Evaluating Neonatal Nutrition Protocols in a Large Urban Hospital Network

Thesis Abstract

Neonatal nutrition practices across a network of seven tertiary urban hospitals influence short- and long-term infant outcomes, yet heterogeneous protocols, inconsistent feeding advancement strategies, and variable documentation hinder optimal growth and clinical outcomes. This study aims to evaluate the alignment, effectiveness, and equity of neonatal nutrition protocols within a large urban hospital network and to identify modifiable determinants of adherence and outcomes. The specific objectives are (1) to map and compare current enteral and parenteral nutrition protocols across hospitals; (2) to quantify associations between protocol adherence and key neonatal outcomes (growth velocity, time to full feeds, incidence of necrotizing enterocolitis, length of stay, and readmission rates); (3) to examine the influence of hospital-level factors (bed capacity, nurse-to-patient ratio, presence of lactation support, and electronic medical record completeness) on protocol adherence; (4) to explore clinician and caregiver perspectives on barriers and facilitators to protocol implementation; and (5) to develop evidence-based recommendations to harmonize practices and improve outcomes. A mixed-methods design is employed, integrating a retrospective cohort analysis with qualitative exploration grounded in complexity theory and the socio-ecological model. The quantitative component analyzes data from 3,500 neonates admitted to the network between 2018 and 2023, including birth weight, gestational age, feeding type, time to first and full feeds, nutritional intake records, and clinical outcomes. Multivariable hierarchical regression models evaluate the association between adherence to standardized nutrition protocols and outcomes, adjusting for confounders such as illness severity and prematurity. Time-to-event analyses (Cox models) assess time to reach full feeds, and logistic regression estimates odds of necrotizing enterocolitis and sepsis. The qualitative component includes semi-structured interviews with 40 clinicians (neonatologists, nurses, dietitians) and 30 parent caregivers across the seven sites, analyzed via thematic analysis to identify perceived barriers, enablers, and contextual determinants. Data triangulation integrates quantitative findings with qualitative insights to explain observed variability. Key expected findings include (a) substantial inter-hospital variation in protocol components, such as initiation thresholds for enteral feeds and advancement rates, with higher adherence associated with standardized, order-set–driven protocols; (b) faster attainment of full feeds and reduced hospital length of stay in units with higher adherence, without increased rate of feeding intolerance or NEC, after adjusting for confounders; (c) identification of modifiable hospital-level factors—adequate lactation support, robust documentation in electronic health records, and continuous education—that predict better adherence; (d) barriers such as workload pressure, inconsistent documentation, and communication gaps between teams, with facilitators including protocol champions, structured training, and parent engagement initiatives. The analysis will be informed by theoretical frameworks including Donabedian’s structure-process-outcome model and the Normalization Process Theory to interpret implementation dynamics. The study’s contribution to knowledge lies in providing comprehensive, network-level evidence on how standardized neonatal nutrition protocols influence outcomes in diverse urban hospital settings, clarifying how organizational context shapes implementation, and offering actionable, scalable recommendations for harmonizing care pathways. Methodologically, it demonstrates the value of combining robust statistical modeling with qualitative implementation science to elucidate causal pathways and practical barriers. The anticipated conclusion emphasizes the necessity of harmonized, evidence-based nutrition pathways supported by interoperable electronic documentation, dedicated lactation resources, and continuous audit-and-feedback mechanisms. Policy and practice recommendations include the adoption of uniform insulin-free feeding advancement schedules where appropriate, mandatory protocol checklists integrated into computerized order sets, enhanced multidisciplinary rounds focusing on nutrition, and ongoing training for staff, with periodic outcome monitoring to sustain improvements across the network.