The Evidence-based Practice for Improving Quality method has greater impact on improvement of outcomes than dissemination of practice change guidelines and quality improvement training in neonatal intensive care units.

Objectives: To determine whether outcome improvements achieved by neonatal intensive care units (NICUs) in the Evidence-based Practice for Improving Quality (EPIQ) trial could be reproduced in other NICUs by providing quality improvement (QI) training and practice change guidelines developed during the EPIQ trial; and to examine whether the results of the EPIQ trial were sustained.

Methods: The present prospective before-after study included 5812 infants born at ≤32 weeks' gestation and admitted to 19 level 3 NICUs in the Canadian Neonatal Network between October 1, 2005 and December 31, 2007. During a three-month baseline period, multi-disciplinary teams received general training in QI techniques at a two-day workshop, and practice change guidelines targeting nosocomial infection (NI) and bronchopulmonary dysplasia (BPD) developed during the EPIQ trial were provided to all participants. Outcome data collected during the intervention period were compared with data from the baseline period and reported quarterly.

Results: In NICUs that had not previously participated in the EPIQ trial (non-EPIQ NICUs), there were no significant changes in the incidence trends of NI or BPD. However, within NICUs that had previously participated in the EPIQ trial (EPIQ NICUs) there was a continued reduction in the incidence trend of NI and BPD among EPIQ NICUs randomized during the trial to reduce NI and BPD, respectively.

Conclusions: Providing NICUs with QI training and practice change guidelines developed during a successful QI initiative in other units is not effective. The authors speculate that successful practice change involves organizational culture and behaviour change, and should be driven by context-specific evidence.