Home Health Doctors using AI to take patients’ notes failing to...
Health

Doctors using AI to take patients’ notes failing to capture ‘potentially vital information’

Doctors using AI to take patients’ notes failing to capture ‘potentially vital information’
Key Points

Doctors using AI to take patients’ notes failing to capture ‘potentially vital information’ While many medics are ‘excited’ about so-called AI scribes, the ‘experiences of patients are poorly considered, experts said - Bookmark - CommentsGo to comments Artificial intelligence tools used by doctors to generate clinical notes are failing to record "potentially vital information" conveyed through non-verbal cues like facial expressions and gestures, a study suggests. While many clinicians...

Doctors using AI to take patients’ notes failing to capture ‘potentially vital information’ While many medics are ‘excited’ about so-called AI scribes, the ‘experiences of patients are poorly considered, experts said - Bookmark - CommentsGo to comments Artificial intelligence tools used by doctors to generate clinical notes are failing to record "potentially vital information" conveyed through non-verbal cues like facial expressions and gestures, a study suggests. While many clinicians remain "excited" about the ability of AI scribes to drastically cut down paperwork, experts warned that the "experiences of patients are poorly considered". So-called ambient voice technology combines speech recognition and generative AI to automatically transcribe spoken words into structured medical records, with the aim of saving time and improving doctor-patient interactions. However, a review of 27 published papers by researchers at the University of Edinburgh highlights that these systems listen only to audio. Consequently, they miss physical gestures, facial expressions, and vocal tone used by doctors and patients alike. Researchers described the tools as having a “written language bias”. “They convert spoken consultations into text and then process the interaction as a textual record,” they added. “Ambient scribes cannot capture the potentially vital information a clinician can glean from gestures, facial expressions, movement and emotional affect.” Experts also suggest that patients may “adapt their behaviour” in front of AI, and a “lack of trust may lead to less disclosure”. Dr Lucas Seuren, Gail fellow and research fellow at the University of Edinburgh’s Centre for Biomedicine, Self and Society, said: “Many clinicians are excited about ambient AI scribes, because they promise to cut down on paperwork. “But the experiences of patients are poorly considered, and there are real risks that the patients’ stories are lost. “This can further disadvantage people who already face marginalisation in health and social care services.” The findings, published in BMJ Digital Health and AI, highlight that the technology may “require careful adaptation to clinical and health system contexts”, and “their implementation may reshape how health and care work is organised and performed”. Researchers said the “enthusiasm surrounding ambient scribes has driven widespread adoption”. It is estimated that up to four in 10 GPs in the UK are using AI scribes. However, most studies have focused on benefits to productivity and technological performance, while “sociotechnical dimensions of adoption have been neglected”, researchers said. “Ambient scribes are complex interventions with the potential to reshape clinical work, documentation practices and the organisation of care. “Before pursuing large-scale spread, it is important to develop a robust evidence base of the impacts of early deployment and develop a comprehensive understanding of how care with ambient scribes could and should be designed to support patients, clinicians and care systems,” they added. Join our commenting forum Join thought-provoking conversations, follow other Independent readers and see their replies Comments
AI (ORG) the University of Edinburgh (ORG) Dr Lucas Seuren (PERSON) Gail (PERSON) the University of Edinburgh’s (ORG) Self and Society (ORG) BMJ Digital Health (ORG) UK (LOCATION)
Originally published by The Independent UK Read original →