TOPLINE:

All five early warning scores (EWSs) accurately predicted clinical deterioration within 24 hours in patients aged 80 years or older in the emergency department (ED). The Rapid Emergency Medicine Score (REMS) showed the most consistent performance, with fewer false-positive predictions, especially in patients aged 94 years or older.

METHODOLOGY:

• Researchers conducted a single-center, retrospective prognostic study using electronic medical records from 50,645 patients without trauma aged 80 years or older (median age, 85 years; 54.6% women) at an urban teaching hospital in Rome, Italy, between 2015 and 2024.
• At ED admission, five EWSs were used to evaluate patients: National Early Warning Score (NEWS), NEWS2, Modified Early Warning Score (MEWS), REMS, and International Early Warning Score (IEWS).
• The primary outcome was clinical deterioration, defined as death or ICU admission within 24 hours of ED arrival. Secondary outcomes included the comparative contribution of individual score components to predicting deterioration and how these contributions varied with increasing age.
• Discrimination was assessed using the area under the receiver operating characteristic curve (AUROC).

TAKEAWAY:

• Clinical deterioration occurred in 2.4% of patients. All five EWSs showed fair discrimination, with median AUROC ranging from 0.747 for MEWS to 0.782 for NEWS; however, all other scores had significantly lower AUROC than NEWS.
• Its performance improved with age, becoming the best predictor in patients older than 94 years.
• Sensitivity was generally low to moderate across scores and declined further at higher predictive thresholds. Discriminatory performance declined with increasing age beyond 90 years for NEWS, NEWS2, MEWS, and IEWS, whereas REMS showed improved predictive accuracy with age, surpassing the other scores in predictive accuracy for patients aged 94 years or older (P < .001 for all).
• Among the variables, oxygen supplementation, systolic blood pressure, and Glasgow Coma Scale score had the largest increase in predictive contribution to clinical deterioration in patients aged 87 years or older vs those aged 80-86 years.

IN PRACTICE:

"In this prognostic study, all evaluated EWSs demonstrated satisfactory performance in predicting short-term clinical deterioration among patients aged 80 years or older seen in the ED. The REMS showed superior accuracy in the subgroup aged more than 94 years and was particularly effective in identifying a smaller cohort of high-risk patients, albeit with reduced sensitivity," the authors wrote.

"Therefore, the choice of a specific scoring system should be guided by local operational requirements and clinical context. However, given their inherent limitations and potential biases, EWSs should be regarded as supportive tools rather than substitutes for clinical judgment in the population older than 80 years," they added.

SOURCE:

The study was led by Marcello Covino, MD, PhD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Universita Cattolica del Sacro Cuore, in Rome, Italy. It was published online on March 19, 2026, in JAMA Network Open.

LIMITATIONS:

The study was limited by its single-center retrospective design. The analysis was confined to predicting deterioration within the first 24 hours of ED arrival. Exclusion of patients who were critically ill early in their presentation may have affected the results, as EWSs were unlikely to add prognostic value when severe conditions were already evident or when early ICU admission was mandated. Additionally, each encounter was treated as a single observation without adjusting for early readmissions, which may have introduced dependency and limited the generalizability of the findings.

DISCLOSURES:

Funding information for the study was not provided. The authors reported having no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.