TY - JOUR
T1 - Assessing the Ability of a Large Language Model to Score Free-Text Medical Student Clinical Notes
T2 - Quantitative Study
AU - Burke, Harry B.
AU - Hoang, Albert
AU - Lopreiato, Joseph O.
AU - King, Heidi
AU - Hemmer, Paul
AU - Montgomery, Michael
AU - Gagarin, Viktoria
N1 - © Harry B Burke, Albert Hoang, Joseph O Lopreiato, Heidi King, Paul Hemmer, Michael Montgomery, Viktoria Gagarin. Originally published in JMIR Medical Education (https://mededu.jmir.org).
PY - 2024/7/25
Y1 - 2024/7/25
N2 - Background: Teaching medical students the skills required to acquire, interpret, apply, and communicate clinical information is an integral part of medical education. A crucial aspect of this process involves providing students with feedback regarding the quality of their free-text clinical notes. Objective: The goal of this study was to assess the ability of ChatGPT 3.5, a large language model, to score medical students’ free-text history and physical notes. Methods: This is a single-institution, retrospective study. Standardized patients learned a prespecified clinical case and, acting as the patient, interacted with medical students. Each student wrote a free-text history and physical note of their interaction. The students’ notes were scored independently by the standardized patients and ChatGPT using a prespecified scoring rubric that consisted of 85 case elements. The measure of accuracy was percent correct. Results: The study population consisted of 168 first-year medical students. There was a total of 14,280 scores. The ChatGPT incorrect scoring rate was 1.0%, and the standardized patient incorrect scoring rate was 7.2%. The ChatGPT error rate was 86%, lower than the standardized patient error rate. The ChatGPT mean incorrect scoring rate of 12 (SD 11) was significantly lower than the standardized patient mean incorrect scoring rate of 85 (SD 74; P=.002). Conclusions: ChatGPT demonstrated a significantly lower error rate compared to standardized patients. This is the first study to assess the ability of a generative pretrained transformer (GPT) program to score medical students’ standardized patient-based free-text clinical notes. It is expected that, in the near future, large language models will provide real-time feedback to practicing physicians regarding their free-text notes. GPT artificial intelligence programs represent an important advance in medical education and medical practice.
AB - Background: Teaching medical students the skills required to acquire, interpret, apply, and communicate clinical information is an integral part of medical education. A crucial aspect of this process involves providing students with feedback regarding the quality of their free-text clinical notes. Objective: The goal of this study was to assess the ability of ChatGPT 3.5, a large language model, to score medical students’ free-text history and physical notes. Methods: This is a single-institution, retrospective study. Standardized patients learned a prespecified clinical case and, acting as the patient, interacted with medical students. Each student wrote a free-text history and physical note of their interaction. The students’ notes were scored independently by the standardized patients and ChatGPT using a prespecified scoring rubric that consisted of 85 case elements. The measure of accuracy was percent correct. Results: The study population consisted of 168 first-year medical students. There was a total of 14,280 scores. The ChatGPT incorrect scoring rate was 1.0%, and the standardized patient incorrect scoring rate was 7.2%. The ChatGPT error rate was 86%, lower than the standardized patient error rate. The ChatGPT mean incorrect scoring rate of 12 (SD 11) was significantly lower than the standardized patient mean incorrect scoring rate of 85 (SD 74; P=.002). Conclusions: ChatGPT demonstrated a significantly lower error rate compared to standardized patients. This is the first study to assess the ability of a generative pretrained transformer (GPT) program to score medical students’ standardized patient-based free-text clinical notes. It is expected that, in the near future, large language models will provide real-time feedback to practicing physicians regarding their free-text notes. GPT artificial intelligence programs represent an important advance in medical education and medical practice.
KW - Clinical Competence/standards
KW - Education, Medical, Undergraduate/methods
KW - Educational Measurement/methods
KW - Humans
KW - Language
KW - Male
KW - Medical History Taking/methods
KW - Retrospective Studies
KW - Students, Medical
UR - http://www.scopus.com/inward/record.url?scp=85203660250&partnerID=8YFLogxK
U2 - 10.2196/56342
DO - 10.2196/56342
M3 - Article
C2 - 39118469
AN - SCOPUS:85203660250
SN - 2369-3762
VL - 10
SP - e56342
JO - JMIR Medical Education
JF - JMIR Medical Education
M1 - e56342
ER -