Does parental smoking explain impacts of smoke-free public places legislation on youth smoking initiation in the UK

Background Evidence on the impact of the smoke-free public places legislation in the UK on youth smoking initiation is not well established. Changes in parental smoking behaviour may be a major mechanism by which smoke-free legislation impacts on youth smoking. Smoke-free legislation could also displace parental smoking behaviour into the home (by restricting alternative smoking spaces) or out of the home (by increasing awareness of risks to others from second-hand smoke exposure), potentially either strengthening or weakening its influence. We investigated how much of any impact of the legislation on youth smoking initiation could be explained by parental smoking, and whether associations between parental smoking and youth smoking initiation differed before or after the legislation. Methods Longitudinal data from the annual British Household Panel Survey and Understanding Society (1994–2016) were examined with discrete-time event history analyses. 14 992 youths contributed data for up to five observations (67 556 person-years) representing ages 11–15 years, with data right-censored at the year of smoking initiation (or age 15). We examined associations between smoking ban implementation (2006 for Scotland, 2007 for other UK countries) and youth smoking initiation (i.e. trying smoking for the first time), before and after adjustment for parental smoking, and tested for interactions between parental smoking and the ban implementation. Multiple imputation was used to handle missing data. Analyses were adjusted for youth sex, age, UK country, socioeconomic status, and temporal trends in initiation. Results Youth initiation of smoking was less likely after the implementation of the smoke-free legislation than before (OR: 0.16; 95% CI 0.14 to 0.18), and this difference further increased with each year after implementation (OR per year after the ban:0.88; 95% CI 0.82 to 0.94). Adjustment for parental smoking only marginally attenuated the association with ban implementation (OR: 0.20; 95% CI 0.16 to 0.24) and the per-year decrease after the ban (OR: 0.86; 95% CI 0.80 to 0.92) was similar. There was an interaction between the ban implementation and parental smoking (p-value: 0.001) such that parental smoking was more strongly associated with youth smoking initiation before the ban (OR: 1.41; 95% CI 1.26 to 1.58) than after (OR: 0.92; 95% CI 0.65 to 1.32). Discussion Changes in parental smoking behaviour did not seem to be a major explanation for the strong impacts of smoke-free legislation on youth smoking initiation, suggesting other mechanisms may be at work. The influence of parental smoking on youth initiation weakened after the ban which may indicate the displacement of parental smoking behaviour out of the home