FROM REACTIVE TO PREDICTIVE: A HUMAN-IN-THE-LOOP AI FRAMEWORK FOR ANTICIPATORY PICO SCOPING FOR EU JOINT CLINICAL ASSESSMENTS
Author(s)
Aditi Bajpai, PharmD, MHS1, Angeline Babitha Dhas, BS1, Maria Rizzo, BSc, MS2, Lingamaiah Doddolla, B.Pharm, M.Pharm3, Revanth M, B.E.3, Viji Queen V, Sr., PharmD4, Meghan Oates-Zalesky, MSc5.
1MadeAi, Cambridge, MA, USA, 2Independent, Kent, United Kingdom, 3MadeAi, Nagercoil, India, 4MadeAI, Nagercoil, India, 5Chief Marketing Officer, MadeAi, Cambridge, MA, USA.
1MadeAi, Cambridge, MA, USA, 2Independent, Kent, United Kingdom, 3MadeAi, Nagercoil, India, 4MadeAI, Nagercoil, India, 5Chief Marketing Officer, MadeAi, Cambridge, MA, USA.
OBJECTIVES: Under the European Union (EU) Health Technology Assessment Regulation (HTAR), Joint Clinical Assessments (JCAs) must reflect Member States' (MS) needs, often resulting in multiple population, intervention, comparator, and outcome (PICO) frameworks. Failure to anticipate these requirements may create evidence gaps, as shown in the first published JCA (tovorafenib). We aimed to develop a human-in-the-lead artificial intelligence (AI) framework for anticipatory PICO scoping to support transparent, auditable evidence planning for JCAs.
METHODS: EU HTA Coordination Group guidance was reviewed to identify procedural requirements and best practices for JCA PICO scoping. These requirements were synthesized with structured input from HEOR and AI experts to develop a human-in-the-lead framework for anticipatory JCA PICO scoping that pairs AI-assisted evidence planning with expert oversight and provenance-tagged outputs to preserve auditability.
RESULTS: A human-in-the-lead framework was developed to support anticipatory JCA PICO scoping while maintaining expert oversight. It comprised six stages: (1) AI-assisted development of a baseline PICO from regulatory and clinical evidence sources; (2) identification of potential MS variation in populations, comparators, and outcomes; (3) consolidation of anticipated requirements into a defensible set of PICOs with documented rationale; (4) assessment of comparator relevance, including individualized comparators; (5) identification and prioritization of evidence gaps across anticipated PICOs; and (6) generation of protocol-ready outputs for SLRs and ITCs. The framework generated anticipated PICO specifications, comparator assessments, and evidence-gap maps, while maintaining transparency and traceability through expert review at each stage.
CONCLUSIONS: A human-in-the-lead framework was developed by operationalizing principles from EU JCA scoping and evidence synthesis guidance within an AI-assisted workflow. Combining automated evidence processing with expert review at predefined decision points provides a transparent, traceable approach to anticipatory PICO planning and supports downstream SLR and ITC activities. Future validation using published JCAs is warranted.
METHODS: EU HTA Coordination Group guidance was reviewed to identify procedural requirements and best practices for JCA PICO scoping. These requirements were synthesized with structured input from HEOR and AI experts to develop a human-in-the-lead framework for anticipatory JCA PICO scoping that pairs AI-assisted evidence planning with expert oversight and provenance-tagged outputs to preserve auditability.
RESULTS: A human-in-the-lead framework was developed to support anticipatory JCA PICO scoping while maintaining expert oversight. It comprised six stages: (1) AI-assisted development of a baseline PICO from regulatory and clinical evidence sources; (2) identification of potential MS variation in populations, comparators, and outcomes; (3) consolidation of anticipated requirements into a defensible set of PICOs with documented rationale; (4) assessment of comparator relevance, including individualized comparators; (5) identification and prioritization of evidence gaps across anticipated PICOs; and (6) generation of protocol-ready outputs for SLRs and ITCs. The framework generated anticipated PICO specifications, comparator assessments, and evidence-gap maps, while maintaining transparency and traceability through expert review at each stage.
CONCLUSIONS: A human-in-the-lead framework was developed by operationalizing principles from EU JCA scoping and evidence synthesis guidance within an AI-assisted workflow. Combining automated evidence processing with expert review at predefined decision points provides a transparent, traceable approach to anticipatory PICO planning and supports downstream SLR and ITC activities. Future validation using published JCAs is warranted.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
MSR289
Topic
Health Technology Assessment, Methodological & Statistical Research, Study Approaches
Topic Subcategory
Artificial Intelligence, Machine Learning, Predictive Analytics
Disease
No Additional Disease & Conditions/Specialized Treatment Areas