Hypermasculinity and Sexual Risk Among Black and White Men Who Have Sex With Men and Women

Abstract

Men who have sex with men and women (MSMW), particularly Black MSMW, are at high risk for HIV. However, few studies have focused on factors that influence Black MSMW’s specific HIV risk behaviors, and there are no evidence-based interventions specifically targeting this population. Some studies have suggested that masculine ideals are associated with high-risk sex practices and partners. Norms around masculinity in the social environments in which MSMW live may prohibit nonheterosexual identities and behaviors, may lead to internalized homophobia, and may promote high-risk strategies to seek male partners. Using data collected from 180 Black and 101 White MSMW recruited for a study to develop strategies for recruiting MSMW for research and services and to inform the content of HIV prevention messages, we examined the association between hypermasculinity ideals and sexual behaviors that may contribute to increased HIV risk among Black MSMW and a comparison group of White MSMW. Comparing Black and White MSMW, we explored how this association may differ by race. Multivariate models, controlling for sociodemographic and other covariates, indicate that hypermasculine ideals are associated with increased numbers of male and female partners among Black MSMW and an increased number of female partners among White MSMW. Hypermasculinity is important to address in programs that aim to reduce HIV risk among Black MSMW.

Keywords

masculinity ideology risk behaviors HIV/AIDS bisexual

Three decades into the HIV epidemic, men who have sex with men (MSM) continue to represent the most severely affected population in the United States, with the highest rates of infection being among Black MSM (Centers for Disease Control and Prevention [CDC], 2011). Black MSM are more likely than MSM of other racial or ethnic groups to identify as bisexual or report bisexual behavior (Millett, Malebranche, Mason, & Spikes, 2005; Montgomery, Mokotoff, Gentry, & Blair, 2003). Among Black MSM, men who have sex with men and women (MSMW) may be more likely to report behaviors that increase HIV risk compared with MSM with male partners only, including recent sex exchange activity, recent substance use, and criminal justice involvement (Wheeler, Lauby, Liu, Van Sluytman, & Murrill, 2008).The high prevalence of HIV infection and increased likelihood of bisexual behavior among Black MSM compared with men of other racial or ethnic groups make an understanding of the HIV risk and prevention needs of Black MSMW imperative. While there is an emerging literature that includes Black MSMW (Siegel, Schrimshaw, Lekas, & Parsons, 2008), few studies have focused specifically on MSMW, and currently, there are no published studies about effective interventions tailored specifically to address the HIV risk of this population. Development of effective HIV prevention interventions specifically targeting Black MSMW is hampered by the paucity of data on factors that influence this population’s HIV risk behaviors. Most research studies to date have only included bisexually active men in samples of MSM, obscuring MSMW’s unique sexual health and HIV risk and prevention needs (Dodge, Jeffries, & Sandfort, 2008; Wheeler et al., 2008).

Researchers have called for the development of HIVprevention interventions that consider more culturally-and contextually-specific approaches to work with Black MSMW (Dodge et al., 2008; Mays, Cochran, & Zamudio, 2004; Siegel et al., 2008; Wheeler et al., 2008). For example, masculinity may play a unique role in the HIV risk behaviors of Black MSMW, intersecting with male gender norms, internalized homophobia, and disclosure of sexual identity or behaviors. The construction of masculinity among Black men in the United States differs historically from that of White men (Ward, 2005). White masculinity is socially constructed on a foundation of power and privilege in this country (Mays et al., 2004). In contrast, some assert that Black masculinity in the United States evolved in response to the historical experience of racism and slavery and is often constructed in hypermasculine terms (Ward, 2005). Hypermasculinity refers to the exaggeration of traditionally masculine traits, extolling male physical strength, aggression, violence, competition, dominance, and sexual prowess, likening the lack of these characteristics to being weak and feminine (Ward, 2005; Wolfe, 2003). Previous research has suggested that hypermasculine ideals among Black men are associated with higher rates of multiple sex partners and an aversion to using condoms (Oparanozie, Sales, DiClemente, & Braxton, 2012; Rhodes et al., 2011; Wolfe, 2003).

Masculine gender norms in the social environments in which many Black men live and interact, including family and religious networks, may prohibit expressions of nonheterosexual identities and behaviors (Millett et al., 2005; Stokes, McKirnan, Doll, & Burzette, 1996; Stokes, Vanable, & McKirnan, 1997). Studies suggest that Black MSM are less likely to disclose their bisexual or homosexual behavior than White MSM (Millett et al., 2005; Wheeler et al., 2008). There have been mixed findings regarding the role of nondisclosure of bisexual or homosexual identity or behavior in HIV risk. For example, nondisclosure of same-sex sexual behavior has been linked to increasing HIV and sexually transmitted infection rates among female partners of Black MSMW (Montgomery et al., 2003), supporting recent research focusing on Black bisexually active men serving as a bridge between the HIV epidemics among Black MSM and Black heterosexual women (Malebranche, Arriola, Jenkins, Dauria, & Patel, 2010; Reback & Larkins, 2013). On the other hand, researchers have also found that nondisclosure of sexual identity is associated with a lower likelihood of HIV infection (Millett, Flores, Peterson, & Bakeman, 2007). Still others have found a lack of a role in disclosure of same-sex behavior and condom use practices among Black bisexual men (Malebranche et al., 2010).

The need to exhibit a masculine public persona, social pressures to establish a family, and pressures to conform to social and sexual norms may lead to internalized homophobia as well as strategies for seeking male partners that may be associated with high-risk sex practices and partners (Miller, Serner, & Wagner, 2005; Reback & Larkins, 2013; Schrimshaw, Downing, & Siegel, 2013; Ward, 2005). Men with more traditional masculine ideologies have been found to be more likely to engage in risk behaviors, such as unprotected vaginal sex and intimate partner violence perpetration (Santana, Raj, Decker, La Marche, & Silverman, 2006) and report lower frequencies of health-promoting behaviors more generally (Mahalik, Burns, & Syzdek, 2007). There is a need for more research examining the influence of masculinity on behaviors that increase HIV risk among Black MSMW. Furthermore, there is a need to explore how internalized homophobia and disclosure of MSM behavior may or may not affect the relationship between masculinity and HIV risk.

In the present study, we examined the influence of hypermasculinity ideals on sexual behavior that may contribute to increased risk for HIV acquisition or transmission among Black and White MSMW and their male and female partners, taking into account internalized homophobia and disclosure of MSM behavior. By focusing on bisexually active Black men and a comparison group of White MSMW, we address gaps in the literature (Siegel et al., 2008) and contribute to the knowledge base on culturally- and contextually- specific factors relevant to HIV risk among Black MSMW. Based on previous research that has found hypermasculine ideals to predict higher rates of multiple partners (Rhodes et al., 2011) and the recognition that having multiple sexual partners may increase HIV risk, we hypothesized that increased levels of hypermasculine ideals would be positively associated with increased sexual risk for HIV through higher numbers of male and female partners in the past 3 months. Based on the unique construction of hypermasculinity among Black men, we also hypothesized that the influence of hypermasculine ideals would be different for Black MSMW than for White MSMW, although we cannot predict whether the influence will be greater or lesser for Black men compared with White men.

Method

Participants

The data were collected as part of a research study with Black and White MSMW designed to test the feasibility of respondent-driven sampling (RDS) in this population and to develop strategies for recruiting bisexually active men for research, HIV testing, and prevention services, as well as to inform the content of HIV prevention messages and interventions. The study was sponsored by the CDC. Men were enrolled in the study from December 2007 through June 2008. To be eligible, participants had to be male, 18 years of age or older, identify as Black or White, reside in the Philadelphia metropolitan area, be proficient in English, and report sex (oral sex, vaginal or anal intercourse) with at least one female and at least one male in the past 12 months. Men who were HIV-negative, HIV-positive, or of unknown serostatus were eligible to enroll.

A total of 237 Black MSMW and a comparison group of 109 White MSMW were recruited through RDS in Philadelphia, PA to participate in a quantitative survey and HIV testing. RDS, a form of chain-referral sampling (Heckathorn, Semaan, Broadhead, & Hughes, 2002), has been used in past research to reach persons from hidden or hard-to-reach populations, including injection drug users and non–gay-identified MSM who cannot always be reached through venue-based sampling, random-digit dialing, or targeted recruitment in the field (Abdul-Quader et al., 2006; Carballo-Diéguez et al., 2011; Heckathorn et al., 2002; Ma et al., 2007; Ramirez-Valles, Heckathorn, Vázquez, Diaz, & Campbell, 2005). Of the 346 MSMW enrolled in the study, 65 men (57 Black and 8 White) who reported being HIV-positive were excluded from this analysis. Previous research has shown that individuals reduce their HIV-risk behaviors once they know they are HIV-positive (Lauby et al., 2008). As mentioned previously, the original study for which these data were collected included men regardless of HIV status. However, since the focus of this analysis was to examine factors influencing sexual risk for HIV and factors affecting sexual risk behaviors may be different for HIV-positive men, we limited our analyses to men who were HIV-negative or did not know their status. In all, 180 Black men and 101 White men reported having never tested, being HIV-negative, or not knowing their HIV status. This subgroup of 281 MSMW was the focus of the present analysis.

Procedure

Enrollment for this study began through the selection of an initial set of participants (called “seeds”) who met study eligibility criteria and who represented different subgroups based on race and age. Selection of seeds occurred prior to the onset of quantitative data collection as well as during the data collection period as needed. Initial seeds were identified in consultation with the study collaborators at community-based organizations where data collection was to take place as well as the study’s Community Advisory Board members. Seeds participated in study protocol and were then given up to three recruitment coupons and encouraged to recruit other eligible individuals from their social networks to participate in the study. Men recruited by the seeds were then asked to recruit the next wave of persons, with the process continuing until the target sample size was achieved. Demographic data from participants were monitored during the course of enrollment to check on the diversity of the sample. Monitoring of demographic data revealed that enrollment of White men into the study was proceeding at a much slower pace than that of Black men. Additional White seeds were then recruited through targeted outreach methods on Internet websites, in chat rooms, and on the street. Once men who were recruited through these methods participated in the study, they were given recruitment coupons following the protocol described above. This study was approved by the Public Health Management Corporation Institutional Review Board.

Men who received a recruitment coupon enrolled in the study by calling the study site to confirm their eligibility and schedule an appointment at one of two community-based organizations that are accessible by public transportation. All study candidates had to present a valid coupon before enrolling. After screening for eligibility and providing written consent, participants completed a survey on a laptop computer. The first half of the survey was administered by a trained interviewer and the second half, which contained sensitive questions about drug use and sexual behavior, was self-administered using audio computer-assisted self-interview. The interview took 30 to 60 minutes to complete. A unique numeric identification code that did not contain any personal identifiers was generated for each participant.

Measures

Hypermasculinity Ideals

Hypermasculinity ideals were measured using an adapted version of the Hypermasculinity Posturing subscale of The Multicultural Masculinity Ideology Scale developed and validated among Anglo-American (α = .88) and African American college-based samples (α = .76) by Doss and Hopkins (1998). This 13-item scale had a high reliability for our sample (α = .83 among Black MSMW; α = .87 among White MSMW). Sample items from this scale included, “A guy should prove his masculinity by having sex with a lot of people,” “To be a guy, you’ve got to be tough,” “A guy should always have a woman he is dating,” and “The best way a man can care for his family is to get the highest paying job he can.” A higher score on the scale from 1 to 4 equals a higher level of hypermasculinity ideals.

Internalized Homophobia

Internalized homophobia was measured using an 8-item scale adapted from two different scales (Ross, Rosser, Neumaier, & The Positive Connections Team, 2008; Wagner, 1998). Four items were from a scale adapted by Ross et al. (2008), including, “Sometimes I dislike myself for being gay/bisexual/attracted to men,” “I feel stress and conflicted within myself over having sex with men,” “Sometimes I wish I were not gay/bisexual/attracted to men,” and “I sometimes feel guilty after having sex with men.” Three items were adapted from Wagner’s Internalized Homophobia Scale (Wagner, 1998). These items were, “A gay man can have just as fulfilling a life as a straight man,” “I have no regrets about my desire for men,” and “I am confident that my desire for men does not make me inferior,” The final item was developed by the research team and was “It is important to me to have gay or bisexual friends.” This 8-item scale, where a higher score from 1 to 4 equals greater internalized homophobia, had a high reliability for our sample (α = .74 among Black MSMW; α = .83 among White MSMW).

Disclosure of MSM Behavior

To measure disclosure of MSM behavior, study participants were asked, “Generally, how important is it for you to keep your sexual relationships with men secret?” to which men responded “very important,” “somewhat important,” “a little important,” or “not important at all.” For this analysis, men who said “a little” or “somewhat” important were combined.

Sexual Risk Behaviors

The sexual risk outcome variables in which we were primarily interested were number of male and number of female sex partners in the past 3 months. Participants were asked, “How many male sex partners have you had in the past 3 months? Include only men with whom you had oral or anal sex, with or without a condom, and with or without ejaculation,” and “How many female sex partners have you had in the past 3 months? Include only women with whom you had oral, vaginal, or anal sex, with or without a condom, and with or without ejaculation.” The range for number of male partners in the past 3 months for the whole sample was 0 to 30, and the range for numbers for female partners in the past 3 months was 0 to 90. Participants with greater than 20 partners were recoded to 20 to produce a more normal distribution. Less than 5% of participants had 20 or more sexual partners in the past 3 months.

Other sexual risk behavior variables included in our analysis were insertive unprotected anal intercourse (UAI) with a male partner, receptive UAI with a male partner in the past 3 months, unprotected vaginal or anal intercourse (UVA) with a female partner in the past 3 months, unprotected sex with both male and female partners in the past 3 months, and exchanging sex with a man for money, drugs, shelter, or other things of value in the past 3 months. To measure insertive UAI in the past 3 months, receptive UAI in the past 3 months, and UVA in the past 3 months, we computed variables based on multiple survey questions. Participants who had engaged in insertive or receptive anal sex with a man or vaginal or anal sex with a woman in the past 3 months and indicated having sex without a condom at least one time during that time period for each of these behaviors were computed as having engaged in insertive UAI, receptive UAI, or UVA, respectively. To measure exchanged sex with a man, we combined two questions. Participants were asked, “In the past 3 months did you give money, drugs, food, a place to stay (shelter), or other things of value in exchange for sex with another man? “ and “In the past 3 months did you receive money, drugs, food, a place to stay (shelter) or other things of value in exchange for sex with another man?” Participants who said yes to either or both questions were computed as having exchanged sex with a man in the past 3 months.

Sociodemographic Measures

Demographic variables such as age, annual income, educational attainment, and employment were measured with standard response formats. To assess living situation, participants were asked to identify the different types of places they lived in the past 12 months and where they currently live. Men who reported that they lived on the street or in a park, car, abandoned house, shelter, or mission currently or in the past 12 months were considered to have been homeless in the past 12 months. Sexual identity was measured with the question, “Do you think of yourself as heterosexual or straight, homosexual or gay, bisexual, unsure/questioning, or other?” Men who identified as unsure/questioning or other were combined for analysis. To assess whether participants had ever been married to a female and whether participants had children, men were asked, “Have you ever been married to a female?” and “How many children do you have, whether they are living with you or not?,” respectively.

Statistical Analysis

Descriptive analyses were conducted using frequencies to describe the sample and chi-square and t tests to test for differences between Black and White MSMW on hypermasculinity ideals, internalized homophobia, sociodemographics (age, income, employment, education, homelessness, incarceration, sexual identity, marriage to a female, children), disclosure of MSM behavior, HIV status, and sexual risk behaviors. Separate analyses were then conducted for Black and White MSMW, in which correlations and one-way analyses of variance were used to explore unadjusted bivariate associations between hypermasculinity ideals and internalized homophobia, sociodemographics, disclosure of MSM behavior, HIV status, and sexual risk behavior variables. Separate multivariate linear regression models were then conducted for Black and White MSMW to assess the influence of hypermasculinity ideals in each racial group on number of male partners in the past 3 months and the number of female partners in the past 3 months. Regression analyses were conducted to control for the effects of possible confounders that were associated with the dependent variables or hypermasculinity in this sample (p < .10) or have been linked in previous research with the dependent variables or traditional masculine ideals. These included age, indicators of socioeconomic status (annual income and homelessness in the past 12 months), disclosure of MSM behavior, internalized homophobia, and exchanging sex with a man. Using a p value of <.10 to detect statistical significance in bivariate analyses reduced the chances that we missed any meaningful relationships in this relatively small sample that would be important to account for in further analyses.

Results

Characteristics of the Sample and Comparison by Race

Table 1 summarizes the characteristics of the sample as a whole and compares Black and White MSMW in our sample. As expected, the majority of the total sample identified as bisexual (69%), or heterosexual, or straight (14%). Overall, the sample was socioeconomically disadvantaged. The majority was unemployed (70%), 34% reported an annual income less than $5,000, and 43% had been homeless in the past 12 months. One quarter (25%) had less than a high school education and half (49%) had a high school diploma or GED. Nearly two thirds (63%) of the sample had been in jail in their lifetime. One third (34%) of men in the total sample had ever been married, and more than half (56%) had children. Nearly half (46%) of men felt it was “very important” for them to keep their MSM behavior a secret.

Table 1.

Sociodemographic Characteristics, Hypermasculinity Ideals, Internalized Homophobia, HIV Status, and Sexual Risk Behaviors, Comparison of Black and White MSMW^a: Philadelphia, 2007-2008 (n = 281).

	Total (n = 281); n (%)	Black (n = 180); n (%)	White (n = 101); n (%)	p
Sociodemographic characteristics
Age (years)				<.001
18-29	45 (16.0)	11 (6.1)	34 (33.7)
30-44	102 (36.3)	61 (33.9)	41 (40.6)
45+	134 (47.7)	108 (60.0)	26 (25.7)
Annual income ($)				<.001
<5,000	96 (34.3)	58 (32.2)	38 (38.0)
5,000-9,999	65 (23.2)	46 (25.6)	19 (19.0)
10,000-19,999	57 (20.4)	47 (26.1)	10 (10.0)
20,000+	62 (22.1)	29 (16.1)	33 (33.0)
Educational attainment				.040
<High school	70 (24.9)	51 (28.3)	19 (18.8)
High school/GED	139 (49.5)	91 (50.6)	48 (47.5)
Some college or more	72 (25.6)	38 (21.1)	34 (33.7)
Current employment
Full-/part-time	75 (26.7)	47 (26.1)	28 (27.7)	.951
Unemployed	198 (70.5)	128 (71.1)	70 (69.3)
Student/retired	8 (2.8)	5 (2.8)	3 (3.0)
Homeless past 12 months	160 (43.1)	87 (48.3)	34 (33.7)	.012
Ever been in jail	177 (63.0)	124 (68.9)	53 (52.5)	.005
Ever married (to a female)	94 (33.5)	70 (38.9)	24 (23.8)	.007
Have children	158 (56.2)	122 (67.8)	36 (35.6)	<.001
Sexual identity				.053
Homosexual or gay	15 (5.3)	11 (6.1)	4 (4.0)
Heterosexual or straight	40 (14.3)	32 (17.9)	8 (7.9)
Bisexual	192 (68.6)	113 (63.1)	79 (78.2)
Questioning/other	33 (11.8)	23 (12.8)	10 (9.9)
Disclosure of MSM behavior
Importance of keeping sex with men secret				.024
Not at all important	61 (21.7)	42 (23.3)	19 (18.8)
A little/somewhat important	91 (32.4)	48 (26.7)	43 (42.6)
Very important	129 (45.9)	92 (50.0)	39 (38.6)
Hypermasculinity ideals
Mean hypermasculinity scale score^b	2.16 (range = 1-3.77)	2.24 (range = 1-3.77)	2.03 (range = 1-3.36)	.001
Internalized homophobia
Mean internalized homophobia scale score^c	2.21 (range = 1-4)	2.25 (range = 1-4)	2.12 (range = 1-3)	.048
HIV status
Self-reported HIV status				.113
Never tested/don’t know^d	40 (14.2)	25 (13.9)	15 (14.9)
HIV-negative	241 (85.8)	155 (86.1)	86 (85.1)
Sexual risk behaviors
Mean number of male sex partners, past 3 months	3.3 (range = 0-20)	3.3 (range = 0-20)	3.2 (range = 0-20)	.803
Mean number of female sex partners, past 3 months	4.0 (range = 0-20)	4.6 (range = 0-20)	3.1 (range = 0-20)	.015
Insertive UAI with male past 3 months^e (n = 237)	93 (39.2)	65 (43.6)	28 (31.8)	.076
Receptive UAI with male past 3 months^e (n = 37)	54 (22.8)	32 (21.8)	22 (24.4)	.636
Unprotected vaginal or anal sex with female past 3 months^e (n = 228)	154 (67.5)	100 (69.9)	54 (63.5)	.380
Unprotected anal sex with male and unprotected vaginal or anal sex with female past 3 months^e (n = 215)	79 (36.7)	52 (38.8)	27 (33.3)	.255
Exchanged sex with a man past 3 months	115 (40.9)	82 (45.6)	33 (32.7)	.043

Note. MSMW = men who have sex with men and women; MSM = men who have sex with men; UAI = unprotected anal intercourse. Entries in boldface indicate statistically significant p values.

Chi-square tests of significance and one-way analyses of variances were conducted to examine differences between Black and White MSMW; a conservative cutoff for detecting statistical significance in bivariate analyses was set to p < .10.

Hypermasculinity measured with a 13-item scale that is scored on a range from 1 to 4, where a higher score equals higher levels of hypermasculinity ideals.

Internalized homophobia measured with an 8-item scale that is scored on a range from 1 to 4 where a higher score equals higher levels of internalized homophobia.

Nineteen Black and 15 White MSMW had never taken an HIV test; 6 Black MSMW did not know the result of any of their HIV tests.

Includes only those participants who had that gender partner in the past 3 months and responded to questions about unprotected sex in the past 3 months.

Statistically significant differences (see Table 1) were observed between Black and White MSMW on several variables. Black MSMW in our study scored significantly higher on both the hypermasculinity ideals and internalized homophobia scales compared with White MSMW. They were also more likely to say it was “very important” to keep their MSM behavior secret and more likely to identify as heterosexual or straight than White men. White men were more likely to identify as bisexual than Black men. Black men were also more likely than White men to have ever been married to a female and to have children. Looking at other sociodemographic characteristics, Black men in the sample were significantly older than White men, were more likely to report lower annual incomes, more likely to have less than a high school education, more likely to have been homeless in the past 12 months, and more likely to have ever been in jail. Black and White MSMW also differed on several sexual risk behaviors; Black men had a significantly higher number of female sex partners in the past 3 months, were more likely to report insertive UAI with a man in the past 3 months, and were more likely to report exchanging sex with a man in the past 3 months. However, no racial differences were observed in mean number of male sex partners in the past 3 months, receptive UAI with a man, UVA in the past 3 months, or unprotected sex with both male and female partners in the past 3 months.

Mean Hypermasculinity Ideals Scale Scores by Characteristics and Behaviors

We conducted one-way analyses of variance and correlations to identify significant relationships between hypermasculinity ideals and factors, including internalized homophobia, sociodemographics, disclosure of MSM behavior, HIV status, and sexual risk behaviors. Table 2 shows variables that were found to be associated with hypermasculinity ideals (at p < .10) at the bivariate level. For the sample as a whole, hypermasculinity was associated with income, education less than high school, homelessness in the past 12 months, ever being in jail, feeling that it is “very important” to keep MSM behavior secret, greater internalized homophobia, and exchanging sex with a man in the past 3 months. Separate analyses were conducted for Black and White men in order to examine how hypermasculinity may operate differently for Black men compared with White men. Among Black men, hypermasculinity was significantly associated with income, being HIV-negative, greater internalized homophobia, and exchanging sex with a man in the past 3 months. Among White men, hypermasculinity was associated with income, employment, homelessness in the past 12 months, ever being in jail, feeling that it is “very important” to keep MSM behavior secret, greater internalized homophobia, and exchanging sex with a man in the past 3 months. Hypermasculinity was not correlated with age, marriage to a female, having children, sexual identity, unprotected receptive or insertive anal sex with a man in the past 3 months or unprotected vaginal or anal sex with a woman in the past 3 months (data not shown).

Table 2.

Mean Hypermasculinity Ideals Scale Scores by Sociodemographic and Risk Characteristics for Black and White MSMW.^a

	Total sample (n = 281)	p	Black (n = 180)	p	White (n = 101)	p
Sociodemographic characteristics
Annual income ($)		.074		.076		.070
<5,000	2.136		2.109		2.170
5,000-9,999	2.266		2.336		2.098
10,000-19,999	2.228		2.290		1.938
20,000+	2.057		2.272		1.877
Educational attainment		.004		.109		.162
<High school	2.253		2.287		2.162
High school/GED	2.204		2.274		2.064
Some college or more	2.002		2.097		1.909
Current employment		.761		.728		.096
Full-/part-time	2.129		2.287		1.863
Unemployed	2.177		2.225		2.088
Student/retired	2.202		2.185		2.231
Homeless past 12 months		.088		.722		.005
Yes	2.223		2.252		2.222
No	2.121		2.227		1.933
Ever been in jail		.030		.581		.072
Yes	2.214		2.253		2.114
No	2.081		2.211		1.938
HIV status
Self-reported HIV status		.136		.032		.710
Never tested/don’t know	2.057		2.057		2.074
HIV-negative	2.183		2.272		2.022
Disclosure of MSM behavior
Importance of keeping sex with men secret		.025		.291		.023
Not at all important	2.028		2.152		1.754
A little/somewhat important	2.156		2.222		2.082
Very important	2.236		2.291		2.107
Sexual risk behaviors
Exchanged sex with a man past 3 months		.003		.099		.028
Yes	2.270		2.305		2.184
No	2.092		2.187		1.956
Internalized homophobia
Correlation between internalized homophobia and hypermasculinity scales	.330	<.001	.170	.028	.520	<.001

Note. MSMW = men who have sex with men and women; MSM = men who have sex with men. Entries in boldface indicate statistically significant p values at p<.10.

A conservative cutoff for detecting statistical significance at the bivariate level was set to p < .10.

Effects of Hypermasculinity on Numbers of Male and Female Partners in the Past 3 Months (Dependent Variables)

Bivariate and multivariate analyses were conducted to examine the effects of hypermasculinity ideals on our dependent variables: numbers of male and female sexual partners in the past 3 months. In bivariate analysis among Black men, hypermasculinity was significantly and positively correlated with number of male (correlation = .239; p < .001) and female (correlation = .303; p < .001) sex partners in the past 3 months. Among White MSMW, hypermasculinity was significantly and positively associated with number of female sex partners in the past 3 months (correlation = .309; p = .002), but was not correlated with number of male sex partners in the past 3 months.

Table 3 shows the results of four multivariate linear regression models examining the impact of hypermasculinity on (a) number of male sex partners in the past 3 months among Black MSMW; (b) number of male sex partners in the past 3 months among White MSMW; (c) number of female sex partners in the past 3 months among Black MSMW; and (d) number of female sex partners in the past 3 months among White MSMW, controlling for internalized homophobia, age, annual income, and exchanging sex with a man in the past 3 months. Each of the variables included in the final regression models was associated with hypermasculinity or one of the risk behavior outcomes at the bivariate level or was an important factor to consider contextually (i.e., age and income). Some factors, including unprotected sex with men and women, ever being in jail, HIV status/testing, and disclosure of MSM behavior, that were significant in bivariate analyses were not included in final regression models because multivariate analyses revealed that they were unrelated to our outcome variables when controlling for other factors, and they had no effect on the regression analyses. All four of the final regression models included the same variables.

Table 3.

Associations Between Hypermasculinity and Number of Male and Female Sex Partners in the Past 3 Months (Linear Regression^a; Standardized Regression Coefficients).

	Male Partners in the Past 3 Months				Female Partners in the Past 3 Months
	Black MSMW (n = 167)		White MSMW (n = 96)		Black MSMW (n = 167)		White MSMW (n = 96)
Variable	β	t	β	t	β	t	β	t
Age	−.126	−1.617	.138	1.331	−.019	−0.250	.206*	2.036
Annual income = $5,000–$9,999^b	.045	0.480	−.202	−1.810	−.099	−1.083	.222*	2.041
Annual income = $10,000-$19,999^b	−.022	−0.236	−.186	−1.724	−.122	−1.313	−.151	−1.438
Annual income = $20,000 or more^b	.030	0.332	−.169	−1.402	.041	0.472	−.289*	2.454
Homeless past 12 months	−.089	−1.068	−.122	−1.151	−.142	−1.732	−.115	−1.113
Exchanged sex with a man in the past 3 months	.177*	2.233	.316**	3.021	.138	1.782	.164	1.604
Hypermasculinity ideals	.174*	2.187	.126	1.057	.258**	3.320	.306*	2.646
Internalized homophobia	−.048	−0.588	−.026	−0.224	.074	0.933	−.101	−0.879
R ²	.080		.169		.122		.209

Note. MSMW = men who have sex with men and women. Entries in boldface indicate statistically significant values.

Because of missing data, samples for regression analyses are smaller than the total sample.

Reference group includes men who reported an annual income less than $5,000.

p < .05. **p < .01.

Controlling for other covariates, regression analyses showed that a higher score on the hypermasculinity ideals scale (i.e., being more hypermasculine) was significantly associated with a higher number of male partners and a higher number of female partners in the past 3 months among Black MSMW. Examining the standardized coefficients of the regression models for Black MSMW, hypermasculinity appears to have a stronger influence on the number of female partners than on the number of male partners (β = .258 vs. .174, respectively). Similar to Black MSMW, White MSMW who were more hypermasculine had significantly higher numbers of female partners in the past 3 months. In order to assess whether the effect of hypermasculinity on the number of female partners differed by race, we combined the total study sample and ran an interaction term (Hypermasculinity × Race). We found no significant interaction effect between race and hypermasculinity (data not shown), indicating that the difference in the effect of hypermasculinity by race is not statistically significant. Hypermasculinity was not found to be associated with the number of male partners among White MSMW.

Exchanging sex with a man in the past 3 months was the only predictor of number of male sexual partners in the past 3 months among White MSMW, where men who exchanged sex had a higher number of male partners. Exchanging sex also predicted a higher number of male sexual partners among Black MSMW. While hypermasculinity was the only predictor of number of female sexual partners in the past 3 months for Black MSMW, predictors of the number of female partners for White MSMW also included increased age and annual income. We found a curvilinear relationship between income and number of female partners, where compared with men who reported an annual income less than $5,000, men with an annual income between $5,000 and $9,999 had the highest numbers of female partners, and men with an annual income of $20,000 or more had the lowest numbers of female partners. Men reporting an annual income between $10,000 and $19,999 had similar numbers of female partners compared with men with less than $5,000. Number of sexual partners in the past 3 months was not associated with homelessness in the past 12 months or levels of internalized homophobia in adjusted multivariate analyses for Black or White MSMW.

Discussion

The present study allowed us to compare Black and White MSMW to examine how hypermasculinity differentially influences sexual risk for HIV through multiple-partnered sex. Unadjusted analyses revealed several important sociodemographic differences between Black and White MSMW in our sample. Black MSMW were older, had lower income, and were more likely to have been homeless in the past 12 months and to have been in jail. Comparing Black and White MSMW, Black MSMW in our study were more hypermasculine, had a higher level of internalized homophobia, were more likely to say it was “very important” to keep their MSM behavior secret, to have ever been married to a female, and to have children compared with White MSMW.

Our findings support our first hypothesis that hypermasculine ideals contribute to increased sexual risk for HIV through higher numbers of sexual partners among MSMW. Multiple regression analyses controlling for sociodemographic factors, exchanging sex, and internalized homophobia revealed that greater hypermasculinity predicted higher numbers of female sexual partners in the past 3 months among both Black and White MSMW and a higher number of male sexual partners in the past 3 months among Black MSMW. Similar to previous studies (Miller et al., 2005; Ward, 2005), which have discussed the interconnections between masculinity, internalized homophobia, and nondisclosure of MSM behavior, our analysis found that hypermasculinity was correlated with internalized homophobia among both Black and White MSMW, and with nondisclosure of MSM behavior among White MSMW. Research may be warranted to further explore the intersections between these factors, and how they may influence HIV risk among MSMW. Our study also supports our second hypothesis that the influence of hypermasculinity ideals on multiple sexual partners is different for Black MSMW compared with White MSMW. Hypermasculinity ideals were significantly associated with higher numbers of male sexual partners among Black MSMW but not among White MSMW. However, our findings do not support a differential influence of hypermasculinity on number of female partners by race. Though we expected to see a racial disparity in the relationship between hypermasculinity and number of female partners, our analysis showed that hypermasculinity was significantly associated with higher numbers of female partners among both Black and White men, with no significant interaction effect between race and hypermasculinity. Researchers have asserted that historical racism and reduced opportunities for economically and politically focused confirmation of masculinity have resulted in Black men being socialized to affirm their manhood through the quest for sexual prowess, including sex with multiple partners, particularly female partners (Bowleg, 2004; Malebranche, Fields, Bryant, & Harper, 2009; Ward, 2005; Wolfe, 2003). While our findings support this assertion, it is important to note that our study suggests that this may also be true for low-income White MSMW such as those in our study. We found that White MSMW who had lower income, were less educated, had been homeless in the past 12 months, ever been in jail, felt that it was very important to keep MSM behavior secret, had greater internalized homophobia, and had exchanged sex with a man in the past 12 months tended to be more hypermasculine. This suggests that similar to Black men, economic factors and homophobia may be important areas to explore in terms of the influence of hypermasculinity on HIV risk among White MSMW.

This study is not without limitations. First, although the RDS method facilitated the recruitment of large numbers of MSMW, it also introduced some possible areas of underrepresentation. The men in our sample were predominantly from lower income groups, with fewer men who were employed full time. The monetary incentives for study participation ($50) and for recruitment ($15 for each eligible participant recruited) were probably more attractive to men with less income, and those who were not employed had more time to participate and recruit their peers. Second, because the sample may not be representative, the findings reported here should not be generalized to the entire population of Black and White MSMW in Philadelphia. A third limitation is the reliance on self-report to assess the prevalence of HIV-risk behaviors. However, audio computer-assisted self-interview was used to collect this information in order to lessen underreporting due to social desirability. A fourth limitation lies with the measure of masculinity that was used for this analysis. Since masculinity was not the main focus of the original study for which these data were collected, the subscale of hypermasculinity was chosen as a brief measure that had been validated among Black and White men. Therefore, for this analysis, we were unable to analyze other facets of masculinity that may have contributed further to an understanding of the relationship between masculinity and sexual risk behaviors among Black and White MSMW. However, a strength of the subscale used for this analysis includes its high reliability among Black and White undergraduates in past studies (Doss & Hopkins, 1998) and among Black and White MSMW in the present study. A fifth limitation is that although we were able to measure whether or not study participants engaged in unprotected sex in the past 3 months, we were unable to determine whether each of the sexual encounters men had with their sexual partners was protected or unprotected. This is an important limitation since it may be asserted that even if one has multiple sexual partners, a recognized HIV risk, if each of their sexual encounters was protected, this would significantly reduce their risk for transmission or acquisition of HIV. However, the fact remains that the greater the number of sexual partners, the higher the likelihood there is for engaging in unprotected sex, using condoms incorrectly, or condom failure.

As far as we know, this is one of the few quantitative studies that have examined the influence of masculine ideals on sexual risk for HIV among Black MSMW, and the first study to examine the influence of masculinity on sexual risk behaviors among White MSMW. This study contributes to an emerging literature focusing specifically on HIV risk among MSMW. Our findings have important implications for HIV prevention for both MSMW and their male and female partners. Black women account for the majority of new HIV infections and AIDS cases among women in the United States (CDC, 2011), and the primary source of infection among women is sex with a man (CDC, 2013). Although the extent to which these infections stem from MSMW or from MSW is unknown, recent studies suggest that Black women are at risk for HIV infection from Black MSMW (Lauby et al., 2008). As Reback and Larkins (2013) suggest in their study of heterosexually identified men who have sex with men or transgender women, HIV prevention messages should target MSMW’s female partners as well as MSMW. Our findings that hypermasculine ideals were associated with increased numbers of both male and female partners among Black MSMW suggest that HIV prevention interventions should address issues of masculinity in efforts to reduce HIV risk through a reduction in multiple-partnered sex. Interventions that incorporate discussions of masculinity should expand and reframe the meaning of “manhood” beyond hypermasculine terms that may promote high-risk sexual practices. For example, Hammond and Mattis (2005) found that some African American men have developed more flexible definitions of masculinity, moving from a framework of African American manhood being seen as problematic “posturing” or defense mechanisms against historical racism and lack of economic opportunities to a construction of masculinity as a developmental and relational process promoting responsibility–accountability to self, family, and community. In addition, though Oparanozie et al. (2012) found that hypermasculine ideals contributed to concurrent sexual partnerships among African American men, strong African American racial identity contributed to a reduction in number of sexual partners, indicating that exploring how to focus on both masculinity and racial identity in efforts to reduce numbers of sexual partners may be a promising area of HIV prevention. A particular strength of this study was its inclusion of White MSMW, which allowed us to compare differences in how hypermasculinity may influence sexual risk among Black and White MSMW. Furthermore, our findings indicate that additional research is warranted to explore hypermasculine ideals among White MSMW. Although we found that hypermasculine ideals had a greater impact on the number of sexual partners among Black MSMW, our findings indicate that hypermasculine ideals are important to consider in HIV prevention for White MSMW, particularly those who may be lower income and socioeconomically disadvantaged, such as those in our study. In sum, our findings suggest that more studies that focus specifically on Black MSMW are needed, and that incorporating issues of masculinity into tailored interventions for Black MSMW is important to continue efforts to reduce HIV risk among this population that continues to experience disproportionate rates of HIV infection.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research was supported by funding from the Centers for Disease Control and Prevention (CDC) under grant number: 5UR6PS000324-02.

References

Abdul-Quader

Heckathorn

McKnight

Bramson

Nemeth

Sabin

. . . Des Jarlais

(2006). Effectiveness of respondent-driven sampling for recruiting drug users in New York City: Findings from a pilot study. Journal of Urban Health, 83, 459-476.

Bowleg

(2004). Love, sex, and masculinity in sociocultural context: HIV concerns and condom use among African American men in heterosexual relationships. Men and Masculinities, 7, 166-185.

Carballo-Diéguez

Balan

Marone

Pando

M. A.

Dolezal

Barreda

. . . Ávila

M. M.

(2011). Use of respondent driven sampling (RDS) generates a very diverse sample of men who have sex with men (MSM) in Buenos Aires, Argentina. PLoS One, 6(11), e27447.

Centers for Disease Control and Prevention. (2011). Fact sheet: Estimates of new HIV infections in the United States, 2006-2009. Atlanta, GA: Author.

Centers for Disease Control and Prevention. (2013). Fact sheet: HIV/AIDS among African Americans. Atlanta, GA: Author.

Dodge

Jeffries

IV, W. L.

Sandfort

T. G. M.

(2008). Beyond the down low: Sexual risk, protection, and disclosure among at-risk black men who have sex with both men and women (MSMW). Archives of Sexual Behavior, 37, 683-696.

Doss

B. D.

Hopkins

J. R.

(1998). The multicultural masculinity ideology scale: Validation from three cultural perspectives. Sex Roles, 38, 719-741.

Hammond

W. P.

Mattis

J. S.

(2005). Being a man about it: Manhood meaning among African American men. Psychology of Men & Masculinity, 6, 114-126.

Heckathorn

D. D.

Semaan

Broadhead

R. S.

Hughes

J. J.

(2002). Extensions of respondent driven sampling: A new approach to the study of injection drug users aged 18-25. AIDS and Behavior, 6, 55-67.

10.

Lauby

J. L.

Millett

G. A.

LaPollo

A. B.

Bond

Murrill

C. S.

Marks

(2008). Sexual risk behaviors of HIV-positive, HIV-negative, and serostatus-unknown black men who have sex with men and women. Archives of Sexual Behavior, 37, 708-719.

11.

Zhang

Sun

Yue

Chen

. . . McFarland

(2007). Trends in prevalence of HIV, syphilis, hepatitis C, hepatitis B, and sexual risk behavior among men who have sex with men: Results of 3 consecutive respondent-driven sampling surveys in Beijing, 2004 through 2006. Journal of Acquired Immune Deficiency Syndromes, 45, 581-587.

12.

Mahalik

J. R.

Burns

S. M.

Syzdek

(2007). Masculinity and perceived normative health behaviors as predictors of men’s health behaviors. Social Science & Medicine, 64, 2201-2209.

13.

Malebranche

D. J.

Arriola

K. J.

Jenkins

T. R.

Dauria

Patel

S. N.

(2010). Exploring the “bisexual bridge”: A qualitative study of risk behavior and disclosure of same-sex behavior among black bisexual men. American Journal of Public Health, 100, 159-164.

14.

Malebranche

D. J.

Fields

E. L.

Bryant

L. O.

Harper

S. R.

(2009). Masculine socialization and sexual risk behaviors among black men who have sex with men: A qualitative exploration. Men and Masculinities, 12, 90-111.

15.

Mays

V. M.

Cochran

S. D.

Zamudio

(2004). HIV prevention research: Are we meeting the needs of African American men who have sex with men. Journal of Black Psychology, 30, 78-105.

16.

Miller

Serner

Wagner

(2005). Sexual diversity among black men who have sex with men in an inner-city community. Journal of Urban Health, 82(Suppl. 1), i26-i34.

17.

Millett

G. A.

Flores

S. A.

Peterson

J. L.

Bakeman

(2007). Explaining disparities in HIV infection among black and white men who have sex with men: A meta-analysis of HIV risk behaviors. AIDS, 21(15), 2083-2091.

18.

Millett

Malebranche

Mason

Spikes

(2005). Focusing down-low: Bisexual black men, HIV risk and heterosexual transmission. Journal of the National Medical Association, 97, 52S-59S.

19.

Montgomery

J. P.

Mokotoff

E. D.

Gentry

A. C.

Blair

J. M.

(2003). The extent of bisexual behaviour in HIV-infected men and implications for transmission to their female partners. AIDS Care, 15, 829-837.

20.

Oparanozie

Sales

J. M.

DiClemente

R. J.

Braxton

(2012). Racial identity and risky sexual behaviors among black heterosexual men. Journal of Black Psychology, 38, 38-51.

21.

Ramirez-Valles

Heckathorn

D. D.

Vázquez

Diaz

R. M.

Campbell

R. T.

(2005). From networks to populations: The development and application of respondent-driven sampling among IDUs and Latino gay men. AIDS and Behavior, 9, 387-402.

22.

Reback

C. J.

Larkins

(2013). HIV risk behaviors among a sample of heterosexually identified men who occasionally have sex with another male and/or a transwoman. Journal of Sex Research, 50, 151-163.

23.

Rhodes

S. D.

Hergenrather

K. C.

Vissman

A. T.

Stowers

Davis

A. B.

Hannah

. . . Marsiglia

F. F.

(2011). Boys must be men, and men must have sex with women: A qualitative CBPR study to explore sexual risk among African American, Latino, and white gay men and MSM. American Journal of Men’s Health, 5, 140-151.

24.

Ross

M. W.

Rosser

B. R. S.

Neumaier

E. R.

, & The Positive Connections Team. (2008). The relationship of internalized homonegativity to unsafe sexual behavior in HIV seropositive men who have sex with men. AIDS Education and Prevention, 20, 547-557.

25.

Santana

M. C.

Raj

Decker

M. R.

La Marche

Silverman

(2006). Masculine gender roles associated with increased risk and intimate partner violence perpetration among young adult men. Journal of Urban Health, 83, 575-585.

26.

Schrimshaw

E. W.

Downing

M. J.

Siegel

(2013). Sexual venue selection and strategies for concealment of same-sex behavior among non-disclosing men who have sex with men and women. Journal of Homosexuality, 60(1).

27.

Siegel

Schrimshaw

E. W.

Lekas

H. M.

Parsons

J. T.

(2008). Sexual behaviors of non-gay identified non-disclosing men who have sex with men. Archives of Sexual Behavior, 37, 720-735.

28.

Stokes

J. P.

McKirnan

D. J.

Doll

L. S.

Burzette

R. G.

(1996). Female partners of bisexual men: What they don’t know might hurt them. Psychology of Women Quarterly, 20, 267-284.

29.

Stokes

J. P.

Vanable

McKirnan

D. J.

(1997). Comparing gay and bisexual men on sexual behavior, condom use, and psychosocial variables related to HIV/AIDS. Archives of Sexual Behavior, 26, 383-397.

30.

Wagner

G. J.

(1998). Internalized homophobia scale. In Davis

C. M.

Yarber

W. L.

Bauserman

Schreer

Davis

S. L.

(Eds.), Handbook of sexuality-related measures (pp. 371-372). Thousand Oaks, CA: Sage.

31.

Ward

E. G.

(2005). Homophobia, hypermasculinity and the US black church. Culture, Health & Sexuality, 7, 493-504.

32.

Wheeler

D. P.

Lauby

J. L.

Liu

Van Sluytman

L. G.

Murrill

(2008). A comparative analysis of sexual risk characteristics of black men who have sex with men or with men and women. Archives of Sexual Behavior, 37, 697-707.

33.

Wolfe

W. A.

(2003). Overlooked roles of African-American males’ hypermasculinity in the epidemic of unintended pregnancies and HIV/AIDS cases with young African-American women. Journal of the National Medical Association, 95, 846-852.