Background: The coronavirus pandemic has emerged as a significant threat to public health, economic stability, and healthcare infrastructure. In this study, we sought to determine the influence of gender, poverty level, unemployment rate, and disability prevalence on positive COVID-19 tests per capita. The COVID-19 pandemic has not created health inequalities but rather exposed the preexisting biological and social factors that influence individual outcomes. This secondary analysis aims to provide a more indepth view of how these confounding variables contribute to COVID-19 incidence in Florida counties, with the hope that future studies can examine this relationship throughout the United States.

Hypothesis: Prevalence of male gender, those with disability, population under 100% of poverty, and civilian labor force unemployed would increase the rate of positive COVID-19 tests per capita in 67 counties in Florida.

Methods: This secondary analysis was conducted through a linear regression line with the best fit for the confounding factors, using the Florida Department of Health’s data from 2015-2019. Each model was evaluated using both adjusted R-squared and Akaike Information Criterion, along with a number of significant predictors. We displayed the models with the best Akaike Information Criterion among those with the optimal number of significant predictors. The normalized set used was “Total Positive Tests (Per Capita)”. The four explanatory variables that resulted in the best model were sex, poverty level, unemployment, and disability.

Results: An increase in the Male population of one percent resulted in an expected increase of 3.35 positive COVID tests (per 1,000 people). An increase of one percent in the population of People under 100% of Poverty resulted in an expected increase of 1.01 positive COVID tests (per 1,000 people). An increase of one percent in the population of People with a Disability resulted in an expected decrease of 1.56 positive COVID tests (per 1,000 people). The adjusted R-squared was 0.4099, indicating that 40.99% of the variance in Positive COVID Tests (per Capita) is explained by the model used. Unemployment rates were not statistically significant.

Discussion: We attributed the lower incidence of positive COVID-19 tests amongst Disabled People to an increased focus on isolation due to a higher incidence of pre-existing conditions in this cohort. Additionally, an increase in positivity rate of those under 100% poverty may be due to a lack of secure housing and an inability to work from home. Unemployment rate did not have a significant effect on positive rates, which differed from our initial hypothesis. We believe this finding may be attributed to an increase in federal funding to support the rapidly growing unemployment workforce, which allowed for individuals to remain at home and social distance. Biological sex has been a known modifier of disease, which is exemplified in this study through an increase in positive COVID-19 rates in the male population. Although this study is a microcosm that explores the confounding variables on positivity rate per capita in Florida counties, these variables may be influential in future studies that apply to a broader scope across the United States