(092) Characterizing Post-COVID Muscle Weakness Using Structured and Unstructured Electronic Health Records

Background: Post-COVID Conditions (PCC) include a wide range of respiratory, neurologic, digestive, and musculoskeletal symptoms and can last for extended periods of time. Identifying and diagnosing PCC and its associated symptoms is an important and growing area of research as the long-term impact of the COVID-19 pandemic continues to be understood.

Objectives: To describe individuals infected with COVID-19 and fatigue-related clinical conditions following COVID-19 diagnosis and assess unstructured clinical notes associated with muscle weakness.

Methods: Electronic health record (EHR) data from 2020 to 2024 from the OMNY Health real-world data platform were accessed. Individuals with a COVID-19 diagnosis [International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM): U07.1] and associated encounters were included in this analysis. Other comorbidities of interest included post-viral fatigue (ICD-10-CM: G93.31, G93.32, G93.39) and long COVID-19 (ICD-10-CM U09.9). Clinical notes were searched for the word “muscle” to identify any reported muscle fatigue.

Results: There were 656,362 individuals diagnosed with COVID-19, with more than 70 million observations. The average age was 48, and more than half were female (57.5%). The predominant race was white (77%), followed by African American (14%) and other races (9%). Eleven percent of the study population identified as Hispanic or Latino. Less than 2% of the study population was diagnosed with long COVID. The term “muscle” appeared among 14% of unique individuals’ clinical notes. Diagnosis codes for post-viral fatigue were present in < 1% of individuals. No concordance between a long COVID-19 diagnosis and the presence of post-viral fatigue or “muscle” in the clinical notes was observed (n < 10 for both scenarios: individuals with long COVID and “muscle”, individuals with post-viral fatigue and “muscle” in the notes field).

Conclusions: Characterizing individuals' post-COVID clinical experience is important, and these results highlight a lack of standardized reporting of these novel outcomes. Using unstructured clinical notes provides a nuanced understanding of sequelae and can be bolstered by demographic and clinical data in the structured EHR. Further investigation into the unstructured clinical notes may help define common diagnoses associated with post-viral experiences and demographic or clinical characteristics differences.