The BCD Triage Sieve outperforms all existing major incident triage tools: Comparative analysis using the UK national trauma registry population

Mapping the processes and information flows of a prehospital emergency care system in Rwanda: a process mapping exercise

OBJECTIVE

A vital component of a prehospital emergency care system is getting an injured patient to the right hospital at the right time. Process and information flow mapping are recognised methods to show where efficiencies can be made. We aimed to understand the process and information flows used by the prehospital emergency service in transporting community emergencies in Rwanda in order to identify areas for improvement.

DESIGN

Two facilitated process/information mapping workshops were conducted. Process maps were produced in real time during discussions and shared with participants for their agreement. They were further validated by field observations.

SETTING

The study took place in two prehospital care settings serving predominantly rural and predominantly urban patients.

PARTICIPANTS

24 healthcare professionals from various cadres. Field observations were done on 49 emergencies across both sites.

RESULTS

Two maps were produced, and four main process stages were described: (1) call triage by the dispatch/call centre team, (2) scene triage by the ambulance team, (3) patient monitoring by the ambulance team on the way to the health facility and (4) handover process at the health facility. The first key finding was that the rural site had multiple points of entry into the system for emergency patients, whereas the urban system had one point of entry (the national emergency number); processes were otherwise similar between sites. The second was that although large amounts of information were collected to inform decision-making about which health facility to transfer patients to, participants found it challenging to articulate the intellectual process by which they used this to make decisions; guidelines were not used for decision-making.

DISCUSSION

We have identified several areas of the prehospital care processes where there can be efficiencies. To make efficiencies in the decision-making process and produce a standard approach for all patients will require protocolising care pathways.

Triage in major incidents: development and external validation of novel machine learning-derived primary and secondary triage tools

BACKGROUND

Major incidents (MIs) are an important cause of death and disability. Triage tools are crucial to identifying priority 1 (P1) patients-those needing time-critical, life-saving interventions. Existing expert opinion-derived tools have limited evidence supporting their use. This study employs machine learning (ML) to develop and validate models for novel primary and secondary triage tools.

METHODS

Adults (16+ years) from the UK Trauma Audit and Research Network (TARN) registry (January 2008-December 2017) served as surrogates for MI victims, with P1 patients identified using predefined criteria. The TARN database was split chronologically into model training and testing (70:30) datasets. Input variables included physiological parameters, age, mechanism and anatomical location of injury. Random forest, extreme gradient boosted tree, logistic regression and decision tree models were trained to predict P1 status, and compared with existing tools (Battlefield Casualty Drills (BCD) Triage Sieve, CareFlight, Modified Physiological Triage Tool, MPTT-24, MSTART, National Ambulance Resilience Unit Triage Sieve and RAMP). Primary and secondary candidate models were selected; the latter was externally validated on patients from the UK military's Joint Theatre Trauma Registry (JTTR).

RESULTS

Models were internally tested in 57 979 TARN patients. The best existing tool was the BCD Triage Sieve (sensitivity 68.2%, area under the receiver operating curve (AUC) 0.688). Inability to breathe spontaneously, presence of chest injury and mental status were most predictive of P1 status. A decision tree model including these three variables exhibited the best test characteristics (sensitivity 73.0%, AUC 0.782), forming the candidate primary tool. The proposed secondary tool (sensitivity 77.9%, AUC 0.817), applicable via a portable device, includes a fourth variable (injury mechanism). This performed favourably on external validation (sensitivity of 97.6%, AUC 0.778) in 5956 JTTR patients.

CONCLUSION

Novel triage tools developed using ML outperform existing tools in a nationally representative trauma population. The proposed primary tool requires external validation prior to consideration for practical use. The secondary tool demonstrates good external validity and may be used to support decision-making by healthcare workers responding to MIs.

Injury modelling for strategic planning in protecting the national infrastructure from terrorist explosive events

Terrorist events in the form of explosive devices have occurred and remain a threat currently to the population and the infrastructure of many nations worldwide. Injuries occur from a combination of a blast wave, energised fragments, blunt trauma and burns. The relative preponderance of each injury mechanism is dependent on the type of device, distance to targets, population density and the surrounding environment, such as an enclosed space, to name but a few. One method of primary prevention of such injuries is by modification of the environment in which the explosion occurs, such as modifying population density and the design of enclosed spaces. The Human Injury Predictor (HIP) tool is a computational model which was developed to predict the pattern of injuries following an explosion with the goal to inform national injury prevention strategies from terrorist attacks. HIP currently uses algorithms to predict the effects from primary and secondary blast and allows the geometry of buildings to be incorporated. It has been validated using clinical data from the '7/7' terrorist attacks in London and the 2017 Manchester Arena terrorist event. Although the tool can be used readily, it will benefit from further development to refine injury representation, validate injury scoring and enable the prediction of triage states. The tool can assist both in the design of future buildings and methods of transport, as well as the situation of critical emergency services required in the response following a terrorist explosive event. The aim of this paper is to describe the HIP tool in its current version and provide a roadmap for optimising its utility in the future for the protection of national infrastructure and the population.

The Field's mass shooting: emergency medical services response

BACKGROUND

Major incidents (MI) happen infrequently in Scandinavia and mass shootings are even less frequently occurring. Case reports and research are called for, as literature is scarce. On 3rd July 2022, a mass shooting took place at the shopping mall Field's in Copenhagen, Denmark. Three people were killed and seven injured by a gunman, firing a rifle inside the mall. A further 21 people suffered minor injuries during the evacuation of the mall. In this case report, we describe the emergency medical services (EMS) incident response and evaluate the EMS´ adherence to the MI management guidelines to identify possible areas of improvement.

CASE PRESENTATION

Forty-eight EMS units including five Tactical Emergency Medical Service teams were dispatched to the incident. Four critically injured patients were taken to two trauma hospitals. The deceased patients were declared dead at the scene and remained there for the sake of the investigation. A total of 24 patients with less severe and minor injuries were treated at four different hospitals in connection with the attack. The ambulance resources were inherently limited in the initial phase of the MI, mandating improvisation in medical incident command. Though challenged, Command and Control, Safety, Communication, Assessment, Triage, Treatment, Transport (CSCATTT) principles were followed.

CONCLUSIONS

The EMS response generally adhered to national guidelines for MI. The activation of EMS and the hospital preparedness program was relevant. Important findings were communication shortcomings; inherent lack of readily available ambulance resources in the initial critical phase; uncertainty regarding the number of perpetrators; uncertainty regarding number of casualties and social media rumors that unnecessarily hampered and prolonged the response. The incident command had to use non-standard measures to mitigate potential challenges.

An assessment of mass casualty triage systems using the Alberta trauma registry

OBJECTIVE

Triage is the process of identifying patients with both the greatest clinical need and the greatest likelihood of benefit in the setting of limited clinical resources. The primary objective of this study was to assess the ability of formal mass casualty incident triage tools to identify patients requiring urgent lifesaving interventions.

METHODS

Data from the Alberta Trauma Registry (ATR) was used to assess seven triage tools: START, JumpSTART, SALT, RAMP, MPTT, BCD and MITT. Clinical data captured in the ATR was used to determine which triage category each of the seven tools would have applied to each patient. These categorizations were compared to a reference standard definition based on the patients' need for specific urgent lifesaving interventions.

RESULTS

Of the 9448 records that were captured 8652 were included in our analysis. The most sensitive triage tool was MPTT, which demonstrated a sensitivity of 0.76 (0.75, 0.78). Four of the seven triage tools evaluated had sensitivities below 0.45. JumpSTART had the lowest sensitivity and the highest under-triage rate for pediatric patients. All the triage tools evaluated had a moderate to high positive predictive value (> 0.67) for patients who had experienced penetrating trauma.

CONCLUSIONS

There was a wide range in the sensitivity of triage tools to identify patients requiring urgent lifesaving interventions. MPTT, BCD and MITT were the most sensitive triage tools assessed. All of the triage tools assessed should be employed with caution during mass casualty incidents as they may fail to identify a large proportion of patients requiring urgent lifesaving interventions.

Development of an Anticipatory Triage-Ranking Algorithm Using Dynamic Simulation of the Expected Time Course of Patients With Trauma: Modeling and Simulation Study

BACKGROUND

In cases of terrorism, disasters, or mass casualty incidents, far-reaching life-and-death decisions about prioritizing patients are currently made using triage algorithms that focus solely on the patient's current health status rather than their prognosis, thus leaving a fatal gap of patients who are under- or overtriaged.

OBJECTIVE

The aim of this proof-of-concept study is to demonstrate a novel approach for triage that no longer classifies patients into triage categories but ranks their urgency according to the anticipated survival time without intervention. Using this approach, we aim to improve the prioritization of casualties by respecting individual injury patterns and vital signs, survival likelihoods, and the availability of rescue resources.

METHODS

We designed a mathematical model that allows dynamic simulation of the time course of a patient's vital parameters, depending on individual baseline vital signs and injury severity. The 2 variables were integrated using the well-established Revised Trauma Score (RTS) and the New Injury Severity Score (NISS). An artificial patient database of unique patients with trauma (N=82,277) was then generated and used for analysis of the time course modeling and triage classification. Comparative performance analysis of different triage algorithms was performed. In addition, we applied a sophisticated, state-of-the-art clustering method using the Gower distance to visualize patient cohorts at risk for mistriage.

RESULTS

The proposed triage algorithm realistically modeled the time course of a patient's life, depending on injury severity and current vital parameters. Different casualties were ranked by their anticipated time course, reflecting their priority for treatment. Regarding the identification of patients at risk for mistriage, the model outperformed the Simple Triage And Rapid Treatment's triage algorithm but also exclusive stratification by the RTS or the NISS. Multidimensional analysis separated patients with similar patterns of injuries and vital parameters into clusters with different triage classifications. In this large-scale analysis, our algorithm confirmed the previously mentioned conclusions during simulation and descriptive analysis and underlined the significance of this novel approach to triage.

CONCLUSIONS

The findings of this study suggest the feasibility and relevance of our model, which is unique in terms of its ranking system, prognosis outline, and time course anticipation. The proposed triage-ranking algorithm could offer an innovative triage method with a wide range of applications in prehospital, disaster, and emergency medicine, as well as simulation and research.

Review: Emerging Eye-Based Diagnostic Technologies for Traumatic Brain Injury

The study of ocular manifestations of neurodegenerative disorders, Oculomics, is a growing field of investigation for early diagnostics, enabling structural and chemical biomarkers to be monitored overtime to predict prognosis. Traumatic brain injury (TBI) triggers a cascade of events harmful to the brain, which can lead to neurodegeneration. TBI, termed the "silent epidemic" is becoming a leading cause of death and disability worldwide. There is currently no effective diagnostic tool for TBI, and yet, early-intervention is known to considerably shorten hospital stays, improve outcomes, fasten neurological recovery and lower mortality rates, highlighting the unmet need for techniques capable of rapid and accurate point-of-care diagnostics, implemented in the earliest stages. This review focuses on the latest advances in the main neuropathophysiological responses and the achievements and shortfalls of TBI diagnostic methods. Validated and emerging TBI-indicative biomarkers are outlined and linked to ocular neuro-disorders. Methods detecting structural and chemical ocular responses to TBI are categorised along with prospective chemical and physical sensing techniques. Particular attention is drawn to the potential of Raman spectroscopy as a non-invasive sensing of neurological molecular signatures in the ocular projections of the brain, laying the platform for the first tangible path towards alternative point-of-care diagnostic technologies for TBI.

Exploring the effectiveness of artificial intelligence, machine learning and deep learning in trauma triage: A systematic review and meta-analysis

Background

The development of artificial intelligence (AI), machine learning (ML) and deep learning (DL) has advanced rapidly in the medical field, notably in trauma medicine. We aimed to systematically appraise the efficacy of AI, ML and DL models for predicting outcomes in trauma triage compared to conventional triage tools.

Methods

We searched PubMed, MEDLINE, ProQuest, Embase and reference lists for studies published from 1 January 2010 to 9 June 2022. We included studies which analysed the use of AI, ML and DL models for trauma triage in human subjects. Reviews and AI/ML/DL models used for other purposes such as teaching, or diagnosis were excluded. Data was extracted on AI/ML/DL model type, comparison tools, primary outcomes and secondary outcomes. We performed meta-analysis on studies reporting our main outcomes of mortality, hospitalisation and critical care admission.

Results

One hundred and fourteen studies were identified in our search, of which 14 studies were included in the systematic review and 10 were included in the meta-analysis. All studies performed external validation. The best-performing AI/ML/DL models outperformed conventional trauma triage tools for all outcomes in all studies except two. For mortality, the mean area under the receiver operating characteristic (AUROC) score difference between AI/ML/DL models and conventional trauma triage was 0.09, 95% CI (0.02, 0.15), favouring AI/ML/DL models (p = 0.008). The mean AUROC score difference for hospitalisation was 0.11, 95% CI (0.10, 0.13), favouring AI/ML/DL models (p = 0.0001). For critical care admission, the mean AUROC score difference was 0.09, 95% CI (0.08, 0.10) favouring AI/ML/DL models (p = 0.00001).

Conclusions

This review demonstrates that the predictive ability of AI/ML/DL models is significantly better than conventional trauma triage tools for outcomes of mortality, hospitalisation and critical care admission. However, further research and in particular randomised controlled trials are required to evaluate the clinical and economic impacts of using AI/ML/DL models in trauma medicine.

Accuracy of prehospital triage systems for mass casualty incidents in trauma register studies - A systematic review and meta-analysis of diagnostic test accuracy studies

BACKGROUND

Prioritising patients in mass casualty incidents (MCI) can be extremely difficult. Therefore, triage systems are important in every emergency medical service. This study reviews the accuracy of primary triage systems for MCI in trauma register studies.

METHODS

We registered a protocol at PROSPERO ID: CRD42018115438. We searched MEDLINE, EMBASE, Central, Web of Science, Scopus, Clinical Trials, Google Scholar, and reference lists for eligible studies. We included studies that both examined a primary triage system for MCI in trauma registers and provided sensitivity and specificity for critically injured vs non-critically injured as results. We excluded studies that used paediatric, chemical, biological, radiological or nuclear MCIs populations or triage systems. Finally, we calculated intra-study relative sensitivity, specificity and diagnostic odds ratio for each triage system.

RESULTS

Triage Sieve (TS) significantly underperformed in relative diagnostic odds ratio (DOR) when compared to START and CareFlight (CF) (START vs TS: 19.85 vs 13.23 (p<0.0001)│CF vs TS: 23.72 vs 12.83 (p<0.0001)). There was no significant difference in DOR between TS and Military Sieve (MS) (p<0.710). Compared to START, MS and CF TS had significantly higher relative specificity (START vs TS: 93.6% vs 96.1% (p=0.047)│CF vs TS: 96% vs 95.3% (p=0.0006)│MS vs TS: 94% vs 88.3% (p=0.0002)) and lower relative sensitivity (START vs TS: 57.8% vs 34.8% (p<0.0001)│CF vs TS: 53.9% vs 34.7% (p<0.0001)│MS vs TS: 51.9% vs 35.2% p<0.0001)). CF had significantly better relative DOR than START (CF vs START: 23.56 vs 27.79 (p=0.043)). MS had significantly better relative sensitivity than CF and START (MS vs CF: 49.5% vs 38.7% (p<0.0001)│MS vs START: 49.4% vs 43.9% (p=0.01)). In contrast, CF had significantly better relative specificity than MS (MS vs CF: 91.3% vs 93.3% (p<0.0001)). The remaining comparisons did not yield any significant differences.

CONCLUSION

As the included studies were at risk of bias and had heterogenic characteristics, our results should be interpreted with caution. Nonetheless, our results point towards inferior accuracy of Triage Sieve compared to START and CareFlight, and less firmly point towards superior accuracy of Military Sieve compared to START, CareFlight and Triage Sieve.

Fifteen-minute consultation: An overview of major incidents

Major incidents are rare but require a large amount of preparation, co-ordination and communication across different emergency services and specialities. This ensures that casualties are efficiently managed within the constraints of limited clinical resources. This article aims to provide a brief understanding of what constitutes as a major incident, how it is declared and the chain of command in communication and action, focusing specifically on the paediatric process. It also aims to highlight important considerations that could potentially be missed (eg, the mental health impact, forensic evidence and so on).

Paediatric major incident triage: UK military tool offers best performance in predicting the need for time-critical major surgical and resuscitative intervention

BACKGROUND

Children are frequently injured during major incidents (MI), including terrorist attacks, conflict and natural disasters. Triage facilitates healthcare resource allocation in order to maximise overall survival. A critical function of MI triage tools is to identify patients needing time-critical major resuscitative and surgical intervention (Priority 1 (P1) status). This study compares the performance of 11 MI triage tools in predicting P1 status in children from the UK Trauma Audit and Research Network (TARN) registry.

METHODS

Patients aged <16 years within TARN (January 2008-December 2017) were included. 11 triage tools were applied to patients' first recorded pre-hospital physiology. Patients were retrospectively assigned triage categories (P1, P2, P3, Expectant or Dead) using predefined intervention-based criteria. Tools' performance in <16s were evaluated within four-yearly age subgroups, comparing tool-predicted and intervention-based priority status.

FINDINGS

Amongst 4962 patients, mortality was 1.1% (n = 53); median Injury Severity Score (ISS) was 9 (IQR 9-16). Blunt injuries predominated (94.4%). 1343 (27.1%) met intervention-based criteria for P1, exhibiting greater intensive care requirement (60.2% vs. 8.5%, p < 0.01) and ISS (median 17 vs 9, p < 0.01) compared with P2 patients. The Battlefield Casualty Drills (BCD) Triage Sieve had greatest sensitivity (75.7%) in predicting P1 status in children <16 years, demonstrating a 38.4-49.8% improvement across all subgroups of children <12 years compared with the UK's current Paediatric Triage Tape (PTT). JumpSTART demonstrated low sensitivity in predicting P1 status in 4 to 8 year olds (35.5%) and 0 to 4 year olds (28.5%), and was outperformed by its adult counterpart START (60.6% and 59.6%).

INTERPRETATION

The BCD Triage Sieve had greatest sensitivity in predicting P1 status in this paediatric trauma registry population: we recommend it replaces the PTT in UK practice. Users of JumpSTART may consider alternative tools. We recommend Lerner's triage category definitions when conducting MI evaluations.