ISQUA16-2735HOSPITAL VARIATIONS IN POSTOPERATIVE SEPSIS AND RELATED OUTCOMES AFTER CORONARY ARTERY BYPASS GRAFTING SURGERY

Abstract

Objectives: To assess hospital variations in the incidence rates of sepsis and sepsis related deaths, sepsis case fatality, overall in-hospital mortality and 30-day post-discharge mortality, as well as readmission within 28 days after discharge among elective surgical patients who underwent coronary artery bypass grafting (CABG) in New South Wales (NSW), Australia. Method(s): We selected elective surgical patients who underwent CABG surgery performed in 9 public acute hospitals (all principal referral hospitals) and in 12 private hospitals (7 major, 2 district and 3 community) in NSW from 2007 to 2013 using the state-wide Admitted Patient Data Collection records linked with the NSW Registry of Births, Deaths, and Marriages. Following the inclusion criteria and the definition of "postoperative sepsis" developed by the Agency for Healthcare Research and Quality, we derived the outcome using 54 non-principal diagnostic fields in patient medical records. We targeted elective surgical patients aged >=18 years and stayed in the hospital more than 3 days, and excluded those who were principally diagnosed with sepsis or infection on admission, cases with any-listed ICD-10-AM diagnosis codes for cancer, or cases with any-listed ICD-10-AM diagnosis codes or any-listed ICD-10-AM procedure codes for immunocompromised state. We used Poisson mixed models to derive rate ratios (RR) for each outcome adjusted for patient and hospital characteristics (age, gender, country of birth, marital status, comorbidity and socio-economic status, hospital district (metropolitan and rural)). The comparisons were performed between the public and private hospital groups, and between the bottom 20% (worst) and top 20% (best) within the two hospital groups. Result(s): Between 2007 and 2013, 10,868 (53.9%) elective CABG surgery were admitted to public hospitals, and 9,312 (46.1%) to private hospitals. Patients in the public hospitals were younger than patients in private hospitals (65 yrs vs 67 yrs, P<0.001), and had less proportion male patients. Between the two hospital groups, both of the incidence rates of sepsis and sepsis related deaths in the public hospitals were twice times the incidence rates in the private hospitals (sepsis: 24.9 vs 12.8 per 1,000 admissions; adjusted RR=1.73, 95%CI: 1.23-2.42; sepsis related deaths: 5.0 vs 2.5 per 1,000 admissions; adjusted RR=1.81, 95%CI: 1.06-3.08). There was no significant difference in sepsis cases fatality between the two hospital groups. The higher rate of overall in-hospital mortality was observed in public hospitals compared with private hospitals (14.7 vs 6.4 per 1,000 admissions, adjusted RR=1.74, 95%CI: 1.00-3.03), but similar rates showed in 30-day post-discharge mortality. Within the public hospital group, between the worst and the best quintiles, there were significant differences in the sepsis incidence rate (42.4 vs 15.5 per 1,000 admissions; adjusted RR=1.18, 95%CI: 1.07-1.30), the incidence rate of sepsis related deaths (7.8 vs 1.2 per 1,000 admissions; adjusted RR=1.34, 95%CI: 1.06-1.69), and overall in-hospital mortality rate (21.2 vs 6.6 per 1,000 admissions; adjusted RR=1.15 (1.00-1.31). Within the private hospital group, there was only significant difference in the incidence rate of sepsis related deaths between the worst and the best quintiles (5.6 vs 1.0 per 1,000 admissions, adjusted RR=1.49, 95%CI: 1.07-2.07). Conclusion(s): Significant hospital variation exists in the incidence of sepsis and sepsis related death, overall in-hospital mortality and 28-day readmission after elective CABG surgery. The public hospital group showed greater variations compared with the private hospital group. Further research is needed to investigate the causes of such variations and to develop necessary policy interventions.