Background/Purpose: Individuals with Systemic lupus erythematosus (SLE) face diagnostic delays that can lead to increased disease activity and organ damage. Using a large electronic health record (EHR)-based incident SLE cohort, we examined if demographic, clinical, and provider factors impact time to SLE diagnosis.

Methods: We identified potential SLE individuals using ≥ 4 instances of SLE ICD-10-CM codes (M32.1, M32.8, M32.9) from a de-identified EHR database. Chart review confirmed diagnosis by a specialist (rheumatologist, nephrologist, or dermatologist) and incident case status as newly diagnosed SLE at our institution. We examined age, sex, race, ethnicity, and insurance type closest to SLE diagnosis as well as time to SLE diagnosis, first outpatient provider seen at our institution, and first SLE-related billing code prior to SLE diagnosis. Time to SLE diagnosis was defined as time from first code for any condition to SLE diagnosis by a specialist. First outpatient provider was categorized as primary care, rheumatology, surgical specialties, or internal medicine subspecialties. SLE-related billing codes capture signs or symptoms from the SLE classification criteria but do not mention SLE in the code name and were grouped into positive ANA or other autoantibody, hematologic, neurologic, serositis, renal, skin, and joints. We calculated mean time from first SLE-related code to SLE diagnosis. We used descriptive statistics for cohort characteristics and Mann-Whitney U and Kruskal-Wallis tests for comparisons.

Results: We identified 976 incident SLE individuals, predominantly female and White, with a mean time to SLE diagnosis of 5 ± 6 years (Table 1). Time to SLE diagnosis did not differ significantly by sex, race, or ethnicity (Table 2). Insurance status was significantly associated with time to diagnosis (p < 0.001). Individuals with military insurance had the shortest time to diagnosis (2 ± 4 years), and individuals with Medicare had the longest time (7 ± 7 years) (Table 2). Provider specialty at first outpatient visit was also significantly associated with time to SLE diagnosis (p < 0.001). Individuals seen first by rheumatologists had the shortest time to diagnosis (3 ± 5 years), while those seen by surgical specialties had the longest time (8 ± 6 years). The first SLE-related code was associated with time to SLE diagnosis (p = 0.004). Individuals with codes for encephalitis, renal disease (nephritis, renal failure, or proteinuria), positive ANA/other autoantibody, and serositis all had shorter times to diagnosis than individuals with first codes for hematologic abnormalities, skin findings, and joint pain (Figure 1).

Conclusion: In a large, real-world cohort at an academic medical center, time to SLE diagnosis was long with a mean time of 5 years. Time to SLE diagnosis varied by insurance, specialty of first provider seen, and clinical presentation. Ongoing studies are examining the trajectory of how SLE individuals interact with the healthcare system prior to SLE diagnosis. Our findings demonstrate that access to specialty care and insurance type can significantly shape the SLE diagnostic timeline and highlight areas to focus future interventions.

Figure 1. Time to SLE Diagnosis from first SLE-related billing code. We examined the first SLE-related billing code for individuals prior to SLE diagnosis. Mean time ± standard deviation from first SLE-related billing code to time of SLE diagnosis is labeled for the billing code categories. Billing codes were categorized into positive ANA/other autoantibody, hematologic manifestations (hemolytic anemia, anemia of chronic disease, pancytopenia, thrombocytopenia, leukopenia), neurologic (non-infectious encephalitis), serositis (pericarditis, pleurisy, pleural effusion), renal (nephritis, renal failure, proteinuria), skin (disorders or skin, alopecia, and dermatitis due to solar radiation), and joints (symptoms and disorders of joints and pain in joint).

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/factors-influencing-time-to-diagnosis-in-systemic-lupus-erythematosus-a-real-world-retrospective-analysis/