Predictors and Risk Factors of Necrotizing Enterocolitis

Background: Necrotizing enterocolitis (NEC) is associated with high morbidity and mortality among preterm infants. Early intervention is critical, but existing predictors and risk factors remain inconsistent. Integrating bedside nursing assessments with machine learning (ML) may enable prediction of NEC before deterioration. Purpose: This integrative review synthesizes evidence on predictors and risk factors of NEC in very low birth weight (VLBW; <1500 g) infants to guide ML-based prediction models, support nursing assessment, and support early clinical decision making. Data Sources: MEDLINE, CINAHL, and Embase. Study Selection: We included quantitative studies published from 2015 to 2025, involving VLBW infants <32 weeks gestation and examining NEC predictors or risk factors. From 405 records, 325 abstracts were screened, and 14 studies met inclusion criteria. Data Extraction: Two reviewers independently assessed full texts, resolving differences through consensus. Data were organized thematically using an inductive approach with a predetermined extraction table. Study quality was evaluated with a Joanna Briggs Institute tool. Results: Across 14 studies, 55 predictors or risk factors were identified. Gestational age, birth weight, and human milk exposure were consistently associated with NEC risk. Other maternal, antenatal, delivery, nutritional, social, and clinical variables varied widely, with only 16 factors appearing in multiple studies. Most studies used retrospective cohort designs, and one incorporated ML methods. Implications for Practice and Research: Nurses are central to NEC prevention through human milk promotion, risk surveillance, and bedside advocacy. Future NEC prediction should integrate dynamic, contextual variables with standardized definitions and validation across diverse settings.