Evaluating Patient and Professional Satisfaction and Documentation Time Reduction through an AI-Driven Automatic Clinical Note Generation in Primary Care: A Proof of Concept (Preprint)

BACKGROUND The workload that stems from writing clinical histories is one of the main sources of stress and overload for primary care professionals, accounting for up to 43% of the working day. The introduction of technology, specifically artificial intelligence, in the field of health could significantly reduce the time spent writing clinical reports without compromising the quality of care. OBJECTIVE The objective of this study was to evaluate the impact of implementing an AI solution for the automatic transcription of consultations in several Primary Care Centres in Catalonia. METHODS A proof of concept of a multi-centre study was carried out with alternating assignment of consultations to the intervention group (use of an artificial intelligence assistant that automatically generates consultation notes) or control group (usual clinical practice). The impact was evaluated through the recorded documentation time, the initial quality of the transcription measured with the Levenshtein distance and corrected words per minute, and the perceived satisfaction of patients and professionals through questionnaires evaluated through a Likert scale. RESULTS For the intervention group, the average processing time was 6.63%, while the review time by the professional amounted to 15.2%. In the control group, the writing time that professionals devoted to it was 30.37%. Therefore, the time savings were in an estimated range of 9% to 15% of the total consultation time. Levenshtein analysis showed that in most cases the review was <24 words per minute, indicating a high-quality initial transcription. The satisfaction surveys were answered by 289 patients and 213 professionals. Patient satisfaction was high (≥ 4/5) with no statistically significant differences between the control and intervention groups. The professionals rated the audio quality at 8.88/10 (SD= 1.31) (medicine) and 7.99/10 (SD= 1.76) (nursing), and the transcription at 8.01/10 (SD= 1.72) and 7.73/10 (SD= 1.76), respectively. CONCLUSIONS The implementation of an AI tool demonstrated significant time savings, good professional acceptance and did not show a negative impact on patient satisfaction. The results suggest the possible use of these tools as a form of support for healthcare professionals. Randomised controlled trials are necessary to confirm the benefits in terms of healthcare quality and system efficiency. CLINICALTRIAL ClinicalTrials.gov NCT06618092; https://clinicaltrials.gov/study/NCT06618092 INTERNATIONAL REGISTERED REPORT RR2-10.2196/66232