This article examines a quality-improvement project to reduce clinical practice variation and measure growth outcomes by standardizing the process for the provision of protein intake for premature infants weighing 1250 g or less. Eighteen infants with a birth weight of 1250 g or less and gestational age of 32 weeks or less born between June and October 2010 were compared with 15 historical controls of the same birth weight and gestation born between January and May 2010 using a prospective cohort design. The prospective group followed a newly implemented standardized protein nutritional guideline. Independent t-tests and repeated measures of analyses of variance (between- and within-subjects) were performed on the difference between the intended and actual protein intakes between the 2 groups. A multivariate analysis of variance computed the difference between the birth and discharge gains. The prospective group showed statistically significant differences between the amount of intended and actual protein intakes (P = .023) when compared with historical controls. Growth outcomes were not significantly different (P = .67) between the 2 groups. Length of stay was not statistically significantly shorter (P = .06) in the prospective group. Reduction in clinical practice variation was associated with the implementation of a protein nutritional guideline. There was no statistical support for an improvement in the growth outcomes or reduction in length of stay.
Neonatal Unit (Dr Strahm) and Pediatrix Medical Group (Drs Mohsini and Nwankwo), Covenant HealthCare, Saginaw, Michigan, and Duke University School of Nursing, Durham, North Carolina (Dr Turner).
Correspondence: Alyse Strahm, DNP, NNP-BC, Covenant HealthCare, NICU, 1447 N Harrison St, Saginaw, MI 48602 (email@example.com).
The authors sincerely thank the Covenant HealthCare staff, including Judith Ruediger RD; Erin Van Loo RNC, MSN, neonatal nurse educator; and Lori Charbonneau MS, NNP-BC, CNS, for lending their support and expertise during this quality improvement project.
The authors declare no conflict of interest.