Using an AI scribe in private practice enhances patient-clinician interactions and reduces administrative burden

Abstract

Background: AI may automate and enhance various processes in healthcare, including clinical documentation. No study has investigated the impact of using AI for clinical documentation in allied health private practice.   Methods: In this mixed-methods study, 119 allied health clinicians recorded their interactions with patients using a natural language processing tool. Participants completed questionnaires at baseline, 6-weeks and 3-months. Clinicians were asked to invite their patients to share their perspectives on AI being used during their appointment. Patients who consented (n=157) completed one questionnaire immediately after their appointment. A subset of clinician (n=15) and patient (n=14) participants took part in a semi-structured interview to explore their perspectives in greater depth.   Results: At 3 months, there was a significant reduction in time spent doing notes (p<0.001), doctors' letters (p<0.001), notes out of hours (p<0.00) and burden of clinical documentation (p=0.02). From the interviews, patients and clinicians reported that AI had positively impacted their interaction (e.g., 'He could really engage with me'; 'I felt heard'; 'I was able to pay 100% of attention to my patient'). Patients expressed high trust in AI, largely because of trust in their clinician (e.g., 'I know my clinician really well and I trust him. There's not much that I wouldn't share with him that I wouldn't feel comfortable with being recorded').   Conclusion: Using AI for clinical documentation was well accepted and had a positive impact on patient-clinician interactions. Using AI holds potential to streamline clinical documentation processes and enhance operational efficiencies in allied-health private practice.