The Relationship Between COVID-19 Infection Rates and Social Determinants of Health in Broward and Miami-Dade Counties, Florida

Objective To determine the relationship between per capita income and COVID-19 cases in Broward and Miami-Dade Counties of Florida, USA. Background Low socioeconomic status predisposes individuals to worse health outcomes. For example, during the 2003 SARS-CoV pandemic and the 2009 H1N1 influenza pandemic disadvantaged individuals were more likely to become infected. More recently, a study found that deaths due to COVID-19 were associated with disadvantaged areas across the United States. South Florida, in particular Broward and Miami-Dade Counties, has experienced a significant burden of coronavirus cases. Investigating the association of income on coronavirus cases in Broward and Miami-Dade Counties may aid in identifying and treating those individuals at increased risk. Methods This retrospective cross-sectional study used data gathered by the Florida Department of Health and 2018 U.S. Census. COVID-19 cases from March 2 - November 1, 2020 were tallied by ZIP code in Florida’s Broward and Miami-Dade Counties and scaled per housing unit. An exhaustive regression analysis using County “Miami-Dade” or “Broward,” sex, race, ethnicity, median age, and estimated per capita income was performed for each combination of independent variables in MATLAB (MathWorks, Natick, USA). Regression models were evaluated using both adjusted R-squared and the Akaike Information Criterion, along with the number of significant predictors. The most optimal model with the highest number of significant predictors was selected. Results Among all other variables, sex, race, and ethnicity as the variables that best predicted COVID-19 cases per housing unit within a certain ZIP code. The adjusted R-squared of this optimal model was 0.5062, indicating that within each ZIP code in Broward and Miami-Dade Counties 50.62% of the variance in COVID-19 cases per housing unit can be explained by these variables. A significant relationship was found between the number of COVID-19 cases and individuals who were Black or African American (p < 0.001), individuals who were Hispanic or Latino (p < 0.001), and male to female ratio (p = 0.016). Per capita income, age, and county were not statistically significant predictors in any model tested. Conclusions Racial and gender disparities may be more significant contributors to COVID-19 cases than per capita income in housing units. Based on the results of this study, investigators may consider applying this model to similar variables in order to inform the management and prevention of cases in the present and future pandemics.