Abstract
AIMS: To measure the associations between racial discrimination and distinct alcohol and other drug use outcomes in minoritised racial/ethnic groups and to explore the moderating roles of demographic and methodological characteristics. METHODS: Quantitative studies including racial discrimination as an exposure (both binary and continuous), an alcohol and/or other drug use outcome and a minoritised racial/ethnic sample were identified via database, citation and journal searching. 130 studies contributing 273 effect sizes, across seventeen distinct outcomes, were included in this analysis. Random-effects meta-analytic models were implemented. Moderation effects were explored using subgroup analyses. RESULTS: Racial discrimination was positively associated with sixteen alcohol and other drug use outcomes. The strongest associations were observed for at-risk/hazardous alcohol use [r = 0.24, 95% confidence interval (CI) = 0.17-0.3, I(2) = 94.8%, m = 29, n = 9445], at-risk/hazardous cannabis use (r = 0.24, 95% CI = 0.18-0.29, I(2) = 0%, m = 4, n = 462) and substance use disorder (r = 0.25, 95% CI = 0.14-0.36, I(2) = 97.7%, m = 5, n = 21 051). Considerable heterogeneity was observed across fourteen outcomes (I(2) = 69.5%-97.7%). Concerning tobacco use, Indigenous North Americans had the largest effect (r = 0.27, 95% CI = 0.2-0.35, I(2) = 0%, m = 2, n = 529), followed by Black Americans (r = 0.06, 95% CI = 0.01-0.12, I(2) = 81.7%, m = 7, n = 5409). Little evidence for an association was found for Latinxs (r = 0.06, 95% CI = -0.02 to 0.14, I(2) = 89.2%, m = 3, n = 5404) or Asian Americans (r = -0.18, 95% CI = -0.8 to 0.43, I(2) = 99%, m = 2, n = 572). Regarding composite substance use, Indigenous North Americans had the strongest associations (r = 0.29, 95% CI = 0.23-0.35, I(2) = 0%, m = 3, n = 778), followed by Black Americans (r = 0.13, 95% CI = 0.09-0.18, I(2) = 62.8%, m = 7, n = 5981) and then Latinxs (r = 0.07, 95% CI = -0.17 to 0.31, I(2) = 91.3%, m = 4, n = 1646). Concerning alcohol use problems, younger samples produced stronger associations (r = 0.28, 95% CI = 0.17-0.38, I(2) = 38.8%, m = 3, n = 483), while older samples showed larger effects in six other outcomes (rs = 0.13-0.26). Regarding at-risk/hazardous alcohol use and alcohol use problems/consequences, cross-sectional studies (rs = 0.23-0.24) produced stronger associations than longitudinal studies (rs = 0.13-0.14). Concerning tobacco and illicit substance use, the strongest associations were identified for lifetime exposure (rs = 0.18-0.32). CONCLUSIONS: Racial discrimination appears to be a consistent correlate of multiple alcohol and other drug use outcomes in minoritised racial/ethnic groups, predominantly based in the United States, yet the magnitude of these associations differs across outcomes. Demographic and methodological characteristics somewhat moderate these associations.