What is the clinical evidence from this study?

Study design: Cohort. Population: Major general surgery operations (n=112497). Intervention: Veterans Affairs (VA) medical centers vs. Private-sector (university) medical centers. Primary outcome: 30-day postoperative mortality (OR 1.23, 95% CI 1.08-1.41, p=0.0002).

June 1, 2007

Comparison of Risk-Adjusted 30-Day Postoperative Mortality and Morbidity in Department of Veterans Affairs Hospitals and Selected University Medical Centers: General Surgical Operations in Men

Key Result

Male patients undergoing general surgery at Veterans Affairs hospitals had a higher risk-adjusted 30-day mortality compared with those at private-sector centers (OR 1.23; 95% CI 1.08-1.41).

Study Design

Type

Cohort (n=112,497)

Multicenter

Yes

Structured PICO

Does treatment at VA hospitals compared to private university medical centers affect 30-day postoperative mortality and morbidity in men undergoing general surgery?

Population

112,497 male patients undergoing major general surgery operations at 128 VA and 14 university medical centers, assessed for 30-day postoperative outcomes.

Exposure

Treatment at Veterans Affairs (VA) hospitals

Comparator

Treatment at private university medical centers

Outcome

30-day postoperative mortality and morbidityhard clinical

Risk-adjusted 30-day mortality after general surgery in men was higher in VA hospitals compared to private university medical centers, particularly for less common, complex operations.

Main Result

Odds Ratio: 1.23 (95% CI 1.08–1.41)

Absolute Event Rate: 2.62% vs 2.03%

p-value: p=0.0002

Limitations

Differences in mortality ascertainment in the two sectors might account for some of the observed effect.

Abstract

BACKGROUND: We used data from the Patient Safety in Surgery Study to compare patient populations, operative characteristics, and unadjusted and risk-adjusted 30-day postoperative mortality and morbidity between the Veterans Affairs (VA) (n = 94,098) and private (n = 18,399) sectors for general surgery operations in men. STUDY DESIGN: This is a prospective cohort study. Trained nurses collected preoperative risk factors, operative variables, and 30-day postoperative mortality and morbidity outcomes in male patients undergoing major general surgery operations at 128 VA medical centers and 14 university medical centers from October 1, 2001, to September 30, 2004. Multiple logistic regression analysis was used to identify preoperative predictors of postoperative mortality and morbidity. An indicator variable for VA versus private-sector medical center was added to the model to determine if risk-adjusted outcomes were significantly different in the two systems. RESULTS: The unadjusted 30-day mortality rate was higher in the VA compared with the private sector (2.62% versus 2.03%, p = 0.0002); unadjusted morbidity rate was lower in the VA compared with the private sector (12.24% versus 13.99%, p < 0.0001). After risk adjustment, odds ratio for mortality for the VA versus private sector was 1.23 (95% CI, 1.08-1.41). For morbidity after risk adjustment, the indicator variable for health-care system just missed statistical significance (p = 0.0585). Thirty-day postoperative mortality was comparable in the VA and private sector for very common operations but was higher in the VA for less common, more complex operations. CONCLUSIONS: In general surgery operations in men, the VA appeared to have a higher risk-adjusted mortality rate compared with the private sector, but differences in mortality ascertainment in the two sectors might account for some of this effect. The higher mortality in the VA could be the result of higher mortality in the less common, more complex operations. There is a trend toward lower risk-adjusted morbidity in the VA compared with the private sector.

Bookmark