Title:  Racial, Gender and Geographic Differences in Systemic Lupus Erythematous and Lupus Nephritis Mortality Rates in the Unites States, 1968-2010

Background/Purpose:  Many epidemiologic studies of systemic lupus erythematosus (SLE) mortality in the United States (US) utilize patient registries that are regional and may not be applicable to the general population.  Our approach is to analyze comprehensive mortality data in the US over the past 43 years.  We aimed to determine the variation of SLE and lupus nephritis (LN) mortality by race, gender, and geographic location.

Methods:  Using county-level national mortality data from the National Center for Health Statistics from the period of 1968-2010 divided into 3 cohorts based on International Classification of Diseases periods for version 8 (1968-1978), 9 (1979-1998) and 10 (1999-2010), we estimated age-adjusted morality rates (AMR) per 100,000 persons and stratified results by gender, race, and geographic locations.  We selected cases where the underlying cause of death was SLE (734.1, 710.0, M32) and further identified LN cases to include glomerular disease and renal failure.

Results:  Of the 93,245,807 death records reviewed, we identified 46,786 cases of SLE.  AMR-SLE were 3-5-fold higher in females than in males:  0.628 (95% CI 0.613-0.644), 0.748 (95% CI 0.737-0.759), and 0.700 (95% CI 0.688-0.712) among females; and 0.162 (95% CI 0.154-0.170), 0.176 (95% CI 0.170-0.182) and 0.134 (95% CI 0.128-0.140) among males for the periods of 1968-1978, 1979-1998, and 1999-2010 respectively.  Intriguingly, while the AMR-SLE decreased among white males (from 0.149 in 1968-1978 to 0.107 in 1999-2010, p<0.0001) and remained stable among white females (0.496 to 0.499, p=0.7) over the last 43 years, it increased among black females and males (1.601 to 1.959 in females, p<0.0001; and 0.270 to 0.335 in males, p=0.004).  Thus, black females had ~6-fold higher AMR-SLE than black males and ~10–18-fold higher than white males.  To test whether differences in LN, a severe common manifestation of SLE, contributed to these differences, we analyzed mortality trends among 3,307 deaths (1999-2010) ascribed to LN.  Black females had the highest AMR-LN at 0.473 (95% CI 0.445-0.501), which was ~5-fold higher than AMR-LN in black males and white females and 36-fold higher than in white males.  Analyses of regional differences in AMR-LN showed even more dramatic racial trends, with black females living in the South census region with the highest AMR-LN of 0.507 (95% CI 0.468-0.547) and white males living in the Northeast census region with the lowest AMR-LN of 0.006, a 85-fold difference.  While no regional differences in AMR-LN were observed for black males, AMR-LN was higher in the South than in the Northeast for white males, white females, and black females.

Conclusion:  Overall SLE mortality has not substantially decreased from the 1968-1978 to the 1999-2010 periods, except in white males.  In fact, SLE mortality has increased ~1.2-fold in blacks.  Black females experience up to 18-fold higher mortality than white males.  Increased deaths ascribed to LN in black females, particularly in southern states, might account for the profound racial differences in SLE outcome.