Systemic lupus erythematosus (SLE) is an autoimmune disease that can be difficult to diagnose because of diverse symptoms that mimic other diseases. Delays in diagnosis can postpone treatment, worsening the disease course and increasing the risk of death.
April Barnado, MD, MSCI, and colleagues screened de-identified electronic health records (EHRs) that are linked to genotyped blood samples stored in Vanderbilt University Medical Center’s BioVU biobank.
Using billing codes, they developed a phenotype risk score (PheRS) that identified previously undiagnosed patients whose symptoms overlapped defined SLE criteria.
Patients’ genetic risk scores, derived from large-scale studies of genetic variations associated with SLE, did not add diagnostic value beyond clinical data. The genetic risk scores had very limited utility in SLE patients who are Black, as these patients are currently underrepresented in genetic studies.
The SLE PheRS “could be deployed within the EHR to … identify individuals who have concerning features for SLE that are misdiagnosed or undiagnosed,” the researchers reported April 25 in the journal Arthritis & Rheumatology.
The paper’s co-authors were Lee Wheless, MD, PhD, Alex Camai, Sarah Green, Bryan Han, and Anish Katta at VUMC, Joshua Denny, MD, MS, Chief Executive Officer of the All of Us Research Program at the National Institutes of Health, and Amr Sawalha, MD, of the University of Pittsburgh.
The research was supported in part by NIH grants AR072757, AR080629, AI097134, RR024975 and TR000445, and by the Department of Veterans Affairs Clinical Science Research and Development Service.
