Identifying Long COVID Definitions, Predictors, and Risk Factors using Electronic Health Records: A Scoping Review

Abstract

Abstract Objective Long COVID, or post-COVID condition, is characterized by a range of physical and psychological symptoms and complications that persist beyond the acute phase of the coronavirus disease of 2019 (COVID-19). However, this condition still lacks a clear definition. This scoping review explores the potential of electronic health records (EHR)-based studies to characterize long COVID. Methods We screened all peer-reviewed publications in the English language from PubMed/MEDLINE, Scopus, and Web of Science databases until September 14, 2023. We identified studies that defined or characterized long COVID based on EHR data, regardless of geography or study design. We synthesized these articles based on their definitions, symptoms, and predictive factors or phenotypes to identify common features and analytical methods. Results We identified only 20 studies meeting the inclusion criteria, with a significant majority (n = 17, 85%) conducted in the United States. Respiratory conditions were significant in all studies, followed by poor well-being features (n = 17, 85%) and cardiovascular conditions (n = 14, 70%). Some articles (n = 8, 40%) used a long COVID-specific marker to define the study population, relying mainly on International Classification of Diseases, Tenth Revision (ICD-10) codes and clinical visits for post-COVID conditions. Among studies exploring plausible long COVID (n = 12, 60%), reverse transcription-polymerase chain reaction and antigen tests were the most common identification methods. The time delay for EHR data extraction post-test varied, ranging from four weeks to more than three months; however, most studies considering plausible long COVID used a waiting period of 28 to 31 days. Conclusion Our findings suggest a limited global utilization of EHR-derived data in defining or characterizing long COVID, with 60% of these studies incorporating a validation step. Future meta-analyses are essential to assess the homogeneity of results across different studies.