Liver Resection in Veterans Affairs and Selected University Medical Centers: Results of the Patient Safety in Surgery Study

Publisher Website
Google Scholar
Abstract
Background A congressional mandate, which led to the formation of the National Surgical Quality Improvement Program, is now being fulfilled with the publication of general and vascular surgical outcomes comparisons between Veterans Affairs (VA) and university medical centers. A series of National Surgical Quality Improvement Program articles evaluate the effect of hospital type (VA versus university hospitals) on procedure-specific outcomes. This article focuses on liver resections. Study Design This is a prospective cohort study of a sample of patients undergoing liver resections at 128 VA medical centers compared with 14 university medical centers from October 1, 2001, to September 30, 2004. Preoperative and intraoperative characteristics were evaluated to identify possible variables related to morbidity and mortality and possible confounders of the hospital effect. These variables were then used to identify the effect that the hospital setting might have on surgical outcomes after liver resections. Results Data from 237 liver resections at VA hospitals were compared with 783 procedures performed at university hospitals. The unadjusted 30-day morbidity rate tended to be higher in the VA (university 22.6% versus VA 27.9%; p = 0.10). After risk adjustment, results were equivalent (odds ratio = 0.94; p = 0.77). Unadjusted 30-day mortality rate was significantly higher in VA hospitals (6.8% versus 2.6%; p = 0.002). After risk adjustment, there was no longer a significant difference in mortality between the two hospital systems (odds ratio = 1.62; p = 0.33). Conclusions For liver resections, the National Surgical Quality Improvement Program and Patient Safety in Surgery Study data suggest that there is no significant difference in risk-adjusted morbidity or mortality rates between VA and the university medical centers.

This publication has 18 references indexed in Scilit: