Developing Best Practice for Generative AI in Health Economic Evaluation: Evidence From a Systematic Review and Stakeholder Engagement
Author(s)
Steve Sharp, BA, MSc1, Kusal Lokuge, PhD2, Jamie Elvidge, BA, MSc2.
1Scientific Adviser, National Institute for Health and Care Excellence, Manchester, United Kingdom, 2National Institute for Health and Care Excellence, Manchester, United Kingdom.
1Scientific Adviser, National Institute for Health and Care Excellence, Manchester, United Kingdom, 2National Institute for Health and Care Excellence, Manchester, United Kingdom.
OBJECTIVES: Generative artificial intelligence (GenAI) has emerged in the current decade as a paradigm-shifting technology with potential to transform the process of health economic evaluation (HEE), a resource-intensive element of health technology assessment. This updated version of the first pioneering systematic literature review on this topic aims to incorporate emerging new evidence of applications of GenAI in HEE and its potential advantages, challenges and limitations.
METHODS: We searched multiple databases for English-language, publicly available literature from the last search date of the original review (March 2025), and hand-searched the ISPOR presentations database (2025) for articles describing or investigating the use of GenAI in HEE. Quantitative data on performance outcomes was collected along with qualitative data on stakeholder opinions and experience. Results are presented as a narrative synthesis of new evidence in 2025 and the overall cumulative evidence published to date.
RESULTS: The original search identified 25 studies, including 16 case studies of GenAI applications such as conceptualisation (2/16), replication (4/16), adaptation (3/16) and reporting (3/16), though most were published in formats that lacked detail and precluded quality assessment. Less-explored applications included validation and updating. In this dynamic topic, this update is expected to identify an increase in full publications across the range of GenAI applications in HEE.
CONCLUSIONS: This updated systematic review provides a further snapshot of the current evidence landscape covering potential benefits of GenAI across multiple applications to HEE. It highlights ongoing research gaps where confirmatory research is still needed to prove a range of use cases and to address perceived barriers to implementation. It also complements international stakeholder engagement by NICE to inform best practice recommendations for HTA agencies to follow in using GenAI in HEE effectively and responsibly.
METHODS: We searched multiple databases for English-language, publicly available literature from the last search date of the original review (March 2025), and hand-searched the ISPOR presentations database (2025) for articles describing or investigating the use of GenAI in HEE. Quantitative data on performance outcomes was collected along with qualitative data on stakeholder opinions and experience. Results are presented as a narrative synthesis of new evidence in 2025 and the overall cumulative evidence published to date.
RESULTS: The original search identified 25 studies, including 16 case studies of GenAI applications such as conceptualisation (2/16), replication (4/16), adaptation (3/16) and reporting (3/16), though most were published in formats that lacked detail and precluded quality assessment. Less-explored applications included validation and updating. In this dynamic topic, this update is expected to identify an increase in full publications across the range of GenAI applications in HEE.
CONCLUSIONS: This updated systematic review provides a further snapshot of the current evidence landscape covering potential benefits of GenAI across multiple applications to HEE. It highlights ongoing research gaps where confirmatory research is still needed to prove a range of use cases and to address perceived barriers to implementation. It also complements international stakeholder engagement by NICE to inform best practice recommendations for HTA agencies to follow in using GenAI in HEE effectively and responsibly.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EE677
Topic
Economic Evaluation
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies
Disease
No Additional Disease & Conditions/Specialized Treatment Areas