Discrimination and alcohol-related problems among college students: A prospective examination of mediating effects

Abstract

Background

Discrimination is a risk factor for health-risk behaviors, including alcohol abuse. Far less is known about the mechanisms through which discrimination leads to alcohol-related problems, particularly during high-risk developmental periods such as young adulthood.

Methods

The present study tested a mediation model using prospective data from a large, diverse sample of 1539 college students. This model hypothesized that discrimination would be associated with established cognitive (positive alcohol expectancies) and affective (negative affect and coping motives) risk factors for alcohol-related problems, which would account for the prospective association between discrimination and alcohol problems.

Results

Structural equation modeling indicated that discrimination was associated cross-sectionally with negative affect and more coping motives for drinking, but not with greater alcohol expectancies. Coping motives mediated the prospective relationship between discrimination and alcohol-related problems. Additionally, results indicated significant indirect effects from discrimination to alcohol-related problems through negative affect and coping motives. These associations were evident for multiple groups confronting status-based discrimination, including women, racial/ethnic minorities, and lesbian/gay/bisexual individuals.

Conclusions

This study identified potential affective mechanisms linking discrimination to alcohol-related problems. Results suggest several avenues for prevention and intervention efforts with individuals from socially disadvantaged groups.

Introduction

The role of stress in the etiology of psychopathology (Brown, 1993, Dohrenwend, 2000) and substance use and abuse (Dawson et al., 2005, Hasin et al., 2007a, Hasin et al., 2007b) has been well documented. Discrimination has been recognized as one of the most important stressful life experiences that serve as risk factors for mental health problems (Kessler et al., 1999). Although the majority of this research has focused on symptoms of psychological distress, there is accumulating evidence that discrimination may also contribute to the development of health-risk behaviors, including alcohol use and problem use, which refers to a pattern of consumption resulting in negative consequences (White and Labouvie, 1989). Indeed, cross-sectional studies of adults have documented associations between discrimination and alcohol use and problem use among several groups, including racial/ethnic minorities (Borell et al., 2007, Gee et al., 2007, Yoo et al., 2009) and lesbian, gay, bisexual (LGB) individuals (McCabe et al., 2010, McKirnan and Peterson, 1988).

These studies have provided important information, but the development of alcohol use/misuse and associated problems often occurs much earlier in the life course. Consequently, prospective studies that focus on developmental periods of heightened risk, particularly young adults and college students (Grant et al., 2004, Wechsler and Isaac, 1992), are needed. In addition to evaluating factors that confer risk, it is also critical to understand the mechanisms explaining this risk. That is, how is it that discrimination “gets under the skin” and leads to the development of alcohol-related problems? Recent review articles on discrimination and health have pointed to the paucity of research examining the pathways through which discrimination leads to the development of health problems, including alcohol use and misuse (Paradies, 2006, Williams and Mohammed, 2009).

To address these research questions, this paper draws on a recently developed theoretical framework for understanding how discrimination results in adverse mental and behavioral health outcomes among members of socially disadvantaged groups, including women and LGB populations (Hatzenbuehler, 2009, Hatzenbuehler et al., 2010). According to this framework, the stress associated with discrimination renders individuals more vulnerable to established psychological processes that confer risk for negative health outcomes. In turn, these processes mediate the discrimination-health association. In order to test the predictive validity of this mediation framework with respect to alcohol-related problems, it is necessary to identify psychological processes that are (1) established cognitive and affective risk factors for alcohol-related problems and (2) probable sequelae of discrimination.

Existing research provides support for both of these criteria. First, there is a large literature on established cognitive and affective processes that contribute to higher rates of alcohol consumption and associated problems among college students (see Sher et al., 2005), including negative affect (Hussong et al., 2001, Kassel et al., 2000), coping motives (Ham and Hope, 2003, Kuntsche et al., 2005), and alcohol expectancies (Goldsmith et al., 2009, Larimer et al., 2004). Second, discrimination is conceptualized as a specific stressor (e.g., Clark et al., 1999, Meyer, 2003a, Williams et al., 2003), and there is a considerable literature on the relations between general life stressors and these established cognitive and affective processes.

For instance, negative affect has been shown to mediate the stress-alcohol use relationship (Hussong and Chassin, 1994), and discrimination is associated with negative affect, including psychological distress (Diaz et al., 2001) and DSM-defined major depression and anxiety disorders (Kessler et al., 1999, Mays and Cochran, 2001). Importantly, one prospective study of African American adults demonstrated that the association between discrimination and substance use (including alcohol, tobacco, and other substances) was mediated by symptoms of depression and anxiety (Gibbons et al., 2004). Another affective risk factor for alcohol use and problem use is coping motives, which refer to the “strategic use of alcohol to escape, avoid, or otherwise regulate negative emotions” (Cooper et al., 1995, p. 991). Stress is associated with stronger coping motives, which in turn account for the relationship between stress and increased alcohol consumption among college students (Ham and Hope, 2003, Park et al., 2004). Coping motives have been hypothesized to be a mechanism through which stress related to having a stigmatized identity leads to alcohol use (Bux, 1996), but no studies have empirically examined this pathway.

In addition to affective risk factors, cognitive processes such as alcohol expectancies may also serve as explanations for higher rates of drinking among those exposed to discrimination. According to alcohol expectancy theory (Goldman et al., 1987), the combination of strong positive outcome expectancies (expectations of positive and negative reinforcement from drinking alcohol such as increased sociability and decreased tension) together with low negative expectancies (e.g., that alcohol will lead to cognitive or behavioral impairment) will lead to increased consumption and problems. Results from two recent studies are consistent with the hypothesis that discrimination may lead to the development of alcohol expectancies that in turn place people at risk for alcohol use and misuse. Hatzenbuehler et al. (2008) found that positive alcohol expectancies mediated the prospective relation between LGB status and alcohol use among LGB young adults. A cross-sectional study of Latino youth showed that perceived discrimination was associated with more positive substance use expectancies (Kulis et al., 2009), but this study did not evaluate whether positive expectancies mediated the association between discrimination and substance use.

Together, multiple lines of research suggest that individual differences in negative affect, drinking motives, and alcohol expectancies may result from experiences of discrimination (Hatzenbuehler, 2009), but no study has tested this combination of risk factors as mediators of the effects of discrimination on alcohol-related problems during a developmental period of risk. The present study used prospective data from a large, diverse sample of college students (N = 1539) to evaluate this hypothesis. The study had four primary aims. First, we sought to establish whether discrimination was a predictor of alcohol-related problems. Second, we examined whether discrimination was associated with established cognitive and affective risk factors for alcohol problems. In particular, drawing on prior research on coping motives (Cooper et al., 1995), it was hypothesized that discrimination would lead to greater positive alcohol expectancies and negative affect, which in turn would predict more coping motives for drinking. Third, we determined whether these cognitive (alcohol expectancies) and affective (negative affect and coping motives) processes predicted greater alcohol-related problems, thereby mediating the prospective association between discrimination and alcohol-related problems. Because women (Kawachi et al., 1999), LGBs (Meyer, 2003a), and racial/ethnic minorities (Williams et al., 2003) have consistently been identified as confronting status-based discrimination, we used multi-group analyses to evaluate the generalizability of the mediating pathways across these different groups.