Confirming prior findings (Carpenter and Hasin, 2001; Epstein et al., 2002; Grant et al., 2004a, 2004b; Kesslet et al., 1997; Moss et al., 2007; Pombo and Lesch, 2009; Penick et al., 1999; Regier et al., 1990; Windle and Scheidt, 2004), we found that the adjusted associations with lifetime alcohol dependence were lowest for IO, higher for EO and highest for BIE. The associations between psychopathology and dependence were not significantly reduced among lifetime drinkers, indicating that the increased risk of dependence associated with internalizing and externalizing psychopathology does not solely reflect an increased probability of being a drinker. The reduction in the ORs for externalizing psychopathology after controlling for volume of consumption suggests that the early and heavy drinking associated with externalizing symptomatology (Englund et al., 2008; King and Chassin, 2007; Merline et al., 2008; Maggs et al., 2008; Sartor et al., 2007) mediates some of the increased risk of alcohol dependence. However, the positive associations that remained even after controlling for volume of consumption suggest possible roles of shared genetic liability or residual associations between externalizing psychopathology and drinking
