Senior Researcher Veradigm Chapel Hill, United States
Background: The COVID-19 pandemic has raised clinical and community awareness of viral and vaccine-associated myocarditis and resulted in a focus on long-term patient outcomes. Routinely captured electronic health records (EHR) and claims data provide the scale needed to study myocarditis, but there are challenges in implementation, including missing data, coding limitations, multifactorial secular trends, and variations in care seeking behavior.
Objectives: The objective of this study was to design a linked real-world data environment for identifying probable and confirmed cases of myocarditis for longitudinal follow-up.
Methods: We built a hybrid data environment linking the Veradigm Cardiology Registry (developed by the American College of Cardiology) to the Veradigm Network EHR and claims data sourced from HealthVerity. Leveraging this environment, we developed a 2-phase process for identification and categorization of myocarditis cases. Phase 1, case ascertainment, will involve the identification of patients with an ICD-10 diagnosis code for myocarditis between December 18, 2020 and October 31, 2026. Phase 2, adjudication of cases identified in Phase 1, will involve a cardiologist committee review of structured data elements, unstructured text, and data gained through healthcare provider interaction enabled by the Registry and EHR platforms to determine a) whether the case meets the CDC criteria for probable or confirmed myocarditis and b) the probable etiology of observed myocarditis. We have also convened a Scientific Advisory Board, who will develop the case adjudication guide for applying the CDC harmonized case definition of myocarditis and determining etiology.
Results: In a preliminary analysis of EHR and claims data, we identified 11,781 individuals with a myocarditis ICD-10 code between December 18, 2020, and September 2, 2022. The mean age was 43.4 (SD:20.8) years, 24.5% were under the age of 25, and 55.0% were male. Based on the initial data review, 12.8% had myocarditis coding that appeared to indicate prevalent myocarditis, and 87.2% had coding indicative of incident myocarditis. In addition, 16.8% had a recent history of COVID-19 infection. Next steps include building the case adjudication tool and enriching the dataset to ensure the capture of patient COVID vaccination status and timing.
Conclusions: This hybrid data environment leverages the scale of routinely captured healthcare data, the provider outreach from Registry and EHR sources, and the longitudinally of claims data to enable the comprehensive study of rare outcomes like myocarditis. The adjudication will serve as clinical confirmation of myocarditis and evaluation of etiology.
