Evaluating Patient and Professional Satisfaction and Documentation Time Reduction Through AI-Driven Automatic...

Area of Science:

• Health Informatics
• Artificial Intelligence in Medicine
• Primary Care Technology

Background:

• Clinical documentation significantly contributes to primary care professional workload and stress.
• Artificial intelligence (AI) offers potential solutions to reduce documentation time without compromising care quality.

Purpose of the Study:

• To evaluate the impact of an AI solution for automatic consultation transcription in primary care settings.
• To assess the feasibility, quality, and user satisfaction of AI-assisted clinical note generation.

Main Methods:

• A multicenter proof-of-concept study with alternating assignment to AI intervention or usual care.
• Evaluation metrics included documentation time, transcription quality (Levenshtein distance), clinician-reported errors, and patient/professional satisfaction surveys.
• Qualitative analysis of errors and Likert scale assessments for satisfaction were employed.

Main Results:

• AI transcription drafts required minimal review (average <24 words/minute), with 26% needing no edits.
• Context/meaning errors were most common; AI hallucinations were infrequent.
• Patient satisfaction remained high and comparable between groups; professionals reported high audio and transcription quality.

Conclusions:

• AI tools are feasible and well-accepted in primary care, with a low transcription review burden.
• The AI solution did not negatively impact patient satisfaction.
• Further randomized studies are needed to confirm comparative time savings and benefits for care quality and efficiency.