Search

Objective: Large language models (LLMs) have demonstrated mixed results in their ability to pass various specialist medical examination and their performance within the field of emergency medicine remains unknown. Methods: We explored the performance of three prevalent LLMs (OpenAI's GPT series, Google's Bard, and Microsoft's Bing Chat) on a practice ACEM primary examination. Results: All LLMs achieved a passing score, with scores with GPT 4.0 outperforming the average candidate. Conclusion: Large language models, by passing the ACEM primary examination, show potential as tools for medical education and practice. However, limitations exist and are discussed. [ABSTRACT FROM AUTHOR]

Using a suite of artificial intelligence technologies, the current study sought to determine the prevalence of agitated behaviors in people with dementia in residential aged care facilities (RACFs) in Australia. Computerized natural language processing allowed extraction of agitation instances from the free-text nursing progress notes, a component of electronic health records in RACFs. In total, 59 observable agitated behaviors were found. No difference was found in dementia prevalence between female and male clients (44.1%), across metropolitan and regional facilities (42.1% [SD = 17.9%]), or for agitation prevalence in dementia (76.5% [SD = 18.4%]). The top 10 behaviors were resisting, wandering, speaking in excessively loud voice, pacing, restlessness, pushing, shouting, complaining, frustration, and using profane language. Four to 17 agitated behaviors coexisted in 53% of people with dementia agitation, indicating high caregiver burden in these RACFs. Improving workforce training and redesigning care models are urgent for sustainability of dementia care in RACFs. [Journal of Gerontological Nursing, 48(4), 57–64.] [ABSTRACT FROM AUTHOR]

To inform the development of automated summarization of clinical conversations, this study sought to estimate the proportion of doctor-patient communication in general practice (GP) consultations used for generating a consultation summary. Two researchers with a medical degree read the transcripts of 44 GP consultations and highlighted the phrases to be used for generating a summary of the consultation. For all consultations, less than 20% of all words in the transcripts were needed for inclusion in the summary. On average, 9.1% of all words in the transcripts, 26.6% of all medical terms, and 27.3% of all speaker turns were highlighted. The results indicate that communication content used for generating a consultation summary makes up a small portion of GP consultations, and automated summarization solutions—such as digital scribes—must focus on identifying the 20% relevant information for automatically generating consultation summaries. [ABSTRACT FROM AUTHOR]

Background The official prescribing information document distributed with a prescription drug is a key source of safety information, but it may include excessive or insufficient details. Objectives To compare prescribing information approved by the US Food and Drug Administration with the UK, Canada and Australia to identify content differences in safety warnings. Methods For 20 top-selling prescription drugs, we used an automated natural language processing tool to calculate the number and severity of reported adverse drug reactions (ADRs). We fit hierarchical Poisson models and included fixed effects for other prescribing information characteristics. Separately, we analysed the appearance and content of 'black box' warnings. Results There was substantial variation in safety content of approved prescribing information. Canada had the highest median ADRs per drug (138 (IQR 86-234)) and the UK had the lowest (84 (IQR 51-111)). The number of ADRs reported was on average 50% higher in Canada compared with the USA (ratio of ADRs/ document: 1.5, 95% CI 1.4 to 1.6, p<0.001). By contrast, there were on average 15% fewer ADRs listed in the UK compared with the USA (ratio of ADRs/document 0.85 (95% CI 0.78 to 0.93, p<0.001), and 21% fewer ADRs listed in Australia compared with the USS (ratio of ADRs/ document 0.79, 95% CI 0.74 to 0.85, p<0.001). There were no variations in ADR severity. The presence and qualitative content of boxed warnings also showed substantial diversity. Conclusions International variations exist in the presentation of safety data in drug prescribing information, which may have important implications for patient safety. Better international coordination is necessary to enhance use of this information for patient decision-making. [ABSTRACT FROM AUTHOR]