Impact of comprehensive quality improvement program on outcomes in very-low-birth-weight infants: A cluster-randomized controlled trial in Japan

Background: Differences in outcomes among neonatal intensive care units (NICUs) in Japan have been noted, prompting the need for quality improvement. Aim: To assess a comprehensive quality improvement program on outcomes in very-low-birth-weight (VLBW) infants. Study design: A cluster-randomized clinical trial. Subjects: Forty hospitals and VLBW infants born in 2012-2014 and admitted to those hospitals were study subjects. Outcome measures: The intervention group (IG) received a comprehensive quality improvement program involving clinical practice guidelines, educational outreach visits, workshops, opinion leader training, audits, and feedback. The control group (CG) was provided only with the guidelines. The primary outcome was survival without neurological impairment at three years of age. Results: IG consisted of 19 hospitals and 1735 infants, while CG included 21 hospitals and 1700 infants. There were no significant differences in gestational weeks, 29.1(26.9-31.3) vs. 29.1(26.7-31.1) or birth weights (g), 1054(789-1298) vs. 1084(810-1309) between the two groups. Both groups showed survival rates without neurological impairment of 67.2 % (1166) and 66.9 % (1137), respectively, without a significant difference. There was no significant difference in mortalities at NICU discharge between the groups, with rates of 4.0 % (70) and 4.2 % (72) respectively. Several clinically relevant improvements were observed in IG, including reduced rates of sepsis, adrenal insufficiency, transfusion for anemia, and a shorter interval to achieve full enteral feeding. However, these did not lead to improvements in the primary outcome. Conclusion: The comprehensive quality improvement program to Japanese NICUs did not result in a significant improvement in survival without neurological impairment in VLBW infants.