Current Challenges of Using Patient-Level Claims and Electronic Health Record Data for the Longitudinal Evaluation of Duchenne Muscular Dystrophy Outcomes.

introductionInsurance claims data and electronic health records (EHRs) have been used to characterize Duchenne muscular dystrophy (DMD) in real-world populations. The ability to assess patient-level DMD disease progression within insurance claims or EHR data infrastructures is unknown. Insurance claims and EHR data were comprehensively examined for availability and reliability of DMD outcomes that describe functional status and disease progression at the individual patient level over time.

methodsMarketScan Commercial and Medicaid claims, and EHR-linked Clarivate open claims datasets were examined for data measuring 54 previously identified DMD-relevant outcomes in patients with DMD. Each outcome was assigned to one of five categories: functional and clinical events, clinical measures, biomarkers, functional measures, or patient-reported outcomes (PROs). Patients were identified using published coding algorithms. Annual 5-year attrition and data availability for each outcome was determined. The ability to distinguish disease severity and identify test results was also considered where applicable.

resultsA total of 1964 (MarketScan Commercial), 2007 (MarketScan Medicaid), and 10,639 (Clarivate) patients were identified. At 5 years, 31.7%, 35.1%, and 59.1% of patients remained in MarketScan Commercial, MarketScan Medicaid, and Clarivate, respectively. Claims were available for five of six functional and clinical events, with 45.5% (MarketScan Commercial), 48.0% (Clarivate), and 48.5% (MarketScan Medicaid) of patients with ≥ 1 claim for the most frequently identified clinical event (cardiomyopathy diagnosis). No data were available to describe frequency of wheelchair use or loss of ambulation. Very limited EHR data (≤ 2% of patients) were available to indicate tests were ordered for clinical measures, biomarkers, or functional assessments. No PRO notes or scores were observed. Data existed for inferring disease severity (e.g., hospitalization for cardiomyopathy); however, it was not apparent whether these events were incident.

conclusionInsurance claims and EHR-linked open claims data are of limited utility for holistically evaluating the progression and burden of DMD in individual patients.