In the present study, we focused on all COGA participants of European ancestry (EA) and African ancestry (AA) with genome-wide association (GWAS) data, AUD and psychiatric comorbidities phenotypes, and marital outcomes information. We defined two analytic samples within COGA for the marriage and divorce analyses. The first sample, the marriage analytic sample, included 4045 EA participants (from 986 extended families; 2253 (56%) female; Mage = 36.26 years, age range = 19–84 years) and 1550 AA participants (from 503 extended families; 863 (56%) female; Mage = 37.35 years, age range = 21–79 years). This analytic sample excluded those <23 years of age at their assessment (or most recent assessment, for those with multiple assessments) who were without a lifetime AUD diagnosis. We implemented the age minimum to ensure that participants had passed through the period of highest risk for onset of AUD before being classified as unaffected, as guided by epidemiological data regarding age of onset for severe AUD (Grant et al., 2015). Additionally, we excluded those <30 years of age at their assessment (or most recent assessment, for those with
