Patterns of childhood adversity and health outcomes in early adolescence: Results from the Generation XXI cohort

This longitudinal population-based birth cohort study aims to identify childhood adversity patterns over the first 13 years of life and to examine its association with health-related behaviours and outcomes in early adolescence. Using data from the Portuguese birth cohort Generation XXI, we performed latent class analysis to explore the underlying patterns of adversity from birth to early adolescence, using 13 adversity items assessed in five time points. Health-related behaviours and outcomes were evaluated at 13 years. Adjusting for parental unemployment, logistic regression models were performed to determine the association between adversity patterns and outcomes. Among 8647 participants, three adversity patterns were identified: "low adversity" (56.1%), "household dysfunction" (17.2%) and "multiple adversities" (26.7%). For the "household dysfunction" pattern, girls and boys showed associations with increased odds of alcohol/tobacco use (adjusted odds ratio [AOR]: 1.78; 95% confidence interval [CI]: 1.32-2.40; AOR:1.84; CI:1.38-2.46, respectively) and depressive symptoms (AOR:2.34; CI:1.58-3.48; AOR:5.45; CI:2.86-10.38, respectively). Boys also presented low consumption of fruits/vegetables (AOR:1.51; CI:1.04-2.19). For the "multiple adversities" pattern, both girls and boys showed an increased probability of alcohol/tobacco use (AOR:1.82; CI:1.42-2.33; AOR:1.63; CI:1.30-2.05, respectively) and depressive symptoms (AOR:3.41; CI:2.46-4.72; AOR:5.21; CI:2.91-9.32, respectively). Boys also revealed increased odds of low consumption of fruits/vegetables (AOR:1.67; CI:1.24-2.23). Childhood adversity patterns are associated with unhealthy behaviours and depressive symptoms in early adolescence. Public policies and early interventions targeting vulnerable children, families and communities can potentially reduce the detrimental effects of adversities on health and promote individual and community resilience.