Influence of ascertainment strategy on finding sex differences in genetic estimates from twin studies of alcoholism

Twin studies have yielded contradictory findings about sex differences in g enetic influences on the etiology of alcoholism. Studies based on populatio n registers or epidemiological samples have yielded similar estimates of he ritability (50-60% of the total variance) for males and females. In contras t, studies of twins identified through treatment settings have found sizeab le genetic contributions to alcoholism in males but usually negligible heri tabilities for females. We investigated this discrepancy by applying a "sim ulated" treatment ascertainment strategy to data on alcohol-related disorde rs collected by structured interviews with a population-based sample of adu lt twins aged 18-56 years from the Mid-Atlantic Twin Registry. Structural m odels were used to estimate heritabilities for two definitions of treatment , and these estimates were compared with those obtained from the population -based sample. In males, heritability estimates mere similar across samplin g methods, but the treatment ascertainment methods yielded higher estimates of common environmental influences. For females, heritability estimates ba sed on a broad definition of treatment were similar to those obtained by us ing the random ascertainment design. However, estimates based on sampling w omen who had been in alcohol-treatment programs were (nonsignificantly) low er than those obtained with the other methods. These results provide partia l support for the hypothesis that differences in sampling method may accoun t for differences in heritability estimates for alcoholism among studies of female twins. (C) 2000 Wiley-Liss, Inc.