There are also concerns related to a digital scribe’s clinical utility, such as the effect on a physician’s workflow. These relate to technical aspects such as the accuracy of current ASR systems for transcription of spontaneous speech 13 and a digital scribe’s ability to extract all the essential information from a non-linear, fragmented conversation 13, 17. Although much needed, there are several concerns about implementing a digital scribe in healthcare. The extracted information could, for instance, be used to create clinical notes, add billing codes, or use the extracted information for diagnosis support.Ĭompanies like Google, Nuance, Amazon, and many startups are creating a digital scribe 14, 15, 16. The proposed structure for a digital scribe includes a microphone that records a conversation, an ASR system that transcribes this conversation, and a set of NLP models to extract or summarize relevant information and present it to the physician. This digital scribe uses techniques such as automatic speech recognition (ASR) and natural language processing (NLP) to automate (parts of) clinical documentation. Two recent perspectives 12, 13 describe the need for a so-called digital scribe. As a result, direct medical costs increase, while the administrative burden remains substantial. Although a medical scribe might seem like the perfect solution, it shifts the burden to other personnel. Studies show positive results for the use of medical scribes, with clinicians spending more face-to-face time with patients and less after-hour time on the EHR 10, 11. Recently, clinicians have hired medical scribes to reduce the administrative burden, i.e., persons who manage administrative tasks, such as summarizing a consultation. Other studies have assessed the relationship between EHR-use and burnout and found that more time spent on the EHR, especially after-hours, was linked to a higher risk of burnout 8, 9. These administrative tasks decrease clinicians’ career satisfaction 6 and negatively affect the clinician–patient relationship 7. Since the introduction of the electronic health record (EHR), the time spent on administrative tasks has increased to approximately half of a clinician’s workday 3, 4, 5. An important reason is the increasing administrative burden 2. The committee’s extensive report, called Taking Action Against Clinician Burnout, describes reasons for clinician burnout. To investigate this problem, the National Academy of Medicine formed a committee focused on improving patient care by supporting clinician well-being. In a 2017 survey among 5000 US clinicians, 44% reported at least one symptom of burnout 1. In the past few years, clinician burnout has become an acknowledged problem in healthcare. Future research should focus on more extensive reporting, iteratively studying technical validity and clinical validity and usability, and investigating the clinical utility of digital scribes. However, the studies on digital scribes only focus on technical validity, while companies offering digital scribes do not publish information on any of the research phases. The most promising models use context-sensitive word embeddings in combination with attention-based neural networks. Two studies examined the system’s clinical validity and usability, while the other 18 studies only assessed their model’s technical validity on the specific NLP task. The other 17 articles presented models for entity extraction, classification, or summarization of clinical conversations. Of 20 included articles, three described ASR models for clinical conversations. We included articles that described the use of models on clinical conversational data, either automatically or manually transcribed, to automate clinical documentation. We performed a literature search of four scientific databases (Medline, Web of Science, ACL, and Arxiv) and requested several companies that offer digital scribes to provide performance data. We reviewed the current status of the digital scribe in development towards clinical practice and present a scope for future research. Automatic speech recognition (ASR) and natural language processing (NLP) techniques may address this issue by creating the possibility of automating clinical documentation with a “digital scribe”. The number of clinician burnouts is increasing and has been linked to a high administrative burden.
0 kommentar(er)
