Application of Artificial Intelligence in Shared Decision Making: Scoping Review.

Area of Science:

• Medical Informatics
• Health Services Research
• Artificial Intelligence in Healthcare

Background:

• Artificial intelligence (AI) demonstrates significant potential across various medical domains.
• AI may enhance shared decision-making (SDM) processes between patients and clinicians.
• A comprehensive understanding of AI's role in facilitating SDM is currently lacking.

Purpose of the Study:

• To systematically identify and critically evaluate existing research on AI applications designed to support SDM.
• To map the landscape of AI interventions aimed at improving SDM in clinical practice.

Main Methods:

• A scoping review was conducted, adhering to established methodological frameworks (Levac et al., Arksey and O'Malley, Joanna Briggs Institute).
• A comprehensive literature search was performed across six electronic databases up to May 2021.
• Inclusion criteria encompassed all populations, AI interventions facilitating SDM, relevant outcomes, and all study types published in English.

Main Results:

• The review identified six peer-reviewed publications meeting the inclusion criteria from an initial 1445 records.
• Studies were geographically diverse, with most published after 2017, utilizing machine learning methods.
• AI primarily supported SDM by generating clinical recommendations or predictions, with limited patient or clinician involvement in AI design.

Conclusions:

• The application of AI in SDM is in its early stages, with current tools offering similar support mechanisms.
• A significant gap exists in addressing patient values, preferences, AI explainability, and end-user involvement in AI development.
• Further research and standardization are crucial to optimize AI's integration into all facets of SDM across diverse settings.