Background Over the past half-century, diagnostic and therapeutic developments for SLE have led to dramatic improvements in the 5- and 10-year survival. Whether these achievements have improved the long-term trends in mortality in SLE is unclear.

Materials and methods We measured temporal trends in age-standardised mortality rates (ASMR) for SLE and non-SLE causes by joinpoint trend analysis using county-level data abstracted from the Centres for Disease Control and Prevention database. We calculated the annual percentage change in mortality over 46 years. Logistic regression was applied to model the association of sex, race and geographic region on SLE deaths. We calculated SLE case-fatality by dividing the SLE-mortality by the estimated SLE prevalence within each demographic variable. Since no national SLE prevalence is available, we estimated these values with weighted visit data from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey.

Results SLE was listed as the primary cause of death in 50,249 individuals from 1968–2013 in the United States. While ASMR for non-SLE causes continuously declined throughout this period, the SLE ASMR showed periods of sustained increase from mid-1970s-1990s followed by a significant decline in 2000s. The higher SLE mortality in the general population was associated with female sex, Black race, and residence in the West or South. However, in the SLE subpopulation, males had a higher mortality. The national estimates for SLE prevalence per 100,000 were 221.17 in females, 20.08 (males), 170.5 (Blacks), 107.44 (Whites), 133.5 (Hispanics), 120.36 (non-Hispanics), and 106.36 (Midwest) to 138.35 (Northeast). Even after adjusting for the prevalence variability, the SLE mortality was higher in Blacks than Whites, and in people living in the South and the West than in the Midwest and the Northeast. Analysis of the trend in SLE case-fatality showed an overall decline in rates from 1999 through 2013. The average annual percent change in SLE case-fatality ranged from −2.5% per year to −3.1% per year in various subpopulations during 1999–2013. Blacks and Hispanics died from SLE at a younger age than Whites and non-Hispanics, respectively.

Conclusions Increased SLE mortality in mid-1970s-1990s may reflect increased diagnoses with the establishment of diagnostic criteria as well as corticosteroid overuse, while the subsequent decrease in 2000s may reflect the effect of new immunosuppressive therapies resulting in an overall decreasing trend in SLE mortality in a half-century. Despite this, gender, racial and ethnic disparities persist in SLE mortality.