mechanism to meet a partner and share love.

Two methodological limitations of this study are the small sample sizes, especially in Phase I, and the use of convenience sampling in participant recruitment. With regard to our small ethnographic sample, it should be noted that while the sample of 15 men was sufficient to obtain an initial understanding of the constructs, the sample is limited in terms of making greater generalizations to the population of HIV-positive gay men. However, these data are supplemented and supported by the quantitative findings in this study.

Further, the use of the Internet in both phases of this investigation is noted as another potential limitation in terms of the generalizability of the study findings. Recent research has suggested that men who frequent the Internet may demonstrate greater levels of sexual compulsivity (Cooper, Delmonico, & Burg, 2000; Cooper, Putnam, Planchon, & Boise, 1999; Halkitis & Parsons, 2003). Therefore, these findings should be viewed with caution in that hypersexual men may be more likely to be represented in our study, and thus their conceptions of masculinity may be appropriate for the specific segment of the gay community in which they socialize and meet sexual partners.

These methodological limitations may have yielded a group of White men from higher socioeconomic and educational backgrounds. Further, men of color, who were underrepresented in the study, may have culturally specific conceptions of masculinity based on their racial identities and cultural worldviews that differ from the conception of masculinity set forth here. Given the fact that current trends demonstrate that Black and Latino MSM have increasing disproportionate rates of HIV (Koblin et al., 2000), the behaviors of these men need to be examined separately and in relation to what may be a culturally specific conception of masculinity. Similar analogies can be drawn along the lines of age. While the men in our studies were primarily in their 30s and 40s, new seroconversions have been most evident among younger MSM (CDC 2003b). Furthermore, generational gaps as well as sociological and cultural shifts toward homosexuality may confound conceptions of masculinity among younger MSM, and future studies need to consider these phenomena.

Although the participants in both phases of the investigation were assured confidentiality, the issue of socially desirable responses could confound the results. Participants were asked to provide information about potentially sensitive areas of their life, and as a result may have experienced discomfort in responding to some of the questions on the survey or qualitative interview that may have resulted in participants providing false responses.

Both phases of the study are limited by the fact that alternative definitions of masculinity were not fully examined. Rather, our work focused on understanding the specific masculine-physical ideal. Furthermore, our data do not truly capture the attitudes of heterosexuals regarding masculinity among HIV-positive men specifically and gay men in general. While our participants described their views of heterosexual attitudes toward gay men and masculinity, it should be noted that these are simply perceptions. Future work should seek to gather data from non-gay men regarding homosexuality and masculinity.

CLINICAL AND RESEARCH IMPLICATIONS

The data presented in this study suggest a conception of masculinity that is evident among some HIV-positive gay men. This physical construction of masculinity appears to be intimately linked to the social and sexual behaviors in which HIV-positive gay men engage. Further, these behaviors appear to be viewed as desirable by certain segments of the HIV-positive gay male community. Because such behaviors, which seem to have permeated the gay community at large, often involve sexual adventurism, sexual risk-taking, and an overemphasis on the body, they may create a social, emotional, and physical burden on HIV-positive gay men. For clinicians working with HIV-positive gay men, it is crucial to examine the synergistic interaction that occurs between the individual's gay, masculine, and HIV-positive identities since these appear to be overlapping realities that have an impact on the decisions these men make and the behaviors in which these men engage. Further, it is imperative to understand that, for some of these men, the decisions to engage in risk behaviors associated with their conception of masculinity may not be a matter of choice but rather driven by the perceived community norms. For marginalized HIV-positive gay men, inclusion and acceptance by the gay community may be essential.

In terms of HIV prevention research, there is a scarcity of literature that has focused on the role of masculinity and body image for HIV-positive gay men as well as a lack of understanding of HIV-positive gay men as sexual beings. We believe that both these elements are central to understanding HIV-positive gay men and how they define their masculine self in relation to their gay self and their "HIV-positive self." Much more elaborate studies are needed in this area, particularly as these constructs may serve as salient contextual factors for HIV prevention. In particular, empirical and qualitative investigations should be undertaken to examine how masculinity for HIV-positive gay men may interface with race/ethnicity, geographical location, and other demographic factors as well as how alternative definitions of masculinity may impact sexual and social behaviors. Masculinity may certainly take many forms among HIV-positive gay men, and future investigations should explore these alternate constructions in relation to the "buff agenda" put forth here.

Table 1
Use of Steroids in Relation to Scores on Conceptions of Masculinity,
Emphasis on Body, and Sexual Adventurism

                                         Steroids (1)

                                      Yes                No

Scale                            Mean      (SD)    Mean      (SD)

Masculinity as Sexual Behavior    9.55    (3.14)    8.75    (2.77)
Masculinity as Appearance *      21.79    (4.05)   20.25    (4.18)
Masculinity as Social Behavior   14.26    (3.61)   13.62    (3.89)
Emphasis on Body ***             17.73    (5.38)   10.58    (4.42)
Sexual Adventurism **            35.08   (11.17)   29.71   (11.15)

(1) Testosterone and/or Deca durabolin

* p = .07

** p = .02

*** p < = .001

Table 2
Interrelationships Between Emphasis on Body, Conceptions of
Masculinity, and Sexual Adventurism

                          A       B         C         D         E

Masculinity as Sexual    --    .562 **   .436 **   .372 **   .359 **
  Behavior (A)
Masculinity as           --      --      .381 **   .371 **   .386 **
  Appearance (B)
Masculinity as Social    --      --        --      .334 **   .224 *
  Behavior (C)
Emphasis on Body (D)     --      --        --        --      .444 **
Sexual Adventurism (E)   --      --        --        --        --

* p = .02

** p <.001

Table 3
Mean Frequencies (Standard Deviation) of Barebacking Acts in Relation
to Use of Viagra and Steroids

                                   Viagra
Barebacking
Behavior                     Yes              No

AI HIV+Partners        9.18 (12.46) *     5.19 (8.49)
AR HIV+ Partners      13.05 (17.00) ***   2.74 (4.77)
AI HIV-/UK Partners     1.95 (4.44)       0.90 (2.81)
AR HIV-/UK Partners     3.75 (5.21) ***   0.47 (l.15)

                                 Steroids (1)
Barebacking
Behavior                     Yes              No

AI HIV+Partners        9.30 (12.61) *     5.29 (8.26)
AR HIV+ Partners      11.94 (17.07) ***   3.77 (6.96)
AI HIV-/UK Partners     1.24 (2.97)       1.31 (3.81)
AR HIV-/UK Partners     3.29 (4.96) **    0.81 (2.26)

(1) Testosterone and/or Deca durabolin

* p<=.05

** p < = .01

*** p < =. .001

REFERENCES

Centers for Disease Control and Prevention. (2003). Primary and secondary syphilis--United States, 2002. MMWR, 52(46), 1118.

Centers for Disease Control and Prevention. (2003). Increases in HIV diagnoses--29 States, 1999-2002. MMWR, 52(47), 1145-1148.

Chapple, M.J., Kippax, S., & Smith, G. (1998). "Semi-straight sort of sex": Class and gay community attachment explored within a framework of older homosexually active men. Journal of Homosexuality, 35(2), 65-83.

Charlin, V., Carrasco, F., Sepulveda, C., Torres, M., & Kehr, J. (2002). Nutritional supplementation according to energy and protein requirements on malnourished HIV-infected patients. Archives of Latinoam Nutrition, 52(3), 267-273.

Clark, J.M. (1997). Doing the work of love: I. Men's studies at the margins. The Journal of Men's Studies, 5, 315-331.

Cooper, A., Delmonico, D.L., & Burg, R. (2000). Cybersex users, abuse, and compulsives: New findings and implications. Sexual Addiction & Compulsivity, 7, 5-29.

Cooper, A., Putnam, D.E., Planchon, L.A., & Boies, S.C. (1999). Online sexual compulsivity: Getting tangled in the net. Sexual Addiction & Compulsivity, 6, 79-104.

Gough, J. (1989). Theories of sexual identity and the masculinization of gay men. In S. Sheperd & M. Wallis (Eds.), Coming on strong: Gay politics and culture (pp. 119-136). London: Unwin Hymen.

Halkitis, P.N. (1999). Masculinity in the age of AIDS: Seropositive gay men and the "buff agenda." In P. Nardi (Ed.), Gay masculinities (pp. 130-151). Newbury Park, CA: Sage.

Halkitis, P.N. (2001). An exploration of perceptions of masculinity among gay men living with HIV. The Journal of Men's Studies, 9, 413-429.

Halkitis, P.N., & Parsons, J.T. (2003). Intentional unsafe sex (barebacking) among gay men who seek sexual partners on the Internet. AIDS Care, 15(3), 367-378.

Halkitis, P.N., Parsons, J.T., & Wilton. L. (2003). Barebacking, among gay and bisexual men in New York City. Archives of Sexual Behavior, 32(4), 351-358.

Halkitis, P.N., & Wilton, L. (in press). The meanings of sex for HIV seropositive gay and bisexual men: Synergistic interactions of emotions, physicality, and affirmations of self. In P.N. Halkitis, R J. Wolitski, & C.Gomez (Eds.), Positive living: The sexual lives of gay and bisexual seropositive men. Washington, DC: APA Publications.

Kippax, S., & Smith, G. (2001). Anal intercourse and power in sex between men. Sexualities, 4(4), 413-434.

Kleinberg, S. (1992). The new masculinity of gay men and beyond: Where have all the sissies gone?: A view from the 1970s. In M.S. Kimmel & M.A. Messner (Eds.), Men's lives (pp. 80-94). New York: Macmillan.

Koblin, B.A., Torian, L.V., Gulin, V., Ren, L., MacKellar, D.A., & Valleroy, L.A. (2000). High prevalence of HIV infection among young men who have sex with men in New York City. AIDS, 14, 1793-1800.

Levine, M.P. (1998). Gay macho: The life and death of the homosexual clone. New York: New York University Press.

Mansergh, G., Marks, G., Colfax, G., Guzman, R., Rader, M., & Buchbinder, S. (2002). Barebacking in a diverse sample of men who have sex with men. AIDS, 16, 653-659.

Miles, M.B., & Huberman, A.M. (1984). Qualitative data analysis. Newbury Park, CA: Sage.

Mocroft, A., Ledergerber, B., Katlama, C., Kirk, O., Reiss, P., d'Arminio, M.A., et al. (2003). Decline in the AIDS and death rates in the EuroSIDA study: An observational study. Lancet, 362, 22-29.

Nardi, P. (2000). Anything for a sis, Mary. In P. Nardi (Ed.), Gay masculinities (pp. 1-11). Newbury Park, CA: Sage.

Plummer, D. (1999). One of the boys; Masculinity, homophobia, and modern man-hood. Binghamton, NY: The Haworth Press.

Shilts, R. (1987). And the band played on. New York: Viking Penguin.

Signorile, M. (1997). Life outside: The Signorile report on gay men: Sex, drugs, muscles, and the passages of life. New York: Harper Collins.

Suarez, T., & Miller, J. (2001). Negotiating risks in context: A perspective on unprotected anal intercourse and barebacking among men who have sex with men--where do we go from here? Archives of Sexual Behavior, 30, 287-300.

Tate, H., & George, R. (2001). The effect of weight loss on body image in HIV-positive men. AIDS Care, 13, 163-169.

Correspondence concerning this article should be address to Perry N. Halkitis, New York University, Department of Applied Psychology, 239 Greene Street, Room 537J, New York, NY 10003. Electronic mail: [e-mail].

PERRY N. HALKITIS, KELLY A. GREEN New York University

LEO WILTON Binghamton University