OBJECTIVETo compare central line use and central line–associated bloodstream infection in newborn infants before and after the introduction of a central line infection prevention bundle in order to determine the effectiveness of the bundle and to identify areas for further improvement.DESIGNRetrospective cohort analysis of prospectively collected data.SETTINGLevel 5 neonatal intensive care unit in Sydney, Australia.PATIENTSNewborn infants admitted to the Royal Prince Alfred Hospital Neonatal Intensive Care Unit who had a central venous catheter (CVC) inserted.METHODSData regarding clinical characteristics, CVC use, and infection were collected before and after the introduction of a bundle of interventions. The bundles encompassed (1) insertion of CVC, (2) maintenance of CVC, (3) an education program, and (4) ongoing surveillance and feedback.RESULTSBaseline and intervention groups were comparable in clinical characteristics. The number of CVCs inserted was reduced in the intervention group (central line utilization rate, 0.16 vs 0.2, P<.0001). Overall CVC dwell time was reduced, resulting from significant reduction in peripherally inserted CVC dwell time (6 days [95% CI, 5.0–11.8 days] vs 7.3 days [4.0–10.4 days], P=.0004). Central line–associated bloodstream infections were significantly reduced, predominantly secondary to decreased peripherally inserted CVC–related bloodstream infections (1.2/1,000 central line–days vs 11.5/1,000 central line–days, P<.0001).CONCLUSIONThis central line infection bundle was effective in reducing CVC use, dwell time, and central line–associated bloodstream infections.Infect Control Hosp Epidemiol 2016;37:1029–1036