Scandinavian Journal of Trauma Resuscitation & Emergency Medicine最新文献

{"title":"AquaAI: development and internal validation of a Danish transformer-based model to identify drowning and aquatic incidents in prehospital medical records.","authors":"Caroline Kamuk Ostenfeldt, Niklas Breindahl, Claus Graff, Kristian Hay Kragholm, Helle Collatz Christensen, Stig Nikolaj Fasmer Blomberg","doi":"10.1186/s13049-026-01644-y","DOIUrl":"10.1186/s13049-026-01644-y","url":null,"abstract":"<p><strong>Background: </strong>Effective prevention of drowning and aquatic incidents requires timely and accurate surveillance supported by high-quality validated data. In Denmark, the use of the free-text fields in the Danish Prehospital Medical Record has proven effective in identifying potentially relevant cases for such surveillance. While these free-text fields contain rich contextual information, manual screening of all records is impractical. This study aimed to develop and internally validate a Danish natural language processing pipeline for identifying drowning and aquatic incidents (AquaAI) from routine prehospital records and prioritizing records for final manual validation by medical experts.</p><p><strong>Methods: </strong>This nationwide retrospective cohort study was conducted using Danish prehospital electronic medical records from 2016 to 2024. Medical records were first retrieved using the Danish Drowning Formula, an iteratively developed trigger-word search algorithm, and expert-labelled as Drowning, Aquatic incident, or Non-relevant. A Danish transformer-based language model was then fine-tuned for three-class sequence classification as the second-stage classifier in this workflow and embedded in a hybrid pipeline with rule-based safeguards. Model development used a temporal split with training data from 2016 to 2021 and validation data from 2022. Final performance was evaluated on a temporally separated hold-out dataset comprising records from 2023 to 2024. Primary outcomes were class-wise sensitivity and a binary Relevant (Drowning + Aquatic incident) versus Non-relevant analysis to quantify case finding and workload reduction.</p><p><strong>Results: </strong>The dataset comprised 40,876 medical records retrieved with the Danish Drowning Formula. On the hold-out dataset, AquaAI identified 91% of relevant cases and correctly filtered 89% of non-relevant records from manual review. Overall, this reduced the number of records requiring manual review by 77.5%. In the three-class analysis, sensitivity was 84% for Drowning, 83% for Aquatic incident, and 89% for Non-relevant. Only one drowning case (0.3%) was classified as non-relevant.</p><p><strong>Conclusions: </strong>A hybrid transformer-based pipeline operating as a second-stage classifier within a two-step retrieval workflow can identify drowning and aquatic incidents from prehospital free-text narratives with high case finding and substantial reduction of manual review. This approach may support national surveillance and targeted prevention initiatives.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13479569/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148213099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stroke detection in medical emergency calls: a retrospective analysis and exploratory evaluation of an AI decision support model.","authors":"Emil Iversen, Hege Ihle-Hansen, Kari Krizak Halle, Alexander Selvikvåg Lundervold, Lars Myrmel, Anders Strand Vestbø, Annette Fromm, Cosimo Damiano Persia, Christian Autenried, Guttorm Brattebø","doi":"10.1186/s13049-026-01645-x","DOIUrl":"10.1186/s13049-026-01645-x","url":null,"abstract":"<p><strong>Background: </strong>Timely identification of acute stroke during medical emergency calls is critical for optimizing patient outcomes. We aimed to (i) characterize the prehospital trajectories of patients with stroke and (ii) explore the ability of an artificial intelligence (AI) model developed within the Artificial Intelligence Support in Medical Emergency Calls project to support decision-making at Emergency Medical Communication Centres (EMCCs).</p><p><strong>Methods: </strong>We conducted a retrospective analysis using data from 1,164 patients diagnosed with stroke from an EMCC (2018-2019), focusing on those primarily assessed by EMCC operators. We categorized patients into optimal and nonoptimal trajectory groups and applied logistic regression to explore the factors associated with an optimal trajectory. An AI model was trained using a dataset of 2,980 emergency calls (2019 and 2022), integrating transcribed audio logs and structured clinical data. The model was evaluated on the full dataset and subgroups that were incorrectly assessed by the EMCC.</p><p><strong>Results: </strong>Aphasia/dysarthria was the only factor associated with optimal trajectories. The AI model achieved a sensitivity of 81.0%, specificity of 79.6%, and F1-score of 64.7%. The subgroup analysis included 14 false-negative and 41 false-positive cases. The model correctly predicted stroke in 9 of 14 false-negative cases and ruled out stroke in 9 of 41 false-positive cases.</p><p><strong>Conclusions: </strong>This explorative evaluation indicates that a multimodal AI pipeline combining audio and structured clinical data may show potential to support decision making in medical emergency calls. Internal results are promising, but studies on larger and external datasets are needed to validate these findings.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13487976/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148206769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Best practices on blood and blood products for a prehospital hemorrhage protocol: consensus from the 2025 Canadian prehospital transfusion summit.","authors":"Michael Peddle, Adam Greene, Jan Trojanowski, Andrew W Shih, Eddie Chang, Susan Nahirniak, Dallas Pearson, Oksana Prokopchuk-Gauk, Doug Martin, John Froh, Charles Musuka, Cindy Seidl, Yulia Lin, Scott MacDonald, Lindsay Richards, Michael Farrell, Damien Miller, Johnathan Mack, Luis Da Luz, Ian R Drennan, Kanwal Singh, Darshan Shingala, Brodie Nolan","doi":"10.1186/s13049-026-01630-4","DOIUrl":"10.1186/s13049-026-01630-4","url":null,"abstract":"<p><strong>Background: </strong>Blood transfusion is increasingly recognized as a critical component of prehospital resuscitation. However, Canada's vast geography, environmental extremes, and regional variation in health systems present unique challenges to standardized implementation of prehospital hemorrhage protocols (PHPs). The 2025 Canadian Prehospital Transfusion Summit (CPTS), convened by the Canadian Prehospital and Transport Transfusion (CAN-PATT) network, sought to develop consensus-driven best practices for prehospital transfusion strategies across diverse transport environments.</p><p><strong>Methods: </strong>A structured roundtable consensus process was held at CPTS 2025, bringing together over 50 subject matter experts from eight provinces, six different services/systems and a variety of operational roles to discuss the past, present and future direction of out-of-hospital blood transfusion in Canada. Key domains included: activation criteria, prioritization of blood components and adjuncts, product logistics, and rural/remote applicability. Recommendations were synthesized through iterative discussion, evidence review, and thematic analysis. Each recommendation was then voted upon until consensus was reached through all participants.</p><p><strong>Results: </strong>Consensus was reached on core components of a national PHP. Red blood cells and tranexamic acid were identified as foundational therapies. Freeze-dried plasma and whole blood offer logistical and clinical advantages, particularly in rural and remote areas, while fibrinogen concentrate and prothrombin complex concentrate are valuable adjuncts when plasma is unavailable. Calcium supplementation and active hypothermia prevention were emphasized as essential. Barriers to implementation included short product shelf-life, storage infrastructure, and regulatory constraints. The group recommended development of a standardized Canadian prehospital transfusion dataset and national data registry.</p><p><strong>Conclusions: </strong>The panel produced 12 consensus statements based on the available evidence and expert opinion including activation criteria, prioritization of blood products and adjuncts, product logistics, and rural/remote applicability. These consensus statements establish the basis for safe and effective PHP practices in Canada.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":"34 Suppl 1","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13235005/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148158736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transfusion and mortality after major trauma in patients on direct oral anticoagulants: a multicentre cohort study.","authors":"Johan Henrik Wersäll, Henrik Hedelin, Alexandra Vidlund, Elin Ekdahl, Jacob Kjellberg, Peter U Larsson, Leo Normark, Dominika Högberg","doi":"10.1186/s13049-026-01640-2","DOIUrl":"10.1186/s13049-026-01640-2","url":null,"abstract":"<p><strong>Background: </strong>Major trauma in patients with direct oral anticoagulants (DOACs) is increasing as treatment with DOACs is becoming more prevalent. DOACs add complexity to trauma due to increased risk of bleeding. The primary aim of this study was to compare the likelihood of packed red blood cell (PRBC) transfusion and PRBC volumes in patients with DOACs to non-anticoagulated controls after major trauma. Secondary aims were to compare mortality rates between these groups.</p><p><strong>Methods: </strong>This was a multicentre retrospective cohort study with four participating hospitals. Inclusion criteria were age ≥ 18 years and admission for trauma with New Injury Severity Score (NISS) > 15 from Jan 1, 2019, to Dec 31, 2023. The Swedish Trauma Registry and medical records were used to identify patients and to retrieve data. For primary weighted analyses, propensity scores for DOACs were estimated using sex, age, injury mechanism, ASA class, antiplatelet treatment, and NISS. Non-weighted adjusted regression models were used for complementary analyses.</p><p><strong>Results: </strong>A total of 1817 patients were admitted for major trauma during the study period, of which 756 patients were included in the analytic cohort. PRBC was administered to 30.8% of patients with DOACs and to 20.7% of controls. PRBC likelihood was higher among patients with DOACs (OR 1.73, 95% CI 1.03-2.91). The number of PRBC units among patients who received transfusions was not significantly higher for DOACs (GMR 1.11, 95% CI 0.79-1.56). Mortality rates within 24 h and 30 days did not differ significantly in primary analyses (OR 1.82, 95% CI 0.78-4.23 and OR 1.34, 95% CI 0.80-2.24, respectively).</p><p><strong>Conclusions: </strong>Patients with DOACs were more likely to receive PRBC transfusion, but PRBC volumes were not significantly different among transfused patients. Mortality did not differ significantly in the primary weighted analyses.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":"34 1","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13222495/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148101998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Challenges faced by prehospital emergency physicians providing emergency care to pediatric patients : Defining entrustable professional activities by identifying self-perceived competency strengths and deficits in prehospital emergency medicine.","authors":"Anne Kamphausen, Julia Steg, Tobias Bexten, Jens Kubitz, Christian Engelen","doi":"10.1186/s13049-026-01637-x","DOIUrl":"10.1186/s13049-026-01637-x","url":null,"abstract":"<p><strong>Background: </strong>Pediatric emergencies constitute a small proportion of prehospital cases yet pose unique diagnostic and therapeutic challenges for emergency physicians due to age-specific physiological differences, limited exposure, and infrequent opportunities for skill acquisition. Competency-based medical education frameworks and entrustable professional activities (EPAs) may offer suitable concepts to structure pediatric emergency training.</p><p><strong>Methods: </strong>This mixed-methods study combined a retrospective analysis of 9,409 prehospital emergency physician protocols (2012-2021) with a survey assessing self-perceived competence and training needs among emergency physicians. Pediatric emergency characteristics, clinical management, and utilized interventions were analyzed. Survey data captured demographic characteristics, perceived competency across defined scenarios, sources of knowledge acquisition, and perceived gaps in pediatric emergency preparedness.</p><p><strong>Results: </strong>Among all prehospital missions, 606 pediatric emergencies (6.45%) were identified, predominantly involving adolescents with any condition, and any age with trauma and neurological conditions. Invasive interventions such as intubation, cardiopulmonary resuscitation, or intraosseous access were rarely needed. Survey responses (n = 437) revealed heterogeneous self-perceived competencies: seizures, anaphylaxis, and hypoglycemia were associated with high confidence, whereas neonatal care, pediatric polytrauma, and cardiopulmonary resuscitation demonstrated markedly lower confidence levels. Knowledge was primarily acquired during the initial emergency physician qualification training period, as well as through subsequent clinical exposure in routine practice, with limited engagement in ongoing pediatric-specific training. More than 70% of participants advocated for structured, mandatory, and simulation-based pediatric training.</p><p><strong>Conclusion: </strong>Pediatric emergencies in the prehospital setting are uncommon yet challenging, revealing gaps between training expectations, clinical exposure, and perceived competencies. Strengthened pediatric emergency training-particularly simulation-based practice and EPA-based curricular elements-may enhance preparedness, standardize entrustment decisions, and improve patient safety. Future work should operationalize nested pediatric EPAs and evaluate competency outcomes prospectively.</p><p><strong>Trial number: </strong>IRB-2023-06.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":"34 1","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13222496/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148101989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identifying research priorities for post-collision care in the United Kingdom: outcomes and methodological adaptations from the final prioritisation workshop.","authors":"Tim Nutbeam, Laura Cottey, Kerry Dungay, Lauren R Rodgers, Emily Foote, Caroline Leech, Claire E Baker, Elizabeth Box, Louise Johnson, Brian Lee, Marilyn MacQueen, Rob Fenwick, Nigel Lang, Ianto Guy, Ian Marritt, Ian Dunbar, Ed B G Barnard","doi":"10.1186/s13049-026-01628-y","DOIUrl":"10.1186/s13049-026-01628-y","url":null,"abstract":"<p><strong>Background: </strong>Road traffic injury remains a leading cause of death and serious injury in the United Kingdom, yet the post-collision phase of care has received comparatively little research attention. The Road Injury Chain of Survival framework identifies five interdependent links where coordinated action can improve outcomes. To address evidence gaps across this pathway, we conducted the first UK Priority Setting Partnership focused specifically on post-collision care, following James Lind Alliance methodology.</p><p><strong>Methods: </strong>A national open survey collected research uncertainties from patients, carers, bystanders, clinicians, emergency responders and policy stakeholders between July and August 2025. This was supplemented by a targeted literature review identifying research uncertainties from clinical guidelines and systematic reviews. All submissions underwent evidence checking using the BestBETs methodology. The Steering Group, comprising patients with lived experience, emergency service representatives, clinicians, and researchers, conducted interim prioritisation to produce a shortlist. A final prioritisation workshop was held in November 2025, using nominal group technique across three facilitated small-group rounds followed by plenary consensus. Methodological adaptations enabled structured remote participation for contributors unable to attend due to injury-related barriers.</p><p><strong>Results: </strong>In total, 179 survey submissions and 73 literature-derived questions were consolidated into 57 indicative uncertainties. Following evidence checking and interim prioritisation, 23 questions proceeded to the final workshop. Thirty-nine participants reached consensus on ten priorities. These emphasised preventable deaths and critical intervention windows, recognition of occult life-threatening injuries, multi-agency coordination, technology-assisted bystander care, automatic crash notification, first aid training effectiveness, emergency call-handler decision support, inequalities in injury patterns and care, treatment of entrapped casualties, and patient-centred recovery outcomes.</p><p><strong>Conclusion: </strong>This national Priority Setting Partnership provides an inclusive, stakeholder-driven foundation for research commissioning and policy development in post-collision care. The priorities highlight significant evidence gaps in the earliest phases of the care pathway and underscore the importance of addressing inequalities and aligning research with outcomes meaningful to survivors.</p><p><strong>Protocol: </strong>Nutbeam, T., Leech, C., Baker, CE. et al Identifying research priorities for post-collision care in the United Kingdom: protocol for a road injury priority setting partnership. Scand J Trauma Resusc Emerg Med 33, 203 (2025). https://doi.org/10.1186/s13049-025-01513-0 .</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13397751/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148044569","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of the trauma criteria from the Norwegian national trauma plan: an observational study from a regional trauma centre.","authors":"Guro Bjørke Buljo, Kenneth Thorsen, Thomas Geisner","doi":"10.1186/s13049-026-01634-0","DOIUrl":"10.1186/s13049-026-01634-0","url":null,"abstract":"<p><strong>Background: </strong>The national trauma plan defines common Norwegian criteria dictating which patients are met by a trauma team. Precise criteria are important to avoid undertriage and overtriage. Previous studies have indicated that both overtriage and undertriage fail to meet the goals of the Norwegian Trauma Plan, and the aim of the study was to investigate the performance of the new national criteria.</p><p><strong>Methods: </strong>This is a retrospective cohort study of patients included in the trauma registry at Haukeland University Hospital between 2017 and 2021. Injury Severity Score (ISS) was extracted from the registry. Prehospital records were manually reviewed to determine which trauma criteria were met. The aim was to analyse how well the national trauma criteria predicted ISS > 15, and if national goals of overtriage and undertriage were met.</p><p><strong>Results: </strong>A total of 2508 patients were eligible for analysis, whereas 384 had an ISS > 15, with a total overtriage of 81.1%. Forty-four patients had an ISS > 15 without meeting the criteria, resulting in an undertriage of 11.5%. If hypothetically 4 Patient factors was analysed as an individual criteria group, undertriage would have been 2.9%. Patients who did not meet any criteria had an overtriage of 97.6%. Mechanism of injury (MOI) predicted ISS > 15 in nine of 333 patients, resulting in an overtriage of 97.3% for this criteria group. Hypothetical removal of the MOI criteria except 3.6 Fall would in this data set eliminate 255 Trauma team activations (TTA) among patients with ISS < 15.</p><p><strong>Conclusions: </strong>Overtriage was high overall and across all criteria groups. Patients who triggered a TTA despite not meeting any criteria also demonstrated substantial overtriage. To optimize resource use, criteria with low positive predictive value could be assigned fewer resources or revised to reduce unnecessary activations. Our findings support re-evaluation of specific criteria, particularly those related to mechanism of injury.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13419021/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148044575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Global inventory and audit of road injury data systems: a descriptive study using the Road Injury Chain of Survival Framework.","authors":"Tim Nutbeam, Emily Foote, Lauren R Rodgers, Rob Fenwick, Willem Stassen","doi":"10.1186/s13049-026-01633-1","DOIUrl":"10.1186/s13049-026-01633-1","url":null,"abstract":"<p><strong>Background: </strong>Road injury causes an estimated 1.19 million deaths annually and tens of millions of non-fatal injuries worldwide. Despite this burden, post-collision data remain fragmented across police, transport, and health sectors, impeding system-level improvement. The recently proposed Road Injury Chain of Survival framework provides a structure for mapping data needs across five sequential links: Early Recognition, Early Rescue, Early Initial Care, Early Transport, and Early Hospital Care & Rehabilitation. This study aims to identify and describe national and major regional road injury datasets worldwide and to assess their coverage against a novel RICS-derived sub-indicator framework.</p><p><strong>Methods: </strong>A structured, descriptive global audit was undertaken to identify and characterise national and major regional road injury datasets. Searches included government and institutional portals, international repositories, and peer-reviewed literature. Eligible datasets provided national or nationally aggregated information on road traffic collisions, injuries, or trauma care. Each dataset was assessed against 20 sub-indicators (four per chain link) using a three-point coding system: 1 (present), 0.5 (partial/proxy), 0 (absent), or U (unverified).</p><p><strong>Results: </strong>A total of 349 national and major regional road injury data systems were identified worldwide, representing 357 datasets after disaggregation of linked systems. Police crash databases were the most common data source across all income groups and were near-universal in low-income countries. Mortality-focused datasets predominated in lower-income countries while higher-income countries demonstrated greater diversity, including emergency medical services, fire and rescue, and trauma registry data. Formal cross-sector data linkage was rare, and most datasets were available only in aggregated form. Mapping to the Road Injury Chain of Survival showed that data were most frequently available for Early Recognition, with substantially less coverage for the remaining links in the chain.</p><p><strong>Conclusions: </strong>This global audit of road injury datasets reveals extensive fragmentation across the post-collision care continuum. Few systems link crash, rescue, prehospital, and hospital data, severely limiting opportunities for quality improvement and outcome evaluation. Development of harmonised, interoperable datasets aligned with the Road Injury Chain of Survival is essential to enable benchmarking, guide investment, and improve outcomes from road injury worldwide.</p><p><strong>Pre-registered: </strong>Open Science Framework: https://doi.org/10.17605/OSF.IO/VXE6S.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13397626/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148044537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feasibility and challenges of large language model (LLM)-generated neonatal resuscitation simulations: a multicenter exploratory study.","authors":"Chenguang Xu, Ming Zhou, Dianna Wang, Qianshen Zhang, Jiang-Qin Liu, Genevieve Fung, Mabel Wong, Yin Xue, Georg M Schmölzer, Po-Yin Cheung","doi":"10.1186/s13049-026-01632-2","DOIUrl":"10.1186/s13049-026-01632-2","url":null,"abstract":"<p><strong>Background: </strong>Simulation-based training (SBT) in neonatal resuscitation has positive impact on educational and neonatal outcomes. However, the implementation of well-designed SBT imposes multifaceted demands on instructors. Large language models (LLMs) might have potential in dynamically generating contextual resuscitation scenarios. But gaps exist regarding the feasibility and challenges of LLM-generated simulation scenarios in neonatal resuscitation.</p><p><strong>Methods: </strong>This is a prospective, multicenter pilot study evaluating the feasibility and challenges of LLM-generated simulation scenarios. Four scenarios including extremely premature infant, placental abruption, born before arrival in hospital and meconium-stained amniotic fluid were generated by ChatGPT-4o and DeepSeek-R1. Four equivalent scenarios were extracted from Neonatal Resuscitation Program® (NRP®) and RETAIN (a serious game platform). Totally these 16 scenarios were written on standardized templates, coded and randomized. Nine independent instructors from 5 centers, who were blinded to the source of the scenarios (ChatGPT, DeepSeek, NRP® and RETAIN), evaluated the performance of scenarios using modified Jeffries Simulation Design Scale (JSDS). AI hallucination and qualitative evaluation were also compared among four groups.</p><p><strong>Results: </strong>When compared with NRP® scenarios, ChatGPT-generated scenarios demonstrated comparable median scores without statistically significant difference in overall evaluation, whereas DeepSeek and RETAIN had lower scores in overall evaluation, problem-solving efficacy and scenario fidelity. DeepSeek exhibited inferior performance in providing appropriate information. In debriefing design, ChatGPT achieved higher scores than NRP® (effect size r = 0.25, P = 0.02). There was no statistical difference in AI hallucination between the two LLMs. ChatGPT demonstrated strengths in establishing clear objectives and providing structured debriefing frameworks but exhibited deficiencies in the consistent provision of dynamic vital signs. DeepSeek manifested violations in NRP® algorithm.</p><p><strong>Conclusions: </strong>ChatGPT-generated simulation scenarios might be feasible in facilitating SBT when supervised by NRP® instructors. NRP® deviations and gaps remain in LLM-generated scenarios, which necessitate objective evaluation prior to implementation. Further research assessing educational outcomes and feedback from target learners are essential for the appropriate integration of LLM-generated simulation into SBT.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13386711/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148018246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Epidemiology and severity of mountain rescue missions in Poland: a retrospective study of 9,477 patients rescued by the Tatra Mountain Rescue Service.","authors":"Łukasz Migiel, Sylweriusz Kosiński, Marcin Waligóra, Jan Krzysztof, Maciej Mikiewicz, Andrzej Górka, Michał Jaros, Wojciech Kącki, Matthieu Pasquier","doi":"10.1186/s13049-026-01631-3","DOIUrl":"10.1186/s13049-026-01631-3","url":null,"abstract":"<p><strong>Background: </strong>Mountain activities have become increasingly popular in recent years, leading to a growing number of mountain rescue missions. Despite this trend, detailed epidemiological analyses of mountain accidents in the Polish Tatra Mountains are lacking. The aim of this study was to describe the epidemiology of mountain rescue missions in the Polish Tatra Mountains, with particular emphasis on the severity of the case mix of injuries and illnesses and their implications for rescue response.</p><p><strong>Methods: </strong>A retrospective analysis of prospectively collected data from the Tatra Mountain Rescue Service (TOPR) covering the period of 2015-2024 was conducted. Demographic, epidemiological, and operational data were extracted from the electronic rescue registry. The severity of the patient's condition was classified using the National Advisory Committee for Aeronautics (NACA) score. The population at risk was estimated on the basis of the number of ticket entries to Tatra National Park during the study period. Seasonal differences in rescue incidence and severity were analysed, and absolute risk increase (ARI) and number needed to harm (NNH) were calculated.</p><p><strong>Results: </strong>A total of 9,477 patients were included in the final analysis. The median age of the rescued individuals was 36 years (IQR 25-48); 48% (n = 4,539) were women, and 10% (n = 998) were minors. The median NACA score was 2 (IQR, 0-3), while 38% of the patients were classified as having a NACA score ≥ 3. Deaths occurred in 2% of the rescue missions. The median duration of rescue missions was 1.75 h (1.25-2.75). For every 100,000 visitors to the Tatra National Park, approximately 24 rescue missions were performed, including one involving severe injury or illness (NACA ≥ 4). Severe cases were significantly more frequent during the winter season (6.3% vs. 4.7%; p = 0.001). Helicopters were used in approximately one-third of the missions, and their use was correlated with higher NACA scores.</p><p><strong>Conclusion: </strong>Mountain rescue missions in the Polish Tatra Mountains were associated with seasonal patterns of mountain activities and the number of visitors to Tatra National Park. Systematic assessment of a patient's severity using the NACA score may support decision-making and resource allocation in mountain rescue. Epidemiological analyses based on comprehensive rescue databases may contribute to improved planning, training, and organization of medical search and rescue systems in mountain environments.</p>","PeriodicalId":49292,"journal":{"name":"Scandinavian Journal of Trauma Resuscitation & Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":3.6,"publicationDate":"2026-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13386807/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148007102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}