JMIR Cancer · Published 2026-03-03 · DOI 10.2196/87347
Abstract BackgroundHemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyperinflammatory syndrome that requires rapid diagnosis and intervention. However, identifying these patients is difficult because the HLH-2004 diagnostic criteria are complex and not always captured systematically in electronic health records (EHRs). Furthermore, it is unclear how clinicians use these criteria to diagnose HLH and make treatment decisions. There is a critical need for validated computable phenotypes to accurately identify patients and study treatment-related outcomes in HLH. ObjectiveThe aim of this study is to compare different approaches to using the EHR to build computable phenotypes of patients with HLH and to evaluate characteristics and outcomes of patients meeting the HLH-2004 diagnostic criteria who received HLH-directed therapies compared to those who did not. MethodsThree approaches to computable phenotype development in the EHR were taken by identifying patients (1) with an HLH-specific International Statistical Classification of Diseases, Tenth RevisionICD-10 ResultsWe identified 388 patients with possible HLH across the three cohorts. An HLH ICD-10 ConclusionsConstructing HLH cohorts from EHR data is challenging, with diagnosis codes, treatment plans, and clinical criteria each capturing distinct but overlapping populations.
Abstract from DOAJ. Public domain (CC0 1.0).
Read the article at the publisher →