ARTIFICIAL INTELLIGENCE (AI) ACROSS THE PRO LIFECYCLE: EVIDENCE AND EXPERT PERSPECTIVES ON OPPORTUNITIES AND CHALLENGES

Author(s)

Jason C. Cole, PhD1, Sheetal Sharma, MSc2, Abhra Roy Choudhury, MSc3, Tricia Dimakunne, -4, Pixy Banerjee, -3, Adam Aljundi, -4, Grainne McElligott, -5.
1President & Chief Science Officer, P3 Research, Torrance, CA, USA, 2PharmaQuant Insights Pvt, Panchkula, India, 3PharmaQuant Insights Pvt, Kolkata, India, 4P3 Research, Torrance, CA, USA, 5PharmaQuant, Dublin, Ireland.
OBJECTIVES: To examine current perspectives on how AI is being used in PRO research by integrating literature and expert insights, focusing on applications, perceived benefits, and challenges.
METHODS: A rapid literature review (2018-2026) was conducted to identify and characterize published applications of AI across the PRO lifecycle. Data were extracted on AI methodologies, use cases, reported benefits, and limitations. In parallel, semi-structured interviews were conducted with 6 key opinion leaders (KOLs), including PRO methodologists from industry/consulting and AI specialists familiar with PROs to capture expert views on uses of AI in PRO research. Interview transcripts underwent thematic analysis to identify common perspectives.
RESULTS: Fifty-five publications were identified across therapeutic areas and all stages of the PRO lifecycle. Overall, the literature suggested that AI is most beneficial in supporting early-stage PRO research activities, including concept elicitation and instrument development, with growing applications throughout the broader PRO lifecycle. AI approaches also supported instrument optimization, with machine learning-based computerized adaptive testing reducing respondent burden by up to 50% while maintaining measurement precision. KOLs agreed that AI integration in PRO research remains early-stage, with greatest near-term value in qualitative and early-lifecycle activities (concept elicitation, literature review, translation) rather than psychometric validation or regulatory decision-making. KOLs had concerns regarding hallucination, auditability, and regulatory risk. Perspectives diverged between KOLs on benefits of synthetic/in-silico patient cohorts. Literature and KOL research revealed similarities: large language models were used for translation, concept simulation, and early-stage validation, while AI-based fit-for-purpose analyses identified content gaps in widely used instruments. Key challenges included bias in digital data sources, need for human oversight, limited generalizability of synthetic data, and lack of regulatory guidance for AI-derived PRO evidence.
CONCLUSIONS: AI affords substantial potential for PRO research but demands human oversight, rigorous validation, and urgent development of PRO-specific industry guidance before broader adoption.

Conference/Value in Health Info

2026-11, ISPOR Europe 2026, Vienna, Austria

Value in Health, Volume 29, Issue 12S

Code

PCR48

Topic

Organizational Practices, Patient-Centered Research, Study Approaches

Topic Subcategory

Instrument Development, Validation, & Translation, Patient-reported Outcomes & Quality of Life Outcomes

Disease

No Additional Disease & Conditions/Specialized Treatment Areas

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×