Predicting in-hospital mortality in adult non-traumatic emergency department patients: a retrospective comparison of the Modified Early Warning Score (MEWS) and machine learning approach

Modified Early Warning Score (MEWS) combined with biomarkers in predicting 7-day mortality in traumatic brain injury patients in the emergency department: a retrospective cohort study

Background

Traumatic brain injury (TBI) is a leading cause of injury-related disability and death globally, which negatively affects individuals, families, and society. Predicting the risk for mortality among TBI patients is crucial in guiding further timely and effective treatment plans. Both the standard risk assessment tools and blood-based biomarkers are helpful in predicting outcomes among TBI patients. However, no studies have compared the predicting performance of the individual and combined indicators from the two major types.

Aim

This study aimed to compare the Modified Early Warning Score (MEWS), Red blood cell distribution width (RDW), and creatine in predicting 7-day mortality among TBI patients.

Methods

A retrospective study was conducted in the emergency department of the First People's Hospital of Changde, China, from January 1, 2023, to June 30, 2023. Data of 1,701 patients with TBI were obtained from the hospital's electronic medical records. A logistic regression model was used to determine independent factors influencing 7-day mortality. The area under the curve (AUC) of the receiver operating characteristic curve (ROC) was calculated to compare the individual and combined effects of MEWS, RDW, and creatine in predicting 7-day mortality based on bootstrap resampling (500 times).

Results

Among the 1,701 patients, 225 died, with a mortality rate of 13.23%. The multivariate analysis showed that the type of TBI lesion, MEWS, SBP, DBP, MAP, SpO2, temperature, RDW, and creatine were significantly associated with 7-day mortality. MEWS (AUC: 0.843) performed better than RDW (AUC: 0.785) and creatine (AUC: 0.797) in predicting 7-day mortality. MEWS+RDW (AUC: 0.898) performed better than MEWS+creatine (AUC: 0.875) and RDW+ creatine (AUC: 0.822) in predicting 7-day mortality. The combination of all three indicators, MEWS+RDW+creatine, showed the best predicting performance (AUC: 0.906).

Conclusion

MEWS performed best in predicting the 7-day mortality of TBI patients, and its predicting performance was improved when combined with blood-based biomarkers such as RDW and creatine. Our findings provide preliminary evidence supporting the combination of MEWS with blood-based biomarkers as a new method for predicting 7-day mortality in patients with TBI.

Soluble Urokinase Plasminogen Activator Receptor: A Promising Biomarker for Mortality Prediction Among Critical ED Patients

Patients admitted to the emergency department (ED) are a highly diverse group in terms of the risk of death. In overcrowded EDs, it becomes crucial to quickly and reliably estimate the risk of death or significant health deterioration. For this purpose, the concentration of soluble urokinase plasminogen activator receptor (suPAR) in plasma has been studied in recent years in various patient populations. In the present study, we tested the hypothesis that measuring suPAR upon the ED admission of critically ill patients can identify those at the highest mortality risk. To verify this hypothesis, we analyzed the relationship between suPAR plasma concentration, other biochemical parameters, and Early Warning Scores (EWSs) on admission and survival to hospital discharge. The study group consisted of 61 ED patients with priority 1 in the Manchester Triage System (MTS), excluding patients with illness caused by environmental factors. Positive correlations between suPAR and inflammatory parameters such as CRP and PCT, as well as the warning scales MEWS, MEDS, and qSOFA, were confirmed. Plasma suPAR concentration on admission was found to be a promising predictor of in-hospital mortality. The study indicated the potential prognostic value of suPAR as the mortality risk predictor for a specific population of critically ill ED patients.

Modified Early Warning Score scores in the emergency department: Factors associated with changing scores to evaluate clinical improvement or deterioration

Background

In the acute care chain, a heterogeneous group of patients seeks medical attention, of whom a small proportion become critically ill. Prediction models, such as the Modified Early Warning Score (MEWS), may assist in the identification of these patients and thereby prevent serious adverse events. The delta score within the emergency room (emergency department [ED]) is associated with outcome. However, it is unknown which factors contribute to these changes in MEWS scores.

Methods

This is a retrospective cohort study at the Amsterdam University Medical Center, which included adult patients presented to the ED by ambulance from March 2022 to October 2022. We collected MEWS at ambulance arrival and 3 h after ED admission, as well as information about diagnostic tests, therapy, and interventions. Our primary outcome was the association of patients' characteristics and acute care actions (diagnostics, therapy, and interventions) with changes in the MEWS score.

Results

A total of 261 patients were included. A higher MEWS at presentation with subsequent improvement was related to better outcomes and they received more therapeutic interventions and the administration of therapy, although these results may have been biased by the need for oxygen supply in respiratory unstable patients. In comparison with patients with normal and stable MEWS scores, they received overall less therapy.

Conclusion

MEWS could be used to predict short-term critical illness in patients presenting to the ED. Further research is needed to evaluate the association of the acute care chains' performance and changes in MEWS scores.

Advancements in Artificial Intelligence in Emergency Medicine in Taiwan: A Narrative Review

The rapid progression of artificial intelligence (AI) in healthcare has greatly influenced emergency medicine, particularly in Taiwan-a nation celebrated for its technological innovation and advanced public healthcare. This narrative review examines the current status of AI applications in Taiwan's emergency medicine and highlights notable achievements and potential areas for growth. AI has wide capabilities encompass a broad range, including disease prediction, diagnostic imaging interpretation, and workflow enhancement. While the integration of AI presents promising advancements, it is not devoid of challenges. Concerns about the interpretability of AI models, the importance of dataset accuracy, the necessity for external validation, and ethical quandaries emphasize the need for a balanced approach. Regulatory oversight also plays a crucial role in ensuring the safe and effective deployment of AI tools in clinical settings. As its footprint continues to expand in medical education and other areas, addressing these challenges is imperative to harness the full potential of AI for transforming emergency medicine in Taiwan.

A comparative study of explainable ensemble learning and logistic regression for predicting in-hospital mortality in the emergency department

This study addresses the challenges associated with emergency department (ED) overcrowding and emphasizes the need for efficient risk stratification tools to identify high-risk patients for early intervention. While several scoring systems, often based on logistic regression (LR) models, have been proposed to indicate patient illness severity, this study aims to compare the predictive performance of ensemble learning (EL) models with LR for in-hospital mortality in the ED. A cross-sectional single-center study was conducted at the ED of Imam Reza Hospital in northeast Iran from March 2016 to March 2017. The study included adult patients with one to three levels of emergency severity index. EL models using Bagging, AdaBoost, random forests (RF), Stacking and extreme gradient boosting (XGB) algorithms, along with an LR model, were constructed. The training and validation visits from the ED were randomly divided into 80% and 20%, respectively. After training the proposed models using tenfold cross-validation, their predictive performance was evaluated. Model performance was compared using the Brier score (BS), The area under the receiver operating characteristics curve (AUROC), The area and precision-recall curve (AUCPR), Hosmer-Lemeshow (H-L) goodness-of-fit test, precision, sensitivity, accuracy, F1-score, and Matthews correlation coefficient (MCC). The study included 2025 unique patients admitted to the hospital's ED, with a total percentage of hospital deaths at approximately 19%. In the training group and the validation group, 274 of 1476 (18.6%) and 152 of 728 (20.8%) patients died during hospitalization, respectively. According to the evaluation of the presented framework, EL models, particularly Bagging, predicted in-hospital mortality with the highest AUROC (0.839, CI (0.802-0.875)) and AUCPR = 0.64 comparable in terms of discrimination power with LR (AUROC (0.826, CI (0.787-0.864)) and AUCPR = 0.61). XGB achieved the highest precision (0.83), sensitivity (0.831), accuracy (0.842), F1-score (0.833), and the highest MCC (0.48). Additionally, the most accurate models in the unbalanced dataset belonged to RF with the lowest BS (0.128). Although all studied models overestimate mortality risk and have insufficient calibration (P > 0.05), stacking demonstrated relatively good agreement between predicted and actual mortality. EL models are not superior to LR in predicting in-hospital mortality in the ED. Both EL and LR models can be considered as screening tools to identify patients at risk of mortality.

Machine Learning Models Versus the National Early Warning Score System for Predicting Deterioration: Retrospective Cohort Study in the United Arab Emirates

BACKGROUND

Early warning score systems are widely used for identifying patients who are at the highest risk of deterioration to assist clinical decision-making. This could facilitate early intervention and consequently improve patient outcomes; for example, the National Early Warning Score (NEWS) system, which is recommended by the Royal College of Physicians in the United Kingdom, uses predefined alerting thresholds to assign scores to patients based on their vital signs. However, there is limited evidence of the reliability of such scores across patient cohorts in the United Arab Emirates.

OBJECTIVE

Our aim in this study was to propose a data-driven model that accurately predicts in-hospital deterioration in an inpatient cohort in the United Arab Emirates.

METHODS

We conducted a retrospective cohort study using a real-world data set that consisted of 16,901 unique patients associated with 26,073 inpatient emergency encounters and 951,591 observation sets collected between April 2015 and August 2021 at a large multispecialty hospital in Abu Dhabi, United Arab Emirates. The observation sets included routine measurements of heart rate, respiratory rate, systolic blood pressure, level of consciousness, temperature, and oxygen saturation, as well as whether the patient was receiving supplementary oxygen. We divided the data set of 16,901 unique patients into training, validation, and test sets consisting of 11,830 (70%; 18,319/26,073, 70.26% emergency encounters), 3397 (20.1%; 5206/26,073, 19.97% emergency encounters), and 1674 (9.9%; 2548/26,073, 9.77% emergency encounters) patients, respectively. We defined an adverse event as the occurrence of admission to the intensive care unit, mortality, or both if the patient was admitted to the intensive care unit first. On the basis of 7 routine vital signs measurements, we assessed the performance of the NEWS system in detecting deterioration within 24 hours using the area under the receiver operating characteristic curve (AUROC). We also developed and evaluated several machine learning models, including logistic regression, a gradient-boosting model, and a feed-forward neural network.

RESULTS

In a holdout test set of 2548 encounters with 95,755 observation sets, the NEWS system achieved an overall AUROC value of 0.682 (95% CI 0.673-0.690). In comparison, the best-performing machine learning models, which were the gradient-boosting model and the neural network, achieved AUROC values of 0.778 (95% CI 0.770-0.785) and 0.756 (95% CI 0.749-0.764), respectively. Our interpretability results highlight the importance of temperature and respiratory rate in predicting patient deterioration.

CONCLUSIONS

Although traditional early warning score systems are the dominant form of deterioration prediction models in clinical practice today, we strongly recommend the development and use of cohort-specific machine learning models as an alternative. This is especially important in external patient cohorts that were unseen during model development.

Modified Early Warning Score: Clinical Deterioration of Mexican Patients Hospitalized with COVID-19 and Chronic Disease

The objective was to evaluate the Modified Early Warning Score in patients hospitalized for COVID-19 plus chronic disease.

METHODS

Retrospective observational study, 430 hospitalized patients with COVID-19 and chronic disease. Instrument, Modified Early Warning Score (MEWS). Data analysis, with Cox and logistic regression, to predict survival and risk.

RESULTS

Of 430 patients, 58.6% survived, and 41.4% did not. The risk was: low 53.5%, medium 23.7%, and high 22.8%. The MEWS score was similar between survivors 3.02, p 0.373 (95% CI: -0.225-0.597) and non-survivors 3.20 (95% CI: -0.224-0.597). There is a linear relationship between MEWS and mortality risk R 0.920, ANOVA 0.000, constant 4.713, and coefficient 4.406. The Cox Regression p 0.011, with a risk of deterioration of 0.325, with a positive coefficient, the higher the risk, the higher the mortality, while the invasive mechanical ventilation coefficient was negative -0.757. By providing oxygen and ventilation, mortality is lower.

CONCLUSIONS

The predictive value of the modified early warning score in patients hospitalized for COVID-19 and chronic disease is not predictive with the MEWS scale. Additional assessment is required to prevent complications, especially when patients are assessed as low-risk.

Artificial intelligence model comparison for risk factor analysis of patent ductus arteriosus in nationwide very low birth weight infants cohort

Despite the many comorbidities and high mortality rate in preterm infants with patent ductus arteriosus (PDA), therapeutic strategies vary depending on the clinical setting, and most studies of the related risk factors are based on small sample populations. We aimed to compare the performance of artificial intelligence (AI) analysis with that of conventional analysis to identify risk factors associated with symptomatic PDA (sPDA) in very low birth weight infants. This nationwide cohort study included 8369 very low birth weight (VLBW) infants. The participants were divided into an sPDA group and an asymptomatic PDA or spontaneously close PDA (nPDA) group. The sPDA group was further divided into treated and untreated subgroups. A total of 47 perinatal risk factors were collected and analyzed. Multiple logistic regression was used as a standard analytic tool, and five AI algorithms were used to identify the factors associated with sPDA. Combining a large database of risk factors from nationwide registries and AI techniques achieved higher accuracy and better performance of the PDA prediction tasks, and the ensemble methods showed the best performances.