A Real-Time Prospective Evaluation of the Prognostic Accuracy of SIRS, MEWS, NEWS2 and qSOFA in Predicting ICU Admission and Mortality in an Emergency Department: Implications for Nursing Practice.

backgroundEarly identification of emergency department (ED) patients with suspected infection who are at risk for adverse outcomes is a core nursing responsibility during triage and early observation. Scoring systems vary in sensitivity and specificity, with limited prospective, real-time comparisons in Greek EDs.

aimTo compare the prognostic accuracy of SIRS, MEWS, NEWS2 and qSOFA for predicting prolonged ICU stay (≥ 3 days) and 28-day in-hospital mortality and to assess performance across clinically relevant subgroups that shape nursing decision-making. STUDY

designProspective observational study in the ED of a public general hospital in Athens, Greece. Consecutive adults (≥ 18 years) with suspected infection were enrolled. Scores were calculated in real time at triage by trained ED nurses using only presentation data. Primary outcomes were ICU stay ≥ 3 days and 28-day in-hospital mortality. Discrimination was assessed by AUROC (95% CI) with DeLong pairwise comparisons; prespecified subgroups included age (≥ 65 years), comorbidities, and COPD.

resultsFive hundred and thirty patients who met the inclusion criteria were analysed (mean age 63.7 ± 18.5 years; 53.96% female). For predicting ICU ≥ 3 days, NEWS2 demonstrated the highest sensitivity, NPV and AUROC (0.72) with no significant differences compared to MEWS or qSOFA, but better than SIRS (p = 0.001). For 28-day in-hospital mortality, qSOFA and NEWS2 achieved the highest AUROC (0.79 and 0.77, respectively), both significantly outperforming SIRS and MEWS. NEWS2 maintained the highest sensitivity and NPV, whereas qSOFA showed the highest specificity and PPV. In subgroup analyses, comorbid patients exhibited elevated risk even with negative NEWS2/MEWS; qSOFA-positive younger patients (< 65) had higher ICU admission; COPD was associated with higher mortality in qSOFA-negative and MEWS-negative strata.

conclusionsNEWS2 and qSOFA outperformed SIRS and MEWS. NEWS2 is well suited for ruling out risk (very high sensitivity/NPV), whereas qSOFA provides more specific identification of high-risk patients. RELEVANCE TO CLINICAL PRACTICE: NEWS2 can help nurses quickly rule out high-risk infection cases, whereas qSOFA highlights patients most likely to need urgent escalation or ICU care. Using both scores together supports faster triage decisions and better allocation of monitoring and critical care resources.