In the AA sample, AUD was not associated with divorce. Moreover, none of the psychiatric comorbidities were associated with divorce. These null effects may partly be attributable to reduced statistical power due the smaller AA sample size. Alternatively, because the divorce analytic sample was a subsample of those who got married, we explored whether the null effects might reflect a thresholding effect of SES among those who married. However, supplemental analyses did not support this possibility. Although those in our AA sample who married had higher educational attainment than those who did not, the difference was negligible (12.64 vs. 12.26 years; Supplement, section S2). AA populations are underrepresented in studies of the social correlates and consequences of AUD and psychiatric disorders, and our findings should be considered as initial evidence. It is also possible that ethnic differences in alcohol use behaviors (Zapolski et al., 2014) and marital outcomes (Bryant et al., 2010; Dixon, 2009; Raley & Sweeney, 2020) could be driving differences in the associations of AUD and other psychiatric comorbidities and divorce observed in this study. Clearly, additional research on AUD and marital outcomes/processes in diverse populations is needed.
