SMOKING, SUICIDE AND MENTAL HEALTH: USING POLICY CHANGE TO PROBE CAUSALITY

Washington University, Saint Louis MO

Investigators

Linked publications & trials

Abstract

The association between smoking and suicide has been repeatedly documented, yet the proposition that smoking contributes directly to suicide risk is controversial and definitive evidence remains elusive. However, a growing body of research suggests that smoking may directly increase risk for psychiatric disorders, which would likely increase suicide risk. The global aim of this proposal is to examine whether smoking is a contributing cause for suicide and more proximal adverse mental health outcomes. We will leverage state tobacco control policy changes as natural experiments that lead to reduced prevalence of smoking. We propose that, as a result of their influence on smoking, these policy interventions also influence risk for suicide and mental health problems, including substance misuse. We will examine whether strengthening of tobacco control policies results in reduced risk for suicide and for adverse mental health outcomes. Furthermore, we will examine whether these associations are specific to smokers, as predicted by our model. We discuss our proof-of-concept study, which shows that strengthening of state tobacco control policies corresponds to reductions in suicide risk. We propose a three-year study that seeks to extend this finding using data on over one million survey participants with smoking measurements and prospective mortality follow-up. This will enable a much a much stronger analytical design than our initial study. We will also examine the associations of tobacco policy changes with psychiatric disorders and other adverse mental health outcomes, including suicidal behavior. This will enable us to identify specific channels through which smoking might influence mental health. We will take advantage of large, nationally representative epidemiological data to conduct these analyses and apply sophisticated analytic techniques that reduce the role of observed and unobserved confounders. These studies would have substantial impact on science by being a part of an emerging body of work that is challenging the traditional interpretation of the smoking-mental health link being due to ?self-medication.? This work will also have impact on public health and policy by providing key parameters to cost/benefit analyses of tobacco control policies. Confirmation of our hypotheses could result in increased awareness of the importance of smoking cessation for mental health patients, and also prompt basic researchers to more closely scrutinize the influence of nicotine on phenotypes relevant to mental health. This latter contribution could be critical as emerging tobacco products grow in popularity.