Alcohol-Related Problems among Sexual Minority Women

Abstract

In this article I describe the historical context for research on sexual minority women’s drinking, including the age-old tendency to link homosexuality and alcoholism; I summarize gaps and limitations that characterized much of the research on sexual minority women’s drinking over the past several decades; and I review recent literature to highlight progress in the field—with a particular focus on my own research related to risk and protective factors for heavy drinking and drinking-related problems among sexual minority women. I conclude with a discussion of barriers to treatment for sexual minority women and recommendations for substance abuse treatment providers.

Homosexuality and alcoholism were closely associated in the scientific literature throughout most of the last century. Drawing on Freudian psychoanalytic theories of oral fixation, oedipal conflicts and incestuous drives, clinicians in the early to mid-1900s commonly linked alcohol consumption and alcoholism with homosexuality. Regarding women, for example, Riggall (1923) stated that “such women as drink will frequently show strong homosexual tendencies” (p. 163). Similarly, Clark (1919) noted that “women who have a strong desire for liquor are likely to prove homosexual” (p. 932). In addition, because most early studies of homosexuality used clinical samples from psychiatric settings, research findings tended to support the prevailing view of homosexuality as a form of mental illness. Work conducted by Evelyn Hooker (1957) in the 1950s challenged the pathology model and stimulated a paradigm shift that resulted in the removal of homosexuality from the Diagnostic and Statistical Manual of Mental Disorders (DSM) in 1973. Following the removal of homosexuality from the DSM, research was increasingly conducted with nonclinical samples. However, because recruitment strategies relied heavily on mailing lists of organizations or on social settings where sexual minority women and men could be found in large numbers, research findings continued to reflect the characteristics of populations from which the samples were drawn. Most research on lesbians’ alcohol use in the 1970s and early 1980s recruited some or all of the samples in gay bars (Fifield, Latham, & Phillips, 1977; Lohrenz, Connelly, Coyne, & Spare, 1978; Saghir & Robbins, 1973). Not surprisingly, these studies found high rates of alcohol use/abuse and related problems.

Research on sexual minority women’s drinking has increased dramatically over the past several decades. Much of this work, however, has focused on comparing lesbian and heterosexual women on prevalence of drinking, heavy drinking, and drinking-related problems. Relatively few studies have examined risk and protective factors for hazardous/harmful drinking among lesbians or other sexual minority women. In addition, only recently have researchers begun to compare prevalence rates or risk factors within sexual minority populations (e.g., comparisons of lesbian and bisexual women or comparisons across a continuum of sexual identities including exclusively lesbian, mostly lesbian, bisexual and mostly heterosexual women). This research has begun to highlight variations in risk—often substantially large ones—across subgroups of sexual minority women.

Over the past 15–20 years my colleagues and I have been conducting work designed to address the limitations in research on sexual minority women’s drinking. In this paper I will summarize some of the findings from this work as well as key findings from the work of other researchers in the field. The theoretical model (Figure 1) that has guided my colleagues and my work over the past several years draws from general stress and coping theories and recent theoretical work on sexual minority stress. Stress theory is a prominent framework for conceptualizing group differences in risk for mental health problems (Aneshensel, 1992; Thoits, 1999; Thoits, 2010). Meyer (2003) builds on general stress theory to show how the stressful social environment (created by stigma, prejudice, and discrimination) in which sexual minorities live increases vulnerability to substance abuse and other stress-sensitive mental health disorders. Meyer’s minority-stress model shares the person-process-context perspective of social ecology theory (Bronfenbrenner, 1979) and portrays stress processes in the contexts of the individual’s and the larger environment’s strengths and vulnerabilities.

Figure 1.

Theoretical Model for Hazardous Drinking in Adult Sexual Minority Women.

The model in Figure 1 includes both general and sexual minority-specific factors believed to be important in influencing sexual minority women’s drinking. The model includes general early stressors/risk factors such as parental drinking problems, childhood sexual abuse and childhood physical abuse; early consensual heterosexual sex; early drinking onset; and general later risk factors, such as adult sexual and physical assault and intimate partner violence; and current roles and relationships (e.g., relationship status and quality). In addition, the model incorporates sexual minority-specific early risk factors—earlier age first questioned, first acknowledged to oneself, and first disclosed minority sexual orientation (sexual- identity-development benchmarks)—as well as sexual minority-specific later risk factors such as anti-gay workplace harassment and other sexual orientation–based discrimination. The later risk factors may build on earlier risk factors in an additive or multiplicative manner and combine with chronic stressors associated with sexual minority status to increase sexual minority women’s risk for hazardous drinking. Psychological distress (depression, anxiety, post-traumatic stress) is viewed primarily as a mediator of the relationship between early/later stressors and hazardous drinking in the model, but we recognize that psychological distress, especially depression, may also result from hazardous drinking. Potential moderators of the relationship between early/later stressors and hazardous drinking are level of sexual orientation disclosure, and level of social support. The model serves as an organizing framework for this review of research related to sexual minority women’s drinking.

Hazardous Drinking among Sexual Minority Women¹

As noted above, research on sexual minority women’s drinking has grown substantially. However, understanding of this health concern has been hampered by a number of methodological limitations that have characterized much of this research, especially work conducted prior to the mid-1990s. Among these limitations are: (1) overreliance on small, homogenous, predominantly White samples from a relatively narrow age range; (2) lack of appropriate heterosexual comparisons; (3) indirect measurement of sexual orientation; (4) limited assessment of alcohol use and alcohol-related problems; and (5) lack of longitudinal data (Hughes, Wilsnack, & Johnson, 2005; Hughes & Eliason, 2002).

The Chicago Health and Life Experiences of Women (CHLEW) study was developed to address these limitations. In 2000 and early 2001, my colleagues and I collected baseline data from a diverse sample of 447 Chicago-area sexual minority women (Hughes et al., 2006). The CHLEW was designed to replicate and extend the National Study of Health and Life Experiences of Women (NSHLEW), a 20-year national probability study of more than 1600 predominantly heterosexual women, and to gather data on risk and protective factors among lesbians. We replicated the NSHLEW study design as closely as possible to permit comparisons between heterosexual women in the national study and sexual minority women in the CHLEW study. Because minority sexual orientation is relatively rare in the general population (Gates, 2010; Laumann, Gagnon, Michael, & Michaels, 1994), even large general population samples often have insufficient numbers of sexual minority women to permit reliable statistical analyses. Thus, a volunteer sample for the CHLEW was recruited using a broad range of recruitment sources and strategies, with strenuous efforts to maximize sample representativeness by including subgroups of women that had been underrepresented in most previous studies of lesbian health (aged under 25 and over 50, high school education or less, racial/ethnic minority). Women were eligible if they were 18 or older, lived in the greater Chicago area, and self-identified as lesbian. Unlike the majority of previous studies, we recruited a sample that represented a wide age range (18–83) and diverse race/ethnicity (more than one-half were racial/ethnic minorities).

Female interviewers were trained by National Opinion Research Center staff to conduct 90-minute face-to-face interviews using a slightly modified version of the NSHLEW questionnaire. The interview questionnaire included in-depth assessments of many early and later risk factors associated with women’s drinking. In addition to small changes in the wording of some questions (e.g., “husband” or “spouse” changed to “partner”), we added new sexual orientation-specific questions (e.g., sexual-identity-development benchmarks, level of sexual- orientation disclosure). These new questions and the revised survey questionnaire were pretested in several pilot studies (Hughes et al., 2005; Hughes, 2003).

At baseline, participants’ mean age was 38 years; 27.5% of the sample was Black, 19.7% Latina, 47.6% White, and 5% other racial/ethnic or mixed racial/ethnic groups. Although recruitment materials explicitly stated that eligibility was limited to women who self-identified as lesbian, and women who identified as bisexual or heterosexual were screened out in the telephone screening interview, in the actual interview 11 women (2.3%) identified as bisexual. The sample closely matched the racial/ethnic composition of Cook County, Illinois, where the majority of CHLEW participants lived (Chicago Fact Finder, 2005). In contrast to census figures, but similar to other lesbian samples, CHLEW participants were well educated: more than half (56%) had a bachelor’s degree or higher. The majority worked full-time (68%), 11% worked part-time, and 20% were not employed. Annual household income varied widely, with one-fourth (25%) having annual incomes below $20,000 and 22% having incomes of $75,000 or more. Approximately two thirds (69%) of participants were in a committed relationship with a female partner (though not all lived with their partners) and one-third (31%) had children. Compared with White women, Black and Latina women tended to be younger (χ² = 54.9, 9df, p < .001), to have less education (χ² = 58.5, 9df, p < .001) and lower household incomes (χ² = 24.7, 9df, p < .01). They were also more likely to have ever had children (χ² = 24.4, 3df, p < .001) and to have children living at home (χ² = 11.8, 3df, p < .01).

In wave 2 of the CHLEW study we successfully contacted and re-interviewed 384 (85.9%) of the study participants. Lost to follow-up were 33 (7.4%) women who could not be located, 10 (2.2%) who were deceased, 10 (2.2%) who refused, nine (2.0%) who were located but unable to participate for various reasons (e.g., family illness or scheduling conflicts), and one participant who transitioned from female to male gender. To assess possible bias due to attrition, nonresponse rates (combining refusals and locating failures) were examined in relation to all major drinking variables and seven demographic variables (age, race/ethnicity, education, income, employment, relationship status, and having children living at home). We fit a logistic regression model to the data, examining possible predictors of attrition. Controlling for demographic and major drinking variables, the only significant predictor of attrition was having a high school education or less.

Reports of indicators of hazardous drinking among CHLEW participants at the baseline (wave 1) assessment were somewhat high (Hughes et al., 2006). Among lifetime drinkers (96% of the sample), 42% had sometimes wondered if they might have a drinking problem, and 18% had received help for a drinking problem. These findings are consistent with those from national probability samples of self-identified lesbians (Drabble, Midanik, & Trocki, 2005) and women who report same-gender sex partners (Cochran & Mays, 2000). Unlike findings from studies of women in the general population in which rates of drinking tend to decrease with age (Johnstone, Leino, Ager, & Ferrer, 1996; Moore et al., 2005), we found much less variation across CHLEW age groups in rates of abstention and levels of drinking. Younger women (age 18–30) were, however, significantly less likely than women in the older groups to report 12-month abstention and more likely to be heavy drinkers. Also in contrast to women in the general population (Caetano & Clark, 1998; Gilbert & Collins, 1997), rates of abstention across the three major racial/ethnic groups did not differ statistically. However, Black lesbians were more likely than Latina or White lesbians to be heavy drinkers (p<.05). Women ages 31–40 (10.3%) and 41–50 (12.4%) were most likely to report being in recovery. Approximately two thirds of the drinkers reported one or more lifetime adverse consequences of drinking (71.0%) or one or more symptoms of potential alcohol dependence (64.5%). (Examples of adverse consequences of drinking include driving while under the influence of alcohol and having fights with a partner or family member when drinking. Examples of alcohol dependence symptoms include memory loss or blackouts, morning drinking, tolerance.) More women aged 40 and younger reported lifetime alcohol dependence symptoms than did women over 40 years old. We fit separate multivariate models for each of four lifetime and 12-month indicators of hazardous drinking (heavy episodic drinking, intoxication, any adverse drinking consequences and any alcohol dependence symptoms). Controlling for demographic characteristics, we found significant age differences in most hazardous drinking indicators (younger women were more vulnerable) but fewer racial/ethnic differences (Hughes et al., 2006).

To better understand how sexual minority women’s drinking behaviors, patterns and problems compare with those of heterosexual women we combined data from the 2000 CHLEW and the 2001 NSHLEW. To maximize comparability of the samples we selected cases based on age and area of residence. The final pooled sample included 953 respondents aged 21–70 who lived in large or medium-sized cities or nearby suburbs. Women in the total sample identified as exclusively heterosexual (N=482, 50.6%), mostly heterosexual (n=42, 4.4%), bisexual (n=22, 2.3%), mostly lesbian (n=111, 11.6%), or exclusively lesbian (n=296, 31.1%). We weighted the NSHLEW cases to reflect selection probabilities and to adjust standard errors to reflect that survey’s complex sampling design. We compared four indicators of hazardous drinking (heavy episodic drinking, intoxication, adverse drinking consequences, and alcohol dependence symptoms) across the five sexual identity groups. Exclusively heterosexual women were significantly lower than all four sexual minority groups on all indicators of hazardous drinking. Bisexual women were significantly more likely than exclusively lesbian women to report three of the four hazardous drinking indicators (all except heavy episodic drinking) and more likely than mostly lesbian women to report adverse drinking consequences and alcohol dependence symptoms. The pattern of results for three risk factors (CSA, early onset of drinking, and lifetime and 12-month depression) was generally the same as for the hazardous drinking indicators—lowest in exclusively heterosexual respondents and highest in bisexual respondents (Wilsnack et al., 2008).

Early Risk Factors for Hazardous Drinking

Childhood sexual abuse (CSA) and childhood physical abuse (CPA) are almost universally viewed as major early life stressors and have been consistently linked with long-term adverse mental health consequences, including hazardous drinking and alcohol-use disorders (Briere, 1988; Dube, Anda, Felitti, Edwards, & Croft, 2002; Kendler et al., 2000; Simpson & Miller, 2002; Wilsnack, Wilsnack, Kristjanson, Vogeltanz-Holm, & Harris, 2004; Wilsnack, Vogeltanz, Klassen, & Harris, 1997). However, these risks vary considerably among individuals (Kendall-Tackett, Williams, & Finkelhor, 1993; Widom, Ireland, & Glynn, 1995) and many questions remain about which subgroups of women are at greatest risk and about the processes by which childhood abuse influences drinking.

Childhood sexual abuse

Accumulating evidence indicates that sexual minority women are more likely than heterosexual women to report CSA (Austin, Roberts, Corliss, & Molnar, 2008; Balsam, Rothblum, & Beauchaine, 2005; D’Augelli, 2003; Hughes, Johnson, & Wilsnack, 2001; Hughes, Johnson, Wilsnack, & Szalacha, 2007; Hughes, McCabe, Wilsnack, West & Boyd, 2010; Tomeo, Templer, Anderson, & Kotler, 2001). In a pilot study for the CHLEW that included 63 lesbians and a demographically matched comparison group of 57 heterosexual women recruited from the Chicago metropolitan area, lesbians were significantly more likely to report self-perceived CSA and more likely to report childhood sexual experiences that met study criteria for CSA. These criteria, based on Wyatt’s (1985) work, included sexual activity before age 18 that was unwanted, that was with a family member five or more years older than the respondent, or that occurred before age 13 with a non-family member five or more years older than the respondent (Hughes et al., 2001). This finding was replicated in comparisons of heterosexual women from the 2001 NSHLEW and sexual minority women from the 2001 CHLEW study, where lesbian and bisexual women were two times as likely as heterosexual women (59% vs. 29%) to report childhood experiences that met study criteria for CSA (Wilsnack et al., 2008).

Results from large general population studies lend support to these findings. For example, using data from the Nurses Health Study II, Austin and colleagues (2008) found that lesbian and bisexual women were more than two times as likely as heterosexual women to report CSA. Similarly, in recent analyses of the National Epidemiologic Survey of Alcohol and Related Conditions (NESARC) we found that lesbian and bisexual women were three times as likely as heterosexual women to report CSA (Hughes, McCabe et al., 2010).

Aside from findings indicating higher rates of CSA among sexual minority women, the relative risk for hazardous drinking associated with this early stressor may differ because of sexual-orientation-related factors. For example, given the shame and secrecy that are often associated with both CSA and early recognition of same-gender attraction, coping with CSA may be particularly difficult for sexual minority women (Finkelhor & Browne, 1985). It is also possible that sexual minority women experience more severe forms of abuse (Austin et al., 2008; Balsam et al., 2005; Hughes, Wilsnack, Kristjanson, & Benson, 2010). In analyses of pooled data from the NSHLEW and the CHLEW study that compared women who identified as exclusively heterosexual with those who identified as exclusively lesbian (N=770), we found that among women meeting study criteria for CSA, lesbians consistently reported indicators of more severe CSA. These indicators included higher rates of genital CSA and vaginal or anal intercourse; younger age at first CSA; greater number of perpetrators; greater frequency and longer duration of CSA; and greater use of physical force or coercion. In addition, lesbians were also more likely than heterosexual women to report six of the eight long-term adverse emotional effects of CSA assessed in the study (Hughes, Wilsnack et al., 2010). My colleagues and I are now conducting analyses to explore whether CSA severity helps to explain sexual minority women’s elevated risk for hazardous drinking.

Childhood physical abuse

A smaller but growing body of research on CPA among sexual minority women also points to elevated rates (Austin et al., 2008; Balsam et al., 2005; Corliss, Cochran, & Mays, 2002; Saewyc, Bearinger, Blum, & Resnick, 1999; Stoddard, Dibble, & Fineman, 2009; Tjaden, Thoennes, & Allison, 1999). Using the National Survey of Midlife Development in the United States (MIDUS), Corliss, Cochran and Mays (2002) found that a combined sample of lesbian/bisexual women (weighted n=37) reported greater major physical maltreatment by their parents or guardians than did heterosexual women (weighted n=1607). Major maltreatment included aggressive acts such as being “kicked, bit, or hit … with a fist,” in contrast to minor physical maltreatment, such as being “pushed, grabbed, or shoved.” Similarly, Tjaden et al. (1999) found that adults who lived with a same-sex partner, a proxy indicator of sexual minority status, were more likely to report childhood physical assault by an adult caretaker than were respondents who were married to or had only ever lived with opposite-sex partners.

Balsam and colleagues (2005) examined childhood abuse in lesbian, gay and bisexual (LGB) participants (n=721) and compared them with their heterosexual siblings (n=533). Using hierarchical linear modeling, these researchers found that sexual minority status was a statistically significant predictor of all childhood abuse measures, including physical and sexual abuse by parents or caretakers. Sexual orientation accounted for significant variance in victimization beyond the effect of family, suggesting that even within the same household, sexual minorities are at greater risk of abuse than their heterosexual siblings. Stoddard et al. (2009), who used a similar sibling-matched design, also found elevated rates of both CPA and CSA in lesbians. Twice as many lesbians (20.4%) as heterosexual sisters (10.0%) reported physical abuse before age 16. (These researchers found similar sexual identity-based differences in reports of CSA.) In contrast, my colleagues and I analyzed the 2004 NESARC data and found statistically significant differences in rates of CPA reported by bisexual and heterosexual women, but no such differences between lesbian and heterosexual women (Hughes, McCabe et al., 2010).

To examine potential mediators of the relationship between childhood abuse and hazardous drinking in sexual minority women we used structural equation modeling with data from the baseline (wave 1) CHLEW sample. Our findings showed that CPA directly predicted lifetime psychological distress and that CSA directly predicted lifetime alcohol abuse (a latent construct representing four lifetime indicators of alcohol-related problem drinking). In addition, CSA indirectly increased the risk of lifetime alcohol abuse through its negative effect on age at first heterosexual intercourse. CPA had only indirect effects on lifetime alcohol abuse through its strong relationship to lifetime psychological distress (Hughes et al., 2007).

Sexual identity development/Sexual identity disclosure

Sexual minorities may be particularly vulnerable to hazardous drinking during the sexual identity development process, or the process of ‘coming out’. Women who ‘come out’ as lesbian or bisexual must learn to cope with a stigmatized identity. In addition, disclosure of a minority sexual identity increases risk of rejection, violence and victimization. The process of sexual identity development and disclosure can be particularly difficult for youth given that they must also grapple with the usual physical, mental, emotional and social changes associated with adolescent development. Women in the CHLEW study (mean age = 38 years at baseline) first acknowledged to themselves that they were lesbian at a mean age of 22.3 years and first disclosed their lesbian identity to another person at a mean of 23.8 years (Parks & Hughes, 2007). This is about six years later than younger women, aged 18–19, who were surveyed in the late 80s or early 90s (Savin-Williams & Diamond, 2000). These findings, and those of other studies (e.g., Herek, Norton, Allen, & Sims, 2010), suggest that sexual minority youth are coming out at ages earlier than their older counterparts. Given evidence that younger age of coming out is associated with hazardous drinking as well as elevated health risks of childhood maltreatment, gay-related harassment, and suicidality (Corliss, Cochran, Mays, Greenland, & Seeman, 2009; Friedman, Marshal, Stall, Cheong, & Wright, 2008; Parks, Hughes, & Kinnison, 2007), contemporary sexual minority youth may be at even greater risk of hazardous drinking than their older sexual minority counterparts.

We examined the longitudinal effects of age of sexual identity disclosure and other early life experiences (parental drinking problems, age of first consensual heterosexual sexual intercourse, age of drinking onset) on hazardous drinking and psychological distress in CHLEW participants interviewed in waves 1 and 2. Covariance structural equation model findings revealed a transactional process suggesting that psychological distress is both a cause and a consequence of hazardous drinking in the CHLEW sample. In addition, disclosing sexual orientation at a younger age was positively associated with risk of adult hazardous drinking and this effect was independent of parental drinking, age of drinking onset and age of first consensual heterosexual intercourse (Hughes, Johnson, Szalacha, & Wilsnack, 2009).

In other analyses of CHLEW data we used retrospective reports about early drinking contexts collected in wave 1 to predict current hazardous drinking in wave 2. We compared women who had early social networks that consisted predominantly of sexual minority members with those whose early social networks consisted of mostly heterosexual women and men, or both heterosexuals and sexual minorities (mixed). Women with mostly sexual minority networks reported greater alcohol availability and higher levels of alcohol consumption than women with heterosexual or mixed networks. Risks associated with early alcohol availability and early alcohol consumption in lesbian-specific social settings varied based on early social network type and age of coming out (i.e., age first disclosed lesbian sexual orientation). Risk for hazardous drinking was greatest among respondents who drank more heavily in settings incongruent with their typical social network (e.g., drinking in heterosexual settings but having predominantly sexual minority social networks). Lesbians who came out before age 21 reported earlier onset of drinking than those who came out later. In addition, earlier age of coming out was positively associated with current (30-day) drinking and with treatment for alcohol-related problems (Parks et al., 2007).

Findings from a recent study conducted by Rosario, Scrimshaw and Hunter (2008) suggest that more important than the age of coming out may be the reactions of those to whom sexual minority youth disclose their minority identity. These researchers found that a greater number of rejecting reactions to sexual orientation disclosure was positively associated with cigarette, marijuana, and alcohol use among sexual minority youth aged 14–21 years—even after controlling for demographic factors and emotional distress. They also found that high numbers of accepting reactions buffered the effects of negative reactions. Among youth with fewer accepting reactions, alcohol frequency and quantity increased as the number of rejecting reactions increased. However, among youth with a high number of accepting reactions, the associations of rejecting reactions with alcohol use were largely attenuated such that those with more accepting reactions showed a consistently low/moderate level of alcohol use, regardless of the number of rejecting reactions. These findings are consistent with those of Ryan and colleagues (Ryan, Huebner, Diaz, & Sanchez, 2009; Ryan, Russell, Huebner, Diaz, & Sanchez, 2010) and point to the importance of addressing the support systems of sexual minority individuals in prevention and treatment efforts.

Early consensual sexual intercourse

Sexual minority youth are believed to engage in consensual heterosexual intercourse at younger ages than their heterosexual counterparts (Blake et al., 2001; Saewyc et al., 1999). Potential reasons include “heterosexual immersion” as a strategy for responding to confusion or anxiety about same-sex attraction, and the risks associated with coming out (e.g., sexual minority teens are more likely than their peers to drop out of school and run away from, or be forced out of, their homes, and consequently are at risk of prostitution as a means of survival.) In analyses of wave 1 CHLEW data, earlier age of first heterosexual sex was associated with higher risk of alcohol abuse in adulthood. However, as we discuss in the paper reporting these findings (Hughes et al., 2007), it is difficult to determine which early sexual experiences were forced or consensual and to disentangle the consequences of each. It is also difficult to determine the timing of early risk factors associated with CSA and hazardous drinking. Does CSA lead to early onset of alcohol use, which then increases risk of early heterosexual intercourse? Or (especially in the case of youth questioning their sexual orientation) does early heterosexual sexual activity occur first and hazardous drinking follow as a way to deal with ambivalence about heterosexual sex?

Early onset of drinking

Earlier onset of drinking has been shown to be a robust predictor of alcohol-use disorders in general population studies (Hingson, Heeren, & Winter, 2006; Pitkänen, Lyyra, & Pulkkinen, 2005). Furthermore, the co-occurrence of early onset of alcohol consumption and early engagement in heterosexual intercourse is common among adolescents (Jessor, Van Den Bos, Vanderryn, Costa, & Turbin, 1995) and sexual abuse survivors (Finkelhor & Browne, 1985)—and may be even more common among sexual-minority youth (Balsam et al., 2005; Garofalo, Wolf, Kessel, Palfrey, & DuRant, 1998; Saewyc et al., 1999). In a U.S. community-based longitudinal cohort study of the children of women participating in the Nurses Health Study II, Corliss, Rosario, Wypij, Fisher and Austin (2008) found that both male and female sexual minority youth reported younger ages of drinking onset than their heterosexual counterparts. Younger age of drinking onset among sexual minority youth was associated with higher frequency of later binge drinking, and sexual orientation differences in drinking outcomes tended to be more pronounced in female than in male youth.

The impact of early drinking onset appears to extend to adulthood. In cross-sectional analysis of early childhood predictors of alcohol abuse among CHLEW participants in wave 1 (Hughes et al., 2007), age of drinking onset was the strongest predictor of lifetime alcohol abuse in adulthood. In addition, this variable mediated the effects of parental drinking problems on lifetime alcohol abuse and on lifetime psychological distress. These findings suggest that preventive interventions aimed at postponing the onset of drinking may be protective against hazardous drinking and alcohol-use disorders as well as other health risks associated with early sexual activity, and that such interventions are especially important for sexual minority youth.

Later Risk Factors for Hazardous Drinking

Adult sexual assault (ASA) and revictimization

Women who are victimized in adulthood are at heightened risk for a range of adverse psychological consequences. Research on ASA among sexual minority women has generally found rates similar to those reported by heterosexual women. In data from our CHLEW pilot work (Hughes et al., 2001) rates of unwanted sexual activity after age 18 did not differ statistically for lesbians (39%) and heterosexual women (42%). Similarly, in analyses of data from the NESARC (Hughes, McCabe et al., 2010) we found no statistically significant differences in ASA among lesbian (8.1%), bisexual (6.7%) and heterosexual (3.3%) women. In contrast, however, in two studies that used heterosexual siblings as the control group (Balsam et al., 2005; Stoddard et al., 2009), sexual minority women were more likely than their heterosexual sisters to report sexual victimization in adulthood.

Like heterosexuals, gay, lesbian and bisexual victims of ASA are at increased risk for both psychological distress (Descamps, Rothblum, Bradford, & Ryan, 2000; Houston & McKirnan, 2007; Morris & Balsam, 2003), and substance use (Descamps et al., 2000; Houston & McKirnan, 2007; Kalichman et al., 2001; Ratner et al., 2003). Further, women who experience CSA are two to three times more likely to be victimized as adults (Balsam, Lehavot, Beadnell, & Circo, 2010; Classen, Palesh, & Aggarwal, 2005; Tjaden & Thoennes, 2000)—and revictimization (sexual assault in both childhood and adulthood) adds to women’s risk of hazardous drinking (Descamps et al., 2000; Dube et al., 2006).

In one of the few studies of revictimization in a sexual minority population, Balsam et al. (2010) compared adult lesbians, gay men, and heterosexual women (N=871) on prevalence and mental health correlates of revictimization. CSA was associated with elevated rates of ASA in all three groups. Compared with participants who reported either CSA or ASA and those who reported neither form of abuse, those who were revictimized had higher levels of alcohol use, psychological distress, suicidality and self-harm behaviors. Similarly, Heidt et al. (2005) found that lesbian and gay participants who were revictimized scored significantly higher on measures of psychological distress, including PTSD symptoms, than did nonvictims or victims of only CSA or ASA. Given that psychological distress is strongly associated with hazardous drinking (Bolton, Robinson, & Sareen, 2009), findings of higher rates of revictimization may help explain sexual minority women’s elevated risk for hazardous drinking.

To compare rates of revictimization in sexual minority and heterosexual women, and to explore variations in rates of victimization across sexual identity subgroups we pooled data from 405 participants in the 2001 (wave 1) CHLEW and 548 participants in the 2001 (wave 5) NSHLEW surveys. We found that women in the four sexual minority groups (exclusively lesbian, mostly lesbian, bisexual, mostly heterosexual) reported higher levels of hazardous drinking and higher rates of CSA and sexual revictimization (both CSA and adult sexual assault) than did exclusively heterosexual women. Revictimization was a strong predictor of hazardous drinking, particularly among women who identified as mostly heterosexual and mostly lesbian (Hughes, Szalacha et al., 2010).

Relationship distress and intimate partner violence (IPV)

Relationship distress

Previous studies on women’s drinking (Covington & Surrey, 1997; Wilsnack, Wilsnack, Kristjanson, & Harris, 1998), as well as relationship theories of women’s development (Jordan, 1991), point to the importance of women’s relationships on their mental health and substance use. Because same-sex relationships develop without the support of social institutions, and often without the support of family of origin (Henry J. Kaiser Family Foundation, 2001; Kurdek, 2001), they are vulnerable to disruption and dissolution (Peplau & Fingerhut, 2007). Relationship dissolution is one of life’s most stressful events (Dohrenwend, Krasnoff, Askenasy, & Dohrenwend, 1978; Hope, Rodgers, & Power, 1999) and has been shown to be an important risk factor for increased drinking and drinking-related problems (Chilcoat & Breslau, 1996; Power, Rodgers, & Hope, 1999), especially in women (Fillmore et al., 1997; Horwitz, White, & Howell-White, 1996; Neff & Mantz, 1998). Relationship dissolution is an important source of psychological distress among sexual minority women (Kurdek, 1997; Rothblum, 1994) and may in turn increase the risk of hazardous drinking.

Intimate partner violence (IPV)

Results from the limited research on IPV in same-sex relationships suggest that rates of IPV among lesbians are similar to or higher than those among heterosexual women (Balsam & Szymanski, 2005; Renzetti, 1992; Renzetti, 1994; Tjaden et al., 1999). For example, in comparisons of sexual minority women and men and their heterosexual siblings, Balsam and colleagues (2010) found that minority sexual identity was not predictive of psychological maltreatment or physical assault by a partner in the past year. However, sexual minority participants were more likely than their heterosexual siblings to report psychological maltreatment and physical assault by a partner since age 18. Similarly, in a recent study of California residents conducted by the UCLA Center for Health Policy Research (Zahnd, Grant, Aydin, Chia, & Padilla-Frausto, 2010), lesbian and gay participants were almost twice as likely as heterosexual participants to report IPV since age 18, but past year IPV did not differ based on sexual orientation. Because studies on IPV often do not ask about the gender of perpetrators, it is possible that some of the IPV reported by sexual minority respondents, particularly when longer time periods were assessed, may have been perpetrated by previous opposite sex partners.

As in heterosexual relationships, alcohol use often plays a role in same-sex IPV (Callanan, 2004; Descamps et al., 2000; Schilit, Lie, & Montagne, 1990) and can be both a predictor and a consequence of IPV. In addition, general population studies demonstrate that women who were sexually victimized as children are more likely than those who were not victimized to experience IPV as adults (Briere & Runtz, 1987; Briere, 1988; Russell, 1986), which may add to sexual minority women’s risk of both IPV and hazardous drinking. Regardless of sexual minority women’s relative risk, the issues associated with IPV in same-sex relationships are likely more complex than those in heterosexual relationships. For example, the isolation associated with belonging to a marginalized group, and society’s tendency to blame female victims of violence, make it difficult for abused women in same-sex relationships to disclose IPV or to seek help. Such isolation and stigmatization may induce or increase hazardous drinking, which may increase vulnerability to further violence (Hughes & Norris, 1995).

Sexual orientation harassment and discrimination

Sexual minorities are frequently exposed to harassment and discrimination based on their sexual orientation. Several studies have found associations between sexual-orientation-based harassment and discrimination and psychological distress (Chae et al., 2010; Descamps et al., 2000; Herek, 1998; Lewis, Derlega, Berndt, Morris, & Rose, 2002; Lewis, Derlega, Griffin, & Krowinski, 2003; Meyer, 1995; Nicholson & Long, 1990; Waldo, 1999). Cochran and Mays (2000) examined sexual minorities’ experiences of discrimination and found that perceived discrimination was associated with lower quality of life and with indicators of psychiatric morbidity in both sexual minority and heterosexual respondents. Compared with heterosexuals, however, lesbian, gay and bisexual respondents more frequently reported both discrete discrimination events (e.g., being fired from a job) and day-to-day discrimination (e.g., being called names or insulted). Experiences of lifetime and daily discrimination were associated with negative mental health outcomes including psychiatric disorder diagnosis, self-rated “fair” or “poor” mental health, and psychological distress. When discrimination events were controlled for, the strength of the relationship between sexual identity and mental health was attenuated, suggesting that mental health disparities among sexual minorities can be explained, in part, by their experiences of discrimination.

Almost no research has examined the impact of harassment or discrimination on drinking behaviors or hazardous drinking among sexual minority women. Using data from a university-based study of workplace harassment and alcohol use (N=2,492) Nawyn, Richman, Rospenda and Hughes (2000) examined exposure to workplace harassment and alcohol-related outcomes for lesbians, gay men, and bisexuals. Although lesbian and bisexual women did not differ significantly from heterosexual women in their experiences of workplace harassment, the associations between harassment and elevated alcohol consumption and problems were stronger for lesbian/bisexual women than for heterosexual women. In recent analyses of data from participants in the 2004 NESARC who identified as lesbian, gay or bisexual (N=577) we examined the associations between sexual orientation- race- and gender-discrimination and substance use disorders. More than two-thirds of the sample reported that they had experienced at least one of the three forms of discrimination in their lifetime and nearly one-half reported lifetime sexual orientation discrimination. The odds of past-year substance-use disorders were nearly four times greater among LGB adults who reported all three forms of discrimination than among LGB adults who did not report discrimination (McCabe et al., 2010).

Psychological distress and minority stress

Numerous studies have found an association between psychological distress and alcohol use disorders in women, and research has documented that sexual minority women are substantially more likely than their heterosexual counterparts to report psychological distress (Bostwick, Boyd, Hughes, & McCabe, 2010; Kerr & Emerson, 2004; Matthews, Hughes, Johnson, Razzano, & Cassidy, 2002; Meyer, 2003).

Using data from wave 1 of the CHLEW we examined the prevalence of depression and alcohol dependence symptoms as well as the co-occurrence of depression and symptoms of alcohol dependence symptoms (Bostwick, Hughes, & Johnson, 2005). More than one-half (57%) of the CHLEW respondents reported lifetime depression, and 22% reported 12-month depression. (Depression was measured by questions and diagnostic criteria from the Diagnostic Interview Schedule [DIS].) Past year alcohol dependence symptoms were significantly associated with both past-year and lifetime depression. Lifetime depression was higher among White and Latina lesbians than among Black lesbians. Younger age was the strongest predictor of the co-occurrence of depression and alcohol dependence symptoms. Women not currently in a committed relationship were also more likely than those in a committed relationship to meet study criteria for past-year depression. Having depression in the year prior to the interview doubled the odds of having at least one alcohol dependence symptom within the same time frame. Further, lifetime depression also almost doubled the odds of past-year alcohol dependence symptoms, suggesting that the influence of depression on hazardous drinking in lesbians is strong and lasting.

Results of analyses using data from the Australian Longitudinal Women’s Health Survey, a national probability sample survey of women in Australia, also found significantly higher stress and poorer mental health, including depressive symptoms, among sexual minority women than among exclusively heterosexual women. Furthermore, each of the mental health measures assessed in this study was significantly associated with at-risk drinking (Hughes, Szalacha, & McNair, 2010).

Beginning with Meyer’s earlier work on sexual minority stress (Meyer, 1995), there has been growing consensus that minority stress accounts for the elevated rates of psychological distress observed among sexual minorities. Meyer’s conceptualization of minority stress is based on the premise that sexual minority people who live in heterosexist societies, such as the U.S., are subjected to chronic stress related to their stigmatization (Meyer, 2003). Stress processes include expectations of rejection and discrimination, and actual experiences of discrimination and violence, that create vigilance (e.g., fear of disclosing minority sexual orientation) and internalization of homophobia (directing society’s negative attitudes toward the self). When these unique and chronic sexual-minority specific stressors combine with general stressors, such as those described in earlier sections of this paper, the cumulative stress load may overburden an individual’s coping resources, resulting in psychological stress and heightened risk of hazardous drinking.

Race/ethnic- sexual-minority women

Social location theory (Jackman, 1999; Jackman, 2001) and minority stress theories (Meyer, 2003) consider multiple and intersecting social identities and posit that each individual has a location within society that differs based on the characteristics and attributes that are valued by the larger, dominant culture. As the number of intersecting oppressed identities increases, so does exposure to stress as each identity is subject to socially dictated values, beliefs, and expectations. Studies of lesbians of color find that multiple stressors—racism in sexual minority communities, heterosexism in their racial/ethnic communities, and discrimination in society at large—can compound the negative impact on the mental health of racial/ethnic minority lesbians (Balsam, Huang, Fieland, Simoni, & Walters, 2004; Bowleg, Huang, Brooks, Black, & Burkholder, 2003; Greene, 1994; Mays, Yancey, Cochran, Weber, & Fielding, 2002; Morris & Balsam, 2003). Whether or how the multiple marginalized and stigmatized statuses experienced by sexual minorities impacts risk of hazardous drinking has not been adequately studied.

Moderators or Protective Factors for Hazardous Drinking

Social support

Despite elevated levels of stress most sexual minority individuals do not show evidence of psychiatric/mental health disorders or hazardous drinking (Cochran & Mays, 2000; Meyer, 2003). Although the majority of research with sexual minorities has focused on risk factors, to fully understand health disparities it is important to identify and assess resilience or protective factors in this population group. Research has shown that social support is one of the most robust predictors of positive health and well-being (Sarason, Sarason, & Gurung, 2001; Thoits, 1995). Individuals who experience high levels of stress but receive adequate social support are less likely to become ill than those with inadequate social support. The importance of social support is magnified among sexual minorities who confront both usual life stressors and the added burden of minority stress associated with their sexual minority status.

Several studies have found that perceived social support from family and friends predicts psychological well-being among adult sexual minority women (Ayala & Coleman, 2000; Jordan & Deluty, 1998; Vincke & van Heeringen, 2002; Yakushko, 2005). However, studies often show that sexual minority women (and men) report substantially lower levels of social support from family than from friends (Kurdek, 1994; Peplau & Fingerhut, 2007; Masini & Barrett, 2007; Mays et al., 1994; Oetjen & Rothblum, 2000). In addition, the impact of social support may vary depending on its source. In a study of lesbian, gay and bisexual adults aged 50 and older, Masini and Barrett (2008) found that social support from friends was associated with higher levels of psychological well-being (higher levels of quality of life and lower levels of depression, anxiety and internalized homophobia), but social support from family was not significantly associated with psychological well-being.

The importance of social support and its sources may also vary depending on the developmental stage of the individual. Whereas peer and partner support are increasingly important beginning in young adulthood, family relationships provide the primary context for adolescent development. Although most research has focused on negative parent-child or family interactions to disclosure or discovery of a child or family member’s minority sexual orientation, findings from a recent study demonstrated the protective effects of family acceptance on health. In analyses examining the associations between family acceptance in adolescence and health status in young adulthood, Ryan and colleagues (2010) found that young adults who reported low levels of family acceptance during adolescence had significantly higher levels of negative health outcomes such as depression, substance abuse and suicidal ideation and attempts. For example, participants who reported low levels of family acceptance were 2 times as likely to report suicidal ideation and suicide attempts as those who reported high levels of family acceptance. In contrast, young adults who reported high levels of family acceptance scored higher on all measures of positive adjustment and health.

Barriers to Treatment

As described in this review, there is ample evidence that sexual minority women are at greater risk than heterosexual women for hazardous drinking and that they are more likely to experience alcohol-related problems and alcohol-use disorders. There is also some evidence that sexual minority women seek help for substance abuse problems at rates disproportionately higher than heterosexual women, though the sources of this help are unclear. For example, Grella, Greenwell, Mays, and Cochran (2009) compared substance use and mental health treatment of sexual minority and heterosexual participants aged 18–64 years (N=2074) in the California Quality of Life Survey, a follow-up of the 2003 California Health Interview Survey. Among women in the full sample, a greater proportion of lesbian and bisexual women (55.3%) than heterosexual women (27.1%) had received mental health or substance use treatment in the past year (p<.001). Similarly, among women with a mental health disorder only, more lesbian/bisexual women (71.6%) than heterosexual women (55.1%) had received treatment. In contrast, among women who had a substance use disorder only, heterosexual women (50%) were more likely than lesbians (40%) to report past treatment, though this difference was not statistically significant. However, among women without either a mental health or substance use disorder lesbians were more likely to report treatment (43.7% vs. 16.9%, p<.001). As the authors note, sexual minority women (and men) may be more likely than heterosexuals to seek treatment—even in the absence of a diagnosable disorder—because of their higher exposure to discrimination, violence and other life stressors, or because of internalized societal stigma. It may also be that some forms of treatment, particularly mental health counseling, is more acceptable among sexual minority women than among heterosexual women (Hughes et al., 2005).

In a CHLEW pilot study of 63 lesbian and 57 heterosexual women (Hughes, 2003) we found that compared with heterosexual women, significantly more lesbians reported past treatment for alcohol-related problems (18% vs. 2%). We found the same proportion of treatment (18%) among sexual minority women in the subsequent larger-scale CHLEW study (Hughes et al., 2006). Given that the CHLEW study does not ask specific questions about treatment settings or providers it is not possible to determine whether women in the study sought help from traditional substance abuse providers (e.g., inpatient or outpatient substance abuse treatment) or from some other provider, such as mental health counselors/therapists.

Despite their potentially greater need for and willingness to seek treatment, sexual minority women face a number of barriers in accessing high-quality substance use and mental health care. For example, in a large study of lesbian and bisexual women, Corliss and colleagues (2006) found that less than one-half of the study participants with a history of substance-use related functional impairment had received any form of treatment; 16% had wanted but had not received treatment. Access to care refers not only to geographic availability of quality health care services but to financial, social, cultural, and structural issues associated with care (Institute of Medicine, 1993). Access issues encountered by sexual minority women include the difficulty in finding providers who are knowledgeable about and sensitive to the unique needs of sexual minority individuals and the greater likelihood that sexual minority women are uninsured or underinsured.

According to the National Survey on Substance Abuse Treatment Services (N-SSATS), 6% of 13,688 facilities surveyed in 2008 offered specialized programs for sexual minority clients. Facilities whose primary focus was a mix of substance abuse and mental health, and those operated by private-for-profit entities, were more likely than other types of facilities to offer specialized services for sexual minority clients (Substance Abuse and Mental Health Administration [SAMHSA], 2010). What constitutes ‘specialized programs’ in these facilities and whether they actually offer such programs has been questioned. In a telephone survey of programs listed in the 2003–2004 NSSATS database, Cochran, Peavy and Robohm (2007) found that 71% of agencies with listings indicating sexual minority-specific services did not in fact provide such services. Only 7.4% provided any kind of specifically tailored treatment.

The argument about whether sexual minorities need specialized treatment is a common one and is similar to the debate about whether women do better in single- or mixed-gender treatment. Based on a review of the literature on substance abuse treatment, Hodgins, El-Guebaly and Addington (1997) argue that women benefit from gender-specific treatment approaches because they differ from men in their substance use histories and patterns, victimization histories, and gender-role expectations and conflicts. In a more recent literature review, Greenfield, et al. (2007) concluded that although women-only treatment may not be more effective than mixed-gender treatment, services that address problems more common to substance abusing women, or that are designed for specific subgroups of women, are more effective. Similarly, Hicks (2000) argues that specialized treatment programs are substantially better equipped to address the unique issues of sexual minorities such as coming out; internalized homophobia; violence and discrimination; socialization, dating and intimacy; family support; and spirituality and religion. Whether these differences influence treatment outcomes is unknown, and this question will remain unanswered until well-designed clinical trials and other evaluation studies of alcohol treatment among sexual minorities are conducted.

Knowledge and attitudes of health care providers

Knowledge and attitudes of health care providers are critical components in the treatment of sexual minority clients. Like race/ethnicity, sexual orientation can increase the complexity of the counselor-client relationship and the likelihood of misunderstandings between counselors and clients. Cultural misunderstandings may lead to difficulties in communications, obscure expectations, affect quality of care, and dramatically alter the sexual minority client’s willingness or ability to continue with treatment or adhere to treatment recommendations. In an early report about alcohol treatment for lesbians, Weathers (1980) described three major types of negative interactions between alcohol-dependent lesbians and counselors. These included refusal of services if the woman’s sexual orientation was known or suspected; provision of services on a limited basis, or with attitudes not conductive to support, growth, self-disclosure, or the maintenance of sobriety; and provision of services directed toward isolating or ‘curing’ the client’s lesbianism, with little or no attention to the alcohol dependency itself (p. 146). Although the literature focusing on alcohol treatment suggests that attitudes of providers have improved over the past several decades, surveys of these providers indicate that they have limited knowledge about how to assess and treat sexual minority clients and they often do not discuss sexual orientation with their clients (Eliason, 2000; Eliason & Hughes, 2004; Hellman, Stanton, Lee, Tytun, & Vachon, 1989). Not surprisingly, research on sexual minority women and men’s experiences in substance abuse treatment has found that they have lower levels of satisfaction with treatment than heterosexuals (Drabble et al., 2005; Senreich, 2009).

Lack of sufficient health care insurance coverage

Even when sensitive, affirming health care providers are available, many sexual minority women lack the financial resources to access these providers. An increasing number of studies are documenting that sexual minorities, especially sexual minority women, are less likely than their heterosexual counterparts to have health insurance (Buchmueller & Carpenter, 2010; Diamant, Wold, Spritzer, & Gelberg, 2000; Heck, Sell, & Gorin, 2006; Owens, Riggle, & Rostosky, 2007). For example, using data from the Behavioral Risk Factor Surveillance System, Buchmueller and Carpenter (2010) compared health insurance coverage, access to care, and cancer screening for individuals in same-sex versus opposite-sex relationships. Those in same-same sex relationships were significantly less likely to have insurance coverage, were less likely to have had a check-up in the previous year, were more likely to report unmet medical needs, and were less likely to have had a recent mammogram or Pap test.

Similarly, in an on-line survey of mental health services access, Owens and colleagues (2007) found that 18% of the lesbian, gay, bisexual and transgender study participants surveyed had no health insurance. Of those who had insurance, 47% reported either that alcohol treatment services were not covered or that they were unsure about coverage. Among participants who reported that they had sought treatment in the past year, less than one-third (32%) reported that they had a choice of an affirmative provider and that the providers’ services were covered by their insurance plan.

Treatment guidelines for working with sexual minority clients

In a comprehensive review of alcohol use among lesbians and gay men published more than five years ago (Hughes, 2005), I noted that there was no evidence-based practice guidelines to assist alcohol treatment counselors in program development and service delivery for sexual minority clients (p. 306). Unfortunately, the situation has changed little since that time. Whereas the Substance Abuse and Mental Health Administration’s Center for Substance Abuse Treatment (CSAT) was a leader in this area in the 1990s and early 2000s (see e.g., Center for Substance Abuse Treatment [CSAT], 2001), relatively little attention has been given to issues related to substance abuse among sexual minorities since then. One exception is the 2006 Treatment Improvement Protocol (TIP) publication (Center for Substance Abuse Treatment [CSAT], 2006) that addresses women and ‘special populations’ such as the homeless, people with disabilities, and sexual minorities. This publication includes useful information on cultural competency and treatment issues of importance to women—such as violence, victimization, and harassment—that are particularly relevant to the treatment of sexual minority women. Another useful resource is the American Psychological Association’s recommendations for professional development and practice related to psychotherapy with lesbian, gay and bisexual clients (American Psychological Association [APA] Division 44, 2000).

The primary reason for the lack of evidence-based treatment guidelines is the fact that almost no research on treatment effectiveness has been done with sexual minorities. However, given the growing recognition of health disparities among sexual minorities at both the National Institutes of Health (NIH) and the U.S. Department of Health and Human Services (USDHHS), the situation may improve in the future. In addition to a growing number of NIH program announcements that include sexual minority health, is the historic inclusion of lesbian, gay, bisexual and transgender health in the two most recent USDHHS Healthy People documents. Healthy People is published at the beginning of each decade and guides the United States’ public health priorities and goals for the subsequent ten years. Sexual minorities were included for the first time in HP2010 as one of six population groups experiencing substantial health disparities.

Without standards of care and sufficiently trained substance abuse treatment staff, sexual minority clients who seek substance abuse treatment are likely to experience the same sort of stereotyping, stigma, harassment, and discrimination that they experience in society. The California LGBT Constituency Committee and LGBT TRISTAR (LGBT Constituency Committee and LGBT TRISTAR, 2008) compiled a document that outlines minimal standards for providing culturally and linguistically appropriate services to sexual minority clients. These standards include assurance that: 1) agency policies and procedures are inclusive of sexual minority clients, staff and communities; 2) staff members receive training about sexual minority-specific issues and that they have appropriate supervision to insure the provision of sexual-minority inclusive services; 3) language used in forms and documents and in verbal assessments and interventions are inclusive and respectful of sexual minorities; 4) agency climates are welcoming and inclusive of sexual minority clients; and 5) substance abuse treatment and prevention agencies have linkages with local sexual minority communities and use appropriate referral sources and resources for their sexual minority clients (pp. 7–10).

Conclusion

Disparities in alcohol-related problems and alcohol-use disorders do not stem from sexual orientation per se, but are a consequence of cultural and environmental factors associated with being part of a stigmatized and marginalized population group. Sexual minority clients who seek substance abuse treatment are a diverse group with a diverse range of strengths and vulnerabilities. As U.S. society becomes more and more diverse the need for culturally competent health care providers continues to grow. Treatment agencies and individual treatment providers must assume responsibility for learning about sexual minority culture and the particular issues related to sexual minority health—including risk and protective factors for hazardous drinking and drinking-related problems—while also understanding that each sexual minority client is a unique individual whose health is shaped by multiple life experiences and social contexts.