Title: Evaluation of repeated smoking cessation studies via propensity score matching approach

A major goal of many empirical studies in the health sciences is to evaluate the effect of treatments or policy changes. The treatment effect estimation using standard statistical approach may be biased if random allocation to treatments of participants is not feasible, like in many smoking cessation studies. The motivating dataset is from an Italian smoking cessation program that enrolls smokers every year since 2001 and participants voluntarily choose one of the two intervention arms. In January 2005, an indoor smoking ban was enacted in Italy, so the post-ban intervention effect is likely to be intertwined with the ban effect. Separating the effect due to this policy change from the intervention effect is of great interest to the scientific community. Using a generalized potential outcome setup, we propose a propensity score matching based strategy to estimate the intervention effect, the ban effect and their potential interaction, which is unbiased, distribution-free, and adapt to unknown time effects. Sensitivity analysis to assess the impact due to unobserved time-dependent confounder will also be discussed.