How are children's perceptions of the home environment associated with a general psychopathology factor across childhood?

Background: Comorbidity and heterogeneity in psychiatric disorders may stem from a general psychopathology (p) factor influenced by both genetic and environmental factors. Although the relative contributions of these influences on psychopathology are established, the longitudinal associations between the p-factor and specific environmental exposures and the aetiology of these associations across development are not well understood. Here, we examine whether twin-rated home environment contributes to changes in the p-factor over time or, conversely, whether the p-factor influences twin-rated home environment, reflecting potential evocative gene-environment processes.

Methods: Data were obtained from the Twins Early Development Study (TEDS). Cross-lagged panel analyses were conducted separately to ascertain the direction of associations between parent-rated p, twin-rated p, and twin-rated home environment (chaos at home and parental discipline) at ages 9, 12, and 16 (N = 6,213). Biometric autoregressive cross-lagged twin models were used to assess the aetiology of these associations, and MZ differences analyses were used to control for familial effects.

Results: Both parent-rated and twin-rated p-factor and twin-rated home environment were stable over time, although the twin-rated p-factor (r = .44 [0.42, 0.46]-.40 [0.37, 0.41]) was more variable than the parent-rated p-factor (r = .72 [0.71, 0.74]-.63 [0.61, 0.64]). Twin-rated home environment was more variable than p-factor in all cross-lagged models (phenotypic and MZ differences). Small, significant bidirectional associations were found between the p-factor and twin-rated home environment, with stronger cross-lagged paths from the p-factor to the twin-rated home environment than vice versa. These longitudinal associations persisted over time, though attenuated for parent-rated p-factor. Genetic analyses revealed that bidirectional cross-lagged paths were largely explained by shared environmental factors, with a smaller proportion explained by genetic factors. This pattern of results was confirmed in MZ difference analyses.

Conclusions: Our findings suggest a dynamic and bidirectional relationship between p-factor and twin-rated home environment across development, predominantly influenced by shared environmental factors. Changes in one can affect the other, highlighting the complexity of psychopathology's environmental influences. This underscores the need for further investigation into gene-environment interplay to inform prevention and intervention strategies for psychopathology.