BACKGROUND AND AIMS: There is evidence to suggest that earlier initiation to alcohol increases the rate and severity of alcohol consumption. However, this often overlooks the fact that earlier initiation also means a longer drinking history. This paper estimated differences in patterns of alcohol use and harm across adolescence and early adulthood, allowing for the fact that the patterns over time may differ depending on the age at which initiation occurred.

DESIGN: Prospective cohort study based in Australia.

PARTICIPANTS: The Australian Parental Supply of Alcohol Longitudinal Study (APSALS), a cohort of n = 1906 adolescents recruited in adolescence (mean age 12.9) from 107 Australian schools and followed up until adulthood (11 annual waves total, 2010-2021).

MEASUREMENTS: We defined age of initiation as the age at which alcohol consumption was first reported in the study. Our outcomes were amount of alcohol consumed, monthly heavy episodic drinking, alcohol-related harm and self-reported symptoms of alcohol use disorder in the years following initiation.

FINDINGS: Those who initiated at age 12 had a lower risk of consumption [risk ratio (RR) 0.01; 95% confidence interval (CI) = 0.01-0.02] in the year following initiation (age 13), compared with those who initiated at age 18. Similarly, those who initiated at age 15 had a lower risk of alcohol use disorder in the year after initiation (RR 0.66; 95% CI = 0.52-0.83), compared with those who initiated at age 18. However, at age 20, those who initiated at age 12 showed higher consumption (RR 1.57; 95% CI = 1.18-2.09), monthly heavy episodic drinking (RR 1.24; 95% CI = 1.02-1.51) and alcohol-related harms [incidence-rate ratio (IRR) 1.73; 95% CI = 1.21-2.46] than those who initiated at age 18. Similar results were seen for symptoms consistent with DSM-IV alcohol dependence (RR 1.20; 95% CI = 1.05-1.38), DSM-IV alcohol abuse (RR 1.54; 95% CI = 1.04-2.29) and DSM-5 alcohol use disorder (RR 1.36; 95% CI = 1.12-1.65). However, there was evidence of ageing out, with risk of heavy episodic drinking and alcohol-related harm peaking around age 20 and then declining, regardless of when initiation occurred.

CONCLUSIONS: Later initiation to alcohol appears to be associated with more rapid escalation in drinking and related harm, but lower 'peak' harm than earlier initiation. The findings support the current guidelines recommending adolescents avoid alcohol until adulthood, and reinforce the need for public health intervention targeting both children and parents.