Patient Concerns Regarding Artificial Intelligence Applications in Health Care: Systematic Review and Meta-Synthesis...

Area of Science:

• Healthcare Technology
• Medical Informatics
• Artificial Intelligence Ethics

Background:

• Growing use of artificial intelligence (AI) in healthcare lacks multilevel research on patient concerns.
• Existing reviews often descriptively summarize patient attitudes without theoretical frameworks.

Purpose of the Study:

• To systematically review and analyze patient concerns regarding healthcare AI applications.
• To apply social ecological theory to understand multilevel interactive mechanisms of patient concerns.

Main Methods:

• Systematic review and meta-synthesis following PRISMA-S guidelines.
• Searched major databases (PubMed, Embase, Web of Science, CINAHL, Scopus) for qualitative studies on clinical AI.
• Assessed quality using JBI-QARI and confidence using GRADE-CERQual.

Main Results:

• Included 25 qualitative studies with 528 participants.
• Identified six themes: privacy/data security, technology reliability, physician-patient relationships, trust/accountability, ethical/equity issues, and workforce replacement.
• Concerns operate at micro (individual), meso (organizational), and macro (societal) levels.

Conclusions:

• First meta-synthesis applying social ecological theory to patient concerns about medical AI.
• Findings reveal interconnected "ecological imbalance" mechanisms, suggesting concerns are fact-based.
• Systemic interventions are needed, including explainable AI, improved communication, institutional accountability, and global ethical norms.