A major problem in combining alcohol consumption data across studies has been the variability in the way these data have been defined and measured (Agrawal et al. 2012). A strength of the present study is that the meta-analysis incorporated two datasets in which the same assessment tool and quantitative measure of excessive alcohol consumption was employed. Despite the same assessment tool, a potential weakness of our study is the differences in ascertainment between the two studies: multiply affected families versus a case control design. These two designs may enrich for different risk factors. In particular, the large families in COGA may be enriched for rare highly penetrant variants, for which we likely have low power to replicate in a case control design.
