Chunk #19 — 3. Results — 3.2 Familial Influences On and Overlap Between Childhood Maltreatment and Cannabis Phenotypes — 3.2.1 Influences on Individual Phenotypes
Based on the twin-pair correlations (Table 2), a trivariate model with A, C, and E factors was selected as the base model, with variance components and thresholds estimated separately for EAs and AAs. Proportions of variance for the final model are presented in Table 3 and Figures 1a and 1b. Supplemental Table S22 contains models tested and fit indices, and Supplemental Figures S1a and S1b2 contain the standardized parameter estimates from the final models. In the trivariate model, there was no evidence for significant genetic influences on childhood maltreatment in either EAs or AAs; remaining estimates of C and E could be equated across racial/ethnic groups (C=73% and E=27% of variance). A influenced cannabis initiation (60%) and problems (46%) to the same degree in EAs and AAs. In EAs, the remainder of the variance was due to C (initiation=18%; problems=23%) and E (initiation=22%, problems=31%), whereas in AAs, only E contributed to non-genetic sources of variance (initiation=40%; problems=54%).
