LARGE-SCALE AI-ASSISTED MAPPING OF HEALTH ECONOMICS AND OUTCOMES RESEARCH EVIDENCE GAPS ACROSS GASTRO-HEPATIC DISEASES...

Author(s)

Corrina Mau, MPH1, Weicheng Ye, MPH2, Thitima Kongnakorn, PhD3.
1Thermo Fisher Scientific, Toronto, ON, Canada, 2Thermo Fisher Scientific, Waltham, MA, USA, 3Thermo Fisher Scientific, Bangkok, Thailand.
OBJECTIVES: To refine an AI-assisted health economics and outcomes research (HEOR) evidence map for gastro-hepatic diseases—a disease area causing morbidity and healthcare burden in Asia—and to identify evidence concentrations and gaps across disease domains and HEOR subtypes.
METHODS: PubMed records (2000-2025; English) were extracted using predefined HEOR/intervention terms, deduplicated, and screened/classified using a two-stage large language model pipeline. Gastro-hepatic HEOR records were normalized using a reproducible taxonomy grouping diseases into 13 domains. The six most prevalent disease domains in Asia were further analyzed across 11 HEOR evidence categories. Record counts and within-disease-area percentages were summarized.
RESULTS: Of 63,223 unique records identified, 45,918 were HEOR-relevant. Broad gastro-hepatic retrieval identified 7,630 records; however, disease-domain classification showed that 4,933 records were unspecified or outside predefined gastro-hepatic domains. Among the classified records (n=2,697), the six priority domains comprised 1,836 studies: liver hepatitis (n=666; 36.3%), inflammatory bowel disease (n=303; 16.5%), colorectal disease (n=292; 15.9%), liver neoplasms (n=255; 13.9%), esophageal/reflux/upper gastrointestinal disease (n=181; 9.9%), and gastric disease (n=139; 7.6%). Among 1,836 records, economic evaluation (n=1,546; 84.2%) dominated, whereas all other HEOR categories—including real-world evidence (RWE) outcomes, HEOR systematic reviews, patient-reported outcomes/health-related quality of life, treatment patterns, cost/economic burden, burden of disease, access/policy/value assessment, healthcare resource utilization, and HEOR methods—each accounted for <4% of studies. Across all six domains, economic evaluation was the predominant evidence category. Within economic evaluations, cost-utility analyses (n=761; 49.2%) and cost-effectiveness analyses (n=725, 46.9%) were most common, followed by budget impact models (n=21, 1.4%).
CONCLUSIONS: AI-assisted mapped priority disease-domain evidence aligned with Asia’s leading gastro-hepatic disease burden. The HEOR evidence base was heavily concentrated in economic evaluations, with notable gaps in domains that may help inform economic evaluations, particularly RWE, patient-centered outcomes, evidence synthesis/methods, and access/policy research. These findings highlight priority areas for future HEOR research.

Conference/Value in Health Info

2026-09, ISPOR Asia Pacific 2026, Bangkok, Thailand

Value in Health, Volume 55, Issue S1

Code

MSR2

Topic

Methodological & Statistical Research

Topic Subcategory

Artificial Intelligence, Machine Learning, Predictive Analytics

Disease

No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Gastrointestinal Disorders

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

×