et al., 2012). In a meta-analysis including 10 studies with alcohol dependence as a main diagnosis, no significant association was found between the polymorphism and alcohol dependence (Arias et al., 2006), whereas a more recent meta-analysis including 12 studies, 5 conducted in Asians and 7 in Caucasians, indicated an association in Asians, but not in Caucasians (Chen et al., 2012). No association was observed in two family-based studies (Franke et al., 2001; Xuei et al., 2007). It has been speculated that the lack of association in many molecular epidemiological studies may result from small sample sizes, ethnic heterogeneity of the study populations or sample selection bias occurring due to the fact that the patients of the case–control studies generally consist of the treatment-seeking subpopulation of alcoholics, which may not be (genetically) representative of alcohol-dependent individuals in general (True et al., 1996; van der Zwaluw et al., 2007). Consequently, large population-based studies in homogeneous populations are warranted.
