A Multiple Linear Regression Analysis of Rural-Urban COVID-19 Risk Disparities in Texas

As the number of COVID-19 cases in the U.S. rises, the differential impact of the pandemic in urban and rural regions becomes more pronounced, and the major factors relating to this difference remain unclear. Using the 254 counties of Texas as units of analysis, we utilized multiple linear regression to investigate the influence of 83 county-level predictor variables including race demographics, age demographics, healthcare and financial status, and prevalence of and mortality rate from COVID-19 risk factors on the incidence rate and case fatality rate from COVID-19 in Texas on September 15, 2020. Here, we report that urban counties experience, on average, 41.1% higher incidence rates from COVID-19 than rural counties and 34.7% lower case fatality rates. Through comparisons between our models, we found that this difference was largely attributable to four major predictor variables: namely, the proportion of elderly residents, African American residents, and Hispanic residents, and the presence of large nursing homes. According to our models, counties with high incidence rates of COVID-19 are predicted to have high proportions of African American residents and Hispanic residents coupled with low proportions of elderly residents. Furthermore, we found that counties with the highest case fatality rates are predicted to have high proportions of elderly residents, obese residents, and Hispanic residents, coupled with low proportions of residents ages 20-39 and residents who report smoking cigarettes. In our study, major variables and their effects on COVID-19 risk are quantified, highlighting the most vulnerable populations and regions of Texas.