Interhospital variation in mortality among patients with systemic lupus erythematosus and sepsis in the USA

Objective. To determine whether the risk of mortality in patients with SLE hospitalized with sepsis varies among hospitals in the USA. Methods. We used the National Inpatient Sample (2002-2011) to obtain national population-based data on outcomes for adults with SLE admitted with sepsis, and compared it with that for patients without SLE admitted with sepsis at the same hospital. We computed expected mortality based on patient demographic characteristics, comorbidities and major organ dysfunction, and calculated observed/expected (O/E) mortality ratios separately for patients with SLE and without SLE for each hospital. We then computed the ratio of these O/E ratios within hospitals to assess relative SLE mortality. We considered hospitals with a risk ratio (RR) of >= 2.0 as having high relative SLE mortality. Results. Among 424 hospitals that treated a total of 4024 patients with SLE and sepsis, the risk of in-hospital mortality varied from 0% to 60% (median 11.1%). The RR ranged from 0 to 9.75, with a median of 0.84, indicating that O/E mortality was similar in patients with and without SLE at the average hospital. Sixty-one hospitals (14.4%) had a RR of >= 2.0, indicating higher mortality among patients with SLE. Hospitals that on average treated >= 3.9 patients with SLE and sepsis annually were less likely to have a RR of >= 2.0 than hospitals that treated fewer patients (10% vs 17%; P = 0.004). Conclusion. Mortality among patients with SLE and sepsis varied widely between hospitals, and was lower at hospitals that treated more of these patients.