Abstract
Background: Prospective data on the natural history of alcohol-use disorders (AUD) from adolescence into middle adulthood are scarce. This study aims to describe the prevalence, incidence, and remission of Diagnostic and Statistical Manual of Mental Disorders (DSM)-5 AUD from adolescence into middle adulthood and thereafter characterise those who do and do not develop AUD symptoms or experience remission.
Methods: In this prospective cohort study, data were drawn from the Australian 11-wave population-based Victorian Adolescent Health Cohort Study (N=1943; 1000 female and 943 male participants). Between wave 7 (1998) and wave 11 (2019-21), we assessed 12-month DSM-5 AUD symptoms at age 21 years and 24 years (late adolescence), age 29 years and 35 years (young adulthood), and age 42 years (early middle adulthood) using the Composite International Diagnostic Interview (CIDI). The CIDI was administered by a trained interviewer, with AUD symptoms scored to align with DSM-5 AUD clinical diagnosis. We estimated incidence and incidence rate using flexible parametric survival models, and symptom prevalence and remission from symptoms using proportions, with and without cessation of heavy drinking. We used risk ratios and relative risk ratios from univariable generalised linear regression models to describe participant characteristics associated with symptoms of AUD and remission. Multiple imputation was used to address missing data.
Findings: At any wave between ages 21 years and 42 years, estimated cumulative incidence of AUD symptoms using multiply imputed data was 58·0% (95% CI 52·3-63·8), and highest in male individuals (71·6% [65·1-78·1]). Incidence and prevalence increased markedly from age 21 years to 24 years, peaking at age 24 years, then decreased and stabilised across the subsequent assessment waves at age 29 years, 35 years, and 42 years. By age 42 years, 25·0% (95% CI 21·2-28·8) of the population had either ongoing or middle-adulthood-onset AUD, 11-13% had persistent AUD symptoms from late adolescence (age 21-24 years), and most experienced remission from AUD (67·0% [61·1-73·0]), with a majority also reporting cessation from heavy drinking (63·6% [58·6-65·5]). Remission from AUD was most common between the assessment waves at age 29 years and age 35 years (43·9% [95% CI 34·5-53·3]), and in female individuals (55·3% [42·6-67·9]), with a distinct reduction in remission for male individuals between the assessment waves at age 35 years and age 42 years (32·4% [24·8-39·9]). In both male and female individuals, remission from AUD was more common in those with higher education, in stable relationships, or without long-term other substance use.
Interpretation: Cumulative incidence of AUD from late adolescence to middle adulthood is high. Although most individuals with AUD remit by age 42 years, greater investment in public health prevention and health service responses is needed for those with persistent AUD, particularly in male individuals.