Assessing the Use of Artificial Intelligence in Publication-Grade Research: A Literature Review of Applications in HEOR
Author(s)
Nandini Hadker, MA, Aishwarya Kulkarni, MS, Brittany Carletha Smith, MBA, MPH, Saumil Jadhav, BS, MS, PhD, Amod Athavale, BS, MS, PhD.
Trinity Life Sciences, Waltham, MA, USA.
Trinity Life Sciences, Waltham, MA, USA.
OBJECTIVES: The use of artificial intelligence (AI) in health economics and outcomes research (HEOR) is expanding rapidly. This study evaluated the applications of AI in publication-grade qualitative and quantitative primary data research, and secondary research.
METHODS: A literature search was conducted in PubMed for English-language, human studies from the past five years using terms including “artificial intelligence,” “patient,” “physician,” “qualitative,” “quantitative,” “mixed methods,” and “health economics and outcomes research (HEOR).” A total of 152 articles were identified and categorized by study type and theme.
RESULTS: AI-related HEOR publications increased from 5 in 2020 to 46 in 2024, with 28 already published in 2025 (mid-year). The majority were qualitative studies (n = 61, 40.1%), which explored stakeholder experiences, perceptions of, and barriers to AI implementation through interviews and focus groups. Systematic reviews and qualitative syntheses comprised 28 studies (18.4%), highlighting the importance of trust, transparency, and explainability for AI adoption, while identifying barriers such as data privacy, integration challenges, and potential for health disparities. There were 5 (3.2%) systematic reviews and meta-analyses studies that synthesized evidence on AI’s effectiveness, highlighting both opportunities and barriers, and underscoring the need for further rigorous, real-world research to support safe and effective AI integration in healthcare. Mixed methods studies accounted for 4 studies (2.6%), revealing that while clinicians and patients recognize AI’s potential to improve accuracy and efficiency, concerns remain regarding reliability, accountability, and professional roles.
CONCLUSIONS: There is increasing interest in AI-related HEOR publications, especially on adoption, integration, and ethical aspects of AI. A few studies demonstrate the deployment and use of AI to generate publication-grade primary or secondary research, but it is not yet a mainstay in published work. This abstract was generated with the assistance of AI, following ISPOR EU abstract guidelines.
METHODS: A literature search was conducted in PubMed for English-language, human studies from the past five years using terms including “artificial intelligence,” “patient,” “physician,” “qualitative,” “quantitative,” “mixed methods,” and “health economics and outcomes research (HEOR).” A total of 152 articles were identified and categorized by study type and theme.
RESULTS: AI-related HEOR publications increased from 5 in 2020 to 46 in 2024, with 28 already published in 2025 (mid-year). The majority were qualitative studies (n = 61, 40.1%), which explored stakeholder experiences, perceptions of, and barriers to AI implementation through interviews and focus groups. Systematic reviews and qualitative syntheses comprised 28 studies (18.4%), highlighting the importance of trust, transparency, and explainability for AI adoption, while identifying barriers such as data privacy, integration challenges, and potential for health disparities. There were 5 (3.2%) systematic reviews and meta-analyses studies that synthesized evidence on AI’s effectiveness, highlighting both opportunities and barriers, and underscoring the need for further rigorous, real-world research to support safe and effective AI integration in healthcare. Mixed methods studies accounted for 4 studies (2.6%), revealing that while clinicians and patients recognize AI’s potential to improve accuracy and efficiency, concerns remain regarding reliability, accountability, and professional roles.
CONCLUSIONS: There is increasing interest in AI-related HEOR publications, especially on adoption, integration, and ethical aspects of AI. A few studies demonstrate the deployment and use of AI to generate publication-grade primary or secondary research, but it is not yet a mainstay in published work. This abstract was generated with the assistance of AI, following ISPOR EU abstract guidelines.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
SA14
Topic
Study Approaches
Topic Subcategory
Literature Review & Synthesis
Disease
No Additional Disease & Conditions/Specialized Treatment Areas