Introduction

The emergency department (ED) receives a large number of critically ill patients, whose physical condition is complex and progresses rapidly and severely. Moreover, there are more adverse outcomes in ED patients [1], including acute deterioration, sometimes unexpected cardiac arrest, and even death [2]. Therefore, a simple, rapid, and effective assessment tool is essential for the triage and referral of ED patients and evaluating whether need to be admitted to the intensive care unit (ICU) to reduce adverse outcomes [3]. In recent years, there has been an increasing amount of patient scoring systems have been used worldwide to assess the severity of the illness to predict the chances of hospitalization, ICU admission, and in-hospital mortality of ED patients [4]. The Modified Early Warning Score (MEWS), the Vital Pac Early Warning Score, the Rapid Emergency Medicine Score, the electronic Cardiac Arrest Risk Triage (eCART) score, the newly designed National Early Warning Score (NEWS) are commonly used [5-8].

The National Early Warning Score (NEWS) and The National Early Warning Score 2 (NEWS2) were designed by the Royal College of Physicians to detect deteriorating patients in hospital wards, specifically those at increased risk of ICU admission, cardiac arrest, or death within 24 hours  [9-11]. The NEWS was initially used to predict illness severity and deterioration in the hospital  [9]. The NEWS enables the early identification and diagnosis of patient deterioration, timely action of the hospital’s rapid response teams (RRTs), and consistent assessment of illness severity; it also provides useful baseline data to evaluate a patient’s clinical progress  [10]. Moreover, the NEWS is useful to standardize the assessment of acute illness and support urgency and competency of the clinical response and used to re-organize hospital services. The implementation of NEWS was associated with a significant improvement in patients’ outcomes in hospital wards, increases in the frequency of vital signs measurements, and an increase in the number of medical reviews following clinical instability  [12]. Recently, a study demonstrated that the NEWS is also a valuable tool for coronavirus disease 2019 (COVID-19) patients and recommended by the World Health Organization (WHO) during the COVID-19 pandemic [13]. NEWS2 support clinical judgment and provide a standardized communication tool that could be practically feasible in a short time scale, and in the context of strained resources and operational pressure faced by hospitals during the emergency phase of the COVID-19 pandemic outbreak.

Previous studies have concentrated on the relationship between the NEWS, mortality, and the prognostic value of using the NEWS with inflammatory blood markers or the Sequential Organ Function Assessment score [14]. Recent trends in NEWS studies [2,3] should focus on a simple, economic, and fast scanning tool to identify critically ill patients quickly and admit them to the ICU to improve the outcomes  [4]. However, to the best of our knowledge, few numbers of cross-sectional studies were performed to determine the association between NEWS score and ICU admissions of ED patients. In this study, we aimed to conduct a secondary cross-sectional study using non-randomized cluster sampling and cross-sectional study and the data from a multinational observational cohort study conducted at three tertiary care hospitals by a public DATADRYAD database to clarify the relation of NEWS score and ICU admissions of ED patients and which may provide a basis for future planning the random grouping, experimental control, and prospective study of NEWS.

Materials and Methods

Collection of data and score assignment

This study was a secondary analysis of data from a multinational observational cohort study conducted at three tertiary care hospitals in Aarau (Switzerland), Paris (France), and Clearwater (Florida, USA), from March 2013 to October 2014. The DATADRYAD database was used to obtain the data. We cited the DRYAD data package following protocols released in the DATADRYAD  [15]. All adult internal medicine patients seeking emergency care at one of the participating hospitals were included in the study, while surgical and pediatric patients were excluded. The protocol and details regarding the study design have been published previously [16,17]. The patients in the original study were 1303, however, 25 patients were eliminated because of the research indicators missing. The final sample size was 1278, the details of the selection of the patients showed in Figure 1. When patients were admitted to the ED, a triage nurse in ED recorded their age, sex, vital signs, main clinical symptoms and complaints, and socio-demographic characteristics and following regular laboratory tests, MR-proADM, and Procalcitonin (PCT) determination and leftover blood samples for further analyses. The NEWS method was calculated as shown in the original study15 based on six parameters: respiratory rate, oxygen saturation, temperature, blood pressure, heart rate, and level of consciousness. As suggested by the Royal College of Physicians, Patients were classified into three NEWS categories: low (0-4points), moderate (5-6 points), high (≥7 points) risk. Additionally, as supplemental oxygen data were not available, the results of this study correspond to the NEWS-potentially minus 2 points, and detailed demographic data are shown in Table 1.

Statistical Analysis

Continuous variables were expressed as mean ±SD (for normally distributed outcomes) or median (IQR) (for non-normally distributed outcomes), and categorical variables were presented as frequencies and percentages. First, the relationship between the NEWS (or other variables) and ICU admission was assessed using multivariable logistic regression models. Non-adjusted, minimally adjusted, and multivariate-adjusted models were employed. In another separate analysis, the NEWS was included as a continuous and categorical variable. Unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. We employed R programming and Empower Stats software18 to conduct the analyses, and the significance level was set at p< 0.05.

Results

The characteristics of selected patients

In this study, we included 1,303 patients in the original database, and out of which 25 patients with missing data were excluded. Finally 1,278 patients were analyzed, of whom 168 (13.15%) were admitted to the ICU, and 1,110 (86.85%) were not. The general information of the patients is shown in Table 2.

Univariate and multivariate analysis of influence factor in ICU admission

Demographic characteristics including gender, age, respiratory rate, saturation of peripheral oxygen(SPO2), systolic blood pressure(SBP), diastolic blood pressure(DBP), heart rate, temperature, MR-proADM, Procalcitonin (PCT), mental status, and NEWS. Univariate analysis showed gender,respiratory rate, SPO2, SBP, heart rate, temperature, MR-proADM, PCT, and NEWS to be possible risk factors for ICU admission (p<0.05). Multivariate analysis showed that gender and NEWS were the two factors associated with ICU admission (p<0.05). However, no statistical differences were observed with the other variables (Table 3).

Multivariable logistic regression model to determine the relation between NEWS score and ICU admission in ED patients

The results of multivariable logistic regression are shown in Table 4. In the non-adjusted model, the OR was 1.25, and the 95% CI was 1.17-1.34. A unit increase in the NEWS could be interpreted as leading to a 25% increase in the incidence of ICU admission. Patients were classified into three NEWS categories: low (0-4points), moderate (5-6 points), high (≥7 points)risk and calculated the p-value for the trend. There was an increased risk for the prevalence of ICU admission as the cutoff point for the NEWS was elevated (p<0.001). Participants who had a higher NEWS experienced an elevated risk of ICU admission (OR= 3.28, 95% CI: 2.06-5.22). Intuitively, supplemental oxygen data have an extraordinary effect on the overall NEWS scores. To explore the effect of supplemental oxygen data on the overall NEWS scores, we had performed an additional adjusted regression model to determine the differences between NEWS scores without supplemental oxygen as non-adjusted model 1 and the NEWS-potentially add 2 points with supplemental oxygen as an adjusted model. In Table 5, the results show that OR (95% CI): 1.25 (1.17, 1.34) in Non-adjusted Model 1, and respectively 1.25(1.17, 1.33). Therefore, we concluded that supplemental oxygen or not in ED patients might not affect OR values in the relationship between ICU admission and NEWS score.

Discussion

Early detection of clinically unstable emergency department patients is the cornerstone in improving the patient, If you do not find in time, and delays in the ED patients associate with unplanned admissions or readmissions to ICU, cardiac arrests, and even unexpected deaths. This study aimed to clarify the relationship between the NEWS and ICU admission in patients in the ED in the setting of cross-sectional analysis. The major finding of our study lies that the NEWS was independently associated with ICU admission in patients in the ED. The multiple regression analysis performed in the current research indicated that the NEWS noticeably influenced ICU admission, and the result is consistent with that of some other studies. One of the results from retrospective cohort studies data with the hospitalized population showed that the NEWS is more accurate at predicting 10-day and 30-day mortality than the qSOFA and Systemic Inflammatory Response Syndrome (SIRS) in patients admitted to the ED with suspected sepsis  [14]. In addition, the NEWS can effectively predict the risk of admission in ICU of sepsis patients [19]. A tertiary hospital in South China including 383 patients clearly showed that the area under the receiver operating characteristic (ROC) curve of the MEWS for in-hospital mortality prediction was 0.83 (95% CI: 0.786–0.881). This study used a prospective study design and provided a new cut-off point for the MEWS using the ROC curve to increase sensitivity in predicting in-hospital mortality and ICU admission  [5]. Alhmoud’s study [20], concluded the same, as it was found through this study that pre-hospital NEWS was associated with death or critical care unit escalation within 48 hours of hospital admission, the NEWS could be used with ED patients to evaluate whether the patients require hospital treatment or rapid transport.

Most interestingly, controversy exists about the NEWS. Some scholars have concluded that the NEWS has poor predictive powers for 28-day mortality in patients who activate the medical emergency team (MET). A new scoring system is needed to compare the diagnosis and prognosis of patients who activate the METs [2]. Aprospective cohort (SNOOPII study) also did not support the use of the NEWS to predict the in-hospital mortality risk of sepsis in medical wards [3]. One study including 85,322 patients (42,402 patients pre-NEWS and 42,920 patients post-NEWS implementation) found that the first outcome of ICU admission rate or death rate did not change after the NEWS implementation [21]. Asystematic review showed that the methodology and quality of validation studies of the NEWS are insufficient to recommend its use in all diseases and all clinical settings despite the good performance of the NEWS in some subgroups. There is an urgent need for consistent studies following consensus guidelines for the NEWS [22].

Many attempts have been made to increase the sensitivity and specificity of the NEWS for predicting mortality and ICU admission. Chiu  [22], conducted a study on a large sample (13,631 individuals) of patients discharged from the ICU after risk-stratified cardiac surgery in four centers and added some vital signs to generate a logistic version of the NEWS, which he found did better at determining deaths due to cardiac surgeries, cardiac arrest, or unplanned readmission to intensive care. Logistic scores also provided an assessment tool for predicting risk factors, in which the original NEWS AUC scored 0.779 (0.771–0.786) vs. 0.754 (0.746–0.761), p < 0.001 for 24 hours and 6 hours, respectively. However, the logistic scores for these time periods were 0.841 (0.829–0.853) vs. 0.813 (0.800–0.825), p< 0.001, respectively. Andreas Eckart’s15research showed that the NEWS is helpful in risk stratification of ED patients and can be further improved by the addition of inflammatory blood markers. The Japanese investigators used the NEWS and the Classification and Regression Tree (CART) machine learning method to analyze the predictive weight of parameters for the outcomes and found that the NEWS was highly associated with 30-day mortality and ICU admission after RRT/MET calls  [23]. Another study suggest that combination the NEWS with the triage system increase the capacity to separate those patients with conditions that had the potential to worsen, despite previously being classified otherwise  [24,25].

In Table 5, we concluded that supplemental oxygen or not in ED patients might not affect OR values in the relationship between ICU admission and NEWS score. The reason for this may lies in the setting of that oxygen supplement in patients with low NEWS scores (severe case) do not affect the key scoring items of the NEWS system, whereas oxygen intake in patients with high scores (mild case) do not determine overall NEWS scores including the decrement of respiration rate, SPO2, SBP, DBP as a categories variable. We could conclude supplemental oxygen is not an independent variable to determine the ED patients’ ICU admission, but overall NEWS score is independent variable.

To the best of our knowledge, we firstly clarified that NEWS score is a rapid, simple, and accurate screening tool to determine the ED patient’s ICU admission by a cross-sectional study which data integrated from three large, independent, multicenter of clinical research. The NEWS or rapid response systems, which are assigned to clinical measurements and used to distinguish patients who need clinical review or resuscitation, are increasingly and commonly used in United Kingdom hospitals and primary health care settings  [25]. Many studies have shown that the NEWS combined with inflammatory blood markers, the triage system, or the CART machine learning method improved sensitivity and specificity for predicting in-hospital mortality and ICU admission. However, some scholars advise that future studies be undertaken to validate this conclusion.

Limitation

This study is a cross-sectional study but not a complete cohort study. The method in this study was non-randomized cluster sampling and cross-sectional study and the data from a multinational observational cohort study. Therefore, the extrapolation of the conclusion of NEWS score as a screening tool to determine the ED patient’s ICU admission in all settings should be carefully adopted. In addition, we could not obtain the exact data on oxygen therapy in the observational cohort study. In the process of data processing, we reduced the overall NEWS score by 2 points in the data processing, resulting in a low score of the NEWS. Therefore, there may be some deviation. Oxygen therapy might not affect OR values in the relationship between ICU admission and NEWS score. In view of this, the random grouping, experimental control, and prospective study will be the future planned work.

Conclusion

The current research clarified that the NEWS is independently associated with ICU admission in patients in ED.A unit increase in the NEWS could be interpreted as leading to a 25% increase in the incidence of ICU admission. The study provides evidence and methods for future research on the NEWS in ED patients.

Data availability statement

The datasets presented in this study can be accessed via the DRYAD data repository at http://datadryad.org/ with the doi:10.5061/dryad.d22q6vh.

Ethics statement

This is an open-access article distributed in accordance with the Creative Commons Attribution Non-Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, and build upon this work non commercially and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made are indicated, and the use is noncommercial. Ethical approval was issued by the Ethics Committee of The First People's Hospital of Changde City (2018-035-09).

Funding

This research was supported by the Hunan Science and Technology Department–Clinical Medical Technology Demonstration Base for Neurosurgery in Hunan Province (2017SK51304).

Contribution

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Figures

Tables

References

1. Emond M, Grenier D, Morin J. (2017) Emergency Department Stay Associated Delirium in Older Patients[J], Can Geriatr J 20: 10-14.
2. Ahn JH, Jung YK, Lee JR. (2020) Predictive powers of the Modified Early Warning Score and the National Early Warning Score in general ward patients who activated the medical emergency team[J], PLoS One, 15: e0233078.
3. Tirotta D, Gambacorta M, La Regina M. (2017) Evaluation of the threshold value for the modified early warning score (MEWS) in medical septic patients: a secondary analysis of an Italian multicentric prospective cohort (SNOOPII study)[J], QJM, 110: 369-373.
4. Nannan Panday R.S, Minderhoud T.C, Alam N. (2017) Prognostic value of early warning scores in the emergency department (ED) and acute medical unit (AMU): A narrative review[J], Eur J Intern Med, 45: 20-31.
5. Xie X, Huang W, Liu Q. (2018) Prognostic value of Modified Early Warning Score generated in a Chinese emergency department: a prospective cohort study[J], BMJ Open, 8: e024120.
6. Dundar ZD, Ergin M, Karamercan MA. (2016)Modified Early Warning Score and VitalPac Early Warning Score in geriatric patients admitted to emergency department[J], Eur J Emerg Med, 23: 406-412.
7. Green M, Lander H, Snyder A. (2018) Comparison of the Between the Flags calling criteria to the MEWS, NEWS and the electronic Cardiac Arrest Risk Triage (eCART) score for the identification of deteriorating ward patients[J], Resuscitation, 123: 86-91.
8. Endo T, Yoshida T, Shinozaki T. (2020) Efficacy of prehospital National Early Warning Score to predict outpatient disposition at an emergency department of a Japanese tertiary hospital: a retrospective study[J], BMJ Open, 10: e034602.
9. Scott L.J, Redmond N.M, Garrett J. (2019) Distributions of the National Early Warning Score (NEWS) across a healthcare system following a large-scale roll-out[J], Emerg Med J, , 36: 287-292.
10. Royal College of Physicians. (2020) National Early Warning Score (NEWS): standardising the assessment of acute-illness severity in the NHS. Report of a working party. London: RCP.
11. Royal College of Physicians. National Early Warning Score (NEWS) 2: Standardising the assessment of acute-illness severity in the NHS. Updated report of a working party. London: RCP.
12. Torri E, Rigoni M, Dorigoni S. (2019)A model based on intensity of medical care may improve outcomes for internal medicine patients in Italy[J], PLoS One, 14: e0211548.
13. Rigoni M, Torri E, Nollo G. (2021) NEWS2 is a valuable tool for appropriate clinical management of COVID-19 patients[J], Eur J Intern Med, 85: 118-120.
14. Brink A, Alsma J, Verdonschot R. (2019) Predicting mortality in patients with suspected sepsis at the Emergency Department; A retrospective cohort study comparing qSOFA, SIRS and National Early Warning Score[J], PLoS One, 14: e0211133.
15. Eckart A, Hauser S.I, Kutz A. (2019) Combination of the National Early Warning Score (NEWS) and inflammatory biomarkers for early risk stratification in emergency department patients: results of a multinational, observational study[J], BMJ Open, 9: e024636.
16. Schuetz P, Hausfater P, Amin D. (2015) Biomarkers from distinct biological pathways improve early risk stratification in medical emergency patients: the multinational, prospective, observational TRIAGE study[J], Crit Care, 19: 377.
17. Schuetz P, Hausfater P, Amin D. (2013) Optimizing triage and hospitalization in adult general medical emergency patients: the triage project[J], BMC Emerg Med, 13: 12.
18. Zhang L, Zhou Q, Shao L.H. (2021) Association of Carotid Atherosclerosis With White Matter Hyperintensity in an Asymptomatic Japanese Population: A Cross-Sectional Study[J], Front Cardiovasc Med, 8: 665573.
19. McLymont N, Glover GW. (2016) Scoring systems for the characterization of sepsis and associated outcomes[J], Ann Transl Med, 4: 527.
20. Alhmoud B, Bonnici T, Patel R. (2021) Performance of universal early warning scores in different patient subgroups and clinical settings: a systematic review[J], BMJ Open, 11: e045849.
21. Bedoya A.D, Clement M.E, Phelan M. (2019) Minimal Impact of Implemented Early Warning Score and Best Practice Alert for Patient Deterioration[J], Crit Care Med, 47: 49-55.
22. Chiu Y.D, Villar S.S, Brand J.W. (2020) Logistic early warning scores to predict death, cardiac arrest or unplanned intensive care unit re-admission after cardiac surgery[J], Anaesthesia, , 75: 162-170.
23. Naito T, Hayashi K, Hsu HC. (2021) Validation of National Early Warning Score for predicting 30-day mortality after rapid response system activation in Japan[J], Acute Med Surg, 8: e666.
24. Abbott T.E.F, Cron N, Vaid N. (2018) Pre-hospital National Early Warning Score (NEWS) is associated with in-hospital mortality and critical care unit admission: A cohort study[J], Ann Med Surg (Lond), 27: 17-21.
25. Abbott TEF, Torrance HDT, Cron N. (2016) A single-centre cohort study of National Early Warning Score (NEWS) and near patient testing in acute medical admissions[J], Eur J Intern Med, 35: 78-82.