Dr. LLM Will See You Now: The Ability of ChatGPT to Provide Geographically Tailored Colorectal Cancer Screening and Surveillance Recommendations
Bibliographic record
Abstract
Background/Objectives: This study evaluates the performance of a large language model (lLm) in providing geographically tailored colorectal cancer screening and surveillance recommendations to gastrointestinal surgeons. Methods: Fifty-four patient cases, varying by age and family history, were developed based on colorectal cancer guidelines. Standardized prompts with predefined query terms were used to query ChatGPT-4.5 on 18 April 2025, from four locations: Canada, Italy, Romania, and the United Kingdom. Responses were classified as “Correct,” “Partially Correct,” or “Incorrect” based on clinical guidelines and expert recommendations for each country. Outcomes were analyzed using descriptive statistics. Results: ChatGPT provided recommendations on screening eligibility, test interpretation, the management of positive results, and surveillance intervals. Correct recommendations were given for 50.0% (27/54) of cases in Canada, 63.0% (34/54) of cases in Italy, 40.7% (22/54) of cases in Romania, and 55.6% (30/54) of cases in the United Kingdom. Queries in Italian yielded correct guidance for 64.8% (35/54) of cases, while Romanian queries were accurate for 40.7% (22/54) of cases. Notably, Romania and Italy lacked detailed guidelines for polyp management and post-test surveillance. A key finding was the inconsistency between ChatGPT-generated titles and corresponding recommendations, which may impact its reliability in clinical decision-making. Conclusions: ChatGPT-4.5’s performance varies by country and language, highlighting inconsistencies in geographically tailored recommendations. This study highlights limitations associated with the training data cutoff and the potential biases introduced by model-generated responses. Healthcare professionals should recognize these limitations and the possible gaps in guideline availability, particularly for high-risk screening, polyp management, and surveillance in certain European countries.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.015 | 0.093 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.011 | 0.005 |
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 source (direct Gemma or distilled Codex), 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".