Trends in age of smoking initiation in the Netherlands: a shift towards older ages?

Background and aim As smoking initiation generally occurs in adolescence, smoking prevention is targeted primarily at young adolescents (aged below 16 years). We hypothesize that, with the adoption of increasingly stronger youth access laws, a shift in the age of smoking initiation may have occurred. Design Repeated cross-sectional survey. Setting The Netherlands. Participants A total of 13 163 respondents born between 1980 and 1995. Measurements In the 2010-15 National Health Survey we measured retrospectively self-reported initiation rates per year of age by birth cohort (1980-83, 1984-87, 1988-91, 1992-95). Logistic regression analyses were performed to assess differences between cohorts, per sex and educational level. Findings Of those who started smoking, 67.2% initiated smoking between 12 and 16 years of age. Contrary to our hypothesis, no shift in initiation towards higher ages was observed. The peak age of initiation was 16 years for all cohorts. Initiation at ages 17-21 was lower in younger birth cohorts than in older birth cohorts [e.g. cohort 1993-95 versus 1980-83: odds ratio (OR) for initiating at 19-21 years = 0.45, 95% confidence interval (CI) = 0.31-0.64]. Trends in age of initiation were similar for males and females. Initiation at ages 15-17 decreased across cohorts for respondents in higher education levels, but not for those in lower education levels. Conclusions In the Netherlands, the peak age of smoking initiation did not shift through our subsequent cohorts born between 1980 and 1995, suggesting that young adolescents remain the main target group for smoking initiation prevention policies