In-hospital cardiac arrest (IHCA) is a critical medical issue threatening the survival and prognosis of hospitalized patients, characterized by high incidence, high mortality and poor prognosis. Early warning and intervention for IHCA are urgently needed. The early warning score (EWS) is developed as a point-of-care warning tool for early identification and intervention of hospitalized patients with deteriorating condition. In recent years, EWS has become one of the important methods for early warning of IHCA, especially EWS based on machine learning (ML) has shown great potential. This review mainly focuses on the traditional EWS and ML-based EWS, discusses the research status of EWS worldwide, and focuses on the research progress of EWS in early warning of IHCA.
Humans ; Heart Arrest/therapy* ; Early Warning Score ; Prognosis ; Machine Learning ; Hospitalization

OBJECTIVE: To explore the predictive value of acute physiology and chronic health evaluation II (APACHE II), sequential organ failure assessment (SOFA), quick sequential organ failure assessment (qSOFA) and modified early warning score (MEWS) in evaluating the prognosis of patients in intensive care unit (ICU) of secondary hospitals, and to provide guidance for clinical application. METHODS: The clinical data of adult critical patients admitted to the ICU of Wanzhou District First People's Hospital from October 2022 to April 2023 were retrospectively analyzed. According to the clinical outcome of ICU, the patients were divided into improvement group and death group. The general information, blood routine, heart, liver and kidney function indicators, coagulation indicators, blood gas analysis, APACHE II score, SOFA score, qSOFA score, MEWS score at the time of admission to the ICU, the number of cases of invasive mechanical ventilation (IMV) and continuous blood purification (CBP) were compared between the two groups. Univariate analysis was performed, and multivariate Logistic regression analysis was used to analyze the related factors of death. Receiver operator characteristic curve (ROC curve) was used to analyze the predictive value of the four scores in ICU patients. RESULTS: A total of 126 patients were included, of which 45 patients died in the ICU and 81 patients improved and transferred out. Univariate analysis of death-related critically ill patients showed that procalcitonin (PCT), serum creatinine (SCr), blood urea nitrogen (BUN), albumin (ALB), prothrombin time (PT), activated partial prothrombin time (APTT), D-dimer, pH value, HCO₃^-, blood lactic acid (Lac), number of patients treated with IMV and CBP, APACHE II score, SOFA score, qSOFA score and MEWS score were significantly different between the two groups (all P < 0.05). Multivariate Logistic regression analysis showed that the APACHE II score [odds ratio (OR) = 1.115, 95% confidence interval (95%CI) was 1.025-1.213, P = 0.011], SOFA score (OR = 1.204, 95%CI was 1.037-1.398, P = 0.015), MEWS score (OR = 1.464, 95%CI was 1.102-1.946, P = 0.009), and APTT (OR = 1.081, 95%CI was 1.015-1.152, P = 0.016) were independent risk factors affecting the mortality of critically ill patients in the ICU. ROC curve analysis showed that APACHE II, SOFA, qSOFA, and MEWS scores could predict the prognosis of critically ill ICU patients, among which SOFA score had the strongest predictive effect, and the area under the curve (AUC) was 0.808. There was a statistically significant difference in the time required for the four scores (F = 117.333, P < 0.001), among which the MEWS scoring required the shortest time [(1.03±0.39) minutes], and the APACHE II scoring required the longest time [(2.81±1.04) minutes]. CONCLUSIONS APACHE II, SOFA, qSOFA, and MEWS scores can be used to assess the severity of critically ill patients and predict in-hospital mortality. The SOFA score is superior to other scores in predicting severity. The MEWS is preferred because its assessment time is shortest. Early warning score can help secondary hospitals to detect potentially critical patients early and provide help for clinical rapid urgent emergency decision-making.
Adult ; Humans ; Sepsis/diagnosis* ; ROC Curve ; Retrospective Studies ; Critical Illness ; Early Warning Score ; Organ Dysfunction Scores ; Intensive Care Units ; Prognosis ; Hospitals

INTRODUCTION: The purpose of this study was to assess the application of the early warning score system (EWS-S) and gauge physician awareness, perceptions of necessity and attitudes regarding these tools based on previously experienced unnoticed clinical deterioration (CDET). METHODS: A cross-sectional survey was carried out via an online questionnaire at a large 3,500-bed Class 3A general hospital in China. A total of 299 physicians of adult general wards were asked to answer a translated questionnaire that was localised from the original version. Demographic profiles of patients were included as well as three other sections assessing awareness of CDET/EWS-S and gauging attitudes towards and perceptions of the necessity of EWS-S at our hospital. RESULTS: A high level of physician awareness of the CDET problem was observed. Most physicians knew about the existence of a systematic assessment tool for clinical application. Physicians with previous experience in reanimation, unplanned transfer to intensive care unit (UTICU) and/or death tended to consider EWS-S necessary in attentive and well-trained staff (p < 0.05). Physicians who had previous experience with UTICU were more likely to recommend implementing EWS-S in their wards compared with those without such experience (p < 0.05). CONCLUSION Most physicians have positive attitudes towards EWS-S. However, their awareness should be further heightened. Physicians who had previous experience with CDET/UTICU were more likely to employ EWS-S in their clinical practices. To better facilitate the implementation of EWS-S in Chinese hospitals, existing facilities, policy supports, standardised managements and the development of information systems should be strengthened.
Adult ; Attitude ; Clinical Deterioration ; Cross-Sectional Studies ; Early Warning Score ; Humans ; Physicians