Skip to main content

Identifying who has Long COVID in the USA: A machine learning approach using N3C data

Pfaff, ER; Girvin, AT; Bennett, TD; et al., The Lancet Digital Health

View Full Publication on PubMed

Published

July 2022

Journal

The Lancet Digital Health

Abstract

Background: Post-acute sequelae of SARS-CoV-2 infection, known as long COVID, have severely affected recovery from the COVID-19 pandemic for patients and society alike. Long COVID is characterised by evolving, heterogeneous symptoms, making it challenging to derive an unambiguous definition. Studies of electronic health records are a crucial element of the US National Institutes of Health's RECOVER Initiative, which is addressing the urgent need to understand long COVID, identify treatments, and accurately identify who has it-the latter is the aim of this study. Methods: Using the National COVID Cohort Collaborative's (N3C) electronic health record repository, we developed XGBoost machine learning models to identify potential patients with long COVID. We defined our base population (n=1 793 604) as any non-deceased adult patient (age ≥18 years) with either an International Classification of Diseases-10-Clinical Modification COVID-19 diagnosis code (U07.1) from an inpatient or emergency visit, or a positive SARS-CoV-2 PCR or antigen test, and for whom at least 90 days have passed since COVID-19 index date. We examined demographics, health-care utilisation, diagnoses, and medications for 97 995 adults with COVID-19. We used data on these features and 597 patients from a long COVID clinic to train three machine learning models to identify potential long COVID among all patients with COVID-19, patients hospitalised with COVID-19, and patients who had COVID-19 but were not hospitalised. Feature importance was determined via Shapley values. We further validated the models on data from a fourth site. Findings: Our models identified, with high accuracy, patients who potentially have long COVID, achieving areas under the receiver operator characteristic curve of 0·92 (all patients), 0·90 (hospitalised), and 0·85 (non-hospitalised). Important features, as defined by Shapley values, include rate of health-care utilisation, patient age, dyspnoea, and other diagnosis and medication information available within the electronic health record. Interpretation: Patients identified by our models as potentially having long COVID can be interpreted as patients warranting care at a specialty clinic for long COVID, which is an essential proxy for long COVID diagnosis as its definition continues to evolve. We also achieve the urgent goal of identifying potential long COVID in patients for clinical trials. As more data sources are identified, our models can be retrained and tuned based on the needs of individual studies. Funding: US National Institutes of Health and National Center for Advancing Translational Sciences through the RECOVER Initiative. 

Authors

Emily R Pfaff, Andrew T Girvin, Tellen D Bennett, Abhishek Bhatia, Ian M Brooks, Rachel R Deer, Jonathan P Dekermanjian, Sarah Elizabeth Jolley, Michael G Kahn, Kristin Kostka, Julie A McMurry, Richard Moffitt, Anita Walden, Christopher G Chute, Melissa A Haendel; N3C Consortium

Keywords

Adolescent; Adult; COVID-19/complications/diagnosis/epidemiology; COVID-19 Testing; Humans; Machine Learning; Pandemics; SARS-CoV-2; United States/epidemiology; Post-Acute COVID-19 Syndrome

Short Summary

RECOVER researchers wanted to learn if a computer program could identify if people have Long COVID based on electronic health records (EHRs). They used EHRs from the National COVID Cohort Collaborative (N3C). Researchers created and used a computer program to compare people with Long COVID to those who didn’t have Long COVID (based on whether or not people had a visit to a Long COVID clinic in their EHR). The computer program looked for patterns in people’s symptoms, health conditions, and other data.

The researchers concluded their computer program could be used to identify people with possible Long COVID. The computer program found that people with Long COVID have patterns in their health care visits, age groups, symptoms and health conditions, and the medicines they take. This could help connect people with Long COVID to health care or invite them to join research studies.

Resources

Tags

Findings
Summary
Back to Top