ChatGPT and post-test probability
Abstract: Reinforcement learning-based LLMs, such as ChatGPT, are believed to have potential to aid human experts in many domains, including healthcare. There is, however, little work on ChatGPT's ability to perform a key task in healthcare: formal, probabilistic medical diagnostic reasoning. This type of reasoning is used, for example, to update a pre-test probability to a post-test probability. In this work, we probe ChatGPT's ability to perform this task. In particular, we ask ChatGPT to give examples of how to use Bayes rule for medical diagnosis. Our prompts range from queries that use terminology from pure probability (e.g., requests for a posterior of A given B and C) to queries that use terminology from medical diagnosis (e.g., requests for a posterior probability of Covid given a test result and cough). We show how the introduction of medical variable names leads to an increase in the number of errors that ChatGPT makes. Given our results, we also show how one can use prompt engineering to facilitate ChatGPT's partial avoidance of these errors. We discuss our results in light of recent commentaries on sensitivity and specificity. We also discuss how our results might inform new research directions for LLMs.
- Evaluating artificial intelligence responses to public health questions. JAMA Network Open 6, e2317517–e2317517.
- A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity. arXiv preprint arXiv:2302.04023 .
- Role of chat gpt in public health. Annals of Biomedical Engineering 51, 868–869.
- A categorical archive of chatgpt failures. arXiv preprint arXiv:2302.03494 .
- Language models are few-shot learners. Advances in neural information processing systems 33, 1877–1901.
- Can chatgpt pass the life support exams without entering the american heart association course? Resuscitation 185.
- Accuracy and reliability of chatbot responses to physician questions. JAMA Network Open 6, e2336483–e2336483.
- The (in) validity of sensitivity and specificity. Statistics in medicine 19, 1783–1792.
- Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. volume 608. Springer.
- Factors affecting sensitivity and specificity of exercise electrocardiography: multivariable analysis. The American journal of medicine 77, 64–71.
- Accuracy of a Generative Artificial Intelligence Model in a Complex Diagnostic Challenge. JAMA URL: https://doi.org/10.1001/jama.2023.8288, doi:10.1001/jama.2023.8288.
- Chatgpt-reshaping medical education and clinical management. Pakistan Journal of Medical Sciences 39, 605.
- Chatgpt and the clinical informatics board examination: the end of unproctored maintenance of certification? Journal of the American Medical Informatics Association , ocad104.
- Performance of chatgpt on usmle: Potential for ai-assisted medical education using large language models. PLoS digital health 2, e0000198.
- Reasoning foundations of medical diagnosis: symbolic logic, probability, and value theory aid our understanding of how physicians reason. Science 130, 9–21.
- Can large language models reason about medical questions? arXiv preprint arXiv:2207.08143 .
- Using AI-generated suggestions from ChatGPT to optimize clinical decision support. Journal of the American Medical Informatics Association 30, 1237–1245. doi:10.1093/jamia/ocad072.
- Limitations of sensitivity, specificity, likelihood ratio, and bayes’ theorem in assessing diagnostic probabilities: a clinical example. Epidemiology 8, 12–17.
- Sensitivity and specificity should be de-emphasized in diagnostic accuracy studies. Academic radiology 10, 670–672.
- Large language models propagate race-based medicine. npj Digital Medicine 6, 195.
- OpenAI, 2023. Chatgpt. https://chat.openai.com.
- Understanding and using sensitivity, specificity and predictive values. Indian journal of ophthalmology 56, 45.
- Evaluation of abbott binaxnow rapid antigen test for sars-cov-2 infection at two community-based testing sites—pima county, arizona, november 3–17, 2020. Morbidity and Mortality Weekly Report 70, 100.
- Improving language understanding by generative pre-training .
- Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347 .
- An independent evaluation of chatgpt on mathematical word problems (mwp). arXiv preprint arXiv:2302.13814 .
- Sensitivity, specificity, and predictive values: foundations, pliabilities, and pitfalls in research and practice. Frontiers in public health 5, 307.
- Can chatgpt defend the truth? automatic dialectical evaluation elicits llms’ deficiencies in reasoning. arXiv preprint arXiv:2305.13160 .
- Use of a clinical model for safe management of patients with suspected pulmonary embolism. Annals of internal medicine 129, 997–1005.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
