Conectar
To explore whether large language models (LLMs) Generated Pre-trained Transformer (GPT)-3 and ChatGPT can write clinical letters and predict management plans for common orthopaedic scenarios.
Fifteen scenarios were generated and ChatGPT and GPT-3 prompted to write clinical letters and separately generate management plans for identical scenarios with plans removed.
Letters were assessed for readability using the Readable Tool. Accuracy of letters and management plans were assessed by three independent orthopaedic surgery clinicians.
Both models generated complete letters for all scenarios after single prompting. Readability was compared using Flesch-Kincade Grade Level (ChatGPT: 8.77 (SD 0.918); GPT-3: 8.47 (SD 0.982)), Flesch Readability Ease (ChatGPT: 58.2 (SD 4.00); GPT-3: 59.3 (SD 6.98)), Simple Measure of Gobbledygook (SMOG) Index (ChatGPT: 11.6 (SD 0.755); GPT-3: 11.4 (SD 1.01)), and reach (ChatGPT: 81.2%; GPT-3: 80.3%). ChatGPT produced more accurate letters (8.7/10 (SD 0.60) vs 7.3/10 (SD 1.41), p=0.024) and management plans (7.9/10 (SD 0.63) vs 6.8/10 (SD 1.06), p
This study shows that LLMs are effective for generation of clinical letters. With little prompting, they are readable and mostly accurate. However, they are not consistent, and include inappropriate omissions or insertions. Furthermore, management plans produced by LLMs are generic but often accurate. In the future, a healthcare specific language model trained on accurate and secure data could provide an excellent tool for increasing the efficiency of clinicians through summarisation of large volumes of data into a single clinical letter.
Between 2013 and 2018, there has been a 71% increase in the number of patients who have required wound care in the NHS and such large numbers has placed a significant burden on healthcare systems. However, there is currently no evidence as to whether medical students are equipped with the necessary skills to deal with an increasing number of wound care related issues that patients present with. A total of 323 medical students across 18 UK medical schools completed an anonymous questionnaire evaluating the wound education received at their medical school, encompassing the volume, content, format and efficacy of teaching. 68.4% (221/323) of respondents had received some form of wound education during their undergraduate studies. On average students received 2.25 h of structured, preclinical teaching and only 1 h of clinical based teaching in total. All students that received wound education reported undertaking teaching on the physiology of, and factors affecting wound healing, with only 32.2% (n = 104) of students receiving clinically based wound education There was very weak correlation and no significant association in student's ability to assess wounds (R 2 = 0.190, p = 0.013), manage wounds (R 2 = 0.060, p = 0.37), and prescribe wound care products (R 2 = 0.093, p = 0.18) with their stage of training. Students strongly agreed that wound education is an important part of the undergraduate curriculum and post graduate practice, and do not feel their learning needs have been met. This is the first study to assess the provision of wound education in the United Kingdom, demonstrating a clear deficit in the provision of wound education compared to expectation of junior doctors. Wound education is largely overlooked in the medical curriculum, lacks a clinical focus and does not prepare junior doctors with the necessary clinical abilities to deal with wound related pathology. Expert opinion to direct changes to future curriculum and further evaluation of teaching methodology is required to address this deficit and ensure students have the necessary clinical skills to excel as newly graduated doctors.
FreshRSS 