Combinatorial Analysis of Phenotypic and Clinical Risk Factors Associated With Hospitalized COVID-19 Patients

Area of Science:

• Infectious Diseases
• Biochemistry
• Medical Informatics

Background:

• Understanding risk factors for severe COVID-19 is crucial for patient outcomes.
• Previous genomic studies suggested a link between calcium/lipid homeostasis and severe COVID-19.
• Clinical and phenotypic data analysis is needed to validate these findings in diverse populations.

Purpose of the Study:

• To identify clinical and phenotypic disease signatures associated with severe COVID-19 hospitalization.
• To analyze combinations of features in a separate patient cohort using clinical data.
• To explore potential shared mechanisms between calcium, lipid, and immune cell signaling in severe COVID-19.

Main Methods:

• Utilized the PrecisionLife combinatorial analytics platform on de-identified health records from the UnitedHealth Group COVID-19 Data Suite.
• Analyzed 836 disease signatures across two cohorts: hospitalized COVID-19 cases vs. mild cases, and a subset with laboratory data.
• Investigated associations between lipid levels, serum calcium, leukocytes, statin use, and vitamin D deficiency.

Main Results:

• Identified multiple disease signatures correlating with increased COVID-19 hospitalization risk.
• Found co-occurrence of lower lipid levels with lower serum calcium and leukocytes in severe cases.
• Observed that hypocalcemia signatures were frequently linked to vitamin D deficiency and often independent of statin use.

Conclusions:

• Combinatorial analysis of clinical data can identify severe COVID-19 risk signatures independently of genomic data.
• Findings support a potential role for calcium and lipid signaling pathways in the pathophysiology of severe COVID-19.
• The identified signatures offer insights into host-pathogen interactions and potential therapeutic targets.