In Reply: Institutional Review of Mortality in 5434 Consecutive Neurosurgery Patients: Are We Improving?

Abstract

BACKGROUND Despite increasing emphasis on quality improvement in neurosurgery, few studies have evaluated the impact of quality initiatives on health-assessment metrics including risk of mortality (ROM), severity of illness (SOI), case mix index (CMI), and mortality index. OBJECTIVE To evaluate the impact of a multifactorial quality initiative on mortality and quality metrics on a neurosurgical service. METHODS Records of 5434 consecutive neurosurgery inpatients and consults including all inpatient mortalities were prospectively collected and reviewed from July 2014 to June 2016 at major academic institution. A multifactorial quality improvement intervention was implemented in July 2015. UHC risk models mortality index, CMI, ROM, SOI present on admission (POA), and at hospital discharge (DC) were compared in the prior 12 mo and the 12 m after implementation. For mortality cases, diagnosis-related group codes, procedure type, and etiology of mortality were collected. RESULTS Compared to the pre-intervention cohort (n = 2793), the postintervention cohort (n = 2641) trended to have a decreased mean-observed monthly mortality (3.08 vs 4.17) and mean-monthly mortality index (0.73 vs 0.98). Additionally, the postintervention cohort had significantly higher CMI (3.14 vs 2.96, P = .02), POA-ROM (1.52 vs 1.46, P = .02), POA-SOI (1.97 vs 1.84, P = .0002), DC-ROM (1.69 vs 1.58, P = .003), and DC-SOI (2.1 vs 1.95, P < .0001). Of 131 mortalities (pre-intervention: n = 70, postintervention: n = 61), the postintervention cohort had a higher proportion of moralities due to emergent and trauma admissions than elective. CONCLUSION Our study suggests that our quality initiative impacted observed mortality, improved documentation, and enhanced overall quality of care on a neurosurgical service.