Institution-Specific Machine Learning Models for Prehospital Assessment to Predict Hospital Admission: Prediction Model Development Study

Prediction of hospitalization and waiting time within 24 hours of emergency department patients with unstructured text data

Overcrowding of emergency departments is a global concern, leading to numerous negative consequences. This study aimed to develop a useful and inexpensive tool derived from electronic medical records that supports clinical decision-making and can be easily utilized by emergency department physicians. We presented machine learning models that predicted the likelihood of hospitalizations within 24 hours and estimated waiting times. Moreover, we revealed the enhanced performance of these machine learning models compared to existing models by incorporating unstructured text data. Among several evaluated models, the extreme gradient boosting model that incorporated text data yielded the best performance. This model achieved an area under the receiver operating characteristic curve score of 0.922 and an area under the precision-recall curve score of 0.687. The mean absolute error revealed a difference of approximately 3 hours. Using this model, we classified the probability of patients not being admitted within 24 hours as Low, Medium, or High and identified important variables influencing this classification through explainable artificial intelligence. The model results are readily displayed on an electronic dashboard to support the decision-making of emergency department physicians and alleviate overcrowding, thereby resulting in socioeconomic benefits for medical facilities.

Artificial intelligence and machine learning in prehospital emergency care: A scoping review

Our scoping review provides a comprehensive analysis of the landscape of artificial intelligence (AI) applications in prehospital emergency care (PEC). It contributes to the field by highlighting the most studied AI applications and identifying the most common methodological approaches across 106 included studies. The findings indicate a promising future for AI in PEC, with many unique use cases, such as prognostication, demand prediction, resource optimization, and the Internet of Things continuous monitoring systems. Comparisons with other approaches showed AI outperforming clinicians and non-AI algorithms in most cases. However, most studies were internally validated and retrospective, highlighting the need for rigorous prospective validation of AI applications before implementation in clinical settings. We identified knowledge and methodological gaps using an evidence map, offering a roadmap for future investigators. We also discussed the significance of explainable AI for establishing trust in AI systems among clinicians and facilitating real-world validation of AI models.

The Ability of Emergency Medical Service Staff to Predict Emergency Department Disposition: A Prospective Study

Purpose

Paramedics' decision to notify receiving hospitals and transport patients to an appropriate healthcare facility is based on the Prediction of Intensive Care Unit (ICU) and Hospital Admissions guide. This study aimed to assess the paramedics' gestalt on both ward and ICU admission.

Patients and Methods

A prospective study was conducted at King Abdulaziz Medical City between September 2021 and March 2022. Paramedics were asked several questions related to the prediction of the patient's hospital outcome, including emergency department (ED) discharge or hospital admission (ICU or ward). Additional data, such as the time of the ambulance's arrival and the staff years of experience, were collected. The categorical characteristics are presented by frequency and percentage for each category.

Results

This study included 251 paramedics and 251 patients. The average age of the patients was 62 years. Of the patients, 32 (12.7%) were trauma, and 219 (87.3%) were non-trauma patients. Two-thirds of the patients (n=171, 68.1%) were predicted to be admitted to the hospital, and 80 (31.8%) of the EMS staff indicated that the patient do not need a hospital or an ambulance. The sensitivity, specificity, PPV, and NPV of the emergency medical service (EMS) staffs' gestalt for patient admission to the hospital were, respectively (77%), (33%), (16%), and (90%). Further analysis was reported to defend the EMS staffs' gestalt based on the level of EMS staff and the nature of the emergency (medical vs trauma), are reported.

Conclusion

Our study reports a low level of accurately predicting patient admission to the hospital, including the ICU. The results of this study have important implications for enhancing the accuracy of EMS staff predictive ability and ensuring that patients receive appropriate care promptly.

Machine Learning-Based Models for Prediction of Critical Illness at Community, Paramedic, and Hospital Stages

Overcrowding of emergency department (ED) has put a strain on national healthcare systems and adversely affected the clinical outcomes of critically ill patients. Early identification of critically ill patients prior to ED visits can help induce optimal patient flow and allocate medical resources effectively. This study aims to develop ML-based models for predicting critical illness in the community, paramedic, and hospital stages using Korean National Emergency Department Information System (NEDIS) data. Random forest and light gradient boosting machine (LightGBM) were applied to develop predictive models. The predictive model performance based on AUROC in community stage, paramedic stage, and hospital stage was estimated to be 0.870 (95% CI: 0.869-0.871), 0.897 (95% CI: 0.896-0.898), and 0.950 (95% CI: 0.949-0.950) in random forest and 0.877 (95% CI: 0.876-0.878), 0.899 (95% CI: 0.898-0.900), and 0.950 (95% CI: 0.950-0.951) in LightGBM, respectively. The ML models showed high performance in predicting critical illness using variables available at each stage, which can be helpful in guiding patients to appropriate hospitals according to their severity of illness. Furthermore, a simulation model can be developed for proper allocation of limited medical resources.

Effects of evidence-based ICU care on long-term outcomes of patients with sepsis or septic shock (ILOSS): protocol for a multicentre prospective observational cohort study in Japan

INTRODUCTION

Sepsis is not only the leading cause of death in the intensive care unit (ICU) but also a major risk factor for physical and cognitive impairment and mental disorders, known as postintensive care syndrome (PICS), reduced health-related quality of life (HRQoL) and even mental health disorders in patient families (PICS-family; PICS-F). The ABCDEF bundle is strongly recommended to overcome them, while the association between implementing the bundle and the long-term outcomes is also unknown.

METHODS AND ANALYSIS

This is a multicentre prospective observational study at 26 ICUs. All consecutive patients between 1 November 2020 and 30 April 2022, who are 18 years old or older and expected to stay in an ICU for more than 48 hours due to sepsis or septic shock, are enrolled. Follow-up to evaluate survival and PICS/ PICS-F will be performed at 3, 6 and 12 months and additionally every 6 months up to 5 years after hospital discharge. Primary outcomes include survival at 12 months, which is the primary outcome, and the incidence of PICS defined as the presence of any physical impairment, cognitive impairment or mental disorders. PICS assessment scores, HRQoL and employment status are evaluated. The association between the implementation rate for the ABCDEF bundle and for each of the individual elements and long-term outcomes will be evaluated. The PICS-F, defined as the presence of mental disorders, and HRQoL of the family is also assessed. Additional analyses with data up to 5 years follow-up are planned.

ETHICS AND DISSEMINATION

This study received ethics approvals from Saiseikai Utsunomiya Hospital (2020-42) and all other participating institutions and was registered in the University Hospital Medical Information Network Clinical Trials Registry. Informed consent will be obtained from all patients. The findings will be published in peer-reviewed journals and presented at scientific conferences.

TRIAL REGISTRATION NUMBER

UMIN000041433.