In the eight regressions we tested, the interaction between alcohol intake and sex is not significant at the 1% level, except weakly for GMV when including the extended control variables (BMI, weight, and educational attainment). Similarly, none of the interactions between alcohol intake and age are significant at the 10% level. Only the regressions that exclude abstainers were significant at the 10% level (p = 0.034 for GMV, p = 0.0014 for WMV). Consequently, we excluded the interaction terms from the analyses of local IDPs.
