Background Past studies suggest that males with lupus nephritis (LN) may have increased rates of end-stage renal disease (ESRD) and mortality compared to females. However studies included few males and were focused on biological differences, not healthcare use. In a nationwide cohort of SLE patients with incident LN, we investigated LN-related outcomes and utilisation by sex.

Materials and methods We used the Medicaid Analytic eXtract (MAX) with nationwide billing claims to identify individuals 5–65 years with LN (2000–2004) using a validated algorithm (PPV 80%) and required 12 months without any LN codes to define incident cases. MAX data were linked to the U.S. Renal Data System (USRDS) 2000–2006 to identify ESRD onset. Mortality was determined using National and Social Security Death Index Files (2000–2006). We assessed sex-specific incidence rates and adjusted incidence rate ratios (IRRs) for healthcare utilisation, medications, preventive care and renal biopsies using Poisson regression. We used Fine and Grey proportional hazard models to compare the subdistribution hazard ratios (HRsd) of ESRD by sex accounting for the competing risk of death, and Cox models to compare hazard ratios (HR) of death, adjusted for age and race/ethnicity.

Results Of 2576 patients with incident LN, 230 (9%) were male. Mean follow-up was 2.8 (SD 1.5) years for both sexes. Mean age was 30 (SD 16) years among males and 34 (SD 14) years among females (p < 0.001). 31% of males and 36% of females underwent renal biopsy (p = 0.06). Other than azathioprine use, which was more frequent among females (p = 0.02), were no differences in medications or preventive care. Adjusted rates of outpatient and emergency department (ED) visits were lower for males compared to females (IRR 0.81, 95% CI: 0.68–0.98 and 0.88, 95% CI: 0.79–0.99, respectively); hospitalizations were comparable. The five-year cumulative incidence of ESRD was 13% and the HRsd of ESRD for males compared to females was 0.86 (95% CI: 0.53–1.38) in the first two years and 2.21 (95% CI: 1.2–4.1) in the subsequent two years. The five-year cumulative incidence of death was 15% with no difference in HR by sex.

Conclusions In this incident LN Medicaid cohort, we found high rates of ESRD and mortality overall, with no differences in ESRD by sex in the first two years but more than twice the risk among males thereafter. Males had lower rates of outpatient and ED visits compared to females. Further studies are needed to understand the relationship between utilisation and long-term outcomes.

Acknowledgements Drs. Feldman and Broder contributed equally.