Abstract
BACKGROUND: There is not yet a consensus on the best way to conceptualise adverse childhood experiences (ACEs). We used data-driven methods across two populations to examine (a) if there were meaningful dimensions underlying ACEs and (b) whether dimensions were differentially associated with increased risk of adolescent psychopathology. METHODS: Participants were 18,539 British children from the UK Millennium Cohort Study (MCS) and 11,876 American children from the US Adolescent Brain Cognitive Development Study (ABCD). A wide range of ACEs (e.g., abuse, neglect, parental psychopathology, peer victimisation) were measured prospectively from infancy to mid-adolescence using interviews and questionnaires. Internalising and externalising symptoms were assessed with child and/or parent reports during adolescence. RESULTS: Our preregistered exploratory factor analysis revealed four latent dimensions in the MCS (parental threat, deprivation, victimisation, and parental discipline) and ABCD (parental threat, deprivation, victimisation, and traumatic events). All dimensions except deprivation were associated with increased risk for internalising and externalising symptoms. Over and above the other dimensions, victimisation was more strongly associated with internalising (MCS β = .34, 95% CI 0.33-0.36; ABCD β = .11, 95% CI 0.10-0.13) and externalising (MCS β = .31, 95% CI 0.30-0.33; ABCD β = .13, 95% CI 0.11-0.15) symptoms. CONCLUSIONS: Across two distinct populations, we found that ACEs can be captured by common underlying dimensions of parental threat, deprivation, and victimisation, as well as additional sample-specific dimensions. Our findings expand dimensional theories of childhood adversity by suggesting that in addition to threat and deprivation, victimisation is a distinct dimension of adversity that has the strongest associations with adolescent psychopathology.