Substance use, generation and time in the United States: the modifying role of gender for immigrant urban adolescents.

Although immigrant youth have lower rates of substance use than US born youth, whether substance use varies by generation and time in the US is unclear. This study examines adolescent alcohol, tobacco and marijuana use by generation/time in US (i.e., first generation, in US ≤4 years; first generation, in US >4 years; second generation; and third generation or higher). Data come from a 2008 survey of Boston, Massachusetts public high school students (n = 1485). Multivariable logistic regression models were used to examine the association between generation/time in the US and risk of past 30-day substance use, adjusting for age and race/ethnicity. To determine whether the associations differed by gender, we fit gender stratified regression models. The prevalence of substance use was lowest among immigrants who had been in the US ≤4 years. Among girls, generation/time in US was not related to alcohol use or to tobacco use. For boys, being an immigrant regardless of number of years in the US, as well as second generation was associated with a significantly lower risk of tobacco use, compared to third generation youth. Additionally, immigrant boys who had been in the US ≤4 years had a significantly lower risk of alcohol use. Among both boys and girls, all first and second generation youth were significantly less likely to report marijuana use compared to third generation youth. Immigrant youth have a lower risk of alcohol, tobacco and marijuana use relative to US born youth; however the protective effect of foreign nativity on alcohol was eroded much more quickly than for tobacco or marijuana. The effects of generation and time in US on substance use differ by gender and the particular substance.