Effects of progression to cigarette smoking on depressed mood in adolescents:: evidence from the National Longitudinal Study of Adolescent Health

Aims To investigate the relationship between smoking status and continuously distributed depressed mood among a cohort of adolescents. Design Quasi-experimental design, selecting the subset of adolescents who reported never having smoked a cigarette at baseline, some of whom progressed subsequently to smoking at follow-up approximately 1 year later. Setting Data were drawn from the National Longitudinal Study of Adolescent Health, an ongoing study designed to assess the health status of adolescents, and explore the causes of adolescent health-related behaviours. Participants Nationally representative sample of adolescents from the USA (n = 12 149), including a subsample who reported never having smoked a cigarette at baseline (n = 5475), aged on average 15 years at baseline and of predominantly European ancestry. Measurements Logistic and linear regression models controlling for potential confounders to explore the relationship between smoking status and depressed mood measured using the Centers for Epidemiological Studies Depression Scale (CES-D). Findings Various relationships between smoking status and depressed mood were observed, with a general trend for these effects to be greater among females. Smoking status at baseline did not significantly predict CES-D score at follow-up, although this effect approached significance in females (P = 0.077). Among never smokers at baseline, level of depressed mood at baseline predicted subsequent progression to smoking initiation (P = 0.022) but not progression to regular smoking (P = 0.229). Among never smokers at baseline, progression to smoking initiation during the follow-up period was associated with higher CES-D scores at follow-up, even after adjusting for baseline depressed mood (P < 0.001), with this effect greater for females than for males. Among those who initiated smoking, progression to regular smoking was associated with higher CES-D score at follow-up among females (P = 0.001), but not males (P = 0.966). Conclusions These data appear to support a complex model of the relationship between depressed mood and smoking status which includes elements of both confounding and causal models. The relationship between cigarette smoking and depression may be a factor in the development of subsequent dependence.