As the zygosity difference indicated the role of overlapping genetic but not individual-specific environmental influences on the covariance between EAU and NSSI, we fit a bivariate twin model in Mx (Neale, 2004) to all available same-sex twin data. EAU was moderately heritable (0.33, 95% C.I. 0.10 – 0.57) with the remainder of the variance attributable to shared (0.22, [95% C.I. 0.02-0.41]) and individual-specific environmental (0.45, [95% C.I. 0.35-0.60]) factors. Consistent with prior work, 54% (95% CI: 0.40 – 0.65) of the variance in NSSI could be attributed to broad sense heritability (A+D; see Maciejewski et al., 2014) with the remainder accounted for by individual-specific environmental factors (0.46 [95% CI: 0.35-0.60]). In line with the findings from the discordant twin modeling, a moderate genetic correlation emerged between EAU and NSSI (rG=0.35, 95% CI: 0.14-0.70). The corresponding environmental correlation was modest (rE=0.24, 95% CI: 0.05-0.42). Power to resolve whether genetic or individual-specific environmental influences exclusively influenced the covariation was limited as both rG and rE could be individually constrained to zero but not both.
