Abstract
Background and aims: There has been a great deal of evidence showing that high novelty seeking (NS) is a risk factor for the development of substance use disorders (SUDs). However, the possible causal role of NS in SUDs is unconfirmed. The aim of this study was to examine the associations between NS at age 16 and SUDs from ages 18 to 35 years, net of a series of covariate factors. Design: Longitudinal study of a birth cohort. Setting: Christchurch, New Zealand. Participants: General community sample with sample sizes ranging from n = 1011 (age 21) to n = 962 (age 35). Measurements: The Composite International Diagnostic Interview was used to derive DSM-IV diagnoses of alcohol, nicotine, cannabis and other illicit SUDs at four time intervals from ages 18 to 35. NS was measured at age 16 using the Tridimensional Personality Questionnaire. Findings: An increase in NS was associated with increases in the prevalence of all four SUDs at age 18–35. Following adjustment for a broad range of covariate factors, estimated effect sizes (odds ratios) were reducing in magnitude, but remained moderate to large. Adjusted odds ratios of SUDs for the highest NS quartile compared to with the lowest were 2.0 [95% confidence interval (CI) = 1.5, 2.7] for alcohol; 1.8 (95% CI = 1.3, 2.7) for nicotine; 3.6 (95% CI = 2.4, 5.6) for cannabis and 5.1 (95% CI = 2.9, 9.2) for other illicit substances. Conclusions: The association between high novelty seeking and substance use disorders is not explained by common underlying individual factors and environmental exposures. This is consistent with the view that novelty seeking may play a causal role in the development of substance use disorders.