Session 2: Tobacco harm reduction models

Assessing the long-term impact of reduced-risk tobacco products (RRPs) on public health is an essential issue and also a requirement of the U.S. Food and Drug Administration (FDA) in the process for modified-risk tobacco product applications.

In the absence of well-defined and valid data on short- and long-term use of products from the still-young RRPs category, scientists from the public health community and tobacco companies have developed various population models that enable them to predict the effects of RRPs on factors such as transition behavior, life expectancy and cost savings in the health care system compared with a counterfactual design implying RRPs did not exist.

Analysis that incorporates transition between use of RRPs and combustible cigarettes and that takes into account the important time dimension is made possible with the help of a more complex model that was introduced during the panel sessions.

Applying a decision-theoretic framework based on a public health approach, another model looked at initiation and examined the effect of transitions to final states of established use among 15-year-olds.

A birth cohort model by age and gender scrutinizing the effects of e-cigarette use on public health—which was started in 2016 and is set to continue until 2100—assumed an optimistic and a pessimistic scenario regarding the excess mortality risk, cessation and initiation rates, and residual cigarette prevalence.

Another model presented includes the time since quitting, as well as dual use and relapsing rates, and allows tipping-point analysis. Under clearly defined assumptions, it requires relatively few input values and enables exact predictions about health care cost and life years savings. Uncertainty in model output is incorporated.

Single-use pattern and risk profiles, such as individual use linked to product use, can be assessed with an individual simulation model.

All studies carried out under the respective modeling regimes came to the conclusion that, even under worst-case assumption of low switching rates, RRPs will have a significant benefit in terms of population health effects. They also made clear how much the future success of novel tobacco products will depend on policy—one panelist asked for a two-pronged approach, with stronger policies against cigarettes on the one hand and policies to encourage the switch to e-cigarettes on the other hand.

While current statistical models are highly elaborate, they nevertheless lack the ability to include the impact of policy on the population. The panel discussion came to the conclusion that, for the future, it will be relevant to enhance models in such a way that a differentiation between various RRPs becomes possible and the impact of, for example, flavors in products can be assessed.

To make the models more realistic, new parameters need to be added. For the time being, tobacco companies have made individual endeavors at designing tobacco harm reduction models.

The panelist suggested that companies join forces and develop a common model, which should be posted for public use.