An Agent-Based Modeling Approach for Tobacco Product Risk Assessments

• Activity Number: 204
• Type: Contributed
• Date/Time: Monday, August 1, 2016 : 11:35 AM to 12:20 PM
• Sponsor: Section on Risk Analysis
• Author(s): Raheema Muhammad-Kah* and Yezdi B. Pithawalla and Maria Gogova and Lai Wei and Edward L. Boone
• Companies: Altria Client Services and Altria Client Services and Altria Client Services and Altria Client Services and Virginia Commonwealth University
• Keywords: Modified Risk Tobacco Product (MRTP) ; Population Modeling ; Agent-Based Modeling ; Tobacco Product Risk Assessments ; Simulations ; Regulations

Abstract

We have developed an Agent Based Model (ABM) to assess the impact of introducing a Modified Risk Tobacco Product (MRTP) into a hypothetical population by estimating changes in tobacco use prevalence and all-cause mortality, against a status quo scenario. MRTP draft guidance from the FDA encourages the use of computational models for estimating overall population benefits and risks. Each individual is represented as an agent within the model. Microsimulations follow the individual actions of each agent over an extended timespan, to assess the final impact on the overall population. The agents transit between different use states, such as Never Tobacco User, Smoker, MRTP User, Dual User, Former Tobacco User, etc., using transition probabilities that are governed by initiation and cessation rates. Statistical models combined with relative risks are used to determine survival probabilities. We will present scenarios discussing the impact on prevalence and all-cause mortality of a simulated population of 1,000,000 individuals. Model assumptions, validation results and outputs from sensitivity analysis on key input parameters will also be shared.