Community-acquired pneumonia (CAP) is a common acute respiratory infection encountered in emergency practice and continues to contribute substantially to hospital admission, intensive care use, and short-term mortality among adults. Its clinical course ranges from a relatively uncomplicated infection suitable for ward-based management to rapidly progressive respiratory failure, septic shock, multiorgan dysfunction, and death. Current CAP guidance therefore emphasizes early assessment of disease severity alongside clinical judgment when deciding the site and intensity of care.1 The emergency department is a particularly important setting for this assessment because physiological deterioration can become evident before definitive microbiological information is available. Several prognostic tools have been developed to reduce variability in bedside assessment. The Pneumonia Severity Index (PSI) uses demographic characteristics, comorbidities, examination findings, and laboratory variables to identify patients at low risk of death.2 Although extensively validated, its number of variables can limit rapid application in a busy emergency environment. CURB-65 offers a simpler pneumonia-specific alternative based on confusion, blood urea, respiratory rate, blood pressure, and age 65 years or older.3 Its simplicity has promoted widespread clinical use, but the score was primarily derived for mortality prediction and does not directly incorporate several dynamic physiological abnormalities that can signal early deterioration. The National Early Warning Score 2 (NEWS2) takes a different approach. It is based on routinely measured physiological variables, including respiratory rate, oxygen saturation, supplemental oxygen requirement, systolic blood pressure, pulse, temperature, and level of consciousness or new confusion.4 Because these variables change with acute illness, NEWS2 is designed to identify physiological instability across diagnostic categories. In CAP, this property is attractive because the immediate clinical concern in the emergency department is often not only mortality risk but also the likelihood of deterioration requiring intensive care or enhanced monitoring. A large prospective CAP cohort showed that NEWS contributed prognostic information for intensive care admission beyond established pneumonia-specific scores.5 Risk scores should, however, be interpreted according to the outcome for which they are used. Reviews of CAP severity tools have shown that mortality prediction and prediction of intensive care requirements are not equivalent, and a score that performs well for one endpoint can be less informative for another.6 Consequently, direct comparison of a general physiological warning score with a disease-specific score is clinically relevant, particularly in emergency settings where rapid escalation decisions are required. The present study was undertaken to evaluate adults with CAP presenting to the emergency department of a tertiary multispeciality hospital in Bengaluru. The primary objective was to compare the performance of NEWS2 and CURB-65 for identifying patients who subsequently experienced a high-risk outcome, defined as intensive care unit admission and/or in-hospital mortality. The secondary objectives were to compare score distributions between outcome groups and to assess the sensitivity, specificity, predictive values, accuracy, and discriminative ability of predefined NEWS2 and CURB-65 thresholds.

Study design and setting: This prospective observational comparative prognostic-accuracy study was conducted in the Department of Accident and Emergency Care, Manipal Hospital, Varthur Road, Bengaluru, Karnataka, India, from February 2026 to June 2026. The hospital is a multispeciality tertiary-care facility serving Varthur Road, Whitefield, and surrounding areas, with round-the-clock emergency services and access to inpatient, critical-care, laboratory, and radiological support. Adult patients were evaluated at their first emergency department presentation, before the occurrence of the study outcomes. Participants and eligibility criteria: Consecutive patients aged 18 years or older with community-acquired pneumonia were screened. CAP was diagnosed when an acute compatible clinical syndrome, such as cough, sputum production, dyspnoea, pleuritic symptoms, or fever, was accompanied by a new pulmonary infiltrate on chest imaging and the infection had been acquired outside a healthcare facility.1 Patients were excluded when pneumonia developed after hospitalization, when an alternative diagnosis better explained the radiological findings, when essential variables required to calculate either score were unavailable, or when the patient was transferred after definitive intensive-care treatment had already begun. Only the first eligible presentation during the study period was analyzed. Data collection and score assessment: Demographic details, comorbidities, initial vital signs, mental status, oxygen supplementation, blood urea, and relevant clinical outcomes were recorded on a standardized data-collection form. NEWS2 was calculated from respiratory rate, oxygen saturation, supplemental oxygen, systolic blood pressure, pulse rate, temperature, and consciousness/new confusion using the established NEWS2 framework.4 CURB-65 assigned one point each for confusion, urea greater than 7 mmol/L, respiratory rate of at least 30 breaths/min, systolic blood pressure below 90 mmHg or diastolic pressure of 60 mmHg or lower, and age 65 years or older.3 For the primary comparison, NEWS2 ≥5 and CURB-65 ≥2 were prespecified as high-risk thresholds. Outcome measures: The primary endpoint was a composite high-risk outcome comprising admission to the intensive care unit and/or in-hospital death. ICU admission and in-hospital mortality were also examined separately. Three patients who died after ICU admission were counted once in the composite outcome. Treatment and disposition decisions remained under the responsibility of the attending clinical team and were not determined by the study analysis. Statistical analysis: Continuous variables were summarized as mean ± standard deviation and categorical variables as frequency and percentage. High-risk and non-high-risk groups were compared using the independent-samples t test for continuous score values and the chi-square test or Fisher exact test, as appropriate, for categorical variables. Diagnostic performance was expressed as sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy. Receiver operating characteristic curves were used to estimate the area under the curve with 95% confidence intervals. A two-sided p value <0.05 was considered statistically significant. Ethical considerations: The study was conducted in accordance with institutional ethical standards and the principles of the Declaration of Helsinki. Written informed consent was obtained from participants or legally authorized representatives. Necessary Permissions were obtained before starting the study.

A total of 100 adult patients with community-acquired pneumonia presenting to the emergency department were included in the final analysis. The mean age was 58.6 ± 16.2 years, with a range of 21-87 years; 39 patients (39.0%) were aged 65 years or older. Sixty-one (61.0%) patients were male and 39 (39.0%) were female. Hypertension was the most common comorbidity, present in 42 (42.0%) patients, followed by diabetes mellitus in 34 (34.0%), chronic obstructive pulmonary disease in 18 (18.0%), chronic kidney disease in 11 (11.0%), and ischemic heart disease in 10 (10.0%). At presentation, the mean respiratory rate was 24.8 ± 5.6 breaths/min, mean oxygen saturation was 91.8 ± 5.1%, and mean systolic blood pressure was 118.7 ± 19.6 mmHg. Baseline characteristics are shown in Table 1. Table 1. Baseline demographic and clinical characteristics of patients with community-acquired pneumonia (n = 100) Characteristic Value Age, years, mean ± SD 58.6 ± 16.2 Age range, years 21-87 Age ≥65 years, n (%) 39 (39.0) Male, n (%) 61 (61.0) Female, n (%) 39 (39.0) Hypertension, n (%) 42 (42.0) Diabetes mellitus, n (%) 34 (34.0) COPD, n (%) 18 (18.0) Chronic kidney disease, n (%) 11 (11.0) Ischemic heart disease, n (%) 10 (10.0) Respiratory rate, breaths/min, mean ± SD 24.8 ± 5.6 Oxygen saturation, %, mean ± SD 91.8 ± 5.1 Systolic blood pressure, mmHg, mean ± SD 118.7 ± 19.6 Temperature, °C, mean ± SD 38.1 ± 0.9 NEWS2 score, mean ± SD 4.3 ± 2.5 CURB-65 score, mean ± SD 1.4 ± 1.0 Note. SD = standard deviation; COPD = chronic obstructive pulmonary disease; NEWS2 = National Early Warning Score 2; CURB-65 = confusion, urea, respiratory rate, blood pressure, and age ≥65 years. The mean NEWS2 score was 4.3 ± 2.5 and the mean CURB-65 score was 1.4 ± 1.0. On NEWS2 classification, 31 (31.0%) patients scored 0-2, 32 (32.0%) scored 3-4, and 37 (37.0%) scored ≥5. CURB-65 scores were 0 in 29 (29.0%) patients, 1 in 41 (41.0%), 2 in 22 (22.0%), and ≥3 in 8 (8.0%). Therefore, 37 patients met the NEWS2 ≥5 threshold and 30 met the CURB-65 ≥2 threshold (Table 2). Table 2. Distribution of NEWS2 and CURB-65 severity scores Severity score/category n (%) NEWS2 0-2 31 (31.0) NEWS2 3-4 32 (32.0) NEWS2 ≥5 37 (37.0) CURB-65 = 0 29 (29.0) CURB-65 = 1 41 (41.0) CURB-65 = 2 22 (22.0) CURB-65 ≥3 8 (8.0) CURB-65 ≥2 30 (30.0) Twenty patients (20.0%) required ICU admission and eight (8.0%) died during hospitalization. Three of the patients who died had been admitted to the ICU. After accounting for this overlap, the composite high-risk outcome of ICU admission and/or in-hospital mortality occurred in 25 patients (25.0%); 75 patients (75.0%) had no composite high-risk outcome. Patients with the composite high-risk outcome had substantially higher severity scores at emergency department presentation. Mean NEWS2 was 7.1 ± 2.1 in the high-risk group and 3.4 ± 1.8 in the non-high-risk group (p<0.001). Mean CURB-65 was 2.3 ± 0.9 and 1.1 ± 0.8, respectively (p<0.001). NEWS2 ≥5 was present in 22 of 25 (88.0%) high-risk patients compared with 15 of 75 (20.0%) patients without the composite outcome. CURB-65 ≥2 was present in 18 of 25 (72.0%) high-risk patients and 12 of 75 (16.0%) non-high-risk patients. Both categorical associations were statistically significant (p<0.001) (Table 3). Table 3. Comparison of severity scores according to composite clinical outcome Variable High-risk outcome (n = 25) Non-high-risk outcome (n = 75) p value NEWS2 score, mean ± SD 7.1 ± 2.1 3.4 ± 1.8 <0.001 CURB-65 score, mean ± SD 2.3 ± 0.9 1.1 ± 0.8 <0.001 NEWS2 ≥5, n (%) 22 (88.0) 15 (20.0) <0.001 CURB-65 ≥2, n (%) 18 (72.0) 12 (16.0) <0.001 Note. High-risk outcome = ICU admission and/or in-hospital mortality. Continuous scores were compared using the independent-samples t test; categorical thresholds were compared using the chi-square test or Fisher exact test, as appropriate. At a threshold of ≥5, NEWS2 achieved a sensitivity of 88.0%, specificity of 80.0%, positive predictive value of 59.5%, negative predictive value of 95.2%, and overall accuracy of 82.0% for the composite high-risk outcome. The area under the receiver operating characteristic curve was 0.89 (95% CI 0.82-0.96). CURB-65 ≥2 showed lower sensitivity at 72.0% but slightly higher specificity at 84.0%; its positive predictive value was 60.0%, negative predictive value 90.0%, and overall accuracy 81.0%. The CURB-65 AUC was 0.82 (95% CI 0.73-0.91) (Table 4). The AUC values were compared descriptively because a formal statistical test for the difference between correlated ROC curves was not part of the supplied analysis. Table 4. Diagnostic performance of NEWS2 and CURB-65 for predicting the composite high-risk outcome Performance measure NEWS2 ≥5 CURB-65 ≥2 Sensitivity, % 88.0 72.0 Specificity, % 80.0 84.0 Positive predictive value, % 59.5 60.0 Negative predictive value, % 95.2 90.0 Overall accuracy, % 82.0 81.0 Area under ROC curve 0.89 0.82 95% CI for AUC 0.82-0.96 0.73-0.91 Note. ROC = receiver operating characteristic; AUC = area under the curve; CI = confidence interval. When individual outcomes were examined, NEWS2 ≥5 identified 18 of 20 patients who required ICU admission (90.0%) and seven of eight patients who died (87.5%). CURB-65 ≥2 identified 15 of 20 ICU admissions (75.0%) and six of eight in-hospital deaths (75.0%). The same pattern was observed for the composite endpoint, for which NEWS2 detected 22 of 25 events and CURB-65 detected 18 of 25 events (Table 5). Overall, NEWS2 provided greater sensitivity and negative predictive value, while CURB-65 provided slightly greater specificity. Table 5. Detection of individual and composite adverse outcomes by high-risk score thresholds Outcome Events, n NEWS2 ≥5, n (%) CURB-65 ≥2, n (%) ICU admission 20 18 (90.0) 15 (75.0) In-hospital mortality 8 7 (87.5) 6 (75.0) ICU admission and/or in-hospital mortality 25 22 (88.0) 18 (72.0)

The present study compared NEWS2 with CURB-65 for early identification of high-risk adults with CAP in the emergency department. One quarter of the cohort experienced the composite endpoint of ICU admission and/or in-hospital mortality. Patients with adverse outcomes had markedly higher admission scores: NEWS2 averaged 7.1 compared with 3.4 in the non-high-risk group, while CURB-65 averaged 2.3 compared with 1.1. At the selected thresholds, NEWS2 provided higher sensitivity and negative predictive value, whereas CURB-65 showed slightly higher specificity. The AUC values of 0.89 for NEWS2 and 0.82 for CURB-65 indicate good discrimination for both tools, with a numerical advantage for NEWS2. These findings are clinically plausible because NEWS2 captures acute physiological instability more directly than CURB-65. CURB-65 was derived as a simple mortality-oriented rule,3 while NEWS2 incorporates oxygenation, supplemental oxygen use, pulse, temperature, and evolving neurological status.4 In a prospective emergency-department CAP cohort, Sbiti-Rohr et al. found that NEWS added prognostic information for ICU admission and complications beyond conventional pneumonia scores.5 The current findings extend that concept by showing a high sensitivity of NEWS2 ≥5 for a composite endpoint relevant to early escalation decisions. Previous work has highlighted limitations of mortality-oriented pneumonia scores when the outcome is intensive care requirement. The SMART-COP study demonstrated that prediction tools built specifically around intensive respiratory or vasopressor support can identify a different high-risk phenotype from mortality scores.7 Ananda-Rajah et al. reported that PSI and CURB-65 differed in their ability to predict mortality and ICU admission,8 while España et al. demonstrated that severity instruments varied in predicting clinically important outcomes beyond death.9 Systematic reviews similarly concluded that no single CAP score is uniformly optimal across mortality, mechanical ventilation, vasopressor support, and ICU admission.6,10 This distinction is important in emergency medicine, where failure to recognize instability can delay escalation. Physiology-based approaches have shown encouraging results in CAP. Jo et al. demonstrated prognostic value for an early warning score combined with lactate,11 and Tajarernmuang et al. found that higher NEWS2 values were associated with early progression to severe CAP.12 More recent validation work has shown that NEWS2 retains useful discrimination among complex pneumonia populations, including kidney transplant recipients.13 However, performance varies by population and endpoint. Thirawattanasoot et al. reported that NEWS2 was less discriminative than A-DROP for in-hospital mortality in an emergency-department pneumonia cohort, underscoring that score performance is context dependent.14 The higher sensitivity and NPV observed for NEWS2 in the present study support its use as an early screening instrument, particularly when the clinical priority is to avoid overlooking deterioration. CURB-65 remains valuable because of its simplicity and slightly greater specificity. Rather than treating the two scores as interchangeable, emergency clinicians can interpret NEWS2 as a rapid physiological warning signal and CURB-65 as complementary pneumonia-specific risk information. Clinical judgment, imaging, comorbidities, laboratory findings, and response to initial treatment should continue to guide final disposition decisions. Limitations This study has several limitations. It was conducted at a single tertiary-care emergency department with a modest sample of 100 patients and a short recruitment period. The composite endpoint combined ICU admission and mortality, outcomes influenced partly by local admission practices and resource availability. Microbiological aetiology, treatment timing, radiographic extent, serial score changes, and post-discharge outcomes were not evaluated. External validation in larger multicentre cohorts is required.

In adults presenting to the emergency department with community-acquired pneumonia, both NEWS2 and CURB-65 provided clinically useful early risk stratification. NEWS2 ≥5 demonstrated higher sensitivity, a higher negative predictive value, and greater overall discrimination for ICU admission and/or in-hospital mortality, whereas CURB-65 ≥2 showed slightly higher specificity. These findings support NEWS2 as a practical bedside screening tool for recognizing patients who require closer observation and early escalation of care. CURB-65 remains useful as complementary pneumonia-specific prognostic information. Integration of physiological warning scores with clinical assessment, imaging, laboratory findings, comorbidity burden, and response to initial treatment can strengthen disposition decisions and improve prioritization of emergency and critical-care resources when necessary.

1. Metlay JP, Waterer GW, Long AC, Anzueto A, Brozek J, Crothers K, et al. Diagnosis and treatment of adults with community-acquired pneumonia: an official clinical practice guideline of the American Thoracic Society and Infectious Diseases Society of America. Am J Respir Crit Care Med. 2019;200(7):e45-e67. 2. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336(4):243-250. 3. Lim WS, van der Eerden MM, Laing R, Boersma WG, Karalus N, Town GI, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58(5):377-382. 4. Smith GB, Redfern OC, Pimentel MAF, Gerry S, Collins GS, Malycha J, et al. The National Early Warning Score 2 (NEWS2). Clin Med (Lond). 2019;19(3):260. 5. Sbiti-Rohr D, Kutz A, Christ-Crain M, Thomann R, Zimmerli W, Hoess C, et al. The National Early Warning Score (NEWS) for outcome prediction in emergency department patients with community-acquired pneumonia: results from a 6-year prospective cohort study. BMJ Open. 2016;6(9):e011021. 6. Chalmers JD, Singanayagam A, Akram AR, Mandal P, Short PM, Choudhury G, et al. Severity assessment tools for predicting mortality in hospitalised patients with community-acquired pneumonia: systematic review and meta-analysis. Thorax. 2010;65(10):878-883. 7. Charles PGP, Wolfe R, Whitby M, Fine MJ, Fuller AJ, Stirling R, et al. SMART-COP: a tool for predicting the need for intensive respiratory or vasopressor support in community-acquired pneumonia. Clin Infect Dis. 2008;47(3):375-384. 8. Ananda-Rajah MR, Charles PGP, Melvani S, Burrell LL, Johnson PDR, Grayson ML. Comparing the pneumonia severity index with CURB-65 in patients admitted with community acquired pneumonia. Scand J Infect Dis. 2008;40(4):293-300. 9. España PP, Capelastegui A, Quintana JM, Diez R, Gorordo I, Bilbao A, et al. Prospective comparison of severity scores for predicting clinically relevant outcomes for patients hospitalized with community-acquired pneumonia. Chest. 2009;135(6):1572-1579. 10. Marti C, Garin N, Grosgurin O, Poncet A, Combescure C, Carballo S, et al. Prediction of severe community-acquired pneumonia: a systematic review and meta-analysis. Crit Care. 2012;16(4):R141. 11. Jo S, Jeong T, Lee JB, Jin Y, Yoon J, Park B. Validation of modified early warning score using serum lactate level in community-acquired pneumonia patients: the National Early Warning Score-Lactate score. Am J Emerg Med. 2016;34(3):536-541. 12. Tajarernmuang P, Sanwirat P, Inchai J, Phinyo P, Limsukon A. The National Early Warning Score 2 (NEWS2) to predict early progression to severe community-acquired pneumonia. Trop Med Infect Dis. 2023;8(2):68. 13. Müller-Plathe M, Osmanodja B, Barthel G, Budde K, Eckardt KU, Kolditz M, et al. Validation of risk scores for prediction of severe pneumonia in kidney transplant recipients hospitalized with community-acquired pneumonia. Infection. 2024;52(2):447-459. 14. Thirawattanasoot N, Chongthanadon B, Ruangsomboon O. Performance of A-DROP, NEWS2, and REMS in predicting in-hospital mortality and mechanical ventilation in pneumonia patients in the emergency department: a retrospective cohort study. Int J Emerg Med. 2024;17(1):198.