as a consequence of societal events in which antigay hostility is demonstrated (e.g., antigay violence, antigay political campaigns). Likewise, these authors suggest that self-stigma may occur among some sexual minorities as a result of accepting society’s negative attitudes toward them (e.g., internalized homophobia, internalized heterosexism, and internalized homonegativity). These negative feelings about one’s own homosexual desires can have negative impacts on mental and physical health and heighten the stress experienced by sexual minorities.

Some population studies have compared rates of anxiety disorders in homosexually and heterosexually active men and women. Using data from the 1996 National Household Survey of Drug Abuse, Cochran and Mays (2000b) examined differences in the prevalence of psychiatric syndromes among sexually active individuals. Behavioral sexual orientation was compared between respondents who reported exclusively other-sex sex partners (n = 9,714) and those who reported any same-sex sex partners (n = 194) in the prior year. While most homosexually active individuals did not meet criteria for any of the syndromes assessed, multivariate logistic regression analyses revealed that homosexually active men were more likely than other men to evidence a panic attack syndrome.

Cochran and Mays (2000a) also examined possible associations between homosexual/bisexual behavior patterns and lifetime prevalence of affective disorders, including mania, major depression, and dysthymia, among men (aged 17–39) using data from the National Health and Nutrition Examination Survey (NHANES) III. Sexual orientation was defined behaviorally based on self-reports of sexual partners and classified into three groups: any male sex partners (n = 108), only female partners (n = 3,208), and no sexual partners (n = 187). The results revealed that homosexually/ bisexually experienced men were no more likely than exclusively heterosexual men to meet criteria for lifetime diagnoses of affective disorders.

Similarly, Gilman and colleagues (2001) examined the risk of psychiatric disorders among individuals with same-sex and different-sex sexual partners based on data from the National Comorbidity Survey. Respondents were asked two sexual behavior questions: the number of women and, separately, men with whom they had engaged in sexual intercourse in the prior 5 years. Based on these responses, the respondents were classified into three groups: any same-sex partners (n = 125), exclusively heterosexual partners (n = 4,785), and no sexual partners (n = 967). Respondents with same-sex sexual partners had a higher 12-month prevalence of anxiety and mood disorders than respondents with different-sex partners only. The authors note the limitation that using a behavioral definition of