Session: Nurse Staffing, Nurse Human Capital And The Quality Of Hospital Care
Room: Phillips 203
Time: Tue 10:15-11:45

Presenter: Jeannette Rogowski (University of Medicine and Dentistry of New Jersey. )

Discussant: Laurence Baker (Stanford University)

Abstract

Objective: There are large, unexplained variations in outcomes for infants born prematurely across NICUs, even after controlling for case mix. Infants with very low birth weights (under 1500 grams), who are cared for in neonatal intensive care units are one of the highest risk pediatric patient populations. The care of these infants requires high levels of nursing intensity. Thus, it is likely that nursing plays an important role in determining patient outcomes in the NICU. There is an increasing body of evidence demonstrating a relationship between nurse staffing, nurse education and experience, and patient outcomes in adult acute-care settings. There is, however, little current evidence on the role of nursing in outcomes in the NICU. This study will address this issue.

Methods Used: The patient population consists of infants with very low birth weights (VLBW) treated in the NICU. Data on nurse characteristics, nurse staffing, education and experience, and work environments (PES scale) were obtained for 104 NICUs in the US through a survey of nurses. Additional data were obtained for a subset of 70 NICUs through primary data collection on the units. Data on VLBW infant outcomes and infant characteristics were obtained from the Vermont Oxford Network Database. Patient outcomes include mortality, nosocomial infection, chronic lung disease, and severe intraventricular hemorrhage. Multilevel models of patient outcomes were estimated as a function of infant characteristics, nursing characteristics (the observed-to-expected staffing ratio, nurse education and experience, nurse work environments) and hospital characteristics. Expected nurse staffing was derived from estimation of acuity weights, determined through estimation of random effects logit models of nurse workloads as a function of infant acuity levels.

Results: The observed-to-expected staffing ratio and percent BSN nurses were significantly associated with mortality. A one standard deviation increase in each reduced the odds of death by 11%. Further, NICU experience was significantly associated with SIVH. A one standard deviation increase in the percent of nurses with five or more years of NICU experience was associated with a reduction of 18% in the odds of SIVH. Infants in units with the best nurse work environments had half the rate of infection of those in units with the worst work environments. Hospitals with better-educated, more NICU-experienced staff had lower staffing levels. Acuity weights, measuring the fraction of a nurse required for the care of infants of each acuity level, were found to be robust to the data collection method.

Conclusions: Nurse staffing, education, experience, and work environments affect patient outcomes for very low birth weight infants treated in the NICU. Minimum staffing levels, based on infant acuity, have the potential to improve morbidity and mortality outcomes for these infants. However, the education and experience of nurses should also be taken into consideration in setting NICU nurse staffing guidelines.

Key Terms

nurse staffing, patient outcomes, hospital quality

Authors:

Eileen Lake (Penn State University) , Jeannette Rogowski (University of Medicine and Dentistry of New Jersey) , Thelma Patrick (University of Pittsburgh) , Jeffery Horbar (University of Vermont) , Douglas Stainger (Dartmouth College) , Mike Kenney (University of Vermont) and Robyn Cheung (University of Pennsylvania)