Aim To examine the extent to which smoking by step-parents and biological parents predicts adolescent smoking.
Design Five-year cohort study.
Setting Thirty-six schools in South London, England.
Participants A subset of 650 students participating in the Health and Behaviour In Teenagers Study (HABITS), who reported living in step-families, were assessed annually from age 11–12 to age 15–16 years.
Measurements Students reported their smoking status, which was cotinine-verified, as well as whether their parents smoked and, if they lived with a step-parent, whether that step-parent smoked. Analyses also controlled for gender, ethnicity and deprivation.
Findings Students who reported that just their step-parent smoked at age 11–12 were significantly more likely to report current smoking at any time-point from age 11–16 than those who reported having neither biological parents nor a step-parent who smoked [odds ratio (OR) 2.72, 95% confidence interval (CI) = 1.36–5.47], as were those with both a parent and a step-parent who smoked (OR 2.23, 95% CI = 1.46–3.41). While the association between smoking in students and smoking in biological parents in this subsample did not reach statistical significance (OR 1.39, 95% CI = 0.88–2.19), these students were no more or less likely to smoke than those with just a step-parent who smoked.
Conclusion Smoking by a non-biological parent appears at least as influential as smoking by biological parents. This confirms the importance of social influence on smoking initiation and suggests that attempts to work with parents in smoking prevention should involve, and perhaps pay particular attention to, step-parents who smoke.