Statistical model for factors correlating with COVID-19 deaths

Abstract

BackgroundThe COVID-19 pandemic has caused major disruption in societies globally. Our aim is to understand, what factors were associated with the impact of the pandemic on death rates. This will help countries to better prepare for and respond in future pandemics. MethodsWe modeled with a linear mixed effect model the impact of COVID-19 with the dependent variable “Daily mortality change” (DMC) with country features variables and intervention (containment measurement) data. We tested both country characteristics consisting of demographic, societal, health related, healthcare system specific, environmental and cultural feature as well as COVID-19 specific response in the form of social distancing interventions. ResultsA statistically significant country feature was Geert Hofstede's masculinity, i.e., the extent to which the use of force is endorsed socially, correlating positively with a higher DMC. The effects of different interventions were stronger that those of country features, particularly cancelling public events, controlling international travel and closing workplaces. ConclusionSocial distancing interventions and the country feature: Geert Hofstede's masculinity dimension had a significant impact on COVID-19 mortality change. However other country features, such as development and population health did not show significance. Thus, the crises responders and scholars could revisit the concept and understanding of preparedness for and response to pandemics.