771 - STEM Evaluation of Neonatal CATCH: 30-day Post-Discharge Remote Management of Complex NICU Graduates

Publication Number: 1747.771

John Chuo, University of Pennsylvania Perelman School of Medicine, doylestown, PA, United States; Kristen Smith, Childrens Hospital of Philadelphia, Horsham, PA, United States; Lori Spotts, Childrens Hospital of Philadelphia, Doylestown, PA, United States; Meredith White, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Jenna Perkins, Children’s Hospital of Philadelphia, Villanova, PA, United States; Jess Boyle, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Sarah Hadley, CHOP, Garnet Valley, PA, United States; Bolu Oluwalade, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Matthew Coyle, Childrens Hospital of Philadelphia, Springfield, PA, United States; Bimal R.. Desai, Childrens Hospital of Philadelphia, Philadelphia, PA, United States

Background:

Background: Medically complex infants (MCI) comprise a fraction of births yet account for disproportionate health-care utilization after NICU discharge. Inadequacies of post-discharge care systems creates an opportune role for remote patient management (RPM). Digital monitoring enables timely, coordinated responses to condition-specific changes, with potential to avert avoidable utilization. Building on prior published work , we expanded the Neonatal CATCH program (Collaborative Assistance and Telehealth for Complex Homecare) in 2023 with a 30-day period of remote therapeutic monitoring (RTM) after discharge.

Objective: The CATCH program co-manages tube-feed-dependent MCIs alongside primary and specialty care teams, delivering coordinated care, guidance and caregiver education. Our objective was to reduce unnecessary healthcare utilization while optimizing integrated family centered care for medically complex infants.

Design/Methods: Methods: In a pragmatic, staged roll-out (March 2023-February 2025), we prospectively followed eligible NICU graduates discharged on tube feeds, yielding two cohorts: enrolled (E) and non-enrolled (NE), plus a historical "would-have-been-eligible" cohort. Interventions included telemedicine visits, 7 ≥ bidirectional text check-ins, and collaboration with primary and subspecialty care. Outcomes included 30-day readmission (RA30), potentially avoidable readmission (ARA30), 30-day ED visit (EDV30), and stable-to-discharge days (STDD), Karitane Parental Confidence Scale, and Net Promoter Score (NPS). STEM Framework guided multi stakeholder centered measurement approach while analyses combined statistical process methods (SPC), bivariate tests, and logistic regression modeling.

Results: We identified 1380 eligible infants - 853 in the historical cohort, 287 and 240 infants were, respectively, enrolled (E) and not-enrolled (NE) during CATCH implementation. Mean weight gain did not differ (18.8 vs. 22.3 g; P=0.28). RA30 was lower in E vs. NE (13% vs. 22%; P< 0.01), as was EDV30 (20% vs. 28%; P=0.03); STDD was shorter in E (P < 0.001). ARA30 favored E (5.4% vs. 10.5%) without significance (P=0.67). SPC complemented inferential statistical findings and provided insight into process patterns. At least one "good catch" intervention occurred in 23% of enrolled infants; NPS was 89.

Conclusion(s): A short, structured RPM program after NICU discharge was associated with lower 30-day readmissions, ED utilization and earlier discharge. Further study needed with attentiveness to social determinants and interpreter access. Synergizing QI and inferential analytics is critical to understanding data

Neonatal CATCH Program Workflow
Within 1 week of discharge, families had a 30-minute telemedicine visit. Over 30 days, caregivers received seven bidirectional text "chats" focused on baby's weight, growth, feeding, and respiratory status, and could request non-emergent call-backs from the CATCH team within one business day.

CATCH Program Outcomes
STEM Framework used to identify a measure set important to neonates, families, providers, health systems, payors, and legislators across three critical domains - Clinical, Healthcare Delivery Quality and Cost, Personal Experience - along with a disparity stratifier.

Run and Statistical Process Control Charts of Readmission Outcomes amongst Historical, E- and NE-Cohorts
Enrolled Cohort showed a non random variation towards lowered readmission rate, particularly in a more complex subgroup of neonates with a history of respiratory failure