458 - Cultural Health Navigators Promote Equitable Developmental Surveillance and Early Intervention Access in a Refugee NICU Population

Publication Number: 4449.458

Ariana Iranmahboub, Creighton University School of Medicine, Phoenix, AZ, United States; Srinidhi Seshadri, Creighton University School of Medicine, Phoenix, AZ, United States; Isheeta Tewari, Valleywise Health, Phoenix, AZ, United States; Erika Anaya, Phoenix Children's Hospital, Phoenix, AZ, United States; Claudia Islas, Valleywise Health Medical Center, Phoenix, AZ, United States; Jason Couch, Creighton University, Phoenix, AZ, United States; Alyssa Bernardi, Creighton University School of Medicine, Phoenix, AZ, United States

Background: Admission to the newborn intensive care unit (NICU) remains a significant risk factor for long-term neurodevelopmental (ND) impairment. Early identification and referral to state Early Intervention Programs (EIP) improve ND outcomes, yet socioeconomic and cultural barriers limit access. Valleywise Health, a Federally Qualified Health Center (FQHC) and safety-net hospital in Phoenix, AZ, serves a diverse population, including refugee families through its Women's Refugee Services. This program employs Cultural Health Navigators (CHNs) to address barriers such as language, transportation, and health-system navigation. However, refugee and immigrant children remain underrepresented in studies examining developmental surveillance and EIP access.

Objective: To determine the incidence of ND diagnoses and EIP referrals among infants born at Valleywise Health who were admitted to the NICU and assess whether an integrated Refugee Health Program with CHN support promotes equitable developmental support for these patients.

Design/Methods: A retrospective chart review was performed for infants born in 2013-2022 who were followed in FQHC clinics at Valleywise Health. Data included demographics, refugee status, NICU comorbidities, ICD-10 developmental delay diagnoses, EIP referrals, and CHN involvement. Associations with social determinants of health were assessed using chi-squared/Fisher's exact tests, and logistic regression identified predictors of EIP referral.

Results: Among 329 patients, 132 (40.1%) had a developmental delay diagnosis and 122 (37.1%) received an EIP referral. Refugee and non-refugee infants had comparable rates of developmental delay diagnoses (37.7% vs 40.6%, p = 0.76) and EIP referrals (41.5% vs 36.2%, p = 0.53; Fig. 1). Outcomes did not differ by insurance type, language, or CHN use (p > 0.4). CHN-supported families showed equivalent rates of both screening and referral. Multivariable logistic regression identified gestational age as the only independent predictor of EIP referral (OR 0.86 per week, 95% CI 0.78-0.95, p = 0.003; Fig 2A, 2B).

Conclusion(s): In this diverse cohort, developmental delay diagnoses and EIP referrals were common, with no disparities by refugee status, language, or CHN involvement. These findings highlight the success of Valleywise Women's Refugee Service and CHN model in promoting equitable developmental surveillance and early intervention access. Embedding CHNs within FQHCs may sustain equitable developmental care for refugee populations and offer a scalable framework for advancing health equity across safety-net systems.

Figure 1: Developmental Delay Diagnosis and EIP Referral by Refugee Status
Across all patients, 37.7% of refugee patients and 40.6% of non-refugee patients were diagnosed with developmental delay (p = 0.76), and 41.5% of refugee patients and 36.2% of non-refugee patients received Arizona Early Intervention Program (AzEIP) referrals (p = 0.53).

Figure 2A: Logistic Regression of Odds Ratios for Predictors of EIP Referral
Predictors included necrotizing enterocolitis (NEC), intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), CHN use, bronchopulmonary dysplasia (BPD), patent ductus arteriosus (PDA), English language preference, AHCCCS insurance, gestational age, small for gestational age (SGA), seizure, and sepsis.

Figure 2B. Table of Logistic Regression Results for Predictors of EIP Referral
Adjusted odds ratios (OR), 95% confidence intervals (CI), and p-values for predictors of Arizona Early Intervention Program (AzEIP) referral. Variables included gestational age, AHCCCS insurance, English language, CHN use, BPD, IVH, seizure disorder, ROP, NEC, PDA, sepsis, and SGA. NEC was very rare, leading to unstable or over-inflated odds ratio estimates.