Safety and accuracy of AI in triaging patients in the emergency department
Bibliographic record
Abstract
BACKGROUND: Artificial Intelligence (AI) has been increasingly explored in healthcare, particularly in emergency department (ED) triage. This study aimed to evaluate the effectiveness of the AI chatbot ChatGPT in triaging patients, focusing on its accuracy, safety, efficiency, and impact on patient care. METHODS: A prospective observational study was conducted at the ED of King Saud Medical City (KSMC) in Riyadh, Saudi Arabia, with a sample size of 138 patients. Patients requiring immediate resuscitation were excluded. ED physicians assigned triage scores using the Canadian Triage and Acuity Scale (CTAS), followed by AI-generated scores for the same patients. In cases of discrepancy, the final decision by the senior ED consultant was considered the gold standard. The study assessed inter-rater reliability between AI and human raters and evaluated the accuracy of each compared to the consultant's assessment. RESULTS: The results indicated a high agreement rate (85.61%) between ChatGPT and ED physicians, with substantial inter-rater reliability (κ = 0.780, 95% Confidence Interval [CI] 0.676-0.884, p < 0.001). Agreement between ED physicians and consultants was at 63.9%, with moderate reliability (κ = 0.406, 95% CI 0.006-0.806, p = 0.018). Consultants assigned lower acuity levels than physicians in most cases. ChatGPT's accuracy compared to the consultant was 42.86%, with slight reliability, showing a tendency to overestimate acuity, particularly in critical cases. However, it performed better in mid-range acuity levels. CONCLUSION: The findings suggested that AI could support ED triage by aligning closely with human decision-making. However, its overestimation of severity could lead to over-triaging and increased resource use. Limitations included a small sample size and the use of a general AI model not specifically trained for medical triage. Future research should focus on AI models tailored for ED triage to improve reliability and clinical applicability.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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".