Improving Clinical Documentation with Artificial Intelligence: A Systematic Review

Area of Science:

• Medical Informatics
• Clinical Documentation Improvement
• Artificial Intelligence in Healthcare

Background:

• Clinicians spend substantial time on clinical documentation, leading to significant opportunity costs.
• Artificial Intelligence (AI) presents a potential solution for enhancing documentation quality and efficiency.

Purpose of the Study:

• To systematically review peer-reviewed AI tools designed to improve clinical documentation.
• To understand how AI applications can mitigate the opportunity costs associated with documentation.

Main Methods:

• A systematic review of PubMed, Embase, Scopus, and Web of Science databases.
• Inclusion of original, English-language studies published up to July 2024 focusing on AI tool development, application, and validation for clinical documentation.
• Extraction and analysis of 129 studies from 673 initial candidates.

Main Results:

• AI tools enhance documentation through data structuring, note annotation, quality evaluation, trend identification, and error detection.
• Real-time AI assistance during patient visits shows moderate accuracy, limiting widespread adoption.
• No end-to-end AI documentation assistant with high accuracy has been reported in peer-reviewed literature.

Conclusions:

• Current AI techniques, particularly data structuring, offer targeted improvements to clinical documentation workflows.
• Further research and development are needed to achieve highly accurate, comprehensive AI solutions for clinical documentation.
• AI holds promise for reducing clinician burden and opportunity costs in healthcare documentation.