A recent report has found that AI-powered scribes have the potential to assist with administrative tasks in healthcare, though it emphasized the need for further research to determine how these tools function in various clinical environments and how their design can best support both patients and medical professionals.

This study follows reports from doctors in the West Midlands who have experienced a notable reduction in their workload thanks to the transcription tool Heidi. At a clinic in Dudley, for example, the implementation of this technology helped shorten a backlog of patient letters from six months down to just 14 days.

One significant insight from researchers at the University of Edinburgh was that, while AI tools can effectively capture patient speech, they sometimes fail to recognize important dimensions of clinical consultations, such as non-verbal cues including facial expressions, gestures, and the emotional tone of interactions. The report also observed that AI-generated summaries often emphasize clinical data, potentially overlooking the richer narratives and personal experiences that patients share about their illnesses.

Additionally, concerns were raised about patient willingness to disclose sensitive information—such as issues related to substance abuse, domestic violence, or mental health—when consultations are recorded and analyzed by AI systems. The report also noted that traditional manual notetaking plays a crucial role in helping clinicians reflect on and reason through patient information. While AI scribes can alleviate some cognitive burden by taking over mental workload, allowing clinicians to focus on deeper conversations or complex tasks, this benefit may come at a cost. For instance, over-reliance on these tools might impair memory recall and professional skill development, leading to situations where clinicians fail to recognize their own notes or recall patients during follow-up visits

Read the full article from The BBC here: Read More