The FDA has recently been given broader authority to regulate the manufacture, product standards, marketing, and distribution of any new or existing product made or derived from tobacco in order to protect the public health. There is little research on the effects of marketing and product regulations being considered by the FDA on the sole and dual use of cigarettes and smokeless tobacco. In these situations, modeling provides a "virtual laboratory" to synthesize existing evidence and compare the impact of multiple regulations on US tobacco- related deaths. In this proposal, we conduct statistical analyses of the transitions to and from cigarette and smokeless tobacco use. The statistical analyses will consider the effect of policies on the initiation, cessation, multiple product use and quantity smoked. We will apply the statistical analyses to three existing models, the SimSmoke tobacco control policy model and two natural history of disease models, the Michigan-Fred Hutchinson (Michigan) and the Massachusetts General Hospital Lung Cancer Policy Model (LCPM) models. These models were developed as part of the National Cancer Institute's Cancer Intervention and Surveillance Modeling Network (CISNET). A key advance in CISNET has been the collaborative use of multiple models to address a common question using shared inputs, an approach cited for best modeling practices. The SimSmoke, Michigan and LCPM models will be applied to the US. In addition to these models, we will develop SimSmoke models for two states with very different policy structures, Minnesota and Kentucky, to gauge how the effects of FDA regulations may vary depending on state policies. The models will be developed with the capacity and flexibility to incorporate new data (e.g., the PATH study) as it becomes available. The SimSmoke, Michigan and LCPM models will project smokeless tobacco and cigarette use, incorporating multiple product use and the initiation, cessation, and switching between products. We will compare the US population impact of regulations particularly relevant to FDA authority: health warnings, retail point-of-sale restrictions and the regulation of product content. The models will consider the impact of FDA regulations on population smokeless tobacco and cigarette prevalence (in total and by age and gender) and on tobacco- attributable deaths.
For this project, we classify the population of cigarettes and smokeless tobacco users, and conduct statistical analyses of the association of that use with public policies aimed at tobacco control. We use those results to model the impact on public health of proposed Food and Drug Administration regulations, including health warnings on cigarette packages, retailer point-of-sale marketing restrictions and the regulation of product content.