BMI, educational attainment, and weight; N = 36,678) yield similar findings, though the variance explained by alcohol intake beyond other control variables is reduced to 0.4% for GMV and 0.1% for WMV when individuals who consume a high level of alcohol are excluded (Supplementary Tables 1 and 2).Table 1Regression analysis with global IDPs as outcome variables.Dependent variable: global GMVDependent variable: global WMVN: 36,678 (df: 36,585), R2: 0.514N: 36,678 (df: 36,585), R2: 0.514VariableRegression coefficient (SE), 95% CIt-stat (p-value)Regression coefficient (SE), 95% CIt-stat (p-value)Intake−0.1095 (0.0058),CI: [−0.1209, −0.0982]−19.0(p < 1.0e−16)−0.0650 (0.0078),CI: [−0.0802, −0.0498]−8.4(p < 1.0e−16)Intake2−0.0651 (0.0037),CI: [−0.0723, −0.0579]−17.7(p < 1.0e−16)−0.0370 (0.0050),CI: [−0.0468, −0.0273]−7.5(p = 7.8e−14)Intake × male0.0174 (0.0080),CI: [0.0018, 0.0330]2.2(p = 2.9e−02)0.0164 (0.0107),CI: [−0.0046, 0.0374]1.5(p = 1.2e−01)Intake × std. age0.0080 (0.0037),CI: [0.0008, 0.0152]2.2(p = 3.0e−02)0.0111 (0.0050),CI: [0.0014, 0.0208]2.2(p = 2.5e−02)Std. age−0.5991 (0.0038),CI: [−0.6066, −0.5916]−157.0(p < 1.0e−16)−0.3213 (0.0051),CI: [−0.3313, −0.3112]−62.6(p < 1.0e−16)Std. age2−0.0378 (0.0034),CI: [−0.0445, −0.0311]−11.0(p < 1.0e−16)−0.0127 (0.0046),CI: [−0.0217, −0.0037]−2.8(p = 5.7e−03)Against model without intake and interactionsDelta R2: 0.0099, F-test: p < 1.0e−16Against model without intake and interactionsDelta R2: 0.0033, F-test: p < 1.0e−16Note. IDPs imaging derived phenotypes, GMV gray matter volume, WMV white matter volume, df degrees of freedom, SE standard error, CI confidence interval, std. age standard age.All regressions include standard
