Implementing a Program To Prevent Neonatal Nosocomial Infections

BACKGROUND: Nosocomial infections constitute a leading cause of morbility and mortality in NICUs. Late-onset sepsis (LOS) is very frequent in VLBWI. A systematic approach using quality improvement techniques and an educational program can reduce the incidence of nosocomial sepsis. OBJECTIVE: 1) To evaluate the usefulness of “Potentially Better Practices” to reduce LOS in NICUs 2) To evaluate the compliance of healthcare workers (HCWs) with the new protocols during the study period. DESIGN/METHODS: Phase 1 (June 2007-May 2008). Two “Potentially Better Practices” were evaluated: a) hand hygiene, promoting handrubbing with an alcohol based solution b) proper management of central lines, implementing hub care and reducing the duration of central venous catheter use. Education programs and promotion campaigns were performed to improve HCWs compliance. Phase 2 (Jun 2008-May 2009). Application of the “Potentially Better Practices” developed. Adherence of HCWs to new protocols was evaluated by internal observers: 1) proper handwashing before entering the ward by cameras 2) proper handwashing during patient care 3) bacterial colony counts on the hands before patient care 4) proper central lines management 5) duration of central lines use. Phase 3 (November 2009-October 2010). Two “Potentially Better Practices” were added: environmental control b) antibiotic policy (adherence to the guidelines defined by the CDC 12-Step Campaign). Statistical analysis was performed to compare the incidence of nosocomial sepsis in the 3 phases of the study. RESULTS: We enrolled 597 neonates (mean GA 34 wks, mean BW 1910 g; VLBWI 120, mean GA 29 wks, mean BW 1040 g) during the phase 1, 574 neonates (mean GA 33 wks, mean BW 1895 g; VLBWI 116, mean GA 29 wks, mean BW 1109 g) during the phase 2 and 586 neonates (mean GA 32.9 wks, mean BW 1860 g; VLBWI 115, mean GA 28.4 wks, mean BW 1030 g) during the phase 3. The incidence of LOS decreased globally from 10% (phase 1) to 7.7% (phase 2) and to 5.1% (phase 3) (P < .05). In VLBWI, the incidence of LOS was 25.8% in phase 1, 19.8% in phase 2 and 16% in phase 3. HCWs adherence to new protocols significantly increased over time. CONCLUSIONS: The application of “Potentially Better Practices” is effective in reducing nosocomial infections in NICUs. A continuous educational program is essential to improve HCWs adherence with infection control measures, resulting in a reduction of nosocomial infections.