Increasing Exclusive Nursery Care of Late Preterm and Low Birth Weight Infants Free

This study occurred at an academic medical center with ∼5000 annual deliveries. Approximately 10% of nursery admissions are LPT or LBW infants. Infants <35 weeks GA or <2000 g are generally triaged to the NICU from the delivery room, although exceptions are occasionally permitted. Infants admitted to the nursery are primarily cared for by newborn hospitalists and pediatric residents, with a nurse-to-mother-infant dyad ratio of 1:4. Infants are encouraged to room in with mothers. Pasteurized human donor milk (HDM) and formula are available at no cost to families.

Before our initiative rollout, LPT and LBW infants received nursing care similar to term infants with weights appropriate for gestational age. Standard nursery care included a bath at 6 hours of life and vitals assessed every 8 hours. A universal feeding protocol existed for all infants admitted to the nursery. However, end-user feedback suggested it was complicated to use given separate yet intersecting pathways for term infants, infants with specific risk factors, such as hypoglycemia, and LPT and LBW infants. The pathway for LPT and LBW infants involved initiating support for ineffective feeding after 24 hours of life. Lactation consultants were not regularly informed of new admissions to prioritize visits with these families. In addition, barriers to human donor milk utilization included a physician authorization process and a lack of generalized awareness of its benefits for LPT and LBW infants.¹⁷  As a result, LPT and LBW infant care was not readily distinguishable from other infants admitted to the nursery.

Inclusion criteria for our initiative were infants 35 to 36 weeks’ GA and term infants with low birth weight (<2500 g) admitted from the delivery room to the nursery. Infants were excluded if they were admitted to the nursery but designated as hospice care, after confirmation by manual chart review, or if discharge from the birth hospital was not to home (ie, discharge to an off-site hospital).

A multidisciplinary team convened to improve nursery care for LPT and LBW infants. Stakeholders included newborn hospitalists, nurses, lactation consultants, occupational therapists, and nutritionists. Using the Define-Measure-Analyze-Improve-Control quality framework, the team reviewed relevant published literature to understand the best evidence-based practices. With our baseline exclusive nursery care for LPT and LBW infants being 84%, higher than reports in the literature, we set a goal to increase exclusive nursery care by 5%.⁶  Although small, we felt this change to be meaningful and achievable. A key driver diagram was developed to highlight opportunities for improvement (Fig 1). Three areas identified to drive change were feeding protocol, knowledge of LPT/LBW special care needs, and LPT/LBW care standardization.

A review of published best practices on LPT and LBW infant nutrition and feeding guided our development of a feeding protocol specific to these populations.^12–14  A new LPT and LBW feeding protocol was created on the basis of multidisciplinary team member input and integration of locally available resources (Supplemental Fig 5). The new protocol prioritized human breast milk with specifics on early hand expression, pumping, and supplementation. In addition, explicit recommendations to consult lactation consultants or occupational therapists for infants not feeding well were added to encourage early maternal support and mitigate the challenges of breastfeeding LPT and LBW infants.

Educational material specific to the care of LPT infants was developed and disseminated. An informational folder provided to families on admission included handouts on cue-based feeding, LPT infant stress cues, and paced bottle feeds. An informative video on hand expression was uploaded to each patient’s bedside tablet to enhance staff teaching of breastfeeding mothers. Knowledge of the new feeding protocol, patient educational materials, and changes to care practice were disseminated to staff 1 month before our designated launch date to streamline the implementation of the feeding protocol and changes to care practice. Weekly e-mail newsletters, strategically posted signage in high-traffic staff areas throughout the unit, and verbal communication by stakeholder leaders during monthly staff meetings with reminders at shift change huddles facilitated the dissemination of knowledge.

We sought to provide standardized yet distinct care for LPT and LBW infants to identify and mitigate known risk factors earlier. Vital sign frequency was changed from every 8 to every 4 hours, and baths were delayed to after 12 hours of life to promote adequate feeding and glucose homeostasis. Acknowledging how delayed baths and occlusive hats have improved hypothermia among LPT and LBW infants, yellow hats sized for LPT and LBW infants were purchased, providing staff with a quick visual cue to identify these higher-risk patients.¹⁸  A nurse-driven standing admission order set, to be entered by labor and delivery staff at the time of birth under the name of the admitting newborn hospitalist, was approved by hospital administration. The order set incorporated the changes to nursing care and supported timely care delivery. The go-live date for the order set in our electronic health system, November 15, 2020, became the project launch date, coinciding with the implementation of the feeding protocol and distribution of educational materials to families. Bedside nurses, however, were approved by the administration to obtain authorization for human donor milk in March 2021, 4 months after the project launch. In addition, separate from our planned quality improvement project, the unit received support to improve staffing to a nurse-to-mother-infant dyad ratio of 1:3, 8 months after our go-live date.

To account for potential variations related to the first year of the coronavirus disease 2019 pandemic, the baseline period was extended to 2 years, from November 15, 2018 to November 14, 2020. The postintervention period ended after 12 months on November 14, 2021.

Data were extracted from electronic medical records and generated into a report. Baseline data were retrospectively collected. Postintervention data were tracked monthly and reviewed during multidisciplinary team meetings. A manual chart review confirmed missing data, such as gestational age and birth weight. Discrepant data were also manually chart reviewed to confirm inclusion criteria were met. The monthly meetings allowed the team to discuss processes and support each other in reminding our respective teams of the ongoing quality initiative. Six months after our launch date, the multidisciplinary team agreed to continue the initiative as planned due to noted improvements. Reporting of our study was done in accordance with Standards for Quality Improvement Reporting Excellence guidelines.¹⁹ 

The primary outcome measure was the percentage of LPT and LBW infants who received exclusive nursery care. To assess the impact on feeding-related conditions (neonatal hypoglycemia, jaundice, and dehydration) and in alignment with previous studies, secondary measures included rates of hypoglycemia, hyperbilirubinemia requiring phototherapy treatment, and the average of the maximum documented weight change relative to birth weight.⁵  Hypoglycemia was defined as a glucose value of ≤40 mg/dL for infants 0 to 4 hours of age, ≤45 mg/dL for infants 4 to 24 hours of age, or ≤50 mg/dL for infants >24 hours of age.²⁰  Because of concerns that interventions could negatively impact exclusive breast milk feeding with increased supplementation and that an increase in exclusive nursery care for LPT and LBW infants would lead to extended hospitalization to accommodate LPT and LBW infant stabilization needs, our balancing measures included the rate of exclusive breast milk feeding and LOS. Exclusive breast milk feeding rates included the use of HDM and LOS measurement was obtained in days instead of hours to provide a more global understanding of impact.

The team used statistical process control charts to track outcome measures (QI Macros SPC Software for Excel). American Society for Quality rules defined special cause variation as 8 consecutive points or 10 out of 11 consecutive points on the same side of the centerline.²¹  We tracked classification data with p-charts and continuous data with x-bar charts. A 2-proportion z-test was used to compare variables between the baseline and intervention populations (Minitab v. 19.2).

Institutional policy deemed this project was not human subject research and thereby exempt from institutional review board review. Health insurance portability and accountability act requirements were met. The team obtained a data quality release authorization for external dissemination of study results.

Infants meeting inclusion criteria totaled 1336. The 2-year baseline period included 891 infants, and the 1-year postintervention period included 445. Table 1 summarizes demographic information for the included LPT and LBW infants. All categories, including sex, gestational age, birth weight, method of delivery, twin gestation, Apgar scores, race/ethnicity, and financial class were similar between the 2 periods.

During the quality initiative, exclusive nursery care for LPT and LBW infants increased from 83.9% to 88.8%, a 4.9% absolute increase (Fig 2). Special cause variation was detected with 8 consecutive points above the baseline. During the 12-month study period, an upward centerline shift started in December 2020, 1 month after our launch date. This centerline shift was sustained for the remainder of the project.

For our secondary outcome measure of hypoglycemia, overall rates decreased by 6.6%, with a downward centerline shift starting in December 2020 (Fig 3). Special cause variation with 10 out of 11 consecutive points below the baseline was met. Other secondary outcome measures, rates of phototherapy utilization and average maximum weight loss, remained stable with no special cause variation noted on statistical process control charts (Fig 3). Balancing measures of exclusive breast milk feeding rates and LOS remained unaffected (Fig 4). A series of 9 consecutive points below the baseline was appreciated for LOS (Fig 4B, top). The first 2 data points (October and November 2020) were used in establishing the baseline for our balancing measure, discounting special cause variation. One point on the LOS S chart for the month of July 2021 was outside the upper control limit (Fig 4B, bottom). A closer look at the data from that month, and manual chart review, identified relatively fewer data points with relatively more having a LOS beyond 4 days, mostly attributable to high-risk social situations and monitoring for neonatal opiate withdrawal syndrome.

Our quality improvement study of >1300 infants was associated with an increase in the percentage of LPT and LBW infants being cared for exclusively in the nursery after interventions involving a new LPT/LBW nursery-based feeding protocol, dissemination of knowledge, and standardization of care for these higher-risk patient populations. Secondary measures were used to analyze the impact to feeding-related conditions and identify a reduction in hypoglycemia rates with no change in phototherapy utilization rates or weight loss relative to birth weight. The initiative was associated with increased numbers of LPT and LBW infants receiving exclusive nursery care without negatively impacting exclusive breast milk feeding rates or prolonging the length of stay. Overall, this initiative allowed for the preservation of the mother-infant dyad while providing high-value care.

Feeding protocols have been shown to affect feeding-related conditions in higher-risk neonates and facilitate standardization of care. Previous literature has mostly focused on feeding protocols in the NICU.^15,16  A proactive enteral nutrition protocol in moderately preterm LBW infants revealed improvements in hypoglycemia, IV fluid use, weight loss, and LOS.¹⁵  Conversely, a protocol by Burnham et al for LPT infants revealed no change in IV fluid use or breastfeeding, although it did reveal improved satisfaction among the nursing staff and physicians due to fewer IV insertions and decreased variation in practice.¹⁶  Our nursery-based feeding protocol and care standardization for LPT and LBW infants similarly was associated with improvements in hypoglycemia without impacting weight loss, which averaged 4.1% at baseline, a value already within physiologic norms.¹⁴  Exclusive breast milk feeding rates did not increase despite increased retention in the nursery setting of infants rooming in with mothers. This may have reflected the support of proactive enteral nutrition by supplementation endorsed in our feeding protocol. Preserving the mother-infant dyad, however, is known to alleviate birth-related physiologic stress and maintains the potential for improving breastfeeding among LPT infants.^22,23 

Reductions in the incidence of hypoglycemia seen in our study population aligned with the rationale and results published in the following reports.^{12–14,24,25}  Delays and low volumes in enteral feeds can exacerbate LPT infant risk toward hypoglycemia, whereas proactive enteral nutritional support can mitigate hypoglycemia.^12–14  One previous study revealed an increase in mean blood glucose values by 20mg/dL after supplementing breastfeeding with HDM, a value significantly exceeding the mean increase seen with supplementation by formula or expressed breast milk alone.²⁴  Continuous glucose monitoring has also revealed prophylactic administration of dextrose gel to increase blood glucose values of LPT and other at-risk infants.²⁵  Our findings, together with these reports, suggest that standardized approaches to providing early enteral carbohydrates can reduce hypoglycemia in LPT and LBW infants.

The lack of change in phototherapy rates among LPT and LBW infants in our study is consistent with previously reported mixed findings. Zecca et al showed no statistical difference in phototherapy initiation or duration among infants on a proactive feeding regimen.¹⁵  Whereas Burnham et al demonstrated that LPT infants require less phototherapy despite decreasing feed volumes to align with best practices.¹⁶  It remains possible that our phototherapy rates did not decrease despite a proactive feeding regimen due to prematurity being a well-established independent risk factor for neonatal hyperbilirubinemia.

This study had several limitations. Rehospitalization rates, known to be higher in LPT infants, were not measured.²⁶  Birth hospitalization periods >4 days have been associated with fewer rehospitalizations with feeding difficulty in exclusively breastfeeding infants cited as one of the most common causes for readmission.^26–28  Our data revealed an unchanged mean LOS of ∼2 days. Although there was a period of 9 months with a shorter LOS than the baseline, it is difficult to separate this finding from the shorter birth hospitalization period that paralleled the rise in hospitalizations during the coronavirus disease 2019 pandemic. It was reassuring, however, that these higher-risk newborn populations were discharged with feeding regimens inclusive of supplementation and ongoing lactation support options. Our single-site quality improvement study may limit generalizability beyond high-volume nurseries. In addition, despite efforts to consolidate the rollout of our initiatives on a single date aligned with the release of updates to the order set in our electronic health system, HDM authorization by bedside nurses occurred 4 months after project initiation due to administrative delays. Additional extrinsic factors, such as changes to the unit staff-to-patient ratio, may have also influenced our study outcomes; however, special cause variation was detected before this change in staffing.

Our quality initiative was associated with a prompt and sustained increase in the percentage of LPT and LBW infants receiving exclusive nursery care by implementing a nursery-based feeding protocol, disseminating knowledge, and improving the standardization of care for these higher-risk infants. Additional benefits included improvements in feeding-related conditions, such as hypoglycemia, without adversely affecting phototherapy rates, weight loss, exclusive breast milk feeding rates, or length of stay. An attribute of this study was its ability to safely preserve the mother-infant dyad by using high-value care.