What is the impact of modeling scenarios?

Why ask about the evaluation of modeling scenarios? Decision-making during the covid-19 pandemic is marked by uncertainty: uncertainty about the evolution of the epidemic, and uncertainty about the impact of measures taken in response. Modeling scenarios are used to inform decision makers (scientific council, health defense council…). Therefore, in order to make good decisions, it is important to have modelling scenarios that anticipate reality.

It may be thought that it is better to overestimate reality, and thus lead decision makers to react with more measures to limit the spread of the epidemic. In reality, most measures to curb the epidemic (containment, curfews, school closures, public place closures) have negative health, social and economic impacts (see the CollateralGlobal project for detailed examples). Therefore, overreacting to the epidemic is not a good thing (nor is underreacting). Therefore, one should not “systematically plan for the worst”, but plan for the best so that decision-makers can find the best balance between limiting the impact of the epidemic and limiting the impact of curbing measures.

If the scenarios underestimate the evolution of the epidemic (A), the risk is that decision-makers underestimate the evolution of the epidemic and do not react, or do not react enough. Conversely, if the scenarios overestimate the evolution of the epidemic (B), there is a risk of overreacting and, in particular, of taking too many measures with deleterious effects without justification.