BMC Emergency Medicine最新文献

Introduction: Determining the triage level of patients upon their arrival at the hospital emergency department is highly important for identifying high-risk patients and allocating resources to them. This issue can be of even greater importance in patients presenting with cardiac-related symptoms. This study was conducted to predict the triage level of patients presenting with cardiac-related symptoms using machine learning methods and to compare the performance of different approaches.

Methods: This prospective study was conducted in 2024 in three main steps. In the first step, a literature review was performed, and the factors influencing patient triage levels were extracted from previous studies. Then the identified factors from the literature review were presented to experts for their opinion, and the final influential factors were determined based on their feedback. In the second step, patient data were collected from the triage unit of a specialized cardiac hospital. In the third and final step, the collected data were preprocessed and then analyzed using Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Gradient Boosting (GB) machine learning methods.

Results: After reviewing the literature and surveying experts, the confirmed factors were finally identified, and data related to 1862 patients were collected. 52% of the participants in this study were male. RF achieved the highest performance with a test accuracy of 93.57%, Cohen's Kappa of 0.82, and a weighted F1-score of 0.93, followed by GB (accuracy = 90.08%, Kappa = 0.73) and SVM (accuracy = 86.6%, Kappa = 0.66). The most influential factors on patients' triage levels included: the type of high-risk condition that elevates a patient to Level 2, need for life-saving intervention, having high-risk conditions (what condition), chief complaint, level of consciousness, and diseases.

Conclusion: In this study, five machine learning models were utilized for the triage of patients presenting with cardiac-related symptoms. The results of the study indicated that these algorithms had a good ability to discriminate between patients with different triage levels. The Random Forest method performed slightly better than the other techniques. These techniques can be used to differentiate between low-risk and high-risk patients and to allocate resources to high-risk patients.

Background: Injuries within the emergency services populations are unfortunately common. Effective injury reduction programs need to be designed based on the profile of common injuries. Therefore, this study aimed to profile the injuries suffered within a state Emergency Services Agency which comprised Ambulance, Fire and Rescue, Rural Fire, volunteer emergency State Emergency Services (SES), and the Communication Centre (CC).

Methods: A retrospective cohort analysis was conducted on the entirety of an Australian State Emergency Service Agency injury database over a ten-year period (2012-2022). Records were extracted with details including (a) the total number; (b) the bodily site; (c) the nature; (d) and mechanism of injury. Total injuries were converted into injuries per 1000 full-time equivalent (FTE) years of service and incidence rate (IR) and ratios (IRR) were calculated per service.

Results: In total, there were 2,703 physical injuries reported by the agency, with Ambulance Services sustaining over half the injuries (57.5%). The most common body site, nature of injury, and mechanism was the lower back (27.9%), soft tissue (59.7%), and body stressing (45.5%), respectively. Based on 1000 FTE years, SES had the highest IR of 2054.2 followed by Rural Fire (IR = 1295.1), Communications (IR = 753.7), Ambulance Services (IR = 566.3), and Fire and Rescue (IR = 183.7). State Emergency Services sustained the highest IRR of 3.64 [3.13-4.22] when compared to Ambulance Services. The age group most likely to be injured were 45-49 years of age (17%), with males suffering the majority (65%) of injuries.

Conclusion: State Emergency Services present with injury rates above those of other emergency services personnel. These findings lay the groundwork for customised injury prevention strategies to promote better occupational safety across emergency service populations. Tailored injury prevention strategies may decrease subsequent time off due to injury.

Background: Understanding patient transfer patterns in medically underserved areas is crucial for healthcare system optimization. This study analyzed the differences between transferred and direct-visit patients at a tertiary emergency department in a medically underserved region.

Methods: We conducted a retrospective cohort study of 35,254 adult patients (≥ 18 years) visiting a tertiary care hospital Emergency Department in South Korea from January to December 2023. Patients were categorized into transferred (n = 7,069) or direct-visit (n = 28,185) groups. Demographics, clinical characteristics, resource utilization, and outcomes were compared using appropriate statistical tests. Logistic regression was used to analyze factors associated with clinical outcomes, including main diagnosis categories. Sensitivity analyses were performed excluding out-of-hospital cardiac arrest patients.

Results: Transferred patients were older (63.7 vs. 58.4 years, p < 0.001), more often male (59.4% vs. 49.7%, p < 0.001), and presented with higher acuity (KTAS 1-3: 87.5% vs. 31.9%, p < 0.001). Resource utilization was significantly higher in transferred patients, including hospitalization (55.1% vs. 23.4%, p < 0.001) and ICU admission (6.1% vs. 2.5%, p < 0.001). Paradoxically, in-hospital mortality was lower in transferred patients (0.47% vs. 0.75%, p = 0.012). After multivariable adjustment including diagnosis categories, transferred status remained associated with hospitalization (OR 3.42, 95% CI 3.22-3.64), ICU admission (OR 1.98, 95% CI 1.73-2.27), and 30-day revisits (OR 1.31, 95% CI 1.11-1.55), but inversely associated with mortality (OR 0.48, 95% CI 0.31-0.74). However, after excluding out-of-hospital cardiac arrest patients, the mortality difference became non-significant (p = 0.604).

Conclusions: Transferred patients in medically underserved areas present with higher acuity and consume more resources, but show lower mortality rates that may reflect effective triage and survivorship bias rather than superior care delivery. These findings highlight the importance of interfacility transfer networks while acknowledging selection effects in regional healthcare delivery.

Background: Sepsis poses a significant threat in emergency settings, necessitating tools for early and interpretable risk assessment. This study aimed to develop a robust explainable boosting machine (EBM) model, one of the explainable artificial intelligence (XAI) technologies, to construct a predictive model that balances high accuracy and clinical interpretability for use in emergency departments (EDs) and to examine candidate biomarkers.

Methods: The study identified a significant class imbalance problem in the sepsis distribution among 560 sepsis and 1012 non-sepsis patients. To address the imbalance issue, SMOTE-NC was applied in the training data. The data was divided into two parts, 80% training and 20% testing. To ensure the reliability of the models and to report unbiased results, this process was repeated 100 times and the average performance was reported. To determine the best model for sepsis prediction, five different models (AdaBoost, Gradient Boosting, CatBoost, LightGBM, and EBM) were trained, and their performances were evaluated. In the last stage, we presented local and global explanations of EBM.

Results: The EBM model achieved the highest success by reaching 79.1% F1-score, 80.9% sensitivity, and 84.8% AUC after resampling. In the global explanations, the variables with the highest weights in the model's decision process were identified as positive blood culture, oxygen saturation, and procalcitonin, respectively.

Conclusion: The EBM model accurately predicts sepsis risk based on clinically relevant biomarkers. The model's high performance and inherent transparency can foster trust among clinicians and facilitate its integration into emergency department workflows for real-time decision support.

Background: Patients presenting to the ED with headache are placed in a quiet, darkened room to minimize external stimuli; however, creating and maintaining such an environment can be challenging in the context of a busy ED setting. This study evaluated whether adding noise-reduction earmuffs and sleep eye masks to standard metoclopramide therapy improves pain relief in adult patients presenting with acute primary headache in the emergency department.

Methods: This single-center, open-label, randomized, controlled, parallel-group trial was conducted in the urban emergency department of a tertiary care hospital. Adult patients diagnosed with primary headache were randomized (1:1:1:1) to receive: Standard treatment alone (10 mg metoclopramide in 150 mL normal saline over 10 min). Standard treatment plus earmuffs (3 M Peltor Optime III, SNR 35 dB). Standard treatment plus a disposable sleep eye mask (> 99.9% light blockage). Standard treatment plus both earmuffs and an eye mask. Pain intensity was recorded on a 100 mm visual analogue scale (VAS) at baseline (VAS0), 30 min (VAS30), and 60 min (VAS60). Primary outcomes were the differences ΔVAS30 and ΔVAS60 versus baseline. Before the study, we calculated that 34 patients in each group would be sufficient to detect a 13 mm difference between the ΔVAS scores between the groups that would be clinically significant.

Results: Of the 194 screened patients, 140 were randomized (n = 35 per group) and analyzed by intention to treat. At 30 min, the combination group exhibited a mean ΔVAS30 reduction 23 mm greater than standard treatment alone (P < .05), exceeding the 13 mm minimal clinically important difference. No significant intergroup differences were observed in ΔVAS60, patient preference, or rescue analgesia rates. No adverse events were reported.

Conclusions: Earmuffs combined with sleep eye masks as an adjunct to metoclopramide significantly enhance early headache relief in the emergency department and represent a safe, low-cost complementary therapy. Participants were not blinded to the intervention due to practical constraints, and for the same reason, placebo control was not used.

Trial registration: The study protocol was established before starting and was registered at clinicaltrials.gov (Clinical Trials Identifier: NCT04178252, Date: 10.08.2019).