Performance of ChatGPT-3.5 and GPT-4 on the United States Medical Licensing Examination With and Without Distractions
This addresses the reliability of LLMs in real-world medical settings where casual conversations occur, highlighting a key limitation for older models like ChatGPT-3.5.
The study investigated how small talk mixed with medical data affects the accuracy of ChatGPT-3.5 and GPT-4 on USMLE Step 3 questions, finding that ChatGPT-3.5's performance dropped significantly with distractions (e.g., from 61.5% to 44.3% on open questions), while GPT-4's accuracy remained unaffected (83.6% and 66.2% with and without distractions).
As Large Language Models (LLMs) are predictive models building their response based on the words in the prompts, there is a risk that small talk and irrelevant information may alter the response and the suggestion given. Therefore, this study aims to investigate the impact of medical data mixed with small talk on the accuracy of medical advice provided by ChatGPT. USMLE step 3 questions were used as a model for relevant medical data. We use both multiple choice and open ended questions. We gathered small talk sentences from human participants using the Mechanical Turk platform. Both sets of USLME questions were arranged in a pattern where each sentence from the original questions was followed by a small talk sentence. ChatGPT 3.5 and 4 were asked to answer both sets of questions with and without the small talk sentences. A board-certified physician analyzed the answers by ChatGPT and compared them to the formal correct answer. The analysis results demonstrate that the ability of ChatGPT-3.5 to answer correctly was impaired when small talk was added to medical data for multiple-choice questions (72.1\% vs. 68.9\%) and open questions (61.5\% vs. 44.3\%; p=0.01), respectively. In contrast, small talk phrases did not impair ChatGPT-4 ability in both types of questions (83.6\% and 66.2\%, respectively). According to these results, ChatGPT-4 seems more accurate than the earlier 3.5 version, and it appears that small talk does not impair its capability to provide medical recommendations. Our results are an important first step in understanding the potential and limitations of utilizing ChatGPT and other LLMs for physician-patient interactions, which include casual conversations.