Abstract: SA-PO018
Health 360x Registry (H360xR): Artificial Intelligence (AI)-Enabled Equitable Access to Point-of-Care Decentralized Clinical Trials
Session Information
- Augmented Intelligence, Large Language Models, and Digital Health
October 26, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Obialo, Chamberlain I., Accuhealth Technologies Inc, Atlanta, Georgia, United States
- Lee, Eva, Accuhealth Technologies Inc, Atlanta, Georgia, United States
- Ofili, Elizabeth O., Morehouse School of Medicine, Atlanta, Georgia, United States
Background
The lack of diversity in clinical trial population and non-involvement of community physicians are major barriers to recruitment and retention of representative populations in industry sponsored and National Institute of Health [NIH] research studies.
Methods
AH received an NIH Fast Track Small Business Innovation Research (SBIR) Award to implement a scalable access to point of care decentralized clinical trials (DCT), by systematically addressing gaps in the practice infrastructure of resource limited small-medium practices. The conceptual model is the Practical, Robust Implementation and Sustainability (PRISM) that targets barriers at the patient, provider and practice levels. We developed H360x Clinical Research Platform and Registry as an artificial Intelligence [AI] enabled digital health platform (DHT) that uses machine learning to analyze electronic health records [EHR] and identify suitable clinical trial participants.
Primary outcome: Adoption and implementation progress, with performance indicators based on the H360xR protocol; IRB/informed consent; PhenX surveys for social determinants; Recruitment and Retention rates. Method of measurement: surveys; observation; key informant interviews; focus groups.
Results
H360xR includes data from 35 community-based practices, who have conducted 3 NIH and 1 industry sponsored study. Over a 3- year period, H360xR had 7513 participants. Median age 70 yr., 90% black, 7% white, 58% female, 15% rural residents while 51% have annual income <$25,000. H360xR supplements patient clinical and EMR data with surveys and questionnaires on social determinants. Qualitative analysis demonstrated that EMR AI/ML recruitment resource and Training of site teams are dominant themes; 75% references to H360x patient engagement, 85% references to training of site teams and research workflows (p<0.02 compared to baseline). H360x system usability scale was 95%.
Conclusion
H360xR resource support of trusted healthcare providers, who are research naïve, enabled successful participation in NIH and industry sponsored studies. The DCT/DHT model can effectively scale the participation of underserved populations in clinical trials. Ongoing analysis will generate a full representation of multi-dimensional patient characteristics, with human-in-the-loop AI/ML, and adapted to practice workflows.
Funding
- Other NIH Support – AMGEN