The Influence of Strain and Social Support on Substance Use and Depression in Urban Racial Ethnic and Sexual Minority Youth

Purpose: Lesbian, gay, and bisexual (LGB) youth experience higher rates of depression and substance use than their non-LGB peers (Savin-Williams, 1998; Safren & Heimberg, 1999). Strains such as LGB-related discrimination and traumatic life events increase their risk. According to strain theory, support from family and peers is hypothesized to condition strain's effect in reducing negative outcomes (Agnew, 1992), though the few studies that have explored the impact of social support on LGB youth report mixed results (Wright & Perry, 2006; Williams et al., 2005). In addition, LGB youth from different racial/ethnic backgrounds are not at equal risk and same-gender attraction may function as a protective factor for minority LGBT youth (Cionsolacio et al., 2004; Battle et al. 2006). This study examined the link between sources of strain, depression and substance use and how such relationships may vary as a function of social support and race/ethnicity.

Methods: Two hundred and seventy-three primarily Latino/Hispanic, African American and Black Caribbean LGB youth in South Florida were identified using venue based sampling in a large-scale survey study conducted at 15 high schools. Demographic measures of age, race/ethnicity and sexual orientation as well as family and peer support scales were utilized. Strain measures included traumatic life events (Eitle, 2002; Turner & Wheaton, 1995) and perceived discrimination (Kessler, Mikelson & Williams, 1999). Structural equation modeling (SEM) techniques were used to test multivariate relationships of variables, resulting in a model for understanding the influence of strain and social support on depression and substance use. SPSS, AMOS and MPlus software assisted in analysis to obtain the best fitting model.

Results: The majority of participants identified as Hispanics (55%), Caribbean Black (22%) and African-American (20%) with a mean age of 16. Despite the presence of strains such as traumatic life events and discrimination based on perceived sexual orientation, lower than expected levels of substance use and suicidal ideation were reported. The best fitting model resulting from the SEM analysis had values of x2 (14, p = > .725) = 10.49; CFI = 1.00; GFI =.987; RMSEA = .000 with a p-value of close-fit of .905. Family support partially mediated both depression and substance abuse for all groups. Stressful life events were found to correlate with higher levels of both substance abuse and depression. Experiences of discrimination based on perceived sexual orientation impacted levels of depression for Latinos but not for African-Americans or Caribbean Blacks. The presence of peer support mediated depression for bisexuals but was not as critical for other populations indicating that certain strains may not exacerbate negative outcomes as much as predicted.

Conclusions and Implications: The minimal amount of substance abuse and suicidal ideation found in this sample contrasts with other studies of primarily white LGB youth. The findings from this study indicate that research and practice efforts should address the cumulative impact of strain, including discrimination, and strive to build on strengths within communities of color.