Development and Validation of a Prediction Model for Prehospital Triage of Trauma Patients

Traumatic Brain Injury as an Independent Predictor of Futility in the Early Resuscitation of Patients in Hemorrhagic Shock

This review explores the concept of futility timeouts and the use of traumatic brain injury (TBI) as an independent predictor of the futility of resuscitation efforts in severely bleeding trauma patients. The national blood supply shortage has been exacerbated by the lingering influence of the COVID-19 pandemic on the number of blood donors available, as well as by the adoption of balanced hemostatic resuscitation protocols (such as the increasing use of 1:1:1 packed red blood cells, plasma, and platelets) with and without early whole blood resuscitation. This has underscored the urgent need for reliable predictors of futile resuscitation (FR). As a result, clinical, radiologic, and laboratory bedside markers have emerged which can accurately predict FR in patients with severe trauma-induced hemorrhage, such as the Suspension of Transfusion and Other Procedures (STOP) criteria. However, the STOP criteria do not include markers for TBI severity or transfusion cut points despite these patients requiring large quantities of blood components in the STOP criteria validation cohort. Yet, guidelines for neuroprognosticating patients with TBI can require up to 72 h, which makes them less useful in the minutes and hours following initial presentation. We examine the impact of TBI on bleeding trauma patients, with a focus on those with coagulopathies associated with TBI. This review categorizes TBI into isolated TBI (iTBI), hemorrhagic isolated TBI (hiTBI), and polytraumatic TBI (ptTBI). Through an analysis of bedside parameters (such as the proposed STOP criteria), coagulation assays, markers for TBI severity, and transfusion cut points as markers of futilty, we suggest amendments to current guidelines and the development of more precise algorithms that incorporate prognostic indicators of severe TBI as an independent parameter for the early prediction of FR so as to optimize blood product allocation.

The predictive value of serum lactate to forecast injury severity in trauma-patients increases taking age into account

BACKGROUND

Two-tier trauma team activation (TTA)-protocols often fail to safely identify severely injured patients. A possible amendment to existing triage scores could be the measurement of serum lactate. The aim of this study was to determine the ability of the combination of serum lactate and age to predict severe injuries (ISS > 15).

METHODS

We conducted a retrospective cohort study in a single level one trauma center in a 20 months study-period and analyzed every trauma team activation (TTA) due to the mechanism of injury (MOI). Primary endpoint was the correlation between serum lactate (and age) and ISS and mortality. The validity of lactate (LAC) and lactate contingent on age (LAC + AGE) were assessed using the area under the curve (AUC) of the receiver operating characteristics (ROC) curve. We used a logistic regression model to predict the probability of an ISS > 15.

RESULTS

During the study period we included 325 patients, 75 met exclusion criteria. Mean age was 43 years (Min.: 11, Max.: 90, SD: 18.7) with a mean ISS of 8.4 (SD: 8.99). LAC showed a sensitivity of 0.82 with a specificity of 0.62 with an optimal cutoff at 1.72 mmol/l to predict an ISS > 15. The AUC of the ROC for LAC was 0.764 (95% CI: 0.67-0.85). The LAC + AGE model provided a significantly improved predictive value compared to LAC (0.765 vs. 0.828, p < 0.001).

CONCLUSIONS

The serum lactate concentration is able to predict injury severity. The prognostic value improves significantly taking the patients age into consideration. The combination of serum lactate and age could be a suitable Ad-on to existing two-tier triage protocols to minimize undertriage.

LEVEL OF EVIDENCE

Level IV, retrospective cohort study.

Prospective validation of a hospital triage predictive model to decrease undertriage: an EAST multicenter study

Background

Tiered trauma team activation (TTA) allows systems to optimally allocate resources to an injured patient. Target undertriage and overtriage rates of <5% and <35% are difficult for centers to achieve, and performance variability exists. The objective of this study was to optimize and externally validate a previously developed hospital trauma triage prediction model to predict the need for emergent intervention in 6 hours (NEI-6), an indicator of need for a full TTA.

Methods

The model was previously developed and internally validated using data from 31 US trauma centers. Data were collected prospectively at five sites using a mobile application which hosted the NEI-6 model. A weighted multiple logistic regression model was used to retrain and optimize the model using the original data set and a portion of data from one of the prospective sites. The remaining data from the five sites were designated for external validation. The area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) were used to assess the validation cohort. Subanalyses were performed for age, race, and mechanism of injury.

Results

14 421 patients were included in the training data set and 2476 patients in the external validation data set across five sites. On validation, the model had an overall undertriage rate of 9.1% and overtriage rate of 53.7%, with an AUROC of 0.80 and an AUPRC of 0.63. Blunt injury had an undertriage rate of 8.8%, whereas penetrating injury had 31.2%. For those aged ≥65, the undertriage rate was 8.4%, and for Black or African American patients the undertriage rate was 7.7%.

Conclusion

The optimized and externally validated NEI-6 model approaches the recommended undertriage and overtriage rates while significantly reducing variability of TTA across centers for blunt trauma patients. The model performs well for populations that traditionally have high rates of undertriage.

Level of evidence

2.

Studying the social mind: An updated summary of findings from the Vietnam Head Injury Study

Lesion mapping studies allow us to evaluate the potential causal contribution of specific brain areas to human cognition and complement other cognitive neuroscience methods, as several authors have recently pointed out. Here, we present an updated summary of the findings from the Vietnam Head Injury Study (VHIS) focusing on the studies conducted over the last decade, that examined the social mind and its intricate neural and cognitive underpinnings. The VHIS is a prospective, long-term follow-up study of Vietnam veterans with penetrating traumatic brain injury (pTBI) and healthy controls (HC). The scope of the work is to present the studies from the latest phases (3 and 4) of the VHIS, 70 studies since 2011, when the Raymont et al. paper was published (Raymont et al., 2011). These studies have contributed to our understanding of human social cognition, including political and religious beliefs, theory of mind, but also executive functions, intelligence, and personality. This work finally discusses the usefulness of lesion mapping as an approach to understanding the functions of the human brain from basic science and clinical perspectives.

Spanish vs USA cohort comparison of prehospital trauma scores to predict short-term mortality

BACKGROUND

This study aimed to evaluate three prehospital early warning scores (EWSs): RTS, MGAP and MREMS, to predict short-term mortality in acute life-threatening trauma and injury/illness by comparing United States (US) and Spanish cohorts.

METHODS

A total of 8,854 patients, 8,598/256 survivors/nonsurvivors, comprised the unified cohort. Datasets were randomly divided into training and test sets. Training sets were used to analyse the discriminative power of the scores in terms of the area under the curve (AUC), and the score performance was assessed in the test set in terms of sensitivity (SE), specificity (SP), accuracy (ACC) and balanced accuracy (BAC).

RESULTS

The three scores showed great discriminative power with AUCs>0.90, and no significant differences between cohorts were found. In the test set, RTS/MREMS/MGAP showed SE/SP/ACC/BAC values of 86.0/89.9/89.6/87.1%, 91.0/86.9/87.5/88.5%, and 87.7/82.9/83.4/85.2%, respectively.

CONCLUSIONS

All EWSs showed excellent ability to predict the risk of short-term mortality, independent of the country.

Predicting the complexity and mortality of polytrauma patients with machine learning models

We aim to develop machine learning (ML) models for predicting the complexity and mortality of polytrauma patients using clinical features, including physician diagnoses and physiological data. We conducted a retrospective analysis of a cohort comprising 756 polytrauma patients admitted to the intensive care unit (ICU) at Pizhou People's Hospital Trauma Center, Jiangsu, China between 2020 and 2022. Clinical parameters encompassed demographics, vital signs, laboratory values, clinical scores and physician diagnoses. The two primary outcomes considered were mortality and complexity. We developed ML models to predict polytrauma mortality or complexity using four ML algorithms, including Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN) and eXtreme Gradient Boosting (XGBoost). We assessed the models' performance and compared the optimal ML model against three existing trauma evaluation scores, including Injury Severity Score (ISS), Trauma Index (TI) and Glasgow Coma Scale (GCS). In addition, we identified several important clinical predictors that made contributions to the prognostic models. The XGBoost-based polytrauma mortality prediction model demonstrated a predictive ability with an accuracy of 90% and an F-score of 88%, outperforming SVM, RF and ANN models. In comparison to conventional scoring systems, the XGBoost model had substantial improvements in predicting the mortality of polytrauma patients. External validation yielded strong stability and generalization with an accuracy of up to 91% and an AUC of 82%. To predict polytrauma complexity, the XGBoost model maintained its performance over other models and scoring systems with good calibration and discrimination abilities. Feature importance analysis highlighted several clinical predictors of polytrauma complexity and mortality, such as Intracranial hematoma (ICH). Leveraging ML algorithms in polytrauma care can enhance the prognostic estimation of polytrauma patients. This approach may have potential value in the management of polytrauma patients.

Merging of two level-1 trauma centers in Amsterdam: premerger demand in integrated acute trauma care

PURPOSE

Availability of adequate and appropriate trauma care is essential. A merger of two Dutch academic level-1 trauma centers is upcoming. However, in the literature, volume effects after a merger are inconclusive. This study aimed to examine the premerger demand for level-1 trauma care on integrated acute trauma care and evaluate the expected demand on the system.

METHODS

A retrospective observational study was conducted between 1-1-2018 and 1-1-2019 in two level-1 trauma centers in the Amsterdam region using data derived from the local trauma registries and electronic patient records. All trauma patients presented at both centers' Emergency Departments (ED) were included. Patient- and injury characteristics and data concerning all prehospital and in-hospital-delivered trauma care were collected and compared. Pragmatically, the demand for trauma care in the post-merger setting was considered a sum of care demand for both centers.

RESULTS

In total, 8277 trauma patients were presented at both EDs, 4996 (60.4%) at location A and 3281 (39.6%) at location B. Overall, 462 patients were considered severely injured patients (Injury Severity Score ≥ 16). In total, 702 emergency surgeries (< 24 h) were performed, and 442 patients were admitted to the ICU. The sum care demand of both centers resulted in a 167.4% increase in trauma patients and a 151.1% increase in severely injured patients. Moreover, on 96 occasions annually, two or more patients within the same hour would require advanced trauma resuscitation by a specialized team or emergency surgery.

CONCLUSION

A merger of two Dutch level-1 trauma centers would, in this scenario, result in a more than 150% increase in the post-merger setting's demand for integrated acute trauma care.

Pre-hospital tranexamic acid administration in patients with a severe hemorrhage: an evaluation after the implementation of tranexamic acid administration in the Dutch pre-hospital protocol

PURPOSE

To evaluate the pre-hospital administration of tranexamic acid in ambulance-treated trauma patients with a severe hemorrhage after the implementation of tranexamic acid administration in the Dutch pre-hospital protocol.

METHODS

All patients with a severe hemorrhage who were treated and conveyed by EMS professionals between January 2015, and December 2017, to any trauma-receiving emergency department in the eight participating trauma regions in the Netherlands, were included. A severe hemorrhage was defined as extracranial injury with > 20% body volume blood loss, an extremity amputation above the wrist or ankle, or a grade ≥ 4 visceral organ injury. The main outcome was to determine the proportion of patients with a severe hemorrhage who received pre-hospital treatment with tranexamic acid. A Generalized Linear Model (GLM) was performed to investigate the relationship between pre-hospital tranexamic acid treatment and 24 h mortality.

RESULTS

A total of 477 patients had a severe hemorrhage, of whom 124 patients (26.0%) received tranexamic acid before arriving at the hospital. More than half (58.4%) of the untreated patients were suspected of a severe hemorrhage by EMS professionals. Patients treated with tranexamic acid had a significantly lower risk on 24 h mortality than untreated patients (OR 0.43 [95% CI 0.19-0.97]).

CONCLUSION

Approximately a quarter of the patients with a severe hemorrhage received tranexamic acid before arriving at the hospital, while a severe hemorrhage was suspected in more than half of the non-treated patients. Severely hemorrhaging patients treated with tranexamic acid before arrival at the hospital had a lower risk to die within 24 h after injury.

Identifying Trauma Patients in Need for Emergency Surgery in the Prehospital Setting: The Prehospital Prediction of In-Hospital Emergency Treatment (PROPHET) Study

Prehospital field triage often fails to accurately identify the need for emergent surgical or non-surgical procedures, resulting in inefficient resource utilization and increased costs. This study aimed to analyze prehospital factors associated with the need for emergent procedures (such as surgery or interventional angiography) within 6 h of hospital admission. Additionally, our goal was to develop a prehospital triage tool capable of estimating the likelihood of requiring an emergent procedure following hospital admission. We conducted a retrospective observational study, analyzing both prehospital and in-hospital data obtained from the Lombardy Trauma Registry. We conducted a multivariable logistic regression analysis to identify independent predictors of emergency procedures within the first 6 h from admission. Subsequently, we developed and internally validated a triage score composed of factors associated with the probability of requiring an emergency procedure. The study included a total of 3985 patients, among whom 295 (7.4%) required an emergent procedure within 6 h. Age, penetrating injury, downfall, cardiac arrest, poor neurological status, endotracheal intubation, systolic pressure, diastolic pressure, shock index, respiratory rate and tachycardia were identified as predictors of requiring an emergency procedure. A triage score generated from these predictors showed a good predictive power (AUC of the ROC curve: 0.81) to identify patients requiring an emergent surgical or non-surgical procedure within 6 h from hospital admission. The proposed triage score might contribute to predicting the need for immediate resource availability in trauma patients.

On Scene Injury Severity Prediction (OSISP) model for trauma developed using the Swedish Trauma Registry

BACKGROUND

Providing optimal care for trauma, the leading cause of death for young adults, remains a challenge e.g., due to field triage limitations in assessing a patient's condition and deciding on transport destination. Data-driven On Scene Injury Severity Prediction (OSISP) models for motor vehicle crashes have shown potential for providing real-time decision support. The objective of this study is therefore to evaluate if an Artificial Intelligence (AI) based clinical decision support system can identify severely injured trauma patients in the prehospital setting.

METHODS

The Swedish Trauma Registry was used to train and validate five models - Logistic Regression, Random Forest, XGBoost, Support Vector Machine and Artificial Neural Network - in a stratified 10-fold cross validation setting and hold-out analysis. The models performed binary classification of the New Injury Severity Score and were evaluated using accuracy metrics, area under the receiver operating characteristic curve (AUC) and Precision-Recall curve (AUCPR), and under- and overtriage rates.

RESULTS

There were 75,602 registrations between 2013-2020 and 47,357 (62.6%) remained after eligibility criteria were applied. Models were based on 21 predictors, including injury location. From the clinical outcome, about 40% of patients were undertriaged and 46% were overtriaged. Models demonstrated potential for improved triaging and yielded AUC between 0.80-0.89 and AUCPR between 0.43-0.62.

CONCLUSIONS

AI based OSISP models have potential to provide support during assessment of injury severity. The findings may be used for developing tools to complement field triage protocols, with potential to improve prehospital trauma care and thereby reduce morbidity and mortality for a large patient population.

Derivation and validation of a blood biomarker score for 2-day mortality prediction from prehospital care: a multicenter, cohort, EMS-based study

Identifying potentially life-threatening diseases is a key challenge for emergency medical services. This study aims at examining the role of different prehospital biomarkers from point-of-care testing to derive and validate a score to detect 2-day in-hospital mortality. We conducted a prospective, observational, prehospital, ongoing, and derivation-validation study in three Spanish provinces, in adults evacuated by ambulance and admitted to the emergency department. A total of 23 ambulance-based biomarkers were collected from each patient. A biomarker score based on logistic regression was fitted to predict 2-day mortality from an optimum subset of variables from prehospital blood analysis, obtained through an automated feature selection stage. 2806 cases were analyzed, with a median age of 68 (interquartile range 51-81), 42.3% of women, and a 2-day mortality rate of 5.5% (154 non-survivors). The blood biomarker score was constituted by the partial pressure of carbon dioxide, lactate, and creatinine. The score fitted with logistic regression using these biomarkers reached a high performance to predict 2-day mortality, with an AUC of 0.933 (95% CI 0.841-0.973). The following risk levels for 2-day mortality were identified from the score: low risk (score < 1), where only 8.2% of non-survivors were assigned to; medium risk (1 ≤ score < 4); and high risk (score ≥ 4), where the 2-day mortality rate was 57.6%. The novel blood biomarker score provides an excellent association with 2-day in-hospital mortality, as well as real-time feedback on the metabolic-respiratory patient status. Thus, this score can help in the decision-making process at critical moments in life-threatening situations.

Trauma Patient Volume and the Quality of Care: A Scoping Review

BACKGROUND

Healthcare stakeholders in the Netherlands came to an agreement in 2022 to deal with present and future challenges in healthcare. Among others, this agreement contains clear statements regarding the concentration of trauma patients, including the minimal required number of annual severe trauma patients for Major Trauma Centers. This review investigates the effects of trauma patient volumes on several domains of the quality of healthcare.

METHODS

PubMed was searched; studies published during the last 10 years reporting quantitative data on trauma patient volume and quality of healthcare were included. Results were summarized and categorized into the quality domains of healthcare.

RESULTS

Seventeen studies were included with a total of 1,517,848 patients. A positive association between trauma patient volume and survival was observed in 11/13 studies with adjusted analyses. Few studies addressed other quality domains: efficiency (n = 5), safety (n = 2), and time aspects of care (n = 4). None covered people-centeredness, equitability, or integrated care.

CONCLUSIONS

Most studies showed a better survival of trauma patients when treated in high-volume hospitals compared to lower volume hospitals. However, the ideal threshold could not be determined. The association between trauma volume and other domains of the quality of healthcare remains unclear.

Implementation of Prediction Models in the Emergency Department from an Implementation Science Perspective-Determinants, Outcomes, and Real-World Impact: A Scoping Review

STUDY OBJECTIVE

Prediction models offer a promising form of clinical decision support in the complex and fast-paced environment of the emergency department (ED). Despite significant advancements in model development and validation, implementation of such models in routine clinical practice remains elusive. This scoping review aims to survey the current state of prediction model implementation in the ED and to provide insights on contributing factors and outcomes from an implementation science perspective.

METHODS

We searched 4 databases from their inception to May 20, 2022: MEDLINE (through PubMed), Embase, Scopus, and CINAHL. Articles that reported implementation outcomes and/or contextual determinants under the Reach, Effectiveness, Adoption, Implementation Maintenance (RE-AIM)/Practical, Robust, Implementation, and Sustainability Model (PRISM) framework were included. Characteristics of studies, models, and results of the RE-AIM/PRISM domains were summarized narratively.

RESULTS

Thirty-six reports on 31 implementations were included. The most common prediction models implemented were early warning scores. The most common implementation strategies used were training stakeholders, infrastructural changes, and using evaluative or iterative strategies. Only one report examined ED patients' perspectives, whereas the rest were focused on the experience of health care workers or organizational stakeholders. Key determinants of successful implementation include strong stakeholder engagement, codevelopment of workflows and implementation strategies, education, and usability.

CONCLUSION

Examining ED prediction models from an implementation science perspective can provide valuable insights and help guide future implementations.

Looking back to inform the future: a review of published paramedicine research

OBJECTIVE

Paramedicine has evolved in ways that may outpace the science informing these changes. Examining the scholarly pursuits of paramedicine may provide insights into the historical academic focus, which may inform future endeavors and evolution of paramedicine. The objective of this study was to explore the existing discourse in paramedicine research to reflect on the academic pursuits of this community.

METHODS

We searched Medline, Embase, CINAHL, Google Scholar and Web of Science from January, 2006 to April, 2019. We further refined the yield using a ranking formula that prioritized journals most relevant to paramedicine, then sampled randomly in two-year clusters for full text review. We extracted literature type, study topic and context, then used elements of qualitative content, thematic, and discourse analysis to further describe the sample.

RESULTS

The initial search yielded 99,124 citations, leaving 54,638 after removing duplicates and 7084 relevant articles from nine journals after ranking. Subsequently, 2058 articles were included for topic categorization, and 241 papers were included for full text analysis after random sampling. Overall, this literature reveals: 1) a relatively narrow topic focus, given the majority of research has concentrated on general operational activities and specific clinical conditions and interventions (e.g., resuscitation, airway management, etc.); 2) a limited methodological (and possibly philosophical) focus, given that most were observational studies (e.g., cohort, case control, and case series) or editorial/commentary; 3) a variety of observed trajectories of academic attention, indicating where the evolution of paramedicine is evident, areas where scope of practice is uncertain, and areas that aim to improve skills historically considered core to paramedic clinical practice.

CONCLUSIONS

Included articles suggest a relatively narrow topic focus, a limited methodological focus, and observed trajectories of academic attention indicating where research pursuits and priorities are shifting. We have highlighted that the academic focus may require an alignment with aspirational and direction setting documents aimed at developing paramedicine. This review may be a snapshot of scholarly activity that reflects a young medically directed profession and systems focusing on a few high acuity conditions, with aspirations of professional autonomy contributing to the health and social well-being of communities.

The Safety INdEx of Prehospital On Scene Triage (SINEPOST) study: The development and validation of a risk prediction model to support ambulance clinical transport decisions on-scene

One of the main problems currently facing the delivery of safe and effective emergency care is excess demand, which causes congestion at different time points in a patient’s journey. The modern case-mix of prehospital patients is broad and complex, diverging from the traditional ‘time critical accident and emergency’ patients. It now includes many low-acuity patients and those with social care and mental health needs. In the ambulance service, transport decisions are the hardest to make and paramedics decide to take more patients to the ED than would have a clinical benefit. As such, this study asked the following research questions: In adult patients attending the ED by ambulance, can prehospital information predict an avoidable attendance? What is the simulated transportability of the model derived from the primary outcome? A linked dataset of 101,522 ambulance service and ED ambulance incidents linked to their respective ED care record from the whole of Yorkshire between 1st July 2019 and 29th February 2020 was used as the sample for this study. A machine learning method known as XGBoost was applied to the data in a novel way called Internal-External Cross Validation (IECV) to build the model. The results showed great discrimination with a C-statistic of 0.81 (95%CI 0.79–0.83) and excellent calibration with an O:E ratio was 0.995 (95% CI 0.97–1.03), with the most important variables being a patient’s mobility, their physiological observations and clinical impression with psychiatric problems, allergic reactions, cardiac chest pain, head injury, non-traumatic back pain, and minor cuts and bruising being the most important. This study has successfully developed a decision-support model that can be transformed into a tool that could help paramedics make better transport decisions on scene, known as the SINEPOST model. It is accurate, and spatially validated across multiple geographies including rural, urban, and coastal. It is a fair algorithm that does not discriminate new patients based on their age, gender, ethnicity, or decile of deprivation. It can be embedded into an electronic Patient Care Record system and automatically calculate the probability that a patient will have an avoidable attendance at the ED, if they were transported. This manuscript complies with the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement (Moons KGM, 2015).

Changing from a two-tiered to a one-tiered trauma team activation protocol: a before-after observational cohort study investigating the clinical impact of undertriage

BACKGROUND

The aim of this study was to compare the effect of the change in TTA protocol from a two-tier to one-tier, with focus on undertriage and mortality.

MATERIAL AND METHODS

A before-after observational cohort study based on data extracted from the Stavanger University Hospital Trauma registry in the transition period from two-tier to a one-tier TTA protocol over two consecutive 1-year periods (2017-2018). Comparative analysis was done between the two time-periods for descriptive characteristics and outcomes. The main outcomes of interest were undertriage and mortality.

RESULTS

During the study period 1234 patients were included in the registry, of which 721 (58%) were in the two-tier and 513 (42%) in the one-tier group. About one in five patients (224/1234) were severely injured (ISS > 15). Median age was 39 in the two-tier period and 43 years in the one-tier period (p = 0.229). Median ISS was 5 for the two-tier period vs 9, in the one-tier period (p = 0.001). The undertriage of severely injured patients in the two-tier period was 18/122 (15%), compared to 31/102 (30%) of patients in the one-tier period (OR = 2.5; 95% CI 1.8-4.52). Overall mortality increased significantly between the two TTA protocols, from 2.5 to 4.7% (p = 0.033), OR 0.51 (0.28-0.96) CONCLUSION: A protocol change from two-tiered TTA to one-tiered TTA increased the undertriage in our trauma system. A two-tiered TTA may be beneficial for better patient care.

Machine learning-based prediction of emergency neurosurgery within 24 h after moderate to severe traumatic brain injury

Background

Rapid referral of traumatic brain injury (TBI) patients requiring emergency neurosurgery to a specialized trauma center can significantly reduce morbidity and mortality. Currently, no model has been reported to predict the need for acute neurosurgery in severe to moderate TBI patients. This study aims to evaluate the performance of Machine Learning-based models to establish to predict the need for neurosurgery procedure within 24 h after moderate to severe TBI.

Methods

Retrospective multicenter cohort study using data from a national trauma registry (Traumabase®) from November 2011 to December 2020. Inclusion criteria correspond to patients over 18 years old with moderate or severe TBI (Glasgow coma score ≤ 12) during prehospital assessment. Patients who died within the first 24 h after hospital admission and secondary transfers were excluded. The population was divided into a train set (80% of patients) and a test set (20% of patients). Several approaches were used to define the best prognostic model (linear nearest neighbor or ensemble model). The Shapley Value was used to identify the most relevant pre-hospital variables for prediction.

Results

2159 patients were included in the study. 914 patients (42%) required neurosurgical intervention within 24 h. The population was predominantly male (77%), young (median age 35 years [IQR 24–52]) with severe head injury (median GCS 6 [3–9]). Based on the evaluation of the predictive model on the test set, the logistic regression model had an AUC of 0.76. The best predictive model was obtained with the CatBoost technique (AUC 0.81). According to the Shapley values method, the most predictive variables in the CatBoost were a low initial Glasgow coma score, the regression of pupillary abnormality after osmotherapy, a high blood pressure and a low heart rate.

Conclusion

Machine learning-based models could predict the need for emergency neurosurgery within 24 h after moderate and severe head injury. Potential clinical benefits of such models as a decision-making tool deserve further assessment. The performance in real-life setting and the impact on clinical decision-making of the model requires workflow integration and prospective assessment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13017-022-00449-5.

Need for Emergent Intervention within 6 Hours: A Novel Prediction Model for Hospital Trauma Triage

Objective: A tiered trauma team activation system allocates resources proportional to patients' needs based upon injury burden. Previous trauma hospital-triage models are limited to predicting Injury Severity Score which is based on > 10% all-cause in-hospital mortality, rather than need for emergent intervention within 6 hours (NEI-6). Our aim was to develop a novel prediction model for hospital-triage that utilizes criteria available to the EMS provider to predict NEI-6 and the need for a trauma team activation.Methods: A regional trauma quality collaborative was used to identify all trauma patients ≥ 16 years from the American College of Surgeons-Committee on Trauma verified Level 1 and 2 trauma centers. Logistic regression and random forest were used to construct two predictive models for NEI-6 based on clinically relevant variables. Restricted cubic splines were used to model nonlinear predictors. The accuracy of the prediction model was assessed in terms of discrimination.Results: Using data from 12,624 patients for the training dataset (62.6% male; median age 61 years; median ISS 9) and 9,445 patients for the validation dataset (62.6% male; median age 59 years; median ISS 9), the following significant predictors were selected for the prediction models: age, gender, field GCS, vital signs, intentionality, and mechanism of injury. The final boosted tree model showed an AUC of 0.85 in the validation cohort for predicting NEI-6.Conclusions: The NEI-6 trauma triage prediction model used prehospital metrics to predict need for highest level of trauma activation. Prehospital prediction of major trauma may reduce undertriage mortality and improve resource utilization.

Development and evaluation of a machine learning-based in-hospital COVID-19 disease outcome predictor (CODOP): A multicontinental retrospective study

New SARS-CoV-2 variants, breakthrough infections, waning immunity, and sub-optimal vaccination rates account for surges of hospitalizations and deaths. There is an urgent need for clinically valuable and generalizable triage tools assisting the allocation of hospital resources, particularly in resource-limited countries. We developed and validate CODOP, a machine learning-based tool for predicting the clinical outcome of hospitalized COVID-19 patients. CODOP was trained, tested and validated with six cohorts encompassing 29223 COVID-19 patients from more than 150 hospitals in Spain, the USA and Latin America during 2020-22. CODOP uses 12 clinical parameters commonly measured at hospital admission for reaching high discriminative ability up to 9 days before clinical resolution (AUROC: 0·90-0·96), it is well calibrated, and it enables an effective dynamic risk stratification during hospitalization. Furthermore, CODOP maintains its predictive ability independently of the virus variant and the vaccination status. To reckon with the fluctuating pressure levels in hospitals during the pandemic, we offer two online CODOP calculators, suited for undertriage or overtriage scenarios, validated with a cohort of patients from 42 hospitals in three Latin American countries (78-100% sensitivity and 89-97% specificity). The performance of CODOP in heterogeneous and geographically disperse patient cohorts and the easiness of use strongly suggest its clinical utility, particularly in resource-limited countries.

Evaluating pre-hospital triage and decision-making in patients who died within 30 days post-trauma: A multi-site, multi-center, cohort study

INTRODUCTION

Evaluating pre-hospital triage and decision-making in patients who died post-trauma is crucial to decrease undertriage and improve future patients' chances of survival. A study that has adequately investigated this is currently lacking. The aim of this study was therefore to evaluate pre-hospital triage and decision-making in patients who died within 30 days post-trauma.

MATERIALS AND METHODS

A multi-site, multi-center, cohort study was conducted. Trauma patients who were transported from the scene of injury to a trauma center by ambulance and died within 30 days post-trauma, were included. The main outcome was undertriage, defined as erroneously transporting a severely injured patient (Injury Severity Score ≥ 16) to a lower-level trauma center.

RESULTS

Between January 2015 and December 2017, 2116 patients were included, of whom 765 (36.2%) were severely injured. A total of 103 of these patients (13.5%) were undertriaged. Undertriaged patients were often elderly with a severe head and/or thoracic injury as a result of a minor fall (< 2 m). A majority of the undertriaged patients were triaged without assistance of a specialized physician (100 [97.1%]), did not meet field triage criteria for level-I trauma care (81 [78.6%]), and could have been transported to the nearest level-I trauma center within 45 min (93 [90.3%]).

CONCLUSION

Approximately 14% of the severely injured patients who died within 30 days were undertriaged and could have benefited from treatment at a level-I trauma center (i.e., specialized trauma care). Improvement of pre-hospital triage is needed to potentially increase future patients' chances of survival.

Individual risk factors predictive of major trauma in pre-hospital injured older patients: a systematic review

Background

Older adults with major trauma are frequently under-triaged, increasing the risk of preventable morbidity and mortality. The aim of this systematic review was to identify which individual risk factors and predictors are likely to increase the risk of major trauma in elderly patients presenting to emergency medical services (EMS) following injury, to inform future elderly triage tool development.

Methods

Several electronic databases (including Medline, EMBASE, CINAHL and the Cochrane Library) were searched from inception to February 2021. Prospective or retrospective diagnostic studies were eligible if they examined a prognostic factor (often termed predictor or risk factor) for, or diagnostic test to identify, major trauma. Selection of studies, data extraction and risk of bias assessments using the Quality in Prognostic Studies (QUIPS) tool were undertaken independently by at least two reviewers. Narrative synthesis was used to summarise the findings.

Results

Nine studies, all performed in US trauma networks, met review inclusion criteria. Vital signs (Glasgow Coma Scale (GCS) score, systolic blood pressure, respiratory rate and shock index with specific elderly cut-off points), EMS provider judgement, comorbidities and certain crash scene variables (other occupants injured, occupant not independently mobile and head-on collision) were identified as significant pre-hospital variables associated with major trauma in the elderly in multi-variable analyses. Heart rate and anticoagulant were not significant predictors. Included studies were at moderate or high risk of bias, with applicability concerns secondary to selected study populations.

Conclusions

Existing pre-hospital major trauma triage tools could be optimised for elderly patients by including elderly-specific physiology thresholds. Future work should focus on more relevant reference standards and further evaluation of novel elderly relevant triage tool variables and thresholds.

An economic evaluation of triage tools for patients with suspected severe injuries in England

BACKGROUND

Many health care systems triage injured patients to major trauma centres (MTCs) or local hospitals by using triage tools and paramedic judgement. Triage tools are typically assessed by whether patients with an Injury Severity Score (ISS) ≥ 16 go to an MTC and whether patients with an ISS < 16 are sent to their local hospital. There is a trade-off between sensitivity and specificity of triage tools, with the optimal balance being unknown. We conducted an economic evaluation of major trauma triage tools to identify which tool would be considered cost-effective by UK decision makers.

METHODS

A patient-level, probabilistic, mathematical model of a UK major trauma system was developed. Patients with an ISS ≥ 16 who were only treated at local hospitals had worse outcomes compared to being treated in an MTC. Nine empirically derived triage tools, from a previous study, were examined so we assessed triage tools with realistic trade-offs between triage tool sensitivity and specificity. Lifetime costs, lifetime quality adjusted life years (QALYs), and incremental cost-effectiveness ratios (ICERs) were calculated for each tool and compared to maximum acceptable ICERs (MAICERs) in England.

RESULTS

Four tools had ICERs within the normal range of MAICERs used by English decision makers (£20,000 to £30,000 per QALY gained). A low sensitivity (28.4%) and high specificity (88.6%) would be cost-effective at the lower end of this range while higher sensitivity (87.5%) and lower specificity (62.8%) was cost-effective towards the upper end of this range. These results were sensitive to the cost of MTC admissions and whether MTCs had a benefit for patients with an ISS between 9 and 15.

CONCLUSIONS

The cost-effective triage tool depends on the English decision maker's MAICER for this health problem. In the usual range of MAICERs, cost-effective prehospital trauma triage involves clinically suboptimal sensitivity, with a proportion of seriously injured patients (at least 10%) being initially transported to local hospitals. High sensitivity trauma triage requires development of more accurate decision rules; research to establish if patients with an ISS between 9 and 15 benefit from MTCs; or, inefficient use of health care resources to manage patients with less serious injuries at MTCs.

The Safety INdEx of Prehospital On Scene Triage (SINEPOST) study: the development and validation of a risk prediction model to support ambulance clinical transport decisions on-scene-a protocol

BACKGROUND

Demand for both the ambulance service and the emergency department (ED) is rising every year and when this demand is excessive in both systems, ambulance crews queue at the ED waiting to hand patients over. Some transported ambulance patients are 'low-acuity' and do not require the treatment of the ED. However, paramedics can find it challenging to identify these patients accurately. Decision support tools have been developed using expert opinion to help identify these low acuity patients but have failed to show a benefit beyond regular decision-making. Predictive algorithms may be able to build accurate models, which can be used in the field to support the decision not to take a low-acuity patient to an ED.

METHODS AND ANALYSIS

All patients in Yorkshire who were transported to the ED by ambulance between July 2019 and February 2020 will be included. Ambulance electronic patient care record (ePCR) clinical data will be used as candidate predictors for the model. These will then be linked to the corresponding ED record, which holds the outcome of a 'non-urgent attendance'. The estimated sample size is 52,958, with 4767 events and an EPP of 7.48. An XGBoost algorithm will be used for model development. Initially, a model will be derived using all the data and the apparent performance will be assessed. Then internal-external validation will use non-random nested cross-validation (CV) with test sets held out for each ED (spatial validation). After all models are created, a random-effects meta-analysis will be undertaken. This will pool performance measures such as goodness of fit, discrimination and calibration. It will also generate a prediction interval and measure heterogeneity between clusters. The performance of the full model will be updated with the pooled results.

DISCUSSION

Creating a risk prediction model in this area will lead to further development of a clinical decision support tool that ensures every ambulance patient can get to the right place of care, first time. If this study is successful, it could help paramedics evaluate the benefit of transporting a patient to the ED before they leave the scene. It could also reduce congestion in the urgent and emergency care system.

TRIAL REGISTRATION

This study was retrospectively registered with the ISRCTN: 12121281.

Identifying trauma patients with benefit from direct transportation to Level-1 trauma centers

Background

Prehospital triage protocols typically try to select patients with Injury Severity Score (ISS) above 15 for direct transportation to a Level-1 trauma center. However, ISS does not necessarily discriminate between patients who benefit from immediate care at Level-1 trauma centers. The aim of this study was to assess which patients benefit from direct transportation to Level-1 trauma centers.

Methods

We used the American National Trauma Data Bank (NTDB), a retrospective observational cohort. All adult patients (ISS > 3) between 2015 and 2016 were included. Patients who were self-presenting or had isolated limb injury were excluded. We used logistic regression to assess the association of direct transportation to Level-1 trauma centers with in-hospital mortality adjusted for clinically relevant confounders. We used this model to define benefit as predicted probability of mortality associated with transportation to a non-Level-1 trauma center minus predicted probability associated with transportation to a Level-1 trauma center. We used a threshold of 1% as absolute benefit. Potential interaction terms with transportation to Level-1 trauma centers were included in a penalized logistic regression model to study which patients benefit.

Results

We included 388,845 trauma patients from 232 Level-1 centers and 429 Level-2/3 centers. A small beneficial effect was found for direct transportation to Level-1 trauma centers (adjusted Odds Ratio: 0.96, 95% Confidence Interval: 0.92–0.99) which disappeared when comparing Level-1 and 2 versus Level-3 trauma centers. In the risk approach, predicted benefit ranged between 0 and 1%. When allowing for interactions, 7% of the patients (n = 27,753) had more than 1% absolute benefit from direct transportation to Level-1 trauma centers. These patients had higher AIS Head and Thorax scores, lower GCS and lower SBP. A quarter of the patients with ISS > 15 were predicted to benefit from transportation to Level-1 centers (n = 26,522, 22%).

Conclusions

Benefit of transportation to a Level-1 trauma centers is quite heterogeneous across patients and the difference between Level-1 and Level-2 trauma centers is small. In particular, patients with head injury and signs of shock may benefit from care in a Level-1 trauma center. Future prehospital triage models should incorporate more complete risk profiles.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12873-021-00487-3.

Development and external validation of a clinical prediction model for functional impairment after intracranial tumor surgery

OBJECTIVE

Decision-making for intracranial tumor surgery requires balancing the oncological benefit against the risk for resection-related impairment. Risk estimates are commonly based on subjective experience and generalized numbers from the literature, but even experienced surgeons overestimate functional outcome after surgery. Today, there is no reliable and objective way to preoperatively predict an individual patient's risk of experiencing any functional impairment.

METHODS

The authors developed a prediction model for functional impairment at 3 to 6 months after microsurgical resection, defined as a decrease in Karnofsky Performance Status of ≥ 10 points. Two prospective registries in Switzerland and Italy were used for development. External validation was performed in 7 cohorts from Sweden, Norway, Germany, Austria, and the Netherlands. Age, sex, prior surgery, tumor histology and maximum diameter, expected major brain vessel or cranial nerve manipulation, resection in eloquent areas and the posterior fossa, and surgical approach were recorded. Discrimination and calibration metrics were evaluated.

RESULTS

In the development (2437 patients, 48.2% male; mean age ± SD: 55 ± 15 years) and external validation (2427 patients, 42.4% male; mean age ± SD: 58 ± 13 years) cohorts, functional impairment rates were 21.5% and 28.5%, respectively. In the development cohort, area under the curve (AUC) values of 0.72 (95% CI 0.69-0.74) were observed. In the pooled external validation cohort, the AUC was 0.72 (95% CI 0.69-0.74), confirming generalizability. Calibration plots indicated fair calibration in both cohorts. The tool has been incorporated into a web-based application available at https://neurosurgery.shinyapps.io/impairment/.

CONCLUSIONS

Functional impairment after intracranial tumor surgery remains extraordinarily difficult to predict, although machine learning can help quantify risk. This externally validated prediction tool can serve as the basis for case-by-case discussions and risk-to-benefit estimation of surgical treatment in the individual patient.

Field-Triage, Hospital-Triage and Triage-Assessment: A Literature Review of the Current Phases of Adult Trauma Triage

ABSTRACT

Despite major improvements in the United States trauma system over the past two decades, prehospital trauma triage is a significant challenge. Undertriage is associated with increased mortality, and overtriage results in significant resource overuse. The American College of Surgeons Committee on Trauma benchmarks for undertriage and overtriage are not being met. Many barriers to appropriate field triage exist, including lack of a formal definition for major trauma, absence of a simple and widely applicable triage mode, and emergency medical service adherence to triage protocols. Modern trauma triage systems should ideally be based on the need for intervention rather than injury severity. Future studies should focus on identifying the ideal definition for major trauma and creating triage models that can be easily deployed. This narrative review article presents challenges and potential solutions for prehospital trauma triage.

External validation of the Dutch prediction model for prehospital triage of trauma patients in South West region of England, United Kingdom

IMPORTANCE

This paper investigates the use of a major trauma prediction model in the UK setting. We demonstrate that application of this model could reduce the number of patients with major trauma being incorrectly sent to non-specialist hospitals. However, more research is needed to reduce over-triage and unnecessary transfer to Major Trauma Centres.

OBJECTIVE

To externally validate the Dutch prediction model for identifying major trauma in a large unselected prehospital population of injured patients in England.

DESIGN

External validation using a retrospective cohort of injured patients who ambulance crews transported to hospitals.

SETTING

South West region of England.

PARTICIPANTS

All patients ≥16 years with a suspected injury and transported by ambulance in the year from February 1, 2017.

EXCLUSION CRITERIA

1) Patients aged ≤15 years; 2) Non-ambulance attendance at hospital with injuries; 3) Death at the scene and; 4) Patients conveyed by helicopter. This study had a census sample of cases available to us over a one year period.

INTERVENTIONS OR EXPOSURES

Tested the accuracy of the prediction model in terms of discrimination, calibration, clinical usefulness, sensitivity and specificity and under- and over triage rates compared to usual triage practices in the South West region.

MAIN OUTCOME MEASURE

Major trauma defined as an Injury Severity Score>15.

RESULTS

A total of 68799 adult patients were included in the external validation cohort. The median age of patients was 72 (i.q.r. 46-84); 55.5% were female; and 524 (0.8%) had an Injury Severity Score>15. The model achieved good discrimination with a C-Statistic 0.75 (95% CI, 0.73 - 0.78). The maximal specificity of 50% and sensitivity of 83% suggests the model could improve undertriage rates at the expense of increased overtriage rates compared with routine trauma triage methods used in the South West, England.

CONCLUSIONS AND RELEVANCE

The Dutch prediction model for identifying major trauma could lower the undertriage rate to 17%, however it would increase the overtriage rate to 50% in this United Kingdom cohort. Further prospective research is needed to determine whether the model can be practically implemented by paramedics and is cost-effective.

ISS is not an appropriate tool to estimate overtriage

PURPOSE

The aim of this work is to study a cohort of patients of ISS < 15 admitted to a TC, and to determine the number of patients that ultimately benefited from the skills and resources specific of a level 1 trauma center.

METHODS

Retrospective study from a prospective cohort of patients admitted to TC (Beaujon Hospital, APHP) for suspected severe trauma from January 2011 to December 2017. The main outcome criterion was the use of surgery or interventional radiology within the first 24 h after admission of patients with ISS < 15. The secondary outcomes were stratified into severe (mortality, resuscitation care, length of stay in intensive care units) and non-severe criteria (mild head injury, hospital discharge or transfer within 24 h).

RESULTS

Of 3035 patients admitted during the study period, 1409 with an ISS < 15 were included, corresponding to a theoretical overtriage rate of 46.4%. Among these, 611 patients (43.4%) underwent emergency intervention within the first 24 h (586 surgical interventions, 19 direct transfers to the operating theater and 6 acts of interventional radiology), 238 (16.9%) of patients presented with severe and 531 (38%) with non-severe outcome criteria.

CONCLUSION

This work demonstrates that in a cohort of patients classified as ISS < 15 admitted to a TC, a considerable amount of TC-specific resources are required, and patients present with severe outcome criteria despite being classified as overtriaged. These results suggest that triage of trauma patients should be based on resource use and clinical outcome rather than anatomic criteria.

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

Background

Although multiple prediction models have been developed to predict hospital admission to emergency departments (EDs) to address overcrowding and patient safety, only a few studies have examined prediction models for prehospital use. Development of institution-specific prediction models is feasible in this age of data science, provided that predictor-related information is readily collectable.

Objective

We aimed to develop a hospital admission prediction model based on patient information that is commonly available during ambulance transport before hospitalization.

Methods

Patients transported by ambulance to our ED from April 2018 through March 2019 were enrolled. Candidate predictors were age, sex, chief complaint, vital signs, and patient medical history, all of which were recorded by emergency medical teams during ambulance transport. Patients were divided into two cohorts for derivation (3601/5145, 70.0%) and validation (1544/5145, 30.0%). For statistical models, logistic regression, logistic lasso, random forest, and gradient boosting machine were used. Prediction models were developed in the derivation cohort. Model performance was assessed by area under the receiver operating characteristic curve (AUROC) and association measures in the validation cohort.

Results

Of 5145 patients transported by ambulance, including deaths in the ED and hospital transfers, 2699 (52.5%) required hospital admission. Prediction performance was higher with the addition of predictive factors, attaining the best performance with an AUROC of 0.818 (95% CI 0.792-0.839) with a machine learning model and predictive factors of age, sex, chief complaint, and vital signs. Sensitivity and specificity of this model were 0.744 (95% CI 0.716-0.773) and 0.745 (95% CI 0.709-0.776), respectively.

Conclusions

For patients transferred to EDs, we developed a well-performing hospital admission prediction model based on routinely collected prehospital information including chief complaints.

Trauma-team-activation in Germany: how do emergency service professionals use the activation due to trauma mechanism? Results from a nationwide survey

BACKGROUND

Trauma team activation (TTA) requires significant human and financial resources. The implemented German guidelines reduced the mortality of severe injured patients significantly over the last decade. Up to now there is no two-tier trauma team activation protocol in Germany. A two-tier TTA [often activated due to trauma mechanism (TM)] is thought to be a reasonable way to maintain patient safety while increasing cost efficiency.

METHODS

We created an online survey addressed at the Emergency Medical Service in Germany to conduct a cross-sectional study. Both physicians and rescue service professionals (RSPs) were included. A minimum of 1550 participants answered questions in 4 different categories concerning the aspects of limited-TTA (L-TTA). Case studies were presented to evaluate the usage of TTA due to TM in the daily routine.

RESULTS

Eighty percent (n:1233) of the respondents wish for a possibility to activate a limited trauma team. Seventy-two percent (n: 1109) of the participants consider a L-TTA due to TM to be adequate. There were significant differences (p < 0.05) in the assessment and opinion on L-TTA among physicians and RSPs as well as different medical professions. The evaluated case studies showed diverse answers: depending on the profession, the same patient was ranked as severely injured by 54% and as minorly injured by 46% of the 1550 participants.

CONCLUSIONS

Members of the German Emergency Medical Service call for a two-tier TTA-protocol. Up to now we cannot fully recommend an automatic reduction of the trauma team when activated due to TM in Germany with the guidelines implemented. The profession might affect the L-TTA-behavior. Criteria for a L-TTA in Germany have to be defined and evaluated.

LEVEL OF EVIDENCE

IV, cross-sectional study.

The impact of the Trauma Triage App on pre-hospital trauma triage: design and protocol of the stepped-wedge, cluster-randomized TESLA trial

BACKGROUND

Field triage of trauma patients is crucial to get the right patient to the right hospital within a particular time frame. Minimization of undertriage, overtriage, and interhospital transfer rates could substantially reduce mortality rates, life-long disabilities, and costs. Identification of patients in need of specialized trauma care is predominantly based on the judgment of Emergency Medical Services professionals and a pre-hospital triage protocol. The Trauma Triage App is a smartphone application that includes a prediction model to aid Emergency Medical Services professionals in the identification of patients in need of specialized trauma care. The aim of this trial is to assess the impact of this new digital approach to field triage on the primary endpoint undertriage.

METHODS

The Trauma triage using Supervised Learning Algorithms (TESLA) trial is a stepped-wedge cluster-randomized controlled trial with eight clusters defined as Emergency Medical Services regions. These clusters are an integral part of five inclusive trauma regions. Injured patients, evaluated on-scene by an Emergency Medical Services professional, suspected of moderate to severe injuries, will be assessed for eligibility. This unidirectional crossover trial will start with a baseline period in which the default pre-hospital triage protocol is used, after which all clusters gradually implement the Trauma Triage App as an add-on to the existing triage protocol. The primary endpoint is undertriage on patient and cluster level and is defined as the transportation of a severely injured patient (Injury Severity Score ≥ 16) to a lower-level trauma center. Secondary endpoints include overtriage, hospital resource use, and a cost-utility analysis.

DISCUSSION

The TESLA trial will assess the impact of the Trauma Triage App in clinical practice. This novel approach to field triage will give new and previously undiscovered insights into several isolated components of the diagnostic strategy to get the right trauma patient to the right hospital. The stepped-wedge design allows for within and between cluster comparisons.

TRIAL REGISTRATION

Netherlands Trial Register, NTR7243. Registered 30 May 2018, https://www.trialregister.nl/trial/7038.

Accuracy of pre-hospital trauma triage and field triage decision rules in children (P2-T2 study): an observational study

BACKGROUND

Adequate pre-hospital trauma triage is crucial to enable optimal care in inclusive trauma systems. Transport of children in need of specialised trauma care to lower-level trauma centres is associated with adverse patient outcomes. We aimed to evaluate the diagnostic accuracy of paediatric field triage based on patient destination and triage tools.

METHODS

We did a multisite observational study (P2-T2) of all children (aged <16 years) transported with high priority by ambulance from the scene of injury to any emergency department in seven of 11 inclusive trauma regions in the Netherlands. Diagnostic accuracy based on the initial transport destination was evaluated in terms of undertriage rate (ie, the proportion of patients in need of specialised trauma care who were initially transported to a lower-level paediatric or adult trauma centre) and overtriage rate (ie, the proportion of patients not requiring specialised trauma care who were transported to a level-I [highest level] paediatric trauma centre). The Dutch National Protocol of Ambulance Services and Field Triage Decision Scheme triage protocols were externally validated using data from this cohort against an anatomical (Injury Severity Score [ISS] ≥16) and a resource-based reference standard.

FINDINGS

Between Jan 1, 2015, and Dec 31, 2017, 12 915 children (median age 10·3 years, IQR 4·2-13·6) were transported to the emergency department with injuries. 4091 (31·7%) patients were admitted to hospital, of whom 129 (3·2%) patients had an ISS of 16 or greater and 227 (5·5%) patients used critical resources within a limited timeframe. Ten patients died within 24 h of arrival at the emergency department. Based on the primary reference standard (ISS ≥16), the undertriage rate was 16·3% (95% CI 10·8-23·7) and the overtriage rate was 21·2% (20·5-22·0). The National Protocol of Ambulance Services had a sensitivity of 53·5% (95% CI 43·9-62·9) and a specificity of 94·0% (93·4-94·6), and the Field Triage Decision Scheme had a sensitivity of 64·5% (54·1-74·1) and a specificity of 84·3% (83·1-85·5).

INTERPRETATION

Too many children in need of specialised care were transported to lower-level paediatric or adult trauma centres, which is associated with increased mortality and morbidity. Current protocols cannot accurately discriminate between patients at low and high risk, and highly sensitive and child-specific triage tools need to be developed to ensure the right patient is transported to the right hospital.

FUNDING

The Netherlands Organisation for Health Research and Development, Innovation Fund Health Insurers.

Using machine-learning risk prediction models to triage the acuity of undifferentiated patients entering the emergency care system: a systematic review

BACKGROUND

The primary objective of this review is to assess the accuracy of machine learning methods in their application of triaging the acuity of patients presenting in the Emergency Care System (ECS). The population are patients that have contacted the ambulance service or turned up at the Emergency Department. The index test is a machine-learning algorithm that aims to stratify the acuity of incoming patients at initial triage. This is in comparison to either an existing decision support tool, clinical opinion or in the absence of these, no comparator. The outcome of this review is the calibration, discrimination and classification statistics.

METHODS

Only derivation studies (with or without internal validation) were included. MEDLINE, CINAHL, PubMed and the grey literature were searched on the 14th December 2019. Risk of bias was assessed using the PROBAST tool and data was extracted using the CHARMS checklist. Discrimination (C-statistic) was a commonly reported model performance measure and therefore these statistics were represented as a range within each machine learning method. The majority of studies had poorly reported outcomes and thus a narrative synthesis of results was performed.

RESULTS

There was a total of 92 models (from 25 studies) included in the review. There were two main triage outcomes: hospitalisation (56 models), and critical care need (25 models). For hospitalisation, neural networks and tree-based methods both had a median C-statistic of 0.81 (IQR 0.80-0.84, 0.79-0.82). Logistic regression had a median C-statistic of 0.80 (0.74-0.83). For critical care need, neural networks had a median C-statistic of 0.89 (0.86-0.91), tree based 0.85 (0.84-0.88), and logistic regression 0.83 (0.79-0.84).

CONCLUSIONS

Machine-learning methods appear accurate in triaging undifferentiated patients entering the Emergency Care System. There was no clear benefit of using one technique over another; however, models derived by logistic regression were more transparent in reporting model performance. Future studies should adhere to reporting guidelines and use these at the protocol design stage.

REGISTRATION AND FUNDING

This systematic review is registered on the International prospective register of systematic reviews (PROSPERO) and can be accessed online at the following URL: https://www.crd.york.ac.uk/PROSPERO/display_record.php?ID=CRD42020168696This study was funded by the NIHR as part of a Clinical Doctoral Research Fellowship.

External validation of a prediction model for pain and functional outcome after elective lumbar spinal fusion

OBJECTIVE

Patient-reported outcome measures following elective lumbar fusion surgery demonstrate major heterogeneity. Individualized prediction tools can provide valuable insights for shared decision-making. We externally validated the spine surgical care and outcomes assessment programme/comparative effectiveness translational network (SCOAP-CERTAIN) model for prediction of 12-month minimum clinically important difference in Oswestry Disability Index (ODI) and in numeric rating scales for back (NRS-BP) and leg pain (NRS-LP) after elective lumbar fusion.

METHODS

Data from a prospective registry were obtained. We calculated the area under the curve (AUC), calibration slope and intercept, and Hosmer-Lemeshow values to estimate discrimination and calibration of the models.

RESULTS

We included 100 patients, with average age of 50.4 ± 11.4 years. For 12-month ODI, AUC was 0.71 while the calibration intercept and slope were 1.08 and 0.95, respectively. For NRS-BP, AUC was 0.72, with a calibration intercept of 1.02, and slope of 0.74. For NRS-LP, AUC was 0.83, with a calibration intercept of 1.08, and slope of 0.95. Sensitivity ranged from 0.64 to 1.00, while specificity ranged from 0.38 to 0.65. A lack of fit was found for all three models based on Hosmer-Lemeshow testing.

CONCLUSIONS

The SCOAP-CERTAIN tool can accurately predict which patients will achieve favourable outcomes. However, the predicted probabilities-which are the most valuable in clinical practice-reported by the tool do not correspond well to the true probability of a favourable outcome. We suggest that any prediction tool should first be externally validated before it is applied in routine clinical practice. These slides can be retrieved under Electronic Supplementary Material.

Development and validation of a novel prediction model to identify patients in need of specialized trauma care during field triage: design and rationale of the GOAT study

BACKGROUND

Adequate field triage of trauma patients is crucial to transport patients to the right hospital. Mistriage and subsequent interhospital transfers should be minimized to reduce avoidable mortality, life-long disabilities, and costs. Availability of a prehospital triage tool may help to identify patients in need of specialized trauma care and to determine the optimal transportation destination.

METHODS

The GOAT (Gradient Boosted Trauma Triage) study is a prospective, multi-site, cross-sectional diagnostic study. Patients transported by at least five ground Emergency Medical Services to any receiving hospital within the Netherlands are eligible for inclusion. The reference standards for the need of specialized trauma care are an Injury Severity Score ≥ 16 and early critical resource use, which will both be assessed by trauma registrars after the final diagnosis is made. Variable selection will be based on ease of use in practice and clinical expertise. A gradient boosting decision tree algorithm will be used to develop the prediction model. Model accuracy will be assessed in terms of discrimination (c-statistic) and calibration (intercept, slope, and plot) on individual participant's data from each participating cluster (i.e., Emergency Medical Service) through internal-external cross-validation. A reference model will be externally validated on each cluster as well. The resulting model statistics will be investigated, compared, and summarized through an individual participant's data meta-analysis.

DISCUSSION

The GOAT study protocol describes the development of a new prediction model for identifying patients in need of specialized trauma care. The aim is to attain acceptable undertriage rates and to minimize mortality rates and life-long disabilities.