Factors associated with resistance to SARS-CoV-2 infection discovered using large-scale medical record data and machine learning

Kai-Wen K Yang; Chloé F Paris; Kevin T Gorman; Ilia Rattsev; Rebecca H Yoo; Yijia Chen; Jacob M Desman; Tony Y Wei; Joseph L Greenstein; Casey Overby Taylor; Stuart C Ray

doi:10.1371/journal.pone.0278466

Factors associated with resistance to SARS-CoV-2 infection discovered using large-scale medical record data and machine learning

PLoS One. 2023 Feb 22;18(2):e0278466. doi: 10.1371/journal.pone.0278466. eCollection 2023.

Authors

Kai-Wen K Yang^{1

2}, Chloé F Paris^{1

2}, Kevin T Gorman^{1

2}, Ilia Rattsev^{1

2}, Rebecca H Yoo^{1

2}, Yijia Chen¹, Jacob M Desman^{1

2}, Tony Y Wei^{1

2}, Joseph L Greenstein^{1

2}, Casey Overby Taylor^{1

2

3}, Stuart C Ray⁴

Affiliations

¹ Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, United States of America.
² Institute for Computational Medicine, Johns Hopkins University, Baltimore, MD, United States of America.
³ Department of Medicine, Division of General Internal Medicine, Johns Hopkins School of Medicine, Baltimore, MD, United States of America.
⁴ Department of Medicine, Division of Infectious Diseases, Johns Hopkins School of Medicine, Baltimore, MD, United States of America.

Abstract

There have been over 621 million cases of COVID-19 worldwide with over 6.5 million deaths. Despite the high secondary attack rate of COVID-19 in shared households, some exposed individuals do not contract the virus. In addition, little is known about whether the occurrence of COVID-19 resistance differs among people by health characteristics as stored in the electronic health records (EHR). In this retrospective analysis, we develop a statistical model to predict COVID-19 resistance in 8,536 individuals with prior COVID-19 exposure using demographics, diagnostic codes, outpatient medication orders, and count of Elixhauser comorbidities in EHR data from the COVID-19 Precision Medicine Platform Registry. Cluster analyses identified 5 patterns of diagnostic codes that distinguished resistant from non-resistant patients in our study population. In addition, our models showed modest performance in predicting COVID-19 resistance (best performing model AUROC = 0.61). Monte Carlo simulations conducted indicated that the AUROC results are statistically significant (p < 0.001) for the testing set. We hope to validate the features found to be associated with resistance/non-resistance through more advanced association studies.

Copyright: © 2023 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

COVID-19*
Electronic Health Records
Humans
Machine Learning
Retrospective Studies
SARS-CoV-2

Grants and funding

The data utilized were part of JH-CROWN: The COVID PMAP Registry, which is based on the contribution of many patients and clinicians and is funded by Hopkins inHealth, the Johns Hopkins Precision Medicine Program. Project-specific costs of data extraction were defrayed by funds from the Office of the Dean, JHU School of Medicine. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.