Exploratory Analysis of COVID-19 Case Demographics in Gary, Indiana

Authors

  • Cameron Snapp Indiana University School of Medicine
  • Bill Trimoski Indiana University School of Medicine
  • Martin Brown Gary Health Department and Gary Sanitary District
  • Amy Han Indiana University School of Medicine
  • Tatiana Kostrominova Indiana University School of Medicine

DOI:

https://doi.org/10.18060/24769

Abstract

Background and Hypothesis: 

Health disparities are prevalent in Black populations, and COVID-19 is not an exception. COVID-19 is a pandemic that has been confirmed in 3.8 million Americans and has caused 133,283 deaths in the US (4/20/2020). Recent literature suggests that minoritized and impoverished populations are more severely impacted by COVID-19. Gary, Indiana has a large Black population (80%), high number of residents living below the poverty line (34%), and high unemployment rate (20%). We hypothesized that Black individuals in Gary have a higher rate of positive cases, hospitalizations, and deaths than non-Black individuals. Also, we hypothesized that income (median household income measured by zip code) is negatively correlated with COVID-19 deaths. 

Experimental Design and Project Methods:In collaboration with the Gary Health Department, we analyzed demographic data on all positive cases in the city from 4/16/2020 through 6/19/2020. Case data was de-identified with 16 dimensions including age, race, sex, ethnicity, hospitalization, death, and zip code.  Data was analyzed using Pearson's chi-square test and regression analysis. 

Results: 

Positive cases and hospitalizations are 2-fold and 3-fold more frequent in the Black population compared to the non-Black population in Gary (p<0.0001, P<0.01, age and population-adjusted), respectively. Median household income of a zip code is exponentially and negatively correlated with COVID-19 related deaths in that zip code (R2=0.7450, p=0.0123). 

Conclusion and Potential Impact:  

In Gary, there is a clear health disparity of both income and race, specifically in the context of COVID-19. Health officials can utilize this data to reallocate resources to highly populated, low income, and predominantly Black neighborhoods. In addition, future predictive analysis could be beneficial in developing a model to predict COVID-19 prevalence and severity. Such a model would help local health departments prepare for a second Covid-19 wave, providing for better outcomes for at risk populations through resource allocation. 

Author Biographies

Amy Han, Indiana University School of Medicine

Department of Psychiatry

Tatiana Kostrominova, Indiana University School of Medicine

Department of Anatomy, Cell Biology and Physiology

Downloads

Published

2020-12-15

Issue

Section

Abstracts