Using Electronic Health Record Data to Identify and Assess Algorithms for Detecting Features of Systemic Lupus Erythematosus

A new study assessed the application and utility of algorithms designed to detect attributes (characteristic quality) or features of systemic lupus erythematosus (SLE) using electronic health record (EHR) data from a multisite, urban data network.

Using the Chicago Area Patient-Centered Outcomes Research Network (CAPriCORN), which is a Clinical Data Research Network (CDRN) containing data from multiple healthcare sites, researchers identified people with at least one positively identified attribute or criteria of SLE as set by the American College of Rheumatology (ACR), Systemic Lupus International Collaborating Clinics (SLICC), and European Alliance of Associations of Rheumatology (EULAR). The study found that patients with three or more SLErelated clinical encounters ( medical visits) were identified with roughly 2.5 – 3 times the number of positive SLE attributes or features than patients without an SLE-related clinical encounter (medical visits). People with three or more SLE encounters were more often Black or African American, younger, female, and accessed health care more frequently than those with zero SLE encounters. Additionally, leukopenia was identified at exceptionally higher rates (more than 60%) across SLE encounter groups.

This study highlights the importance of surveillance and population health data management and suggests that algorithms may be useful for surveillance and to help clinicians assess for potential presence of SLE. More studies are needed to further explore EHR data to help with surveillance and shed more light on specific diagnoses and symptoms which could help identify and diagnose SLE. Learn more about patient-powered research.

Read the study