Multiple behavioural impulsivity tasks predict prospective alcohol involvement in adolescents

Abstract

Aims: We investigated reciprocal prospective relationships between multiple behavioural impulsivity tasks (assessing delay discounting, risk-taking and disinhibition) and alcohol involvement (consumption, drunkenness and problems) among adolescents. We hypothesized that performance on the tasks would predict subsequent alcohol involvement, and that alcohol involvement would lead to increases in behavioural impulsivity over time. Design: Cross-lagged prospective design in which impulsivity and alcohol involvement were assessed five times over 2 years (once every 6 months, on average). Setting: Classrooms in secondary schools in North West England. Participants: Two hundred and eighty-seven adolescents (51.2% male) who were aged 12 or 13 years at study enrolment. Measurements: Participants reported their alcohol involvement and completed computerized tasks of disinhibition, delay discounting and risk-taking at each assessment. Cross-sectional and prospective relationships between the variables of interest were investigated using cross-lagged analyses. Findings: All behavioural impulsivity tasks predicted a composite index of alcohol involvement 6 months later (all Ps<0.01), and these prospective relationships were reliable across the majority of time-points. Importantly, we did not observe the converse relationship across time: alcohol involvement did not predict performance on behavioural impulsivity tasks at any subsequent time point. Conclusions: Several measures of impulsivity predict escalation in alcohol involvement in young adolescents, but alcohol use does not appear to alter impulsivity. © 2013 The Authors.