Search

Introduction: During the 2000s the number of adolescents who became new smokers in the United States declined while the number of young adults who did so increased. However, we do not know among which demographic groups these changes occurred.Methods: We analyzed data from the 2006 to 2013 National Survey of Drug Use and Health (n = 180 079). Multivariate linear regression models were used to assess annual trends in smoking onset and log-binomial regression models to assess changes over time in the risk of smoking onset among young adults (18- to 25-years-old) relative adolescents (12- to 17-years-old).Results: From 2006 to 2013, the rate of onset among young adults (6.3%) was greater than among adolescents (1.9%). Time trends demonstrated that annual declines in smoking onset occurred among white young adult males and females. Rates of smoking onset increased among black and Hispanic young adult males with a lower rate of decline among black and Hispanic young adult females. There was a greater risk of smoking onset among young adults relative to adolescents that did not change over time.Conclusions: Smoking onset is becoming more concentrated in the young adult than adolescent years. Despite this trend, there were annual declines in young adult smoking onset but not uniformly across racial/ethnic groups. More effective strategies to prevent young adult smoking onset may contribute to a further decline in adult smoking and a reduction in tobacco-related health disparities.Implications: Smoking onset is becoming more concentrated in the young adult years across sex and racial/ethnic groups. The United States may be experiencing a period of increasing age of smoking onset and must develop tobacco control policies and practices informed by these changes. [ABSTRACT FROM AUTHOR]

This paper reports on the measurement properties of a scale intended to measure nicotine dependence in adolescents using data from the American Legacy Longitudinal Tobacco Use Reduction Study. The Nicotine Dependence Scale for Adolescents (NDSA) is a simple linear sum of items associated with behaviors to avoid withdrawal and items associated with experiences of craving. An exploratory factor analysis indicated that the items formed a single scale. The reliability of the scale was assessed by estimating Cronbach's alpha for the scale items. The estimated alpha reliability was .81. The scale's validity was assessed by estimating the correlation between the scale and measures of smoking and quitting behaviors. As hypothesized, the scale was positively correlated with lifetime number of cigarettes smoked (r = .44, p < .001), days smoked in the past 30 days (r = .66, p < .001), cigarettes smoked per day on days smoked (r = .61, p <.001), and number of quit attempts (r = .10, p <.001) and negatively correlated with the length of the quit attempt (r = - .22, p <.001). We also examined the measurement properties of the NDSA separately for middle and high school students and by gender and race or ethnicity. We found no meaningful differences in the measurement properties of the NDSA across these groups. The six-item self-report NDSA had good measurement properties in our sample. [ABSTRACT FROM AUTHOR]