To estimate the power of our sample to detect a statistical association between the OPRM1 genotype and DSM-IV-based diagnoses and alcohol consumption, a power calculation was performed using The Genetic Power Calculator (http://pngu.mgh.harvard.edu/~purcell/gpc/) (Purcell et al., 2003) and the Quanto 1.2.4 software (http://hydra.usc.edu/gxe/). In all power calculations, type I error probability was set to 0.05 (two tailed). For the diagnosis-based phenotypes, a ≥80% statistical power to detect allele frequency differences was reached in the Health 2000 study with allele relative risk values of 1.4–2.0. For the consumption-related phenotypes, a ≥80% statistical power to detect associations was reached should the variant explain 0.35% (FINRISK with n = 2360) and 0.65% (HBCS with n = 1384) and 0.80% (Health 2000 with n = 1006) of the phenotypic variation.
