Assessing factors of COVID-19 outcomes in the United States based on the ecological framework of population health

Outcomes associated with the coronavirus disease 2019 (COVID-19) pandemic in the United States (U.S.), particularly in terms of infection rates and deaths due to viral infection, has been and continues to be widely discussed. The current study utilized several U.S. county-level datasets representing over 30 predictive variables of the ecological framework of health, a model that includes measures of culture, politics, policy, socioeconomics, lifestyle behaviors, and both chronic disease risk factors and diagnoses. A non-linear artificial intelligence statistical approach was used to assess the ability of these variables (i.e., features) to predict U.S. county-level COVID-19 mortality and case rates. For both COVID-19 deaths and cases per 100,000 persons, the highest R² was achieved using 30 features, representing all areas of the ecological framework for health. Measures of vaccine compliance, hesitancy and concern over a difficult rollout were also retained in the model. The U.S. faces numerous health challenges. Unhealthy lifestyle behaviors, chronic disease risk factors, and diagnoses of chronic conditions have reached crisis levels with few improvements in sight. This is further being compounded by an unprecedented culture clash and politicization of health-related decisions and outcomes. These issues may have been simmering for some time based upon the regional disparities in health characteristics and outcomes and their associations with distinct cultural and political preferences. The COVID-19 pandemic appears to have forcefully brought these issues to the surface.