Markov Model Predicting Impact of Restriction on the Sale of Flavored Tobacco

Flavored tobacco restrictions provide an opportunity to reduce tobacco product initiation. However, this policy raises concerns about the decrease in cessation of combustible tobacco due to decreased transition to e-cigarettes.

We developed a Markov model using transition rates between inhaled tobacco product use states derived from the Population Assessment of Tobacco and Health (PATH) study to assess how a flavored tobacco restriction may impact current users and those at risk of starting tobacco products. This quasi-experimental cohort study assesses the potential impact of flavored restrictions on primary and secondary prevention strategies, providing a cohort and population assessment of flavored tobacco restrictions.

The model predicts that for every million adolescents at risk of starting tobacco, a restrictive flavor policy will have 121,000 fewer people ever using tobacco, with a decrease in every use state. After 10 years of policy for every million people over 21 years old, the model predicts 25,200 fewer tobacco product users and 19,200 fewer combustible tobacco users.

We demonstrated that a flavored tobacco restriction would see a short-term increase in combustible use from a substitution effect, before seeing a long-term downward trend in all tobacco product use is largely driven by a reduction in initiation among the youth cohort. Current users would see little to no long-term change in combustible use rates compared to a permissive flavor policy.

These findings support a restrictive flavor policy by showing that current combustible users face minimal to no harm, with significant improvements in tobacco use rates across the population leading to improved community health.