Unexpected patterns in the global COVID-19 pandemic data

Abstract

A number of public health interventions, including mobility restrictions and vaccination, were undertaken to limit the global impact of the SARS-CoV-2 pandemic. The burden of the associated disease COVID-19 was also expected to be dependent on demographic and socio-economic determinants such as older age and general wellbeing. In this exploratory study, we examine country-level relationships between a selection of these interventions and pre-existing determinants on one hand, and public health outcomes, including COVID-19 cases, intensive-care patients, deaths, and excess mortality, on the other hand. We outline the expected results and highlight countries, continents, and time periods during 2020-2022, where/when unexpected patterns can be found in the data. For example at a global per-country scale, neither mobility restrictions nor school closures were associated with improved outcomes; no intervention or determinant came with lower intensive-care patient rates; and when using aggregation per year and continent, Europe is the only world region where vaccination and the human development index correlated with better outcomes. Between 78% and 92% of the relationships at different scales of analysis either were not statistically significant or went in the wrong direction altogether. This failure to yield expected public health benefits suggests the need for an unbiased, critical reassessment of the global pandemic response with a view to improving preparedness for future emergencies. The articles concludes with a set of research hypotheses to guide this effort.