Our sample size provides us statistical power of 90% to detect effects as small as f2 < 0.00078 at the 5% significance level, after accounting for multiple hypotheses testing (puncorrected < 1.64 × 10−4). Given previous findings, we hypothesized a negative relationship between alcohol intake and global GMV and WMV in individuals who consume large amounts of alcohol (i.e., females who report consuming more than 18 units/week and males who report consuming more than 24 units/week). The large general population sample provided sufficient sensitivity to qualitatively and quantitatively assess how associations vary across the drinking spectrum and test at which threshold associations emerge. Our design also allowed us to explore whether the associations between alcohol intake and GMV and WM microstructure are localized in specific regions or widespread across the brain and compare the associations across various WM integrity indicators.
