Modelling opportunity: An examination of quadratic effects of adolescent Venezuelan substance use.

INTRODUCTION AND

aimsNumerous studies have found a linear relationship between an exposure opportunity and age of first drug use. This study further tests this relationship by exploring whether a quadratic exposure opportunity best fits the data on age of first use and whether gender moderates this relationship. That is, is there a peak age in which the transition to use occurs for male compared to female adolescents? DESIGN AND

methodsA sample of 1716 adolescents from 14 public and private schools, representative of two school districts in Caracas, Venezuela, was examined using a zero-inflated Poisson modelling approach to test for quadratic effects. A series of models were tested for each set of substances (alcohol, tobacco and other drugs) to assess whether a quadratic model has better predictive ability than linear models and to test whether gender moderates the quadratic relationship.

resultsAfter controlling for common covariates, the quadratic models for alcohol indicated the peak age of transition from an exposure opportunity to use was 10-years of age in this Venezuelan sample. Gender did not moderate these models suggesting an important move toward gender equality in substance use. DISCUSSION AND

conclusionsFocused efforts on preventing exposure to alcohol use during these windows of rapid transition is likely to have the greatest impact on delaying age of onset, which will reduce the overall prevalence of alcohol related problems among adolescents.