Using Comorbidities and Demographics to Predict COVID-19 Outcomes in the Midwest

There is a significant need to understand the relationship between preexisting health conditions and COVID-19 disease outcome.

Previous research has suggested that chronic obstructive pulmonary disease, cardiovascular disease, hypertension, and diabetes mellitus can increase the severity of COVID-19. In addition to these comorbidities, patient characteristics such as older age and male sex are associated with an increased risk of COVID-19 mortality.

One study, published in JAMIA Open, sought to evaluate patient demographics and comorbidities as risk factors for severe COVID-19. However, the investigators wanted to perform their analyses in the Midwest of the United States, and specifically in Indiana.

“The Midwest has some of the highest prevalence of comorbidities, specifically hypertension, COPD, and diabetes,” the study authors wrote. “Therefore, analysis of a large, comprehensive cohort of focused Midwest patients is important to better decipher the relationship between comorbidities and COVID-19 outcomes in this region.”

The investigators utilized the COVID-19 Research Data Commons (CoRDaCo) to obtain electronic health record (EHR) data from 776936 COVID-19 patients and 1362545 patients without COVID-19. CoRDaCo data consisted of the Indiana Health Information Exchange, as well as the clinical data warehouses of Indiana University Health and Eskenazi Health.

The study included patient data collected between January 1, 2018 and May 5, 2021. The positive COVID-19 cases were confirmed by laboratory testing, while the negative controls had at least 1 negative COVID-19 test result and no positive test results and had at least 1 Indiana Network for Patient Care encounter between 2018 and 2019.

The 12 comorbidities examined in the study were selected due to their prevalence in Indiana and demonstrated associations with COVID-19 severity. Data regarding county population and per capita income were obtained from the US Census Bureau. The investigators conducted statistical analysis to determine the correlation between demographic and clinical variables with COVID-19 severity. Using predictive analysis, they evaluated the predictive power of CoRDaCo EHR data in determining COVID-19 disease severity.

After their extensive data analyses, the investigators found chronic obstructive pulmonary disease in 3.49%, cardiovascular disease in 2.59%, and type 2 diabetes in 4.76% of COVID-19 patients. COVID-19 patients with these comorbidities had a 10.23%, 14.33%, and 11.11% higher risk of intensive care unit (ICU) admission, respectively. These rates were all significantly higher than the overall COVID-19 patient population, whose rate of ICU admission was 1.94%.

Additionally, patients with chronic obstructive pulmonary disease, cardiovascular disease, and type 2 diabetes had higher mortality rates than the overall COVID-19 patient population.

Considering disease severity by demographics, Black COVID-19 patients in Indiana experienced more adverse outcomes than White patients. These findings align with racial disparities in COVID-19 morbidity and mortality in the US. Potential health disparities in Indiana indicated differences in health care accessibility, as well as representation in CoRDaCo.

“Overall, COVID-19 patients with any of our considered comorbidities had higher death rates than controls (ie, patients without COVID-19) with that comorbidity,” wrote the study authors, adding, “ICU death rate was also higher for COVID-19 patients with comorbidities than for controls with that comorbidity.”

From their predictive analyses, the investigators concluded that CoRDaCo EHR data can sufficiently estimate a patient’s risk of COVID-19 infection, as well as whether a COVID-19 patient will require ICU admission.