AUTOMATING BRIEFING BOOK DEVELOPMENT FOR NICE HEALTH TECHNOLOGY ASSESSMENT SUBMISSIONS USING GENERATIVE AI

Author(s)

Sukriti Sharma, MSc1, Sumeet Attri, MPharm1, Barinder Singh, RPh2, Shubhram Pandey, MSc1, Rajdeep Kaur, PhD1, Nicola Waddell, HNC2, Michael Marentette, MBA3;
1Pharmacoevidence, Mohali, India, 2Pharmacoevidence, London, United Kingdom, 3Pharmacoevidence, Montreal, QC, Canada
OBJECTIVES: Briefing book (BB) documents facilitate scientific dialogue, which informs product development strategies, strengthening future health technology assessment (HTA) submissions and supporting reimbursement potential. BB development is resource-intensive, requiring substantial time and expert input. This study examined the feasibility of applying generative artificial intelligence (GenAI) within a Retrieval‑Augmented Generation (RAG) framework, with structured human oversight, to automate the generation of relevant, accurate and traceable BB content, for HTA submission.
METHODS: A structured workflow was designed to automate the generation of BB content for NICE submission with integrated human oversight. Source documents were processed and standardized into structured markdown format indexed within a RAG pipeline and specialized agents generated relevant sections aligned with the NICE template. Section-specific agents generated BB content, including disease background, current management pathways, unmet needs, study design, and key questions. Previous submissions for appropriate comparators served as reference materials to frame key questions to seek HTA advice.
RESULTS: Relevant sections of the BB generated using the GenAI- and RAG-based framework were validated by a subject matter expert (SME) for relevance, accuracy, and traceability. The SME endorsed AI-generated content for disease background and management sections with minimal inputs required for the unmet needs section. The majority of GenAI-generated questions were deemed appropriate with additional questions added through human review. The RAG‑powered framework also enabled accurate cross-referencing of the content. Overall, the hybrid AI-human workflow achieved an estimated 80-85% reduction in development time compared with traditional manual approaches.
CONCLUSIONS: This study demonstrates the feasibility of GenAI-generated BB documents for HTA submissions, achieving substantial reductions in time and effort while maintaining relevance, accuracy and traceability. Further research is warranted to evaluate generalizability across different HTA bodies and broader application areas.

Conference/Value in Health Info

2026-05, ISPOR 2026, Philadelphia, PA, USA

Value in Health, Volume 29, Issue S6

Code

MSR42

Topic

Methodological & Statistical Research

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

×