The association of alcohol dependence and consumption during adolescence with depression in young adulthood, in England: a prospective cohort study

Background: The role of alcohol use in the development of depression is unclear. We aimed to investigate whether alcohol dependence, but not high frequency or quantity of consumption, during adolescence increased the risk of depression in young adulthood. Methods: In this prospective cohort study, we included adolescents who were born to women recruited to the Avon Longitudinal Study of Parents and Children in Avon, UK, with delivery dates between April 1, 1991, and Dec 31, 1992. Alcohol dependence and consumption were measured at about age 16 years, 18 years, 19 years, 21 years, and 23 years using the self-reported Alcohol Use Disorders Identification Test, and at about age 18 years, 21 years, and 23 years using items corresponding to DSM-IV symptoms. The primary outcome was depression at age 24 years, assessed using the Clinical Interview Schedule Revised. Analyses were probit regressions between growth factors for alcohol dependence and consumption and depression, before and after adjustments for confounders: sex, housing tenure, maternal education, maternal depressive symptoms, parents' alcohol use, conduct problems at age 4 years, being bullied from age 12–16 years, and frequency of smoking cigarettes or cannabis. Adolescents were included in analyses if they had data from at least one timepoint for alcohol use and confounders. Findings: We included 3902 adolescents (2264 [58·0%] female; 1638 [42·0%] male) in our analysis, and 3727 (96·7%) of 3853 participants with data on ethnicity were White. After adjustments, we found a positive association between alcohol dependence at 18 years of age (latent intercept) and depression at 24 years of age (probit coefficient 0·13 [95% CI 0·02 to 0·25]; p=0·019), but no association between rate of change (linear slope) and depression (0·10 [–0·82 to 1·01]; p=0·84). There was no evidence of an association between alcohol consumption and depression (latent intercept probit coefficient –0·01 [–0·06 to 0·03]; p=0·60; linear slope 0·01 [–0·40 to 0·42]; p=0·96) after adjustments. Interpretation: Psychosocial or behavioural interventions that reduce the risk of alcohol dependence during adolescence could contribute to preventing depression in young adulthood. Funding: UK Medical Research Council and Alcohol Research UK (grant number MR/L022206/1).