Artificial intelligence (AI) has shown promise in generating multiple-choice questions (MCQs) for medical education, yet the psychometric quality of such items remains underexplored. This study aimed to compare the psychometric properties of MCQs created by ChatGPT-4o and those written by emergency medicine clinicians. Eighteen emergency medicine residents completed a 100-item examination comprising 50 AI-generated and 50 clinician-authored questions across core emergency medicine topics. Each item was analyzed for difficulty (P_index), discrimination (D_index), and point-biserial correlation (PBCC). Items were also categorized based on standardized index classifications. ChatGPT-4o-generated questions exhibited a higher mean difficulty index (P_index: 0.76 ± 0.23) compared to those created by clinicians (0.65 ± 0.24; p = 0.02), indicating that the AI-generated items were generally easier. Participants achieved significantly higher scores on AI-generated items (76.8 ± 8.18) than on clinician-authored questions (67.3 ± 9.65; p = 0.003). The mean discrimination index did not differ significantly between AI-generated (0.172 ± 0.23) and clinician-generated items (0.196 ± 0.26; p = 0.634). Likewise, the mean point-biserial correlation coefficient (PBCC) was nearly identical between the two groups (AI: 0.23 ± 0.28; clinicians: 0.23 ± 0.25; p = 0.99), suggesting similar internal consistency. Categorical analysis revealed that 56% of AI-generated items were classified as “easy,” compared to 36% of clinician-designed items. Furthermore, based on PBCC values, 36% of AI-generated items and 24% of clinician items were identified as “problematic” (p = 0.015), indicating a higher rate of psychometric concerns among AI-generated questions. The findings suggest that AI-generated questions, while generally easier and associated with higher participant scores, may pose psychometric limitations, as evidenced by a greater proportion of items classified as problematic. Although the overall internal consistency and discrimination indices were comparable to clinician-authored items, careful quality control and validation are essential when integrating AI-generated content into assessment frameworks.

Background/Objectives: Trauma scoring systems are essential tools for predicting clinical outcomes in patients with multiple injuries. This study aimed to compare the performance of various anatomical and physiological scoring systems in predicting mortality among patients admitted to the emergency department following traffic accidents. Methods: In this prospective observational study, trauma patients presenting with traffic-related injuries were evaluated using seven scoring systems: ISS, NISS, AIS, GCS, RTS, TRISS, and APACHE II. Demographic data, clinical findings, and laboratory values were recorded. The prognostic performance of each score was assessed using ROC curve analysis, and diagnostic metrics including sensitivity, specificity, and likelihood ratios were calculated. Results: Among 554 patients included in the study, the overall mortality rate was 2%. The TRISS and GCS scores demonstrated the highest predictive performance, each with an AUC of 0.98, sensitivity of 100%, and specificity exceeding 93%. APACHE II followed closely with an AUC of 0.97, also achieving 100% sensitivity. NISS (AUC = 0.92) and ISS (AUC = 0.91) were effective anatomical scores, while RTS showed moderate predictive value (AUC = 0.90). Strong correlations were noted between ISS, NISS, and AIS (Rho > 0.85), while RTS was negatively correlated with these anatomical scores. All scoring systems showed statistically significant associations with mortality. Conclusions: TRISS, GCS, and APACHE II were the most effective trauma scoring systems in predicting mortality among emergency department patients. While complex models offer higher accuracy, simpler scores such as RTS and GCS remain valuable for rapid triage. The integration of both anatomical and physiological parameters may enhance early risk stratification and support timely decision-making in trauma care.

Background: This study aimed to evaluate the clinical and forensic characteristics of cases admitted to a high-volume tertiary emergency department, focusing on severity-based classification using treatability with simple medical intervention (SMI) and life-threatening status. Methods: We retrospectively analyzed 3014 forensic cases over one year. Patients were classified based on injury severity, anatomical region, and clinical outcomes. Documentation practices and report types were also reviewed. Results: Among all the cases, 60.4% were treatable with SMI, and 10.5% were identified as life threatening. Notably, all patients who died (1.3% mortality) were in the life-threatening group, and none of the SMI-treated patients died, underscoring the accuracy of early triage and alignment between documentation and outcomes. Road traffic accidents were the leading cause of life-threatening injury and hospitalization, while assault cases were predominantly minor and managed conservatively. Seasonal variation peaked in July, and sex-based differences revealed a higher SMI eligibility among female patients. Final forensic reports were more frequently issued in SMI cases, while preliminary reports were predominant in severe trauma. Conclusions: Severity-based classification using SMI and life-threatening categories offers valuable insight for clinical decision-making and forensic documentation. Integrating structured triage, anatomical injury mapping, and standardized report templates can enhance both patient safety and legal reliability.

Background: Gastrointestinal (GI) bleeding is a common and potentially life-threatening condition frequently encountered in emergency departments (EDs). The optimal strategy for red blood cell suspension (RBCS) transfusion, including timing and location, remains unclear. This study aimed to evaluate the impact of transfusion location (ED vs. inpatient units) on mortality and hospital stay in GI bleeding patients. Methods: A cross-sectional descriptive study was conducted in the ED of a tertiary care hospital. Patients admitted with GI bleeding between June 1, 2021, and June 1, 2023, who received RBCS transfusion were included. Data on demographics, laboratory parameters, transfusion details, and clinical outcomes were collected from the hospital information system. Logistic regression was used to identify mortality predictors. Results: A total of 244 patients were included. Patients transfused in the ED had a significantly shorter hospital stay compared to those transfused in inpatient units. However, mortality did not differ between groups. Logistic regression identified age, albumin, hemoglobin, creatinine, and hospital stay as independent mortality predictors, while transfusion location was not significant. Conclusions: Early RBCS transfusion in the ED may reduce hospital stay but does not significantly impact mortality. Identifying mortality-associated factors is crucial for optimizing patient management. Further prospective studies are needed to clarify the role of transfusion location in GI bleeding outcomes.