Our objective was to determine if there is effect modification by home smoking bans in the association between parental smoking and cigarette smoking onset in children.Methods:
Data on smoking onset, number of parents who smoke, and home smoking rules were collected from children who had never smoked in self-report questionnaires in grades 5, 7, 9, and 11. The association between number of parents who smoke and smoking onset in children was tested in pooled logistic regression in 2 groups defined by the presence or absence of a complete home smoking ban.Results:
In homes without a complete ban and relative to participants with no parents who smoke, the odds ratio (95% confidence interval [OR (95% CI]) for smoking onset was 1.5 (1.1–1.9) when one parent smoked and 1.4 (1.0–2.1) when both parents smoked. In homes with a complete ban, the OR (95% CI) was 1.6 (1.1–2.3) if one parent smoked, but 4.9 (2.4–9.9) if both parents smoked.Conclusion:
The association between number of parents who smoke and smoking onset in children was modified by the presence of a complete home smoking ban. In homes with a complete smoking ban in which both parents smoke, it may be prudent those parents communicate clearly with their children about their reasons for implementing the ban as well as about their reasons for continuing to smoke.