This study builds upon previous epidemiological studies that have shown higher prevalences of AOD and MH disorders among sexual minority populations [7
]; we extend these findings by showing that treatment utilization for these disorders varies by both gender and sexual orientation. The study findings are strengthened by the use of a population-based sample and a theoretically guided model of health services utilization. In the broader literature it is well known that health services utilization is greater among women generally. Here we have shown that minority sexual orientation is also an important explanatory variable in understanding treatment seeking among women. Lesbians and bisexual women appear to be approximately twice as likely as heterosexual women to report having received recent treatment for mental health or substance use disorders, after controlling for the presence of either type of disorder and other predisposing and enabling characteristics. Indeed, more than half of the lesbians and bisexual women in the study indicated that they had received services in the past year for mental health or substance use-related problems. Further, this sexual-orientation-related effect was also seen among gay and bisexual men who were significantly more likely than both heterosexual men and women to report having received recent treatment, after controlling for other factors.
The greater propensity for treatment use among those possessing a minority sexual orientation may be related to several factors. These include differential norms that promote help-seeking among sexual minorities in general, particularly among lesbians and bisexual women, as well as higher exposure to discrimination, violence, and other stressful life events [8
]. Further, the pervasive and historically rooted societal pathologizing of homosexuality [53
], particularly among racial/ethnic minorities by their communities, may contribute to this propensity for treatment by construing homosexuality and issues associated with it as mental health problems. This cultural definition may indirectly function as a predisposing factor that encourages the seeking of professional help for problems that are assumed to derive from individual distress, or from the internalization of the stigma ascribed to homosexuality by some [58
]. Further, the culture of gay and lesbian communities may increase the social norms and expectations that therapeutic services are appropriate places for coping with the stresses associated with being a sexual minority.
As anticipated, rates of receiving treatment varied by severity of the disorders that occurred during the period of interest. It is reassuring, for example, that nearly three quarters of individuals meeting criteria for both substance use and mental health disorders indicated that they had received at least some services in the past year. At the same time, nearly 20% of individuals who did not have a diagnosable disorder in the past year reported having received some form of mental health and/or substance abuse-related services. This finding is consistent with national surveys showing that many individuals who receive mental health treatment do not have a diagnosable disorder [60
], but may have other symptoms, such as psychological distress or impairments in functioning, that lead them to seek care [60
]. Moreover, these findings have called into question the criteria that should be used to indicate "need for treatment," apart from diagnostic criteria, as well as the basis for determining the adequacy of the treatment system in providing treatment to those who feel they need it (including those with and without diagnosed disorders) [63
]. This is a particularly salient issue for understanding treatment utilization among sexual orientation minorities, many of whom in this study sought services in the absence of evidence of either a mental health or substance use disorder. Why this is so is unclear but suggests either an over-utilization of care or that estimates of unmet need in this population are less dependent on the presence of diagnosed disorders. Moreover, this finding has implication for estimating need for health services, which is typically based on prevalence estimates of disorders.
None of the enabling characteristics that have been associated with treatment seeking in other studies (i.e., employment, insurance, education, social support, and marital/partner status) were significantly related to treatment use in the multivariate models. It is possible that the effects of disorder and sexual orientation cancelled out any effects associated with these factors. However, we observed that ethnic/racial minorities were less likely to utilize mental health or substance use related services. This effect was found after controlling for differences in morbidity and other predisposing and enabling characteristics, including health insurance, which have been associated with underutilization of these services among ethnic minorities in prior research [64
]. African Americans and Hispanics may underutilize services for mental health and substance use problems for a variety of reasons, including a lack of familiarity with the types of services available [67
]; prior negative experiences with service providers [68
]; or because of greater stigma attached to use of these services by their families and communities [69
]. Further, there are differences among women in utilization of these services by both race/ethnicity and sexual orientation [71
]. Exploration of the interactions among gender, sexual orientation, and race/ethnicity on treatment use is beyond the scope of the present paper, but is an area in need of more investigation.
This study encountered several limitations typical of telephone-based follow-back surveys. The California Quality of Life Survey sample was recruited by recontacting those 2003 CHIS respondents who had agreed to be recontacted, using the telephone number associated with the original interview. Loss to follow-up was most often due to mobility from the original residence and was associated with younger age. Thus our estimates of the relationship between age and treatment received may be imprecise; other factors associated with lack of contact for the follow-up survey may also have influenced the estimates derived from the study sample. Although the follow-back survey oversampled for sexual minorities, the cell sizes for groups defined by sexual orientation and type of disorder (particularly among those with an AOD disorder only or with both MH and AOD disorders) were small (approximately 78 cases). Hence, statistical power was somewhat limited and may have failed to detect some relationships among sexual orientation, type of disorder, and treatment received. Lastly, although the study findings may be generalized to the general population in California, the dependent variable of interest, treatment seeking, may be particularly influenced by the cultural context of California, in which therapeutic interventions are consistent with an overall "therapy culture" [73
], thus limiting generalizability to other locations that differ in this regard.