Effect of comparative data feedback on intensive care unit infection rates in a Veterans Administration Hospital Network System

Background Infection control surveillance is not performed with standardized methodology within the Veterans Affairs (VA) health system. The purposes of this study were (1) to provide network hospitals with a standardized intensive care unit (ICU) surveillance system developed by the Centers for Disease Control and Prevention (CDC); (2) to compare ICU infection rates in hospitals that receive comparative data with those that do not; and (3) to compare network device-associated infection trends to national trends. Methods One VA Medical Center served as the central coordination site where surveillance data were analyzed with CDCʹs criteria and reported back to the sites. During 1999, the experimental group received risk-adjusted infection rates with national comparative data, and the control group received only risk-adjusted infection rates without comparative data. In 2000, hospitals in both groups received risk-adjusted infection rates accompanied by national data. Results In 1999, the device-associated infection rates were significantly higher in the control group compared with the experimental group. In the control group, the device-associated infection rates were significantly higher than the national comparative CDC rates; in the experimental group, the device-associated infection rates were not significantly different from the national comparative CDC rates. In 2000, the control group device-associated infection rates were not significantly different from the experimental group. The observed rates in both groups were not significantly different from the CDC rates. Conclusions Study results suggest that infection rate outcomes may be reduced when national comparative data are provided. The study may serve as an infection control surveillance model for VA hospital networks.