Archives of Sexual Behavior

, Volume 36, Issue 4, pp 569–578 | Cite as

Minority Stress and Sexual Problems among African-American Gay and Bisexual Men

Original Paper


Minority stress, such as racism and gay bashing, may be associated with sexual problems, but this notion has not been examined in the literature. African-American gay/bisexual men face a unique challenge in managing a double minority status, putting them at high risk for stress and sexual problems. This investigation examined ten predictors of sexual problems among 174 African-American gay/bisexual men. Covarying for age, a forward multiple regression analysis showed that the measures of self-esteem, male gender role stress, HIV prevention self-efficacy, and lifetime experiences with racial discrimination significantly added to the prediction of sexual problems. Gay bashing, psychiatric symptoms, low life satisfaction, and low social support were significantly correlated with sexual problems, but did not add to the prediction of sexual problems in the regression analysis. Mediation analyses showed that stress predicted psychiatric symptoms, which then predicted sexual problems. Sexual problems were not significantly related to HIV status, racial/ethnic identity, or gay identity. The findings from this study showed a relationship between experiences with racial and sexual discrimination and sexual problems while also providing support for mediation to illustrate how stress might cause sexual problems. Addressing minority stress in therapy may help minimize and treat sexual difficulties among minority gay/bisexual men.


African-American Black Minority Discrimination Gay Sexual dysfunction 


Few studies on sexual dysfunction have been conducted with gay/bisexual men or lesbian/bisexual women. The dearth of literature in this area may reflect a heterosexual bias in sex therapy research (Masters & Johnson, 1979; Sandfort & de Keizer, 2001). Other reasons might include assumed similarities between heterosexual and homosexual sexual functioning and various barriers to accessing health care for gay/bisexual individuals (e.g., fear of prejudice or discrimination; fear of social exposure during treatment).

As noted  in Sandfort and de Keizer's (2001) review, past studies of sexual dysfunction among gay/bisexual men lack strong empiricism, have variations in sample size and composition, and use varying methodological approaches (e.g., Everaerd et al., 1982; McWhirter & Mattison, 1984; Paff, 1985; Reece, 1982). Consequently, summarizing the literature is difficult. Similar to studies of heterosexual men (see Rosen, 2000), early studies suggested that erectile disorder was possibly the most common sexual dysfunction among gay/bisexual men (Bell & Weinberg, 1978; McWhirter & Mattison, 1984; Paff, 1985; Reece, 1982). A large, recent study supports the idea that erection difficulties are common among gay/bisexual men (Bancroft, Carnes, Janssen, Goodrich, & Long, 2005). A notable number of gay/bisexual men reported orgasmic disorder (i.e., delayed ejaculation) in the earlier studies, and this problem may be more common among gay/bisexual men than heterosexual men (Bell & Weinberg, 1978; McWhirter & Mattison, 1984; Paff, 1985; Reece, 1982). The measurement and reported frequency of premature ejaculation and hypoactive sexual desire varied with each study, but generally the percentages of gay/bisexual men complaining of these sexual dysfunctions were somewhat smaller relative to erection and delayed ejaculation difficulties. Premature ejaculation appears to be more common than low sexual desire. Two separate, more recent studies have highlighted the issue of painful receptive anal sex as a frequent sexual problem among gay/bisexual men (Rosser, Metz, Bockting, & Buroker, 1997; Rosser, Short, Thurmes, & Coleman, 1998).

In an early review paper, Reece (1987) noted that the etiology and treatment of sexual dysfunction among gay/bisexual men was akin to sex therapy with other men in many respects. For example, gay/bisexual men may experience communication difficulties, performance concerns, psychopathology, sexual knowledge deficits, and negative sexual attitudes that contribute to sexual dysfunction (Bancroft et al., 2005; McWhirter & Mattison, 1978; Paff, 1985; Reece, 1987). Potentially unique etiological factors for sexual dysfunction among gay/bisexual male clients may include concerns about one's gender role or identity, internalized homophobia, and concern about HIV prevention (McWhirter & Mattison, 1978; Paff, 1985; Reece, 1982, 1987; Rosser et al., 1998). These etiological factors have been hypothesized, but not examined. In one exception, Rosser et al. (1997) found that high levels of sexual satisfaction were significantly associated with less internalized homophobia, greater comfort with same-sex attraction, and less sexual dysfunction in a sample of 197 predominantly (94%) white gay/bisexual men. Bancroft et al. (2005) found that increases in age predicted greater concerns about erection difficulties and rapid ejaculation in a large sample of gay men (N=1, 196), but mood, trait anxiety, and depression were not related to these sexual concerns. Gender roles were not examined in either of these studies. In their comprehensive review, Sandfort and de Keizer (2001) suggested that more research was needed to examine risk factors for sexual problems among gay/bisexual men, with attention to gender roles and sexual identity development.

Minority stress, sexual problems, and African-American gay/bisexual men

Stress and trauma have been shown to predict sexual problems (Letourneau, Resnick, Kilpatrick, Saunders, & Best, 1996). Minority stress, such as racism and heterosexism, may be associated with sexual dysfunction, but this notion has not been examined in the literature. Brooks (1981) asserted that sexual minorities experience more stress than non-minorities due to various forms of discrimination. Research on minority stress among gay men has shown that discrimination and experiences of negative treatment in society are associated with more mental health problems (Meyer, 1995) and emotional distress (Clark, Anderson, Clark, & Williams, 1999; Ross, 1990). African-American gay/bisexual men face a unique challenge in managing a double minority status (i.e., facing racism and heterosexism), putting them at risk for negative life events (e.g., loss of employment, home, or custody of children) and chronic daily hassles due to discrimination (Brooks, 1981; Crawford, Allison, Zamboni, & Soto, 2002; Icard, 1986; Loiacano, 1993). In the literature on sexual dysfunction among gay/bisexual men, African-American gay/bisexual men have generally been neglected. Demographic details were not always specified by prior studies, but the samples were apparently predominantly Caucasian. A notable exception was Bell and Weinberg (1978), who studied 111 African-American and 575 Caucasian gay/bisexual men.

The notion that experiences of discrimination adversely affect sexual functioning seems intuitive, but the process of how this might occur has not been examined via any conceptual framework. Stress has been viewed as leading to psychiatric symptoms (Wheatley, 2000) and stress has been associated with sexual dysfunction (Letourneau et al., 1996). Thus, psychiatric symptoms may mediate the relationship between stress and sexual dysfunction. More experiences of discrimination based on sexual orientation and race would predict higher levels of psychiatric symptoms which, in turn, should predict higher levels of sexual problems among African-American gay/bisexual men.

Sexual dysfunction and HIV

The current study does not aim to provide a thorough review of sexual dysfunction and HIV-related issues, but no examination of sexual dysfunction among gay/bisexual men would be complete without addressing this topic. Sexual dysfunction has been shown to be associated with HIV infection. Gay/bisexual men who were HIV positive reported more sexual dysfunction than gay/bisexual men without HIV (e.g., Jones, Klimes, & Catalan, 1994). Erectile dysfunction and lack of sexual desire were some of the most common sexual dysfunctions reported among HIV positive gay/bisexual men in the limited research on this topic (Catalan & Meadows, 2000; Dupras & Morisset, 1993; Jones et al., 1994). Problems with orgasmic disorder (i.e., delayed ejaculation) and, to a lesser extent, premature ejaculation, have been reported with other samples of HIV positive gay men (Dupras & Morisset, 1993; Jones et al., 1994).

Psychological and organic factors have been examined as causes of sexual dysfunction among HIV positive gay/bisexual men. Psychological factors include reactions to being HIV positive (e.g., shame, feeling a loss of sexuality, fear of infecting others; Gochros, 1992). Dupras and Morisset (1993) found that gay/bisexual men with sexual dysfunction had greater fear of sexuality, greater sexual depression and anxiety, lower sexual self-esteem, and lower sexual satisfaction when compared to men without sexual dysfunction. Organic etiological factors include biological predispositions toward sexual dysfunction prior to HIV infection (e.g., low testosterone) and side effects from HIV treatment (Catalan & Meadows, 2000).


The few existing studies on sexual dysfunction among gay/bisexual men are largely descriptive, consist of small samples, and include mostly Caucasian samples. Rather than describing frequencies of sexual problems, the current study aimed to expand research in this area by examining predictors of sexual problems among a large sample of African-American gay/bisexual men. Understanding predictors of sexual difficulties will help highlight possible etiological or concomitant factors of sexual problems and provide guidance in determining appropriate intervention strategies.

Early research has cited gender role stress, internalized homophobia, and concern about HIV prevention as relevant etiological factors for sexual problems among gay/bisexual men (McWhirter & Mattison, 1978; Paff, 1985; Reece, 1982, 1987; see also Zilbergeld, 1992). African-American gay/bisexual men have been disproportionately affected by the HIV epidemic in the gay community (Centers for Disease Control and Prevention, 2001). Degree of racial/ethnic identity and degree of gay identity have been shown to be related to the psychosocial well-being of African-American gay/bisexual men (Crawford et al., 2002) and these variables may be related to their sexual well-being (Rosser et al., 1997).

Accordingly, this study included measures that assessed experiences with discrimination based on racial/ethnic background and sexual orientation, measures of racial/ethnic identity and gay identity, and scales of male gender role stress and HIV prevention self-efficacy. Measures of self-esteem, life satisfaction, and psychiatric symptoms were included to assess overall psychosocial functioning. A measure of social support was included to test the idea that social support buffers the effect of stress on sexual problems.


  1. 1.

    Based on the previous literature, African-American gay/bisexual men who were HIV positive were hypothesized to report higher levels of sexual problems than men who were HIV negative.

  2. 2.

    Higher levels of sexual problems were hypothesized to be associated with lower levels of life satisfaction, self-esteem, self-efficacy for HIV prevention, racial/ethnic identity, and gay identity, as well as greater levels of male gender role stress, more psychiatric symptoms, and greater levels of experience with discrimination based on race and sexual orientation.

  3. 3.

    Psychiatric symptoms were hypothesized to mediate the relationship between minority stress variables and sexual problems (e.g., more racial discrimination predicts greater psychiatric symptoms which, in turn, predicts higher levels of sexual problems).

  4. 4.

    Social support was hypothesized to moderate the relationship between stress-related variables (e.g., racial discrimination) and sexual problems (i.e., higher levels of racial discrimination should predict more sexual problems, except when high levels of social support exist).




The data from this study were taken from a larger study investigating the social and sexual behaviors of African-American gay/bisexual men. The study was approved by a university IRB, and an informed consent procedure was followed. Participants were recruited in Chicago, Illinois and Richmond, Virginia through print advertisements in weekly newspapers, outreach by recruiters in community settings (e.g., cafes or coffeeshops, street fairs, parks, bars), and snowball sampling techniques (i.e., referrals from men who participated in the study). Prior studies sponsored by the National Institutes of Health and the Centers for Disease Control and Prevention have successfully used these methods (e.g., Stokes, Damon, & McKirnan, 1997; Stokes, Vanable, & McKirnan, 1996).

Print advertisements and field workers invited men to call a local university about a confidential study concerning the social and sexual experiences of “African-American men who have sex with men.” The advertisement stated that the study would last about 60 min and that volunteers would be paid $20. Men who called were screened for eligibility (i.e., they had to be 18 years or older, identify as engaging in sexual activity with other men, have completed at least 8 years of formal education, and be in fair or good medical condition). Individuals who met these criteria were given an appointment at the local university. Participants were greeted by a research assistant and escorted to a testing room. They were told that all information gathered for the study would be held in strict confidence and that they could discontinue participation at any time. The research assistant checked each survey for completion. Four men were excluded from the study because of incomplete or unreliable data. Of the 174 participants (M age=35 years), most identified themselves as gay (71%), Baptist (40%), and single (63%). Fifty-one percent were HIV negative and 68% had never received alcohol or drug treatment. Table 1 presents a summary of the demographic characteristics of the participants.
Table 1





Highest Level of Education

 High school diploma or less



 1 year or more of college



 Junior College degree



 4-year college degree



 Professional degree



Type of Employment

 Executive, doctor, lawyer



 Manager/owner large business






 Clerical, retail, technical



 Semi-skilled laborer



 Unskilled laborer



Self-defined Sexual Orientation




 Bisexual/mostly homosexual



 Bisexual/equally homosexual/heterosexual



 Bisexual/mostly heterosexual






 Uncertain/Don't Know



Living Arrangement




 With Partner/Lover



 Same-Sex roommate



 Opposite-Sex roommate



 One or Both Parents






Religious Preference



















 Other (unspecified)



HIV Status







 Don't Know




After completing a basic demographics form, participants were asked to complete several scales. The Coopersmith Self-Esteem Inventory (CSEI; Coopersmith, 1981) is a 25-item questionnaire designed to measure evaluative attitudes toward the self in social, academic, family, and personal areas of experience. Participants were asked to respond “like me” or “unlike me” to each item. Internal consistency estimates have ranged from .87 to .92 (Maloney, Cheney, Spring, & Kanusky, 1986). The number of self-esteem items answered positively was summed. The following is an item from the CSEI: “Things usually don't bother me.” Higher scores of the CSEI indicate more self-esteem. Cronbach's alpha was .80 in this study.

The Symptom Checklist 90-Revised (SCL-90-R; Derogatis, 1993) is a 90-item self-report symptom inventory designed to assess various psychiatric symptoms. The Global Severity of Index scale score, a measure of overall psychological distress, was used in this study. Each item was rated on a 5-point Likert-type scale (0=not at all to 4=extremely). Higher scores reflect greater levels of psychiatric symptoms. Internal consistency estimates have ranged from .78 to .90 and test-retest reliability estimates have ranged from .78 to .90 (Derogatis, 1993). Cronbach's alpha was .90 in this study.

The Sexual Problems Scale (SPS) is a 12-item instrument designed specifically for this study to assess sexual problems in an individual's current sex life. To minimize heterosexual biases, sexual dysfunction was conceptualized as traditional sexual disorders as well as sexual problems important to gay men (Sandfort & de Keizer, 2001). Thus, items on the scale covered a variety of sexual problems, including insufficient frequency of sex, responding to and receiving sexual requests, finding a suitable sex partner, maintaining affection for a sex partner, being good enough sexually, and engaging in activities that seem wrong or sinful based on one's upbringing. The SPS also included concerns about sexual desire, arousal, premature ejaculation, and lack of orgasm (for self and partner). Participants were asked, “To what extent are the following problems for you in your sex life?” Ratings were based on a Likert-type scale (from 1=not at all to 4=very much). Higher scores indicate greater levels of sexual problems. Cronbach's alpha was .77 in this study.

The Life Satisfaction Scale (LSS; Bryant & Veroff, 1984) is a 7-item scale that provides a measure of subjective mental health. The scale yields a single, global satisfaction score, although several aspects of individual functioning were assessed (e.g., leisure, work/school, home, romantic relationships, etc.) Factor loadings for items on the LSS were high, and internal consistency measures lie consistently above .90 across several studies (Bryant & Veroff, 1984). The items were rated on a Likert-type scale in which 1=no satisfaction and 4=great satisfaction. The following is an item from the LSS: “How much satisfaction have you gotten out of your work on the job or school?” Higher scores indicate greater life satisfaction. Cronbach's alpha equaled .73 in this study.

The HIV Prevention Self-Efficacy Scale (HPSES; Smith, McGraw, Costa, & McKinley, 1996) is a 9-item measure that assesses the degree to which an individual feels he can engage in HIV preventive behaviors. Ratings were based on a Likert-type scale (1=not at all sure; 5=very sure). An example item from the HPSES is, “How sure were you that you can…buy condoms at a drug store?” This measure has demonstrated adequate internal reliability (Cronbach's alpha=.68 to .78) and validity (Smith et al., 1996). High scores on the HPSES indicate a greater level of self-efficacy in preventing HIV infection. Cronbach's alpha was .73 in this study.

The Male Gender Role Stress (MGRS; Eisler & Skidmore, 1987) is a 40-item self-report scale designed to measure the level of male gender role stress an individual experiences in his day-to-day existence. Participants were asked to rate each item in terms of the impact it has on them. Items were rated on a 7-point Likert-type scale (0=not stressful; 6=extremely stressful). The MGRS provides a total score and five subscale scores, with higher scores reflecting more male gender role stress. For the purpose of this study, only total scores on the MGRS were computed. The following is an item from the MGRS: “Please rate the following according to how stressful the situation would be for you—telling your significant other that you love him/her.” The MGRS has adequate reliability (Cronbach's α=.90) and validity (Copenhaver, Lash, & Eisler, 2000; Eisler, Skidmore, & Ward, 1988). Cronbach's alpha was .93 in this study.

The Schedule of Racist Events (SRE; Landrine & Klonoff, 1996) is an 18-item self-report inventory that assesses the frequency of racist discrimination an individual has experienced in the past year, in one's lifetime, and the extent to which this discrimination was evaluated as stressful. Each of the 18 items was completed 3 times (on a 6-point scale, 1=not at all and 6=extremely) with reference to the aforementioned constructs. Landrine and Klonoff (1996) reported the following reliability coefficients for the three subscales: .95 (recent racist events), .95 (lifetime racist events), and .94 (appraised racist events). Higher scores indicate more frequent encounters with racism and experiencing these events as stressful. The following is an item from the SRE: “How many times have you been treated unfairly by your employers, bosses, or supervisors because you were Black?” Cronbach's alpha was .95 in this study.

The Gay Bashing Scale (GBS) is a 9-item scale developed specifically for this study and assesses lifetime discrimination based on sexual orientation. Participants were asked how often they had experienced 9 events (e.g., verbal insults) because someone knew or assumed the individual was gay or bisexual. A Likert-type scale (0=never; 4=more than three times) response format was used. Higher scores of the GBS indicate more frequent encounters of discrimination based on sexual orientation. Cronbach's alpha was .86 in this study.

The Social Support Questionnaire (SSQ; Sarason, Levine, Basham, & Sarason, 1982) is a 15-item abbreviated version of the original questionnaire that asks participants to list the people to whom they can turn for support in various situations, and to indicate their perceived level of satisfaction with the social support received in each case. The measure yields two scores: the average number of persons listed as supportive and the average degree of satisfaction (1=very dissatisfied to 6=very satisfied) with received social support: “Whom can you count on to be dependable when you need help? How satisfied?” Sarason et al. (1982) reported a reliability index of .94 for the SSQ. Higher scores indicate more social supports and more satisfaction with the support the individual receives. The degree of social support satisfaction was the only index utilized in this study. Cronbach's alpha was .95 in this study.

The Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992) was employed to assess ethnic identity. The 21-item MEIM consists of four scales that assess ethnic identity: Affirmation and belonging, Ethnic Identity Achievement, Ethnic Behaviors, and other-group orientation. Participants were asked to indicate their level of agreement with a series of statements (1=strongly disagree to 5=strongly agree). An example item is “I am active in organizations or groups that include mostly members of my own ethnic group.” The overall MEIM has shown good internal reliability (Cronbach's alpha=.90; Cuellar, Nyberg, Maldonado, & Roberts, 1997). Cronbach's alpha was .85 in this study.

The Gay Identity Scale (GIS) is a 15-item scale developed for this study based on prior research (see Waldo, Hesson-McInnis, & D-Augelli, 1998). Participants were asked to rate a series of statements using a Likert-type response format (1=not all comfortable to 4=very comfortable). Items on the GIS address positive gay/lesbian attitudes, disclosure of one's gay identity, and participation in gay/lesbian organizations. An example item is “How comfortable do you feel disclosing your sexual orientation to the majority of the people in your community?” Higher scores reflect a more positive adjustment to one's sexual identity. Cronbach's alpha was .81 in this study.


Distributions of the data for each measure were examined to detect skewness that may violate statistical assumptions. The data were normally distributed for all measures. To control for Type I error, the level for significance was set at .01 for all analyses. When grouped by region of recruitment (Richmond vs. Chicago), there were no significant differences between groups on any demographic variables. There were no significant relationships between sexual problems and the demographic variables, except for a significant correlation with age (r= .28, p < .01). Thus, age was covaried in subsequent analyses. Means and SDs for each measure are listed in Table 2. According to the SCL-90-R manual (Derogatis, 1993), the mean Global Severity Index score in this study was more than one SD above the mean, indicating a notable level of psychiatric symptoms in this sample.
Table 2

Descriptive statistics and correlations with the Sexual Problems Scale





r with SPS


Sexual Problems










Psychiatric Symptoms






Life Satisfaction






HIV Prevention Self-Efficacy






Male Gender Role Stress






Lifetime Racism






Gay Bashing






Social Support






Ethnic Identity






Gay Identity






Note: **p < .01. Range refers to the possible range of scores. SPS=Sexual Problems Scale. Higher scores on each scale reflect higher levels of the construct being measured.

To test Hypothesis 1, a one-way analysis of covariance (ANCOVA) was conducted to determine possible differences in level of sexual problems based on HIV status, covarying for age. Participants who did not know their HIV status (N=18) were excluded from this analysis. There was no statistically significant difference in levels of sexual problems between individuals who were HIV positive (N=68) versus individuals who were HIV negative (N=89), F(1, 153)=.29, p=.59, ES=.01.

Correlations and forward multiple regression

Pearson correlation coefficients were computed between sexual problems and the other measures in this study to examine Hypothesis 2. Of the 10 correlations, 8 were statistically significant. These correlations are displayed in Table 2. As the table indicates, the correlations were generally significant, but modest in magnitude. All of the significant correlations were in the hypothesized direction at the .01 level. Higher levels of sexual problems were significantly associated with lower levels of self-esteem, life satisfaction, and HIV prevention self-efficacy. Higher levels of social support were associated with lower levels of sexual problems. In addition, higher levels of sexual problems were significantly correlated with more experiences of lifetime racial discrimination and gay bashing, greater levels of psychiatric symptoms, and greater male gender role stress. In contrast to these results, sexual problems were not related to level of gay identity or level of racial/ethnic identity.

A forward multiple regression analysis was conducted in which the independent variables with prior significant correlations were entered into the equation with age as a covariate. Accordingly, the predictors in the regression included self-esteem, life satisfaction, HIV prevention self-efficacy, social support, lifetime racial discrimination, gay bashing, psychiatric symptoms, and male gender role stress. The sexual problems scale was the dependent variable. The results of this analysis are shown in Table 3.
Table 3

Summary of forward multiple regression of variables predicting sexual problems



















Male Gender Role Stress






HIV Prevention Self-Efficacy






Lifetime Racism






Life Satisfaction






Gay Bashing






Social Support






Psychiatric Symptoms






Note. R2=total variance explained by the predictor and any preceding predictors. The p value shows the significance of the change in R2 with each added predictor.

Self-esteem was the strongest predictor of sexual problems, accounting for 14% of the variance in sexual problems, F(2, 163)=22.97, p < .001, ES=.15. Male gender role stress, HIV prevention self-efficacy, and lifetime racial discrimination were additional significant predictors, in that order. Each independent variable accounted for a significant amount of additional variance in overall sexual problems beyond self-esteem, although the change in R2 for HIV prevention self-efficacy was marginal (p=.021). With age as a covariate, these four predictors explained 32.8% of the variance in overall sexual problems, F(5, 160)=15.63, p < .001, ES=.04. The change in R2 was not significant for any of the subsequent predictors, indicating that no other independent variable accounted for additional variance above and beyond the other predictors (see Table 3).

Mediation and moderation analyses

Multiple regression was used to explore the hypothesis that psychiatric symptoms would mediate the relationship between lifetime racial discrimination and sexual problems. To examine mediation via regression, each pair of variables must have a statistically significant relationship (Baron & Kenny, 1986; Holmbeck, 1997). The simple bivariate correlation between psychiatric symptoms and lifetime racial discrimination was significant (r=.27, p < .01, df=173), and Table 2 shows that the other two correlations were significant. The latter two relationships can also be shown via regression analyses with age as a covariate. Racial discrimination significantly predicted sexual problems, F(2, 169)=13.13, p < .001, and psychiatric symptoms significantly predicted sexual problems, F(2, 169)=19.68, p < .001. In the final step, for full mediation, lifetime racial discrimination should no longer predict sexual problems when psychiatric symptoms were covaried with age. This was the case (p=.042), and there was a drop in the beta-weight for lifetime racial discrimination from Step 1 (.24) to Step 4 (.16). Thus, lifetime experiences with discrimination predicted psychiatric symptoms which, in turn, predicted sexual problems, and psychiatric symptoms mediated the relationship. This mediation analysis is illustrated in Fig. 1.
Fig. 1

Illustration of Statistically Significant Mediation Analyses

A second analysis examined psychiatric symptoms as a mediator of gay bashing and sexual problems. The simple bivariate correlation between psychiatric symptoms and gay bashing was significant (r=.30, p < .01, df=173), and Table 2 shows that the other two correlations were significant. Gay bashing significantly predicted sexual problems with age as a covariate, F(2, 170)=10.79, p=.001, and psychiatric symptoms were already shown to be significantly associated with sexual problems with age as a covariate. When the psychiatric symptoms variable was covaried with age, gay bashing was no longer a significant predictor of sexual problems (p=.174). Thus, full mediation was demonstrated. Experiences of gay bashing predicted psychiatric symptoms which, in turn, predicted sexual problems, and psychiatric symptoms fully mediated the relationship. Figure 1 illustrates this analysis of mediation.

The final analysis of mediation examined psychiatric symptoms as a mediator of male gender role stress and sexual problems. The simple bivariate correlation between psychiatric symptoms and male gender role stress was significant (r=.44, p < .01, df=172), and Table 2 shows that the other two correlations were significant. Male gender role stress significantly predicted sexual problems with age as a covariate, F(2, 168)=21.97, p < .001. Male gender role stress still predicted sexual problems when psychiatric symptoms were covaried (p=.001). The drop in the beta-weight for male gender role stress from Step 1 (.37) to Step 4 (.27) and the Sobel test indicated a partial mediation effect (Baron & Kenny, 1986). Male gender role stress predicted psychiatric symptoms which, in turn, predicted sexual problems, but psychiatric symptoms only partially mediated the relationship.

Additional regression analyses were conducted to determine if social support moderated the relationship between stress-related variables (i.e., racial discrimination, gay-bashing, and male gender role stress) and sexual problems. This hypothesis was tested by conducting a series of regression analyses in which each moderator-predictor interaction term was an independent predictor entered after the moderator and the stress-related main effects (Baron & Kenny, 1986; Holmbeck, 1997). None of the analyses examining social support as a moderator were significant.


Causal inferences cannot be made based on this study, but the results suggest possible antecedents of sexual problems. Racial discrimination, gay bashing, and male gender role stress may increase the chances that sexual problems develop among African-American gay/bisexual men or these factors may occur in conjunction with other life stressors faced by these individuals. Although the current study may be the first to show statistically significant relationships among these variables, the findings fit with recent research that emphasizes the importance of mental health and cultural factors among minority groups when examining the etiology of sexual dysfunction (DeFronzo Dobkin, Leiblum, Rosen, Menza, & Marin, 2006).

The mediational analyses suggested that perceived experiences with discrimination based on race/ethnicity and sexual orientation create stress as evidenced by psychiatric symptoms which, in turn, adversely affect sexual functioning. Male gender role stress was associated with sexual functioning in a similar fashion. Because full mediation was not evident in that analysis, there were clearly other variables unspecified by this study that predict both psychiatric symptoms and also sexual problems. For example, problems in one's romantic relationship may predict sexual problems or psychiatric symptoms (Heiman, 2002). Moreover, social support did not moderate the relationship between stress-related variables and sexual problems. These results indicate that the etiology of sexual problems cannot be oversimplified. Despite the significance of the current study, any given predictor may not have a direct effect on sexual problems. Addressing minority stress in the treatment of sexual problems among African-American gay/bisexual men may be important in therapy, but not sufficient for the amelioration of any sexual difficulties they experience. Other risk factors must be identified so that appropriate preventative and therapeutic measures can be taken.

It is important to note that experiencing sexual problems could lead to psychiatric symptoms (e.g., a man might feel depressed in response to erection difficulties), rather than psychiatric symptoms functioning merely as an antecedent to sexual problems. Analyses of mediation must be defined by a specific conceptual framework and corresponding hypotheses (Baron & Kenny, 1986; Holmbeck, 1997). Although the current study followed these tenants of inquiry, clearly the relationship between sexual problems and psychiatric symptoms can be more complex. Reverse causal effects could be explored in future studies (Smith, 1982) or structural equation modeling (Holmbeck, 1997) could be used to illustrate a more complex model of relationships between sexual problems, psychiatric symptoms, and other variables.

Sexual problems were not related to level of racial/ethnic and gay identity, possibly because these constructs were not proximal influences on an African American's sexual functioning. Extensive psychometric validation of the gay identity measure used in this study has not been undertaken. Thus, the gay identity instrument may lack sufficient construct validity and therefore it could mask a possible existing relationship between gay identity and sexual problems.

There was not a significant difference in sexual problems between participant groups based on HIV status (positive versus negative). This was surprising, given that HIV positive gay/bisexual men have reported more sexual dysfunction than gay/bisexual men without HIV (e.g., Jones et al., 1994). This finding suggests one's HIV status is not necessarily associated with disturbances in sexual functioning. Support for individuals who were HIV positive has increased over time, and more people were living longer with HIV due to advanced treatments for the infection, possibly leading to reduced concerns about HIV infection (Vanable, Ostrow, McKirnan, Taywaditep, & Hope, 2000) and decreasing its impact on sexual functioning. The current study did not assess how long each individual had been HIV positive. It was possible that more sexual problems were evident among individuals who have been recently diagnosed as HIV positive. Participants in the current study who were HIV positive may have had sufficient time to adjust to their HIV status, eroding any relationship between HIV status and sexual problems. Future studies should include a thorough assessment of a person's health history and current health status when examining HIV status and sexual functioning. Older gay/bisexual men may be more likely to experience sexual problems (Bancroft et al., 2005) because of hypertension or other health conditions, regardless of when they became HIV positive. Future research might also assess HIV-related stress or possible feelings of stigmatization associated with HIV infection to see if these factors are associated with sexual problems.

The findings of this study may be limited by a combination of participants' self-serving biases, demand characteristics, and the artificial nature of the research setting. The participants were urban African-American gay/bisexual men who were comfortable enough to complete a survey at a major university in their city. Many participants may have been motivated by the monetary compensation. In short, the participants may not be representative of all African-American men who have sex with men. Finally, the data were self-reported and therefore subject to retrospective bias and common method error.

Despite these limitations, the current investigation provides an important empirical investigation into a neglected topic. Sexual difficulties among African-American gay/bisexual men occur in the context of the minority stress they experience in their lives. Researchers and clinicians alike need to consider such factors that affect the psychosocial well-being of marginalized groups. A sex therapist who focuses upon traditional schools of thought when conducting sex therapy (e.g., systems theory or sensate focus exercises) with an African-American gay/bisexual man may neglect to consider how discrimination and other sociocultural factors may relate to his sexual difficulties. The incidence of sexual problems among African American gay/bisexual persons may be related to the stress they experience as a member of two minority groups that are often stigmatized in the U.S. (i.e., racial-ethnic and sexual minority groups). Addressing such stress and promoting a positive self-esteem may help minimize sexual difficulties among minority gay/bisexual men.


  1. Bancroft, J., Carnes, L., Janssen, E., Goodrich, D., & Long, J. S. (2005). Erectile and ejaculatory problems in gay and heterosexual men. Archives of Sexual Behavior, 34, 285297.PubMedCrossRefGoogle Scholar
  2. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.PubMedCrossRefGoogle Scholar
  3. Bell, A. P., & Weinberg, M. S. (1978). Homosexualities: A study of diversity among men & women. New York: Simon and Schuster.Google Scholar
  4. Brooks, V. R. (1981). Minority stress and lesbian women. Lexington, MA: D. C. Heath.Google Scholar
  5. Bryant, F. B., & Veroff, J. (1984). Dimensions of subjective mental health in American men and women. Journal of Health and Social Behavior, 25, 116135.PubMedCrossRefGoogle Scholar
  6. Catalan, J., & Meadows, J. (2000). Sexual dysfunction in gay and bisexual men with HIV infection: Evaluation, treatment and implications. AIDS Care, 12, 279286.PubMedCrossRefGoogle Scholar
  7. Centers for Disease Control and Prevention. (2001, June 1). HIV incidence among young men who have sex with men–Seven U.S. cities, 1994–2000. Morbidity and Mortality Weekly Report, 50, 440444.Google Scholar
  8. Clark, R., Anderson, N. B., Clark, V. R., & Williams, D. R. (1999). Racism as a stressor for African Americans: A biopsychosocial model. American Psychologist, 54, 805–816.PubMedCrossRefGoogle Scholar
  9. Coopenhaver, M. M., Lash, S. J., & Eisler, R. M. (2000). Masculine gender-role stress, anger, and male intimate abusiveness: Implications for men's relationships. Sex Roles, 42, 405414.CrossRefGoogle Scholar
  10. Coopersmith, S. (1981). Self-esteem inventories. Palo Alto, CA: Consulting Psychologists Press.Google Scholar
  11. Crawford, I., Allison, K. W., Zamboni, B. D., & Soto, T. (2002). The influence of dual-identity development on the psychosocial functioning of African-American gay and bisexual men. Journal of Sex Research, 39, 179189.PubMedGoogle Scholar
  12. Cuellar, I., Nyberg, B., Maldonado, R. E., & Roberts, R. E. (1997). Ethnic identity and acculturation in a young adult Mexican-origin population. Journal of Community Psychology, 25, 535549.CrossRefGoogle Scholar
  13. DeFronzo Dobkin, R., Leiblum, S. R., Rosen, R. C., Menza, M., & Marin, H. (2006). Depression and sexual functioning in minority women: Current status and future directions.Journal of Sex & Marital Therapy, 32, 23–36.CrossRefGoogle Scholar
  14. Derogatis, L. R. (1993). Symptom checklist-90-R. Minneapolis: National Computer Systems, Inc.Google Scholar
  15. Dupras, A., & Morisset, R. (1993). Sexual dysfunction among HIV-positive gay males. Sexual & Marital Therapy, 8, 3746.Google Scholar
  16. Eisler, R. M., & Skidmore, J. R. (1987). Masculine gender role stress: Scale development and component factors in the appraisal of stressful situations. Behavior Modification, 11, 123136.PubMedCrossRefGoogle Scholar
  17. Eisler, R. M., Skidmore, J. R., & Ward, C. H. (1988). Masculine gender-role stress: Predictor of anger, anxiety, and health-risk behaviors. Journal of Personality Assessment, 52, 133141.PubMedCrossRefGoogle Scholar
  18. Everaerd, W., Dekker, J., Dronkers, J., van der Rhee, K., Staffeleu, J., & Wiselius, G. (1982). Treatment of homosexual and heterosexual sexual dysfunction in male-only groups of mixed sexual orientation. Archives of Sexual Behavior, 11, 110.PubMedCrossRefGoogle Scholar
  19. Gochros, H. L. (1992). The sexuality of gay men with HIV infection. Social Work, 37, 105109.PubMedGoogle Scholar
  20. Heiman, J. R. (2002). Sexual dysfunction: Overview of prevalence, etiological factors, and treatments. Journal of Sex Research, 39, 73–78.PubMedGoogle Scholar
  21. Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599610.PubMedCrossRefGoogle Scholar
  22. Icard, L. (1986). Black gay men and conflicting social identities: Sexual orientation versus racial identity. Journal of Social Work & Human Sexuality, 4, 83–93.Google Scholar
  23. Jones, M., Klimes, I., & Catalan, J. (1994). Psychosexual problems in people with HIV infection: Controlled study of gay men and men with haemophilia. AIDS Care, 6, 587593.PubMedGoogle Scholar
  24. Landrine, H., & Klonoff, E. A. (1996). The schedule of racists events: A measure of racial discrimination and a study of its negative physical and mental health consequences. Journal of Black Psychology, 22, 144168.CrossRefGoogle Scholar
  25. Letourneau, E. J., Resnick, H. S., Kilpatrick, D. G., Saunders, B. E., & Best, C. L. (1996). Comorbidity of sexual problems and posttraumatic stress disorder in female crime victims. Behavior Therapy, 27, 321–336.CrossRefGoogle Scholar
  26. Loiacano, D. K. (1993). Gay identity issues among Black Americans: Racism, homophobia, and the need for validation. In L. Garnets & D. Kimmel (Eds.), Psychological perspectives on lesbian and gay male experiences (pp. 364–375). New York: Columbia University Press.Google Scholar
  27. Maloney, J. P., Cheney, R., Spring, W., & Kanusky, J. (1986). The physiologic and psychological effects of a 5-week and a 16-week physical fitness program. Military Medicine, 151, 426433.PubMedGoogle Scholar
  28. Masters, W. H., & Johnson, V. E. (1979). Homosexuality in perspective. Boston: Little, Brown.Google Scholar
  29. McWhirter, D., & Mattison, A. (1978). The treatment of sexual dysfunction among gay male couples. Journal of Sex & Marital Therapy, 4, 213218.Google Scholar
  30. McWhirter, D., & Mattison, A. (1984). The male couple: How relationships develop. Engelwood Cliffs, NJ: Prentice Hall.Google Scholar
  31. Meyer, I. H. (1995). Minority stress and mental health in gay men. Journal of Health and Social Behavior, 36, 38–56.PubMedCrossRefGoogle Scholar
  32. Paff, B. A. (1985). Sexual dysfunction in gay men requesting treatment. Journal of Sex & Marital Therapy, 11, 318.Google Scholar
  33. Phinney, J. S. (1992). The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7, 156176.CrossRefGoogle Scholar
  34. Reece, R. (1982). Group treatment of sexual dysfunction in gay men. Journal of Homosexuality, 7(2–3), 113–129.CrossRefGoogle Scholar
  35. Reece, R. (1987). Causes and treatments of sexual desire discrepancies in male couples. Journal of Homosexuality, 14(1–2), 157–172.PubMedCrossRefGoogle Scholar
  36. Rosen, R. C. (2000). Medical and psychological interventions for erectile dysfunction. In S. R. Leiblum & R. C. Rosen (Eds.), Principles and practice of sex therapy (pp. 276–304). New York: Guilford Press.Google Scholar
  37. Ross, M. W. (1990). The relationship between life events and mental health in homosexual men. Journal of Clinical Psychology, 46, 402411.PubMedCrossRefGoogle Scholar
  38. Rosser, B. R. S., Metz, M. E., Bockting, W. O., & Buroker, T. (1997). Sexual difficulties, concerns, and satisfaction in homosexual men: An empirical study with implications for HIV prevention. Journal of Sex & Marital Therapy, 23, 6173.Google Scholar
  39. Rosser, B. R. S., Short, B. J., Thurmes, P. J., & Coleman, E. (1998). Anodyspareunia, the unacknowledged sexual dysfunction: A validation study of painful receptive anal intercourse and its psychosexual concomitants in homosexual men. Journal of Sex & Marital Therapy, 24, 281292.CrossRefGoogle Scholar
  40. Sandfort, T. G. M., & de Keizer, M. (2001). Sexual problems in gay men: An overview of empirical research. Annual Review of Sex Research, 12, 93120.PubMedGoogle Scholar
  41. Sarason, I. G., Levine, H. M., Basham, R. B., & Sarason, B. R. (1992). Assessing social support: The social support questionnaire. Journal of Personality and Social Psychology, 44, 127139.CrossRefGoogle Scholar
  42. Smith, E. (1982). Beliefs, attributions, and evaluations: Nonhierarchical models of mediation in social cognition. Journal of Personality and Social Psychology, 43, 248259.CrossRefGoogle Scholar
  43. Smith, K. W., McGraw, S. A., Costa, L. A., & McKinley, J. B. (1996). A self-efficacy scale for HIV risk behaviors: Development and evaluation. AIDS Education and Prevention, 8, 97105.PubMedGoogle Scholar
  44. Stokes, J. P., Damon, W., & McKirnan, D. J. (1997). Predictors of movement toward homosexuality: A longitudinal study of bisexual men. Journal of Sex Research, 34, 304312.Google Scholar
  45. Stokes, J. P., Vanable, P. A., McKirnan, D. J. (1996). Ethnic differences in sexual behavior, condom use, and psychosocial variables among black and white men who have sex with men. Journal of Sex Research, 33, 373381.CrossRefGoogle Scholar
  46. Vanable, P. A., Ostrow, D. G., McKirnan, D. J., Taywaditep, K. J., & Hope, B. A. (2000). Impact of contribution therapies on HIV risk perceptions and sexual risk among HIV-positive and HIV-negative gay and bisexual men. Health Psychology, 19, 112.CrossRefGoogle Scholar
  47. Waldo, C., Hesson-McInnis, M., & D’Augelli, A. R. (1998). Antecedents and consequences of victimization of lesbian, gay, and bisexual young people: A structural model comparing rural university and urban samples. American Journal of Community Psychology, 26, 307334.PubMedCrossRefGoogle Scholar
  48. Wheatley, D. (2000). The role of stress in psychiatric disorders. Primary Care Psychiatry, 6, 93–100.CrossRefGoogle Scholar
  49. Zilbergeld, B. (1992). The man behind the broken penis: Social and psychological determinants of erectile failure. In R. C. Rosen & S. R. Leiblum (Eds.), Erectile disorders: Assessment and treatment (pp. 27–51). New York: Guilford Press.Google Scholar

Copyright information

© Springer Science&#x002B;Business Media, Inc. 2006

Authors and Affiliations

  1. 1.Program in Human Sexuality, Department of Family Medicine and Community HealthUniversity of MinnesotaMinneapolisUSA
  2. 2.Department of PsychologyLoyola University ChicagoChicagoUSA

Personalised recommendations