Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Jun 1, 2026
Date Accepted: Jun 30, 2026
Preferences for Artificial Intelligence-Enabled Healthcare Technologies: A Systematic Review of Discrete Choice Experiments and Reporting Quality Assessment Using the DIRECT Checklist
ABSTRACT
Background:
Artificial intelligence (AI) is increasingly being integrated into healthcare, making it important to understand stakeholder preferences for AI-enabled technologies. Although discrete choice experiments (DCEs) are widely used to elicit preferences, evidence on preference attributes, willingness-to-pay, and reporting quality in AI-related DCEs has not been systematically synthesized.
Objective:
This systematic review aims to synthesize stakeholder preferences for AI-enabled healthcare technologies elicited through discrete choice experiments (DCEs) and assess reporting quality using the DCE Reporting Checklist (DIRECT). The goal is to identify key preference attributes and methodological gaps to guide the development of AI technologies aligned with real-world needs.
Methods:
A systematic review was conducted following PRISMA guidelines, with databases searched from inception to March 2026. DCE methodological characteristics, including experimental design and econometric models, were summarized descriptively. Extracted attributes were categorized using a structured approach informed by established health technology assessment frameworks, and stakeholder preferences were synthesized narratively. Reporting quality was assessed using the DIRECT checklist.
Results:
A total of 27 studies (28 DCEs) were included. Attributes were grouped into five domains, with AI studies incorporating features like explainability and decision uncertainty. Usability attributes, particularly presentation format (36.81%), were most frequently included but often ranked least important (47.83%). In contrast, performance attributes, particularly effectiveness (62.22%), were most valued and frequently commanded the highest willingness-to-pay. Conditional and mixed logit models predominated, but preference and scale heterogeneity were minimally addressed. Reporting completeness was high (84.47%), though key items related to design effects (37.04%) and randomization (51.85%) were inconsistently reported.
Conclusions:
DCE evidence reveals a mismatch between frequently included attributes and stakeholder priorities, with effectiveness consistently driving preferences. Aligning attributes with stakeholder needs and improving reporting transparency, particularly in experimental design, could enhance the interpretability of DCE evidence and support the development of AI technologies that better reflect real-world needs. Clinical Trial: CRD420261333555
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.