Large language models for generating key-feature questions in medical education
Bibliographic record
Abstract
In this study, we conducted a descriptive study to evaluate the quality of KFQs generated by OpenAI's o3 model. We developed a reusable generic prompt for KFQ generation, designed in alignment with the Medical Council of Canada's KFQ development guidelines. We also created an evaluation metric to systematically assess the quality of the KFQs based on the KFQ development guideline. Twenty unique cardiology-focused KFQs were created using recent European Society of Cardiology guidelines as reference. Each KFQ was independently assessed by two cardiology experts using the quality checklist, with disagreements resolved by a third reviewer. Descriptive statistics were used to summarize checklist compliance and final acceptability ratings. Of the 20 KFQs, 3 (15%) were rated 'Accept as is' and 17 (85%) 'Accept with minor revisions'; none required major revisions or were rejected. The overall compliance rate across checklist criteria was 93.7%, with perfect scores in domains such as key feature definition, scenario plausibility, and alignment between questions and scenarios. Lower performance was observed for inclusion of genuinely harmful 'killer' responses (50%), plausibility of distractors (77.8%), and active language use in phrasing the question (80%). The findings showed that an LLM, guided by a structured prompt, can generate KFQs that closely adhere to established quality standards, with most requiring only minor refinements. While expert review remains essential to ensure clinical accuracy and patient safety, AI-assisted workflows have strong potential to streamline KFQ development and enhance the scalability of CDM assessment in medical education.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.010 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".