We further tested if CAD subjective measures were associated with OPRM1 and DAT1 polymorphisms. We constructed mixed-effect models with random intercept. The dependent variables in these models were BAES stimulation subscale, BAES sedation subscale, positive effects on DEQ VAS, and negative effects on DEQ VAS, SHAS score, and POMS-TA subscale. The independent variables included DAT1 genotype (A10 vs A9), OPRM1 genotype (AA vs *G), and time, as categorical variables; 2-order and 3-order interactions between DAT1 genotype, OPRM1 genotype, and time were also included in the model. Since there were genotype differences on pre-assessment TLFB drinking (see results), we included number of drinking days in all subsequent models as a covariate. Other covariates included were sex and AIM component values (C1, C2). We also compared the BAC peak and area under the curve (AUC) between the genotypes using GLRM.
