Association of adverse childhood experiences with educational attainment and adolescent health, and the role of socioeconomic factors: analysis of a prospective cohort study

Abstract

BackgroundThere is increasing policy emphasis on adverse childhood experiences (ACEs). We aimed to assess the relative contributions of these experiences and socioeconomic factors to educational attainment and health. MethodsUsing data collected between 1991 and 2014, from a UK cohort (ALSPAC), we assessed associations of having had ACEs between birth and 16 years of age (sexual, physical, emotional abuse; emotional neglect; parental substance abuse, mental illness, criminal conviction, or separation; violence between parents; and bullying) with obtaining five or more good GCSEs (grades A* to C, n=9959, assessed through linkage to the National Pupil Database) and health at age 17 (depression, obesity, harmful alcohol use, smoking, and illicit drug use, n=4917, assessed through research clinic measurements and self-completed computer questionnaires). ACEs were assessed using both prospective (reports by main caregiver at time of ACE occurrence or very soon afterwards) and retrospective (reports by 23 year old adults looking back at their childhood) questionnaires and an ACE score was calculated by summing the number of ACE an individual experienced. We adjusted for a wide range of socioeconomic and family factors. We calculated population attributable fractions (PAF) for each outcome for four or more ACEs and key socioeconomic and demographic indicators. FindingsMost participants (84%) experienced at least one adverse experience (24% ≥4). ACE score was associated with lower educational attainment and worse adolescent health. The association with educational attainment reduced by about half after adjustment for socioeconomic and family factors; associations with adolescent health also attenuated, but to a lesser degree. We assessed PAF for each outcome according to experiencing four or more ACEs (24%), low maternal education (30%), manual social class (24%), and maternal smoking during pregnancy (27%). For depression and illicit drug use, PAFs were highest for four or more ACEs (14% and 15%, respectively). For these outcomes, the PAF for maternal smoking during pregnancy was also high (both 10%). For all other outcomes, PAFs were higher for socioeconomic factors or maternal smoking than for ACEs. InterpretationA key limitation of our study is that it is restricted to families in a population-based cohort; vulnerable groups are under-represented. Nevertheless, our analysis suggests that interventions targeted at children experiencing four or more ACEs will not necessarily focus on the groups at highest risk of adverse outcomes. Our results underline the importance of the broader socioeconomic context for policies focused on ACEs. FundingSupported by a grant from the UK Economic and Social Research Council (ES/N000382/1).