Abstract
Aims: To estimate associations between age of first drinking (AFD) and alcohol use disorder, nicotine dependence, cannabis dependence, illicit drug dependence, major depression and anxiety disorder in adulthood, net of a series of covariate factors. Design: Data were obtained from a longitudinal birth cohort. Setting: Christchurch, New Zealand. Participants: The Christchurch Health Development Study (CHDS), a longitudinal study of a cohort born in 1977 and studied to age 35 years. Analysis samples ranged in size from 1056 (ages 11-13 years) to 962 (age 35 years); 50.2% of the total sample was male. Measurements: A measure of AFD (ages 5-13+ years) was generated using latent class analysis. Outcome measures included: major depression, anxiety disorders, alcohol use disorder, nicotine dependence, cannabis dependence and other illicit drug dependence during the period 15-35 years. Covariate factors measured during childhood included family socio-economic status, family functioning, parental alcohol-related attitudes/behaviours and individual factors. Findings: Earlier AFD was associated significantly (P < 0.05) with increased risk of later alcohol use disorders, nicotine dependence and illicit drug dependence, and was associated marginally (P < 0.10) with cannabis dependence, but not depression or anxiety disorder. After controlling for covariate factors, the associations between AFD and outcomes were no longer statistically significant [alcohol use disorder: B = -0.07, 95% confidence interval (CI) = -0.22, 0.08; nicotine dependence: B = -0.15, 95% CI = -0.34, 0.04; illicit drug dependence: B = -0.29, 95% CI = -0.73, 0.15; cannabis dependence: B = -0.05, 95% CI = -0.31, 0.22]. Conclusions: The associations between age of first drinking and later alcohol/drug disorders appear to be accounted for at least to some degree by factors related to characteristics of the individual and family during childhood.