Approximately 20 million adults in the United States meet criteria for any past-year substance use disorder as defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition
]. A growing body of research suggests that non-heterosexual (sexual minority) populations are at particularly high risk for alcohol and other drug use disorders [3
]. Although public health researchers increasingly consider sexual orientation an important area of inquiry [10
], large-scale national epidemiologic studies are relatively rare. Population-based studies on substance use seldom ask about sexual orientation, and broad-based studies of sexual minority populations have only occasionally assessed substance use.
Recent theoretical explanations for heightened risk of substance use disorders among sexual minorities have focused predominantly on cultural and environmental factors, such as fewer roles and responsibilities (e.g., marriage and parenting) that otherwise limit or deter excessive substance use [15
], fewer prohibitions/sanctions against substance use [17
], and social stigma, prejudice, and discrimination associated with sexual minority status [8
]. Despite the proffering of such theoretical explanations, empirical evidence supporting these theories is scant and little attention has been given to within-group variations in substance use and substance dependence. Such information is necessary to better understand heightened risk of substance use and substance dependence among sexual minorities.
Central to understanding the risk of substance use disorders among sexual minority groups are issues related to the measurement and operationalization of the construct “sexual orientation.” Although there is growing consensus that sexual orientation includes at least behavioral, affective, and cognitive dimensions [19
], the few national studies on substance use that have assessed sexual orientation have generally asked about one dimension. Given recent findings suggesting variability in substance use across (and within) the three dimensions of sexual orientation [22
], reliance on a single dimension may miss important differences in health behaviors and hinder theoretical advances [20
Most population-based studies of the relationship between sexual identity and substance use conclude that women and men who self-identify as lesbian/gay or bisexual are at greater risk than heterosexuals for substance use behaviors [3
]. However, these and other studies suggest that risks differ within sexual minority groups and across gender [9
]. In analyses of data from the National Survey of Midlife Development in the United States (MIDUS), Cochran and colleagues found that the one-year prevalence rates of alcohol and other drug dependence did not differ for heterosexual and self-identified sexual minority adults aged 25 to 74 years [3
]. However, this study was limited by small sample sizes that required combining lesbian/gay and bisexual respondents.
In the few studies analyzing data from national probability sample surveys that have included sexual orientation questions, sexual behavior has been most commonly assessed [4
]. In analyses of the 1996 U.S. National Household Survey on Drug Abuse (NHSDA), Cochran and colleagues [4
] combined homosexually active adult women and men who reported only same-gender sexual partners (n=135) and those who reported sexual partners of both genders (n=59). Lifetime drug use was generally higher among respondents with same-gender sexual partners than those with other-gender partners, but very few differences were found for 30-day use.
Furthermore, respondents who reported same-gender partners had significantly higher rates of past-year marijuana dependence symptoms but did not differ significantly in rates of past-year dependence symptoms for eight other drugs (e.g., heroin, inhalants, stimulants, opioid analgesics).
Very few population-based studies have examined associations between sexual attraction and substance use behaviors, with the exception of two studies that focused on adolescents [32
] and young adults [22
]. These two previous studies found that youth attracted to both genders were more likely to report some substance use behaviors and problems, especially among young women. The findings of these two studies highlighted the importance of distinguishing between youths with only same-gender attractions and those attracted to both genders.
A few population-based studies of adults have assessed associations between substance use behaviors and more than one sexual orientation dimension [5
]. In the 2000 National Alcohol Survey, Drabble and colleagues [5
] compared heterosexual, bisexual and lesbian/gay-identified participants with heterosexual-identified participants who reported any same-gender sexual partners. Lesbian and bisexual women had elevated or significantly greater odds of alcohol abuse and dependence than did exclusively heterosexual women, but few significant sexual orientation differences were observed among men. Participants who identified as heterosexual but reported same-gender sexual partners did not differ from heterosexuals with no same-gender sexual partners, suggesting that sexual identity
may be more important than sexual behavior in predicting alcohol abuse and dependence.
Although past research has often combined lesbian/gay and bisexual adults in analyses, emerging evidence suggests that regardless of whether defined by identity [3
], attraction [22
] or behavior [4
], bisexual adults appear to be at higher risk for substance use and related problems [9
]. These findings suggest the usefulness of examining differences within and across dimensions of sexual orientation and sexual minority subgroups separately. Furthermore, findings indicating that the effects of sexual orientation on substance use differ by gender suggest the importance of looking at men and women separately [5
]. Although most studies have found that women who identify as lesbian or bisexual are more likely than heterosexual women to be heavy drinkers [5
], differences in drinking levels based on sexual orientation tend to be smaller among men [5
]; in fact, some studies with younger samples have found significantly lower rates of heavy drinking among gay and bisexual men compared with their heterosexual peers [22
]. Such information can help identify subgroups at greatest risk and aid the development of more effective prevention and intervention strategies.
The current study presents data from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), which is the first national study to provide data on substance use and DSM-IV substance dependence among sexual minorities in the United States based on assessment of all three major dimensions of sexual orientation: identity, attraction and behavior. The number of sexual minorities in Wave 2 of the 2004–2005 NESARC is 3 to 10 times larger than previous national studies [3
], permitting statistical comparisons within and across sexual orientation dimensions. Thus, the main objectives of this exploratory study are to examine the prevalence of substance use behaviors and substance dependence within and across groups defined by sexual identity, attraction and behavior, and to explore potential gender differences in the relationships of the various dimensions of sexual orientation with substance use outcomes.