Abstract :
Summary 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 [greater than or equal to]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. Author Affiliation: (a) Department of Medicine, UNC Chapel Hill School of Medicine, Chapel Hill, NC, USA (b) Palantir Technologies, Denver, CO, USA (c) Section of Informatics and Data Science, Department of Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (d) Section of Critical Care Medicine, Department of Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (e) Colorado Center for Personalised Medicine, Division of Biomedical Informatics & Personalized Medicine, Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (f) Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (g) Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (h) Center for Health AI, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (i) Carolina Health Informatics Program, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA (j) Department of Nutrition, Metabolism, and Rehabilitation Sciences, University of Texas Medical Branch, Galveston, TX, USA (k) The OHDSI Center at the Roux Institute, Northeastern University, Portland, ME, USA (l) Department of Biomedical Informatics, Stony Brook Cancer Center, Stony Brook University, Stony Brook, NY, USA (m) Section of Biomedical Informatics and Data Science, Johns Hopkins University, Baltimore, MD, USA * Correspondence to: Dr Emily R Pfaff, Department of Medicine, UNC Chapel Hill School of Medicine, Chapel Hill, NC 27599, USA (footnote)* Co-first authors (footnote)[Dagger] Members are listed at the end of the Article Byline: Emily R Pfaff, PhD [epfaff@email.unc.edu] (a), Andrew T Girvin, PhD (b), Tellen D Bennett, MD (c,d), Abhishek Bhatia, MS (i), Ian M Brooks, PhD (e), Rachel R Deer, PhD (j), Jonathan P Dekermanjian, MS (f), Sarah Elizabeth Jolley, MD (g), Michael G Kahn, MD (c), Kristin Kostka, MPH (k), Julie A McMurry, MPH (h), Richard Moffitt, PhD (l), Anita Walden, MS (h), Prof Christopher G Chute, MD (m), Prof Melissa A Haendel, PhD (h), Carolyn Bramante, David Dorr, Michele Morris, Ann M Parker, Hythem Sidky, Ken Gersing, Stephanie Hong, Emily Niehaus