COVID-19 Deaths: Which Explanatory Variables Matter the Most?

As Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) spreads around the World, many questions about the disease are being answered; however, many more remain poorly understood. Although the situation is rapidly evolving, with datasets being continually corrected or updated, it is crucial to understand what factors may be driving transmission through different populations. While studies are beginning to highlight specific parameters that may be playing a role, few have attempted to thoroughly estimate the relative importance of these disparate variables that likely include: climate, population demographics, and imposed state interventions. In this report, we compiled a database of more than 28 potentially explanatory variables for each of the 50 U.S. states through early May 2020. Using a combination of traditional statistical and modern machine learning approaches, we identified those variables that were the most statistically significant, and, those that were the most important. These variables were chosen to be fiduciaries of a range of possible drivers for COVID-19 deaths in the USA. We found that population-weighted density (PWD), some "stay at home" metrics, monthly temperature and precipitation, race/ethnicity, and chronic low respiratory death rate, were all statistically significant. Of these, PWD and mobility metrics dominated. This suggests that the biggest impact on COVID-19 deaths was, at least initially, a function of where you lived, and not what you did. However, clearly, increasing social distancing has the net effect of (at least temporarily) reducing the effective PWD. Our results strongly support the idea that the loosening of "lock-down" orders should be tailored to the local PWD. In contrast to these variables, while still statistically significant, race/ethnicity, health, and climate effects could only account for a few percent of the variability in deaths. Where associations were anticipated but were not found, we discuss how limitations in the parameters chosen may mask a contribution that might otherwise be present.