299 - AI-Enhanced Research Management System Accelerates Pediatric Research While Advancing Health Equity: A Multi-Domain Feasibility Study
Friday, April 24, 2026
5:30pm - 8:00pm ET
Publication Number: 1285.299
Allison e. Curry, Children's Hospital of Philadelphia, Green Cove Spings, FL, United States; Ryan Warren, Aims Innovations, Green Cove Springs, FL, United States; Joel D. Hillier, Perelman School of Medicine at the University of Pennsylvania, Eagle, ID, United States; Dominique G.. Ruggieri, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States; Johnathon Ehsani, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States
Professor Children's Hospital of Philadelphia Green Cove Spings, Florida, United States
Background: Pediatric research faces significant operational barriers that delay evidence translation to practice. Traditional approaches require manual coordination across fragmented platforms for recruitment, consent, data collection, and analysis—creating substantial staff burden, restricting research to well-resourced institutions, and excluding non-English-speaking families. These inefficiencies disproportionately affect community practices and underserved populations, ultimately slowing improvements in child health care delivery. Objective: To develop and evaluate an AI-enhanced, HIPAA-compliant Research Management System (RMS) reducing study completion time, staff burden, and enabling multilingual engagement while maintaining rigorous data governance. Design/Methods: We developed a HIPAA-compliant Research Management System consolidating the research lifecycle—planning, recruitment, data collection, analysis, reporting—into unified modules with role-based access, automated audit trails, and IRB-ready documentation. Innovations include: (1) centralized participant management; (2) intelligent multilingual conversational agents enabling concurrent engagement across multiple channels; and (3) AI-assisted data cleaning and analysis. We completed a feasibility pilot at University of Iowa (n=20) and initiated validation studies at University of Washington (firearm prevention), Harvard University (traumatic brain injury), and Johns Hopkins University (teen driving safety). Primary outcomes: study completion time, staff hours, recruitment metrics, and system usability. Results: In the feasibility pilot, the system reduced timeline from 4 weeks to 2 days and staff effort from 50 to 4 hours. Participants reported conversational AI assistants were clear, natural, and easy to understand. Validation study results will be available by conference time.
Conclusion(s): An AI-enabled, HIPAA-compliant unified workflow is feasible and can materially shorten timelines and reduce staff effort while maintaining operational documentation through built-in governance. By compressing timelines and decreasing resource requirements, the system has potential to enable community practices to conduct rigorous studies. Multilingual capabilities may expand research inclusivity. This technology transforms sequential, tool-switching tasks into a coordinated, AI-supported process—reducing administrative burden, improving reproducibility, and enabling scalable operations across diverse teams and settings to support more inclusive and timely pediatric research.
