Natural experiments to assess the effectiveness of Covid-19 containment measures

Abstract

To assess the effectiveness of non-pharmaceutical interventions (NPIs) in controlling infections during the Covid-19 pandemic, natural experiments are useful. This article introduces the principle of natural experiments and their statistical analysis and, in the first step, discusses their implementation in previous NPI studies. The validity of many studies on the effectiveness of NPIs is limited because, on the one hand, there are often no or only weak control conditions and, on the other hand, because there are problems in the specification of the analysis models. It is often implicitly assumed that only NPIs can slow down the infection process because there are no control variables for any other causes. In the second step, a natural experiment is analyzed about the effectiveness of five measures in the second Corona wave. Switzerland (where these measures were not taken) and neighboring Baden-Württemberg serve as comparison regions. It turns out that the closure of close-contact services causes a reduction in the effective reproduction number, i.e., dampens the wave of infections. However, closing schools and daycare facilities as well as hotels is not effective.