UC Berkeley

INTRODUCTION: Sexual minority (e.g., lesbian, bisexual) women experience depression and alcohol use disorder at approximately twice the rates of heterosexual women. Though stigma serves as a common explanation for these disparities, less is known about the mechanisms through which stigma may contribute to these disparities. Past research has found that the strategies that individuals use to regulate their emotions in response to experiences of stigma may influence their depression symptoms and alcohol use symptoms. Among general samples, alcohol use and suppression have been linked to increased risk for depression symptoms whereas seeking social support following a stigma-related stressor has been linked to reduced risk for depression symptoms. A limited body of past research has found constructs similar to suppression and social support to be associated with greater or fewer alcohol use symptoms, respectively, at the same level of alcohol use.

The current dissertation proposes and validates a model to test whether greater levels of alcohol use, greater levels of suppression, and lower levels of social support explain 1) sexual minority women’s disproportionally high rates of depression symptoms/alcohol use symptoms and 2) the relationship between stigma-related stressors and depression symptoms/alcohol use symptoms for sexual minority women. (a) A greater number of stigma-related stressors are hypothesized to be associated with higher levels of suppression, higher levels of alcohol use, and lower levels of social support, (b) which are hypothesized to be associated with a greater number of depression symptoms. (c) Higher levels of alcohol use are hypothesized to be associated with a greater number of alcohol use symptoms; (d) higher levels of suppression are hypothesized to weaken this relationship, whereas higher levels of social support are hypothesized to strengthen this relationship.

METHODS: A national sample of 1,773 sexual minority women participated in an online survey in 2017 through the PRIDE Study. Respondents self-reported demographic information, number of experiences of stigma, levels of alcohol use, suppression, and social support, and number of depression and alcohol use symptoms. Structural equation modeling was used to test hypotheses a to d (detailed above). Chi-square, RMSEA, CFI, and TLI were all employed to assess model fit.

RESULTS: (a) As hypothesized, a greater number of stigma-related stressors was associated with lower levels of social support (standardized coefficient estimate, β = -0.185; p < 0.001), but contrary to hypothesis, the number of stigma-related stressors was not significantly associated with levels alcohol use or suppression. (b) As hypothesized, lower levels of social support (β = -0.210; p < 0.001), higher levels of alcohol use (β = 0.105; p < 0.001), higher levels of suppression (β = 0.169; p < 0.001), and greater number of stigma-related stressors (β = 0.220; p < 0.001) were all significantly associated with a greater number of depression symptoms. (c). As hypothesized, higher levels of alcohol use (β = 0.790; p < 0.001) were significantly associated with a greater number of alcohol use symptoms, but contrary to hypothesis, greater numbers of stigma-related stressors did not predict a greater number of alcohol use symptoms. (d) Contrary to hypothesis, there was no evidence that relationship between alcohol use and alcohol use symptoms was strengthened by higher levels of suppression or weakened by higher levels of social support. All fit statistics exceeded established standards.

DISCUSSION: Stigma-related stressors were found to be associated with depression symptoms, but not alcohol use symptoms, among sexual minority women. Stigma-related stressors may cause sexual minority women to deplete their social support resources, potentially increasing risk for depression symptoms; this mechanism does not appear to drive alcohol use symptoms. This finding suggests that cultivating social support networks that are responsive to experiences of stigma may be an important component of depression treatments for sexual minority women. Though the cross-sectional design of this study limits the potential for causal inference, this study makes an important contribution to the literature by utilizing a largescale, national sample of sexual minority women to evaluate potential mechanisms driving both depression symptoms and alcohol use symptoms among this vulnerable population. Future research should use longitudinal methods to improve potential for causal inference. Future research should also identify mechanisms common to depression symptoms and alcohol use symptoms to facilitate the development of interventions for co-occurring depression and alcohol use disorders for this vulnerable population.