Accuracy of Triage Nurses in Predicting Patient Admissions: Retrospective, Large-sample Evidence from a Community Emergency Department
Bibliographic record
Abstract
Introduction: Emergency department (ED) flow could be improved with quicker disposition decisions. One possible way to expedite decisions is for triage nurses to make predictions about whether patients require admission to hospital. The information contained in these predictions could be useful for disposition planning and for physician decision-making. Previous studies made use of prospective designs that introduced Hawthorne effects and have demonstrated mixed evidence on whether triage nurse predictions are accurate. We examined the accuracy of triage nurse predictions for patient admission in an ED in southeastern Ontario. Methods: We examined a retrospective sample of 134,891 visits to an ED in Ontario from March 2019 – July 2024. Triage nurses made predictions about admission to hospital for these visits, from which we estimated measures of specificity, sensitivity, positive predictive value, negative predictive value, accuracy, and F1 scores. Results: Of 134,891 visits, 13.7% resulted in hospital admission. We found the accuracy of the nurses in predicting admission to be 85.8% (95% confidence interval [CI] 85.7 - 86.1), while overall sensitivity was 36.6% (95% CI 35.9 - 37.3) and specificity was 93.7% (95% CI 93.5 - 93.8). The positive predictive value of admission was 47.9% (95% CI 47.1 - 48.7), and the negative predictive value of admission was 90.3% (95% CI 90.1 - 90.5). F1 scores were 0.415. These results were relatively stable over time, although there was notable variation in prediction ability between nurses. We also report that some presenting conditions lead to relatively higher prediction accuracy than others and that as overall case severity increases, sensitivity increases and specificity decreases. Conclusion: These results suggest that although nursing staff predictions are insufficient to streamline disposition decisions completely, they could be useful in expediting certain decisions related to hospital admission and resource requirement, thereby improving flow in EDs.
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 | 0.024 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 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) | 0.003 | 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".