Impacts of smoke-free public places legislation on inequalities in youth smoking uptake: study protocol for a secondary analysis of UK survey data

INTRODUCTION Smoke-free public places legislation has been introduced in many countries to protect the public from the harmful effects of secondhand smoking. While evaluations of smoke-free policies have demonstrated major public health benefits, the impact on youth smoking and inequalities in smoking remains unclear. This project aims to evaluate how smoke-free public places legislation in the UK has impacted on inequalities in youth smoking uptake, and how much of any impact is via changes in parental smoking behaviour. METHODS AND ANALYSIS The study will constitute secondary analyses of UK data (from the British Household Panel Survey and the Understanding Society study). Merging these datasets gives coverage of the period from 1994 to 2016. Missing data will be handled using multiple imputation. The primary outcomes are the rates and inequalities in initiation, experimentation, escalation to daily smoking and quitting among youths aged 11-15 years. Secondary outcomes include the prevalence of smoking among parents of these youths. Discrete-time event history analysis will be conducted to examine whether changes in the probability of youth smoking transitions are associated with the implementation of the smoke-free public places legislation; and whether any observed effects differ by socioeconomic position and parental smoking. A multilevel logistic regression model will be used to investigate whether there is a step change or change in trend for the prevalence of parental smoking after the policy was implemented. The models will be adjusted for relevant factors (including cigarette taxation, the change in the legal age for purchase of cigarettes and e-cigarette prevalence) that may be associated with the implementation of the legislation. ETHICS AND DISSEMINATION This project will use anonymised survey data which have been collected following independent ethical review. The dissemination of the study findings will adopt multiple communication channels targeting both scientific and non-scientific audiences.