332 - An iterative quality improvement process to establish a Level II nursery in an academic-affiliated community hospital reduces neonatal transfers

Publication Number: 332.3918

Michelle Y. Hamline, University of California Davis, Sacramento, CA, United States; Valerie D. Stump, Adventist Health Lodi Memorial, Lodi, CA, United States; Satyan Lakshminrusimha, UC Davis, Sacramento, CA, United States; Payam Vali, University of California, Davis, School of Medicine, sacramento, CA, United States

Background: Regionalizing pediatric care by consolidating specialized resources at major medical centers enhances outcomes for complex cases but can create access gaps for nearby communities. In our region, this concentration of services has stretched academic neonatal intensive care unit (NICU) capacity, limiting its availability for lower-acuity transfers from community hospitals.

Objective: This quality improvement project aimed to address these capacity and accessibility challenges by reducing the percentage of newborns requiring transfer out of our community hospital newborn nursery by 50% over a 2-year implementation period and 4-year maintenance phase.

Design/Methods: A multidisciplinary team of academic and community pediatric hospitalists, neonatologists, nurses, and hospital administrators identified key drivers for neonatal transfers and developed targeted interventions, including establishing a licensed Level II Nursery, standardizing neonatal policies and protocols, hiring and training neonatal nurses, and reviewing neonatal transfers for improvement opportunities. The primary outcome was percentage of newborns requiring transfer from the community hospital to the academic NICU. The secondary outcomes were average neonatal transfer volume and case mix index (CMI). Additional metrics included delivery volume, Level II nursery volume, and reasons for transfer. The balancing measure was average length of stay (LOS) index. Outcomes were tracked on run charts, comparing Baseline (2017-18), Intervention (2018-2020), and Maintenance (2020-2024) phases. The primary outcome and balancing measure were analyzed using statistical process control.

Results: Average percentage of newborns requiring transfer for higher level of care decreased from 7.8% at baseline to 3.5% with special cause variation detected following the establishment of a Level II nursery at the community hospital site (Figure 1). Average neonatal transfer volume initially rose from 41 to 47 patients/year in the Intervention Phase, but subsequently declined to 37 patients/year in the Maintenance Phase (Figure 2). CMI of neonatal transfers rose from 2.6 to 2.9 in the Intervention Phase and 3.9 in the Maintenance Phase (Figure 2). LOS index remained stable with a median of 0.88 (Figure 3).

Conclusion(s): A multi-year iterative QI process to establish a Level II nursery at a community hospital site, as well as associated policies, procedures, and staff education, reduced neonatal transfers while increasing CMI of neonates requiring transfer.

Figure 1: Statistical process control p-chart showing percentage of neonates requiring transfer over time
Average percentage of neonates requiring transfer for higher level of care decreased from 7.8% in the baseline phase to 3.5% after initiation of a Level II Nursery and associated interventions. This improvement was maintained throughout the maintenance phase. Each data point represents one month of data. Avg = average, UCL = upper control limit, FY = fiscal year.

Figure 2: Volume and case mix index of neonatal transfers for higher level of care
The volume of neonates requiring transfer for higher level of care initially increased from 41 to 47 patients per year in the Intervention Phase, but subsequently declined to 37 patients/year in the Maintenance Phase. Case mix index of neonatal transfers rose from 2.6 to 2.9 in the Intervention Phase and 3.9 in the Maintenance Phase. Each data point represents one fiscal year of data.

Figure 3: X-mR chart showing length of stay index for neonates hospitalized in the Level II Nursery over time
Media length of stay index for neonates remained stable throughout the study period with a median of 0.88. Each data point represents one month of data. Avg = average, UCL = upper control limit, LCL = lower control limit.