How do early-life adverse childhood experiences mediate the relationship between childhood socioeconomic conditions and adolescent health outcomes in the UK?

Abstract

Background: Both adverse childhood experiences (ACEs) and adverse childhood socioeconomic conditions (SECs) in early life are associated with poor outcomes across the life course. However, the complex interrelationships between childhood SECs and ACEs are unclear, as are the consequences for health outcomes beyond childhood. We therefore assessed the extent to which early-life ACEs mediate the relationship between SECs and socioemotional behavioural problems, cognitive disability and overweight/obesity in adolescence.

Methods: We used longitudinal data from the UK Millennium Cohort Study (MSC). Outcomes assessed at age 14 were socioemotional behavioural problems, cognitive disability and overweight/obesity. SECs at birth were measured by maternal education. Potentially mediating ACEs measured up to 5 years were verbal and physical maltreatment, parental drug use, domestic violence, parental divorce, maternal mental illness and high frequency of parental alcohol use. We used counterfactual mediation analysis to assess the extent to which ACEs mediate the association between SECs at birth and behavioural, cognitive and physical outcomes at age 14, estimating total (TE), natural direct and indirect effects, and mediated proportions.

Results: Children with disadvantaged SECs were more likely to have socioemotional behavioural problems (relative risk (RR) 3.85, 95% CI 2.48 to 5.97), cognitive disability (RR 3.87, 95% CI 2.33 to 6.43) and overweight/obesity (RR 1.61, 95% CI 1.32 to 1.95), compared to those with more advantaged SECs. Overall, 18% of the TE of SECs on socioemotional behavioural problems was mediated through all ACEs investigated. For cognitive disability and overweight/obese, the proportions mediated were 13% and 19%, respectively.

Conclusion: ACEs measured up to age 5 years in the MCS explained about one-sixth of inequalities in adolescents behavioural, cognitive and physical outcomes.

Keywords: Health inequalities; child health; cohort studies; public health.

Figure 1

Direct acyclic graph of the natural direct effect of the exposure to outcomes at age 14, and the natural indirect effect throughout mediators by the age 5. Model adjusted for potential baseline confounders.

Figure 2

Counterfactual mediation analysis. Relative risk (95% CI) of natural direct effect (95% CI), natural indirect effect (95% CI) and total effect (95% CI) for relative index of inequality by all adverse childhood experiences for behavioural, cognitive and overweight/obese outcomes at age 14.