Reducing neonatal nosocomial bloodstream infections through participation in a national surveillance system

A national nosocomial surveillance system for neonatal intensive care patients with a very low birthweight was set up in Germany in 2000 (NEO-KISS). Forty-eight neonatal intensive care units (NICUs) participated in the programme, which focused upon nosocomial bloodstream infections (BSIs) and pneumonia. Only data from NICUs participating for at least three years were included and the years compared. The relative risks and their 95% confidence intervals (CIs) were calculated and a multiple logistic regression analysis performed to identify significant risk factors. Twenty-four units that met the selection criteria accumulated data for 3856 patients and 152 437 patient-days in their first three years of participation. The incidence density of BSIs decreased significantly by 24% from 8.3 BSIs per 1000 patient-days in the first year to 6.4 in the third year. In the multiple logistic regression analysis, BSI in the third year of participation was significantly lower than in the first year of participation (odds ratio=0.73, 95% CI 0.60-0.89). The year of participation was an independent risk factor for BSI but not for pneumonia. Our data suggest that participation in ongoing surveillance of nosocomial infections in NICUs, requiring individual units to feedback data, may lead to a reduction in BSI rates.