Correlates of Risk Patterns and Race/Ethnicity Among HIV-Positive Men Who Have Sex With Men

Ann O'Leary; Holly H. Fisher; David W. Purcell; Pilgrim S. Spikes; Cynthia A. Gomez


AIDS and Behavior. 2007;11(5):706-715. 

In This Article


Socio-demographics and Sexual Orientation

Table 1 presents social and demographic characteristics of the sample. Significant differences across race/ethnicity groups were observed for all socio-demographic variables included in these analyses. Relative to all other groups, European Americans were significantly older, F(3, 450) = 9.62, P < .001, and more of them reported making an annual income of $20,000 or higher, X 2 = 25.50 (df = 6), P < .001. European Americans also reported knowing their HIV status significantly longer than African Americans, F(3, 441) = 2.71, P = .05, and they were significantly more likely to have a bachelor's degree or higher, compared to African Americans and Latinos, X 2 = 25.56 (df = 0), P < .01. African Americans were significantly more likely to report being Protestant, compared to all other groups, and Latinos were significantly more likely to report being Catholic than all other groups, X 2 = 111.80 (df = 9), P < .001. Compared to all other groups, African Americans were significantly more likely to report religion/spirituality was very important to them, X 2 = 25.62 (df = 3), P < .001.

Since there were fewer than five men in some of the sexual orientation cells within the race/ethnic groups, the three non-gay categories were combined to represent one category. A chi-square test was conducted to measure the association between race/ethnicity and sexual orientation (gay vs. non-gay). African Americans were significantly less likely to self-identify as gay compared to all other groups and Latinos were significantly less likely than European Americans to self-identify as gay, X 2 = 28.77 (df = 3), P < .001.

Psychosocial Characteristics and Sexual Risk Behavior

As can be seen in Table 2 , significant effects of race/ethnicity were observed for many of the psychosocial and sexual risk variables. African American men expressed significantly higher levels of internalized homophobia than European Americans and men in the Other racial group, F(3, 448) = 5.70, P < .01. European Americans reported a higher level of sexual sensation-seeking than African Americans and Hispanics/Latinos, F(3, 447) = 4.94, P < .01. Regarding disclosure of HIV status, African Americans reported significantly lower self-efficacy to disclose their HIV-positive status to sex partners, relative to European Americans and men in the Other group, F(3, 446) = 4.08, P < .01.

Differences in HIV transmission risk behavior across groups are also evident in Table 2 . Of the 291 men who reported drinking alcohol before or during sex (past 3 months), a significantly greater proportion of African American MSM (69%) reported drinking alcohol at least occasionally before/during sex, compared to European Americans (51%) and Latinos (48%), X 2 = 8.65 (df = 3), P = .03. Significantly more African Americans reported having sex with a woman in the past year, relative to all of the other racial groups, X 2 = 27.57 (df = 3), P < .001 (22% of African Americans versus 4% of European Americans, 6% of Latinos, and 7% of men in Other group). African Americans also reported having sex with significantly more women than any other group, roughly four times the number of women as any of the other groups, F(3, 443) = 8.00, P < .001. Finally, only five men in this sample reported their primary partner was a woman. Although these numbers are too small for statistical comparison, it is noted that four of these five men were African American and the fifth man was Latino.

Participants also reported sexual practices with men ( Table 2 ). Regarding sex with HIV negative/HIV unknown status men, European Americans were significantly more likely to report having unprotected receptive oral sex than African Americans, X 2 = 10.27 (df = 3), P = .02, and they were more likely to report unprotected insertive oral sex than African Americans and Other men, X 2 = 7.73 (df = 3), P = .05. In addition, African American MSM reported having sex with significantly fewer men in the previous year than European American MSM, F(3, 445) = 3.81, P = .01. There were no effects of race/ethnicity on unprotected anal sex, or overall on having any type of unprotected sex, with men of HIV negative or unknown status.

Openness about Sexual Orientation with Others and in the Community

Significant differences across groups were observed for some, but not all, of the items assessing community participation and comfort level discussing homosexual relationships ( Table 3 ). There were no significant group differences for participation in gay-related activities during the past year. However, African Americans reported greater participation in HIV-related activities in the community compared to European Americans, F(3, 450) = 2.79, P = .04. Also, African American and Latino MSM reported greater comfort levels for talking with their mothers about their sexual relationships with men, compared to men in the Other racial group, F(3, 358) = 2.90, P = .04. On the other hand, European Americans reported greater comfort talking with close friends, F(3, 444) = 4.67, P < .01, and acquaintances, F(3, 421) = 3.25, P = .02, compared to the comfort level reported by African Americans.

Regression Models Predicting Sexual Activity with Women

Univariate and multivariate regression results are presented in Table 4 . In the univariate results, younger age, shorter length of time knowing HIV-positive status, higher levels of internalized homophobia, decreased comfort level for talking about homosexual relationships with close friends, and being African American were significantly associated with having sex with a woman. When all variables significant in the univariate analyses were entered into the multivariate regression model, only internalized homophobia and whether or not a man was African American were uniquely associated with having sex with a woman. The odds of having sex with a woman were 2.6 times as great among men with higher levels of homophobia as men with lower levels of homophobia, and 4 times as great among African American men relative to men of other race/ethnic backgrounds.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.