Work

Emergent Health Preferences in Large Language Models: A Time-Trade-Off Analysis Using EQ-5D-3L

Year: 2026

Type: conference-paper

Abstract: We investigate whether Large Language Models (LLMs) exhibit emergent health-state preferences when evaluated with the standard Time Trade-Off (TTO) method and EQ-5D-3L health-state descriptions. We elicit utility values for all 243 EQ-5D-3L states from multiple LLMs under neutral and demographic persona prompts, and compare the resulting value sets with established human population norms. We report model-specific patterns and systematic deviations from human preferences. These findings inform the reliability of using LLMs in value-sensitive healthcare contexts.

Source: IEEE International Conference on Intelligent Engineering Systems (INES)

DOI: 10.1109/ines69513.2026.11661097

Authors: Patrik P. Süli, György Eigner, Zsombor Zrubka

Institution: Obuda University

Language: English

Topic: Health Systems, Economic Evaluations, Quality of Life

Subfield: Economics and Econometrics

Field: Economics, Econometrics and Finance

Domain: Social Sciences

SDG: Good health and well-being

Open Access status: hybrid

OpenAlex ID: W7204184265

FWCI:

Cites: 23

Cited by:

Related to: 10