Utility of Triage Nurses' Quick-Look Assessments of Adults in the Emergency Department for Predicting Hospital Admission.

Abstract

Introduction: Nurses use quick-look assessments, vital signs, and interviews to predict hospital admission, but evidence for the predictive accuracy of quick-look assessments remains limited. This study aimed to evaluate the accuracy of quick-look assessments in predicting hospital admission for adult patients, compare their performance with the National Early Warning Score, and assess the added value of combining quick-look assessments and the National Early Warning Score.

Methods: This prospective, single-center, observational study was conducted in the emergency department of an acute care hospital. During triage, nurses predicted admission likelihood based only on quick-look assessments, without any other data such as vital signs or history of illness, with higher quick-look assessment scores indicating greater admission likelihood. The National Early Warning Score was calculated from vital sign data. Three predictive models (quick-look assessment alone, National Early Warning Score alone, and quick-look assessment + National Early Warning Score) were developed using logistic regression modeling. Model performance was assessed using the area under the curve and calibration plots.

Results: Of 1588 patients, 144 (9.1%) were admitted. Higher quick-look assessment scores were associated with admission, with odds ratios of 83.8 (95% CI, 26.6-263.8) and 145.2 (95% CI, 21.4-986.4) for quick-look assessment scores of 4 and 5, respectively. The quick-look assessment model had an area under the curve of 0.85 (95% CI, 0.81-0.88), outperforming the National Early Warning Score (area under the curve, 0.67; 95% CI, 0.62-0.73). Adding the National Early Warning Score to quick-look assessments led to minimal improvement (area under the curve, 0.87; 95% CI, 0.83-0.90). Calibration showed that quick-look assessments underestimated moderate-to-high-risk predictions.

Discussion: Quick-look assessments allow nurses to make accurate predictions of hospital admissions during ED triage, outperform the National Early Warning Score, and support better patient prioritization.