In a quantitative review of twin studies from the past fifty years, Polderman and colleagues (2015) report that ~50% of the variance in conduct disorder (broadly measured with over 200 phenotypes in 147,974 monozygotic twin pairs and 192,651 dizygotic twin pairs) is attributable to additive genetic influences. Interestingly, and in contrast to other disorders on the externalizing spectrum (Krueger et al., 2002), the results from this meta-analysis also suggest that shared environmental factors account for a significant (14%) proportion of the variance in conduct disorder (Polderman et al., 2015). To focus more narrowly on conduct disorder, we also present heritability estimates obtained from large-scale (n > 1000) population and community-based twin studies that have used reliable and valid measures of conduct disorder symptomatology or diagnoses according to DSM-III-R or DSM-IV criteria. Analyses in community and population-based samples are more likely to provide unbiased estimates of heritability compared to clinically ascertained samples, where affected individuals are over-represented. Conduct disorder is relatively common, and thus there is sufficient variation to provide reliable estimates of its heritability in population and community-based samples.
