About the role of science in COVID-19 policymaking and decision-making under uncertainty: A case-study from the Netherlands

The COVID-19 pandemic fueled a discussion about the role of science in policymaking. Especially in the early phase of the pandemic, policymakers sometimes took decisions against the advice of scientific experts, but mostly they supported their (unpopular) decisions with a reference to science. Despite the crucial role of the scientific information, it remains largely unexplored how policymakers have used scientific information to formulate or implement, more or less, evidence-based COVID-19 policies. The goal of this paper is to understand the decision-situation of policymakers in more depth. Embedded in literature about the policymaking process, the paper analyzes the interactions between those producing knowledge and those taking policy decisions. The hypothesis is that there are different roles in the policymaking process whereby the productive translation from scientific facts into operationalizable action depends on the trustful collaboration between actors with different, yet complementary roles at the science-policy interface. To this end, the paper takes a case from the Netherlands. It focuses on Dutch Safety Regions (regional public agencies specialized in crisis management) and their use of scientific information provided from a mobility-related COVID-19 early warning system developed by a Dutch research consortium. The paper draws upon in-depth structured interviews with actors holding different roles in the policymaking process (operational leader, information manager, data analysts), whereby some respondents are closer to science and others closer to decision-making. Among others, the interviews include quasi-experimental elements through which participants were exposed to model projections of Covid-19 infections. Interviews manipulated the exclusion and inclusion of graphically depicted information about confidence intervals and model validity in order to isolate the effect of uncertainty typical for forecasts during the Covid-19 pandemic. Results indicate that accounting for uncertainty does change some but not all actors’ perceptions of projections’ reliability, or use of (or trust in) scientific information. Perceptions cluster around actors’ role in the policy process whereby those closer to decision-making focus on making sense of scientific information and translating it into actionable policy advice. By contrast, data analysts whose role is to assemble scientific information are concerned with questions of reliability and validity of model projections. Evidence-based policy making is thus a collaborative effort and highly depends on trustful relationships among those producing scientific knowledge and those making sense of it.