RESUMEN

BACKGROUND: In the recent years, National Early Warning Score2 (NEWS2) is utilized to predict early on, the worsening of clinical status in patients. To this date the predictive accuracy of National Early Warning Score (NEWS2), Revised Trauma Score (RTS), and Trauma and injury severity score (TRISS) regarding the trauma patients' mortality rate have not been compared. Therefore, the objective of this study is comparing NEWS2, TRISS, and RTS in predicting mortality rate in trauma patients based on prehospital data set. METHODS: This cross-sectional retrospective diagnostic study performed on 6905 trauma patients, of which 4191 were found eligible, referred to the largest trauma center in southern Iran, Shiraz, during 2022-2023 based on their prehospital data set in order to compare the prognostic power of NEWS2, RTS, and TRISS in predicting in-hospital mortality rate. Patients are divided into deceased and survived groups. Demographic data, vital signs, and GCS were obtained from the patients and scoring systems were calculated and compared between the two groups. TRISS and ISS are calculated with in-hospital data set; others are based on prehospital data set. RESULTS: A total of 129 patients have deceased. Age, cause of injury, length of hospital stay, SBP, RR, HR, temperature, SpO2, and GCS were associated with mortality (p-value < 0.001). TRISS and RTS had the highest sensitivity and specificity respectively (77.52, CI 95% [69.3-84.4] and 93.99, CI 95% [93.2-94.7]). TRISS had the highest area under the ROC curve (0.934) followed by NEWS2 (0.879), GCS (0.815), RTS (0.812), and ISS (0.774). TRISS and NEWS were superior to RTS, GCS, and ISS (p-value < 0.0001). CONCLUSION: This novel study compares the accuracy of NEWS2, TRISS, and RTS scoring systems in predicting mortality rate based on prehospital data. The findings suggest that all the scoring systems can predict mortality, with TRISS being the most accurate of them, followed by NEWS2. Considering the time consumption and ease of use, NEWS2 seems to be accurate and quick in predicting mortality based on prehospital data set.

Asunto(s)

RESUMEN

The advancement of intelligent road systems in developing countries poses unique challenges in identifying risk factors and implementing safety strategies. The variability of factors affecting crash injury severity leads to different risks across levels of roadway smartness, especially in hazardous terrains, complicating the adaptation of smart technologies. Therefore, this study investigates the temporal instability of factors affecting injury severities in crashes across various terrains, with a focus on the evolution of road smartness. Crash data from selected complex terrain regions in Shaanxi Province during smart road adaptation were used, and categorized into periods before, during, and after smart road implementations. A series of mixed logit models were employed to account for unobserved heterogeneity in mean and variance, and likelihood ratio tests were conducted to assess the spatio-temporal instability of model parameters across different topographic settings and smart processes. Moreover, a comparison between partially constrained and unconstrained temporal modeling approaches was made. The findings reveal significant differences in injury severity determinants across terrain conditions as roadway intelligence progressed. On the other hand, certain factors like pavement damage, truck and pedestrian involvement were identified that had relatively stable effects on crash injury severities. Out-of-sample predictions further emphasize the need for modeling across terrain and roadway development stages. These insights are crucial for developing tailored safety measures for smart road retrofitting in different terrain conditions, thereby supporting the transition towards smarter road systems in developing regions.

Asunto(s)

RESUMEN

Street intersection crashes often involve two parties: either two vehicles hitting each other (i.e., a vehicle-vehicle crash) or a vehicle colliding with a pedestrian (i.e., a vehicle-pedestrian crash). In such crashes, the severity of injuries can vary considerably between the parties involved. It is necessary to understand the injuries of both parties simultaneously to identify the causality of a vehicle-pedestrian or two-vehicle crash. While the latent class ordinal model has been used in crash severity studies to capture heterogeneity in crash propensity, most of these studies are univariate, which is inappropriate for crashes involving two parties. This study proposes a latent class parameterized correlation bivariate generalized ordered probit (LCp-BGOP) model to examine 32,308 vehicle-vehicle and vehicle-pedestrian crashes at intersections in Taipei City, Taiwan. The model parameterizes thresholds and within-crash correlations of crash severity involving two parties and classifies these crashes into two distinct risk groups: the "Ordinary Crash Severity" (OCS) group and the "High Crash Severity" (HCS) group. The OCS group is mainly two-vehicle crashes involving motorcycles. The HCS group comprises vulnerable road users such as pedestrians and cyclists, mainly in mixed traffic with heavy volumes. The results also show that the effects of party-specific factors contributing to injury severity are greater than those of generic factors. Our study provides invaluable insight into intersection crashes, helping to reduce the severity of injuries in vehicle-vehicle and vehicle-pedestrian crashes.

Asunto(s)

RESUMEN

Traumatic anorectal injuries are rare in pediatric surgical practice. Only several similar cases are described in the world literature. This causes no generally accepted algorithms and tactics for these patients. We demonstrate successful surgical treatment of combined trauma of the rectum and bladder in a child. A 13-year-old boy was hospitalized after the child sat on the leg of an overturned chair. No evidence of penetrating abdominal injury was revealed. The boy underwent sigmoidoscopy under general anesthesia. We found a lacerated wound of anterior wall of the rectum measuring 1/3 of its diameter with damage to posterior wall of the bladder. Diagnostic laparoscopy revealed intact abdominal cavity. Wall defects were sutured (bladder wound was sutured during traditional cystotomy), and we formed protective separate double-barreled sigmostomy. In 3 months after discharge, the child was hospitalized for cystography and fistulography with subsequent closure of stoma. In long-term postoperative period (6 months), the quality of life is satisfactory. There is no pain and disturbances of urination.

Asunto(s)

RESUMEN

BACKGROUND: There is no standardized grading system for pediatric female genital trauma (PFGT), so patients may have over-utilization of resources relative to injury severity. We described current treatment patterns and outcomes at a high-volume trauma center, developed a novel PFGT grading system, and proposed algorithm for management of PFGT. METHODS: We retrospectively reviewed female patients <19 years presenting with genital trauma to our Level 1 pediatric trauma center between 1/2018-12/2022. A novel grading system developed by pediatric surgery and pediatric gynecology was retrospectively applied to injuries. Patient demographics, injury characteristics, types of intervention, and need for anesthesia were recorded. Outcomes were compared between grades of injury with Kruskal-Wallis tests. RESULTS: Among 353 patients, median age was 6.4 years. Half of patients had grade 1 or 2 injuries, of which 6% required suture repair. 15% of patients had grade 5 or 6 injuries, 75% of whom required suture repair. General anesthesia was used for 83% of all patients undergoing repair. 18% of patients who underwent general anesthesia did not need suture repair. Of patients who were brought to the operating room, median operative duration varied by grade and was 15.0 min for all injuries, 7.0 min for both grade 1 and 2 injuries, and 22.0 and 37.0 min for grade 5 and 6 injuries, respectively (p < 0.0001). CONCLUSIONS: Based on our novel grading system, we propose an algorithm for managing PFGT. Grade 1 and 2 injuries rarely require suture repair and can often be managed without surgical consultation. We recommend surgical consultation for higher grade injuries, however given typically short operative times, repair with bedside sedation should be strongly considered when resources allow. LEVEL OF EVIDENCE: IV.

Asunto(s)

RESUMEN

INTRODUCTION: Mortality due to injuries disproportionately impact low income countries. Knowledge of who is at risk of poor outcomes is critical to guide resource allocation and prioritization of severely injured. Kampala Trauma Score (KTS), developed in 1996 and last modified in 2002 as KTS II, is still widely being used to predict injury outcomes in resource-limited settings with no further revisions in the past two decades, despite ongoing criticism of some of its parameters. The New Trauma Score (NTS), a recent development in 2017, has shown potential in mortality prediction, but a dearth of evidence exist regarding its performance in the African population. OBJECTIVES: To compare NTS to the modified Kampala Trauma Score (KTS II) in the prediction of 30-day mortality, and injury severity amongst patients sustaining road traffic crashes in Ugandan low-resource settings. METHODS: Multi-center prospective cohort study of patients aged 15 years and above. Of the 194 participants, 85.1% were males with a mean age of 31.7 years. NTS and KTS II were determined for each participant within 30-minutes of admission and followed-up for 30 days to determine their injury outcomes. The sensitivity, specificity, and area under receiver operating characteristics curve (AUC) for predicting mortality were compared between the two trauma scores using SPSS version 22. Ethical clearance: Research and Ethics Committee of Kampala International University Western Campus (Ref No: KIU-2022-125). RESULTS: The injury severity classifications based on NTS vs. KTS II were mild (55.7% vs. 25.8%), moderate (29.9% vs. 30.4%), and severe (14.4% vs. 43.8%). The mortality rates for each injury severity category based on NTS vs. KTS II were mild (0.9% v 0%), moderate (20.7% vs. 5.1%), and severe (50% vs. 28.2%). The AUC was 0.87 for NTS (95% CI 0.808-0.931) vs. 0.86 (95% CI 0.794-0.919) for KTS II respectively. The sensitivity of NTS vs. KTS II in predicting mortality was 92.6% (95% CI: 88.9-96.3) vs. 70.4% (95% CI: 63.0-77.8) while the specificity was 70.7% (95% CI: 64.2-77.2) vs. 78.4% (95% CI: 72.1-84.7) at cut off points of 17 for NTS and 6 for KTS II respectively. CONCLUSIONS: NTS was more sensitive but its specificity for purposes of 30-day mortality prediction was lower compared to KTS II. Thus, in low-resourced trauma environment where time constraints and pulse oximeters are of concern, KTS II remains superior to NTS.

Asunto(s)

RESUMEN

Using data from a developing country, the current study develops a copula-based joint modeling framework to study crash type and driver injury severity as two dimensions of the severity process. To be specific, a copula-based multinomial logit model (for crash type) and generalized ordered logit model (for driver severity) is estimated in the study. The data for our analysis is drawn from Bangladesh for the years of 2000 to 2015. Given the presence of multiple years of data, we develop a novel spline variable generation approach that facilitates easy testing of variation in parameters across time in crash type and severity components. A comprehensive set of independent variables including driver and vehicle characteristics, roadway attributes, environmental and weather information, and temporal factors are considered for the analysis. The model results identify several important variables (such as driving under the influence of drug and alcohol, speeding, vehicle type, maneuvering, vehicle fitness, location type, road class, road geometry, facility type, surface quality, time of the day, season, and light conditions) affecting crash type and severity while also highlighting the presence of temporal instability for a subset of parameters. The superior model performance was further highlighted by testing its performance using a holdout sample. Further, an elasticity exercise illustrates the influence of the exogenous variables on crash type and injury severity dimensions. The study findings can assist policy makers in adopting appropriate strategies to make roads safer in developing countries.

Asunto(s)

RESUMEN

Rear-end (RE) crashes are notably prevalent and pose a substantial risk on freeways. This paper explores the correlation between speed difference among the following and leading vehicles (Δν) and RE crash risk. Three joint models, comprising both uncorrelated and correlated joint random-parameters bivariate probit (RPBP) approaches (statistical methods) and a cross-stitch multilayer perceptron (CS-MLP) network (a data-driven method), were estimated and compared against three separate models: Support Vector Machines (SVM), eXtreme Gradient Boosting (XGBoost), and MLP networks (all data-driven methods). Data on 15,980 two-vehicle RE crashes were collected over a two-year period, from January 1, 2021, to December 31, 2022, considering two possible levels of injury severity: no injury and injury/fatality for both drivers of following and leading vehicles. The comparative performance analysis demonstrates the superior predictive capability of the CS-MLP network over the uncorrelated/correlated joint RPBP model, SVM, XGBoost, and MLP networks in terms of recall, F-1 Score, and AUC. Significantly, numerous shared variables influence the injury severity outcomes for the following and leading vehicles across both statistical and data-driven approaches. Among these factors, the following vehicle (a truck) and the leading vehicle (a passenger car) demonstrate contrasting effects on the injury severity outcomes for both vehicles. Furthermore, the SHapley Additive exPlanations (SHAP) values from the CS-MLP network visually show the relationship between Δν and injury severity, revealing non-linear trends unlike the average effects shown by statistical methods. They indicate that the least injury outcomes for both following and leading vehicles occurs at a Δν of 0 to 10 mph, matching observed patterns in RE crash data. Additionally, a marked variation in the trend of SHAP values for the two vehicles is noted as the speed difference increases. Therefore, the findings affirm the superior performance of joint model development and substantiate the non-linear impacts of speed difference on injury outcomes. The adoption of dynamic speed control measures is recommended to mitigate the injury outcomes involved in two-vehicle RE crashes.

Asunto(s)

RESUMEN

Vehicles equipped with automated driving capabilities have shown potential to improve safety and operations. Advanced driver assistance systems (ADAS) and automated driving systems (ADS) have been widely developed to support vehicular automation. Although the studies on the injury severity outcomes that involve automated vehicles are ongoing, there is limited research investigating the difference between injury severity outcomes for the ADAS and ADS equipped vehicles. To ensure a comprehensive analysis, a multi-source dataset that includes 1,001 ADAS crashes (SAE Level 2 vehicles) and 548 ADS crashes (SAE Level 4 vehicles) is used. Two random parameters multinomial logit models with heterogeneity in the means of random parameters are considered to gain a better understanding of the variables impacting the crash injury severity outcomes for the ADAS (SAE Level 2) and ADS (SAE Level 4) vehicles. It was found that while 67 percent of crashes involving the ADAS equipped vehicles in the dataset took place on a highway, 94 percent of crashes involving ADS took place in more urban settings. The model estimation results also reveal that the weather indicator, driver type indicator, differences in the system sophistication that are captured by both manufacture year and high/low mileage as well as rear and front contact indicators all play a role in the crash injury severity outcomes. The results offer an exploratory assessment of safety performance of the ADAS and ADS equipped vehicles using the real-world data and can be used by the manufacturers and other stakeholders to dictate the direction of their deployment and usage.

Asunto(s)

RESUMEN

BACKGROUND: Although road traffic injuries and deaths have decreased globally, there is substantial national and sub-national heterogeneity, particularly in low- and middle-income countries (LMICs). Ghana is one of few countries in Africa collecting comprehensive, spatially detailed data on motor vehicle collisions (MVCs). This data is a critical step towards improving roadway safety, as accurate and reliable information is essential for devising targeted countermeasures. METHODS: Here, we analyze 16 years of police-report data using emerging hot spot analysis in ArcGIS to identify hot spots with trends of increasing injury severity (a weighted composite measure of MVCs, minor injuries, severe injuries, and deaths), and counts of injuries, severe injuries, and deaths along major roads in urban and rural areas of Ghana. RESULTS: We find injury severity index sums and minor injury counts are significantly decreasing over time in Ghana while severe injury and death counts are not, indicating the latter should be the focus for road safety efforts. We identify new, consecutive, intensifying, and persistent hot spots on 2.65% of urban roads and 4.37% of rural roads. Hot spots are intensifying in terms of severity and frequency on major roads in rural areas. CONCLUSIONS: A few key road sections, particularly in rural areas, show elevated levels of road traffic injury severity, warranting targeted interventions. Our method for evaluating spatiotemporal trends in MVC, road traffic injuries, and deaths in a LMIC includes sufficient detail for replication and adaptation in other countries, which is useful for targeting countermeasures and tracking progress.

Asunto(s)

RESUMEN

BACKGROUND: Given the huge impact of trauma on hospital systems around the world, several attempts have been made to develop predictive models for the outcomes of trauma victims. The most used, and in many studies most accurate predictive model, is the "Trauma Score and Injury Severity Score" (TRISS). Although it has proven to be fairly accurate and is widely used, it has faced criticism for its inability to classify more complex cases. In this study, we aimed to develop machine learning models that better than TRISS could predict mortality among severely injured trauma patients, something that has not been studied using data from a nationwide register before. METHODS: Patient data was collected from the national trauma register in Sweden, SweTrau. The studied period was from the 1st of January 2015 to 31st of December 2019. After feature selection and multiple imputation of missing data three machine learning (ML) methods (Random Forest, eXtreme Gradient Boosting, and a Generalized Linear Model) were used to create predictive models. The ML models and TRISS were then tested on predictive ability for 30-day mortality. RESULTS: The ML models were well-calibrated and outperformed TRISS in all the tested measurements. Among the ML models, the eXtreme Gradient Boosting model performed best with an AUC of 0.91 (0.88-0.93). CONCLUSION: This study showed that all the developed ML-based prediction models were superior to TRISS for the prediction of trauma mortality.

Asunto(s)

RESUMEN

This research utilizes data collected in Florida to examine the differentials in injury severities among single-vehicle drivers involved in work zone-related incidents, specifically focusing on the distinctions between rural and urban areas. The study encompasses a four-year period (2016-2019) of crash dataset. A likelihood ratio test was performed to examine model estimate's temporal consistency in datasets from rural and urban areas across several time periods throughout the year. Separate statistical models were estimated for both rural and urban datasets to understand different driver injury severity outcomes (no injury, minor injury, and severe injury) using a mixed logit approach with possible heterogeneity in mean and variance of random parameters. Out-of-sample simulations were conducted to see the effect of different parameter changes on injury severity probabilities in rural and urban work zone crashes. Over multiple years, various years in both rural and urban models have generated statistically significant random factors that effectively capture the presence of heterogeneity in means, accounting for unobservable variations within the data. Clear evidence of factors such as speed limits, work zone type, and traffic volume affecting the work zone injury severities were found to vary significantly between rural and urban work zone areas. However, despite this difference, rural and urban work zones share common safety problems and countermeasures such as driver education, improved signage, and appropriate traffic controls; combining ITS technologies and enhanced law enforcement can help mitigate crash severity in urban and rural work zone areas.

Asunto(s)

RESUMEN

BACKGROUND: The Abbreviated Burn Severity Index (ABSI) is a five-variable scale to help evaluate burn severity upon initial assessment. As other studies have been conducted with comparatively small patient populations, the purpose of this study is to revalidate the prognostic relevance of the ABSI in our selected population (N = 1193) 4 decades after its introduction, considering the progress in the treatment of severe burn injuries over the past decades. In addition, we evaluate whether comorbidities influence the survival probability of severely burned patients. METHODS: This retrospective study presents data from the Center for Severely Burned Patients of the General Hospital in Vienna. We included 1193 patients for over 20 years. Regression models were used to describe the prognostic accuracy of the ABSI. RESULTS: The ABSI can still be used as a prognostic factor for the probability of survival of severely burned patients. The odds of passing increases by a factor of 2.059 for each unit increase in the ABSI with an area under the curve value of 0.909. Over time, the likelihood of survival increased. The existence of chronic kidney disease negatively impacts the survival probability of severely burned patients. CONCLUSION: The ABSI can still be used to provide accurate information about the chances of survival of severely burned patients; however, further exploration of the impact of chronic kidney disease on the survival probability and adding variables to the ABSI scale should be considered. The probability of survival has increased over the last 20 years.

Asunto(s)

RESUMEN

BACKGROUND: Paediatric closed abdominal trauma is common, however, its severity and influence on survival are difficult to determine. No prognostic score integrating abdominal involvement exists to date in paediatrics. OBJECTIVES: To evaluate the severity and short-term and medium-term prognosis of closed abdominal trauma in children, and the performance of severity scores in predicting mortality. DESIGN: Retrospective, cohort, observational study. SETTING AND PARTICIPANTS: Patients aged 0 to 18 years presenting at the trauma room of a French paediatric Level I Trauma Centre over the period 2015 to 2019 with an isolated closed abdominal trauma or as part of a polytrauma. MAIN OUTCOMES: Primary outcome was the six months mortality. Secondary outcomes were related complications and therapeutic interventions, and performance for predicting mortality of the scores listed. Paediatric Trauma Score (PTS), Revised Trauma Score (RTS), Shock Index Paediatric Age-adjusted (SIPA) score, Reverse shock index multiplied by Glasgow Coma Scale score (rSIG), Base Deficit, International Normalised Ratio, and Glasgow Coma Scale (BIG), Injury Severity Score (ISS) and Trauma Score and the Injury Severity (TRISS) score. DATA COLLECTION: Data collected include clinical, biological and CT scan data at admission, first 24 h management and prognosis. The PTS, RTS, SIPA, rSIG, BIG and ISS scores were calculated and mortality was predicted according to BIG score and TRISS methodology. RESULTS: Of 1145 patients, 149 met the inclusion criteria and 12 (8.1%) died. Of the 12 deceased patients, 11 (91.7%) presented with severe head injury, 11 (91.7%) had blood products transfusion and 7 received tranexamic acid. ROC curves analysis concluded that PTS, RTS, rSIG and BIG scores accurately predict mortality in paediatric closed abdominal trauma with AUCs at least 0.92. The BIG score offered the best predictive performance for predicting mortality at a threshold of 24.8 [sensitivity 90%, specificity 92%, negative-predictive value (NPV) 99%, area under the curve (AUC) 0.93]. CONCLUSION: PEVALPED is the first French study to evaluate the prognosis of paediatric closed abdominal trauma. The use of PTS, rSIG and BIG scores are relevant from the acute phase and the pathophysiological interest and accuracy of the BIG score make it a powerful tool for predicting mortality of closed abdominal trauma in children.

Asunto(s)

RESUMEN

PURPOSE: The classification of trauma patients in emergency settings is a constant challenge for physicians. However, the Injury Severity Score (ISS) is widely used in developed countries, it may be difficult to perform it in low- and middle-income countries (LMIC). As a result, the ISS was calculated using an estimated methodology that has been described and validated in a high-income country previously. In addition, a simple scoring tool called the Kampala Trauma Score (KTS) was developed recently. The aim of this study was to compare the diagnostic accuracy of KTS and estimated ISS (eISS) in order to achieve a valid and efficient scoring system in our resource-limited setting. METHODS: We conducted a cross-sectional study between December 2020 and March 2021 among the multi-trauma patients who presented at the emergency department of Imam Reza hospital, Tabriz, Iran. After obtaining informed consent, all data including age, sex, mechanism of injury, GCS, KTS, eISS, final outcome (including death, morbidity, or discharge), and length of hospital stay were collected and entered into SPSS version 27.0 and analyzed. RESULTS: 381 multi-trauma patients participated in the study. The area under the curve for prediction of mortality (AUC) for KTS was 0.923 (95%CI: 0.888-0.958) and for eISS was 0.910 (95% CI: 0.877-0.944). For the mortality, comparing the AUCs by the Delong test, the difference between areas was not statistically significant (p value = 0.356). The diagnostic odds ratio (DOR) for the prediction of mortality KTS and eISS were 28.27 and 32.00, respectively. CONCLUSION: In our study population, the KTS has similar accuracy in predicting the mortality of multi-trauma patients compared to the eISS.

Asunto(s)

RESUMEN

Objective: Many current trauma mortality prediction tools are either too intricate or rely on data not readily available during a trauma patient's initial evaluation. Moreover, none are tailored to those necessitating urgent or emergent surgery. Our objective was to design a practical, user-friendly scoring tool using immediately available variables, and then compare its efficacy to the widely-known Revised Trauma Score (RTS). Methods: The adult 2017-2021 Trauma Quality Improvement Program (TQIP) database was queried to identify patients ≥18 years old undergoing any urgent/emergent operation (direct from Emergency Department to operating room). Patients were divided into derivation and validation groups. A three-step methodology was used. First, multiple logistic regression models were created to determine risk of death using only variables available upon arrival. Second, the weighted average and relative impact of each independent predictor was used to derive an easily calculated Immediate Operative Trauma Assessment Score (IOTAS). We then validated IOTAS using AUROC and compared it to RTS. Results: From 249 208 patients in the derivation-set, 14 635 (5.9%) died. Age ≥65, Glasgow Coma Scale score <9, hypotension (SBP <90 mmHg), and tachycardia (>120/min) on arrival were identified as independent predictors for mortality. Using these, the IOTAS was structured, offering scores between 0-8. The AUROC for this was .88. A clear escalation in mortality was observed across scores: from 4.4% at score 1 to 60.5% at score 8. For the validation set (250 182 patients; mortality rate 5.8%), the AUROC remained consistent at .87, surpassing RTS's AUROC of .83. Conclusion: IOTAS is a novel, accurate, and now validated tool that is intuitive and efficient in predicting mortality for trauma patients requiring urgent or emergent surgeries. It outperforms RTS, and thereby may help guide clinicians when determining the best course of action in patient management as well as counseling patients and their families.

Asunto(s)

Asunto(s)

RESUMEN

Background and Objectives: The Taiwan Triage and Acuity Scale (TTAS) is reliable for triaging patients in emergency departments in Taiwan; however, most triage decisions are still based on chief complaints. The reverse-shock index (SI) multiplied by the simplified motor score (rSI-sMS) is a more comprehensive approach to triage that combines the SI and a modified consciousness assessment. We investigated the combination of the TTAS and rSI-sMS for triage compared with either parameter alone as well as the SI and modified SI. Materials and Methods: We analyzed 13,144 patients with trauma from the Taipei Tzu Chi Trauma Database. We investigated the prioritization performance of the TTAS, rSI-sMS, and their combination. A subgroup analysis was performed to evaluate the trends in all clinical outcomes for different rSI-sMS values. The sensitivity and specificity of rSI-sMS were investigated at a cutoff value of 4 (based on previous study and the highest score of the Youden Index) in predicting injury severity clinical outcomes under the TTAS system were also investigated. Results: Compared with patients in triage level III, those in triage levels I and II had higher odds ratios for major injury (as indicated by revised trauma score < 7 and injury severity score [ISS] ≥ 16), intensive care unit (ICU) admission, prolonged ICU stay (≥14 days), prolonged hospital stay (≥30 days), and mortality. In all three triage levels, the rSI-sMS < 4 group had severe injury and worse outcomes than the rSI-sMS ≥ 4 group. The TTAS and rSI-sMS had higher area under the receiver operating characteristic curves (AUROCs) for mortality, ICU admission, prolonged ICU stay, and prolonged hospital stay than the SI and modified SI. The combination of the TTAS and rSI-sMS had the highest AUROC for all clinical outcomes. The prediction performance of rSI-sMS < 4 for major injury (ISS ≥ 16) exhibited 81.49% specificity in triage levels I and II and 87.6% specificity in triage level III. The specificity for mortality was 79.2% in triage levels I and II and 87.4% in triage level III. Conclusions: The combination of rSI-sMS and the TTAS yielded superior prioritization performance to TTAS alone. The integration of rSI-sMS and TTAS effectively enhances the efficiency and accuracy of identifying trauma patients at a high risk of mortality.

Asunto(s)

RESUMEN

OBJECTIVE: The timely stratification of trauma injury severity can enhance the quality of trauma care but it requires intense manual annotation from certified trauma coders. The objective of this study is to develop machine learning models for the stratification of trauma injury severity across various body regions using clinical text and structured electronic health records (EHRs) data. MATERIALS AND METHODS: Our study utilized clinical documents and structured EHR variables linked with the trauma registry data to create 2 machine learning models with different approaches to representing text. The first one fuses concept unique identifiers (CUIs) extracted from free text with structured EHR variables, while the second one integrates free text with structured EHR variables. Temporal validation was undertaken to ensure the models' temporal generalizability. Additionally, analyses to assess the variable importance were conducted. RESULTS: Both models demonstrated impressive performance in categorizing leg injuries, achieving high accuracy with macro-F1 scores of over 0.8. Additionally, they showed considerable accuracy, with macro-F1 scores exceeding or near 0.7, in assessing injuries in the areas of the chest and head. We showed in our variable importance analysis that the most important features in the model have strong face validity in determining clinically relevant trauma injuries. DISCUSSION: The CUI-based model achieves comparable performance, if not higher, compared to the free-text-based model, with reduced complexity. Furthermore, integrating structured EHR data improves performance, particularly when the text modalities are insufficiently indicative. CONCLUSIONS: Our multi-modal, multiclass models can provide accurate stratification of trauma injury severity and clinically relevant interpretations.

Asunto(s)

RESUMEN

Hereby, we aimed to comprehensively compare different scoring systems for pediatric trauma and their ability to predict in-hospital mortality and intensive care unit (ICU) admission. The current registry-based multicenter study encompassed a comprehensive dataset of 6709 pediatric trauma patients aged ≤ 18 years from July 2016 to September 2023. To ascertain the predictive efficacy of the scoring systems, the area under the receiver operating characteristic curve (AUC) was calculated. A total of 720 individuals (10.7%) required admission to the ICU. The mortality rate was 1.1% (n = 72). The most predictive scoring system for in-hospital mortality was the adjusted trauma and injury severity score (aTRISS) (AUC = 0.982), followed by trauma and injury severity score (TRISS) (AUC = 0.980), new trauma and injury severity score (NTRISS) (AUC = 0.972), Glasgow coma scale (GCS) (AUC = 0.9546), revised trauma score (RTS) (AUC = 0.944), pre-hospital index (PHI) (AUC = 0.936), injury severity score (ISS) (AUC = 0.901), new injury severity score (NISS) (AUC = 0.900), and abbreviated injury scale (AIS) (AUC = 0.734). Given the predictive performance of the scoring systems for ICU admission, NTRISS had the highest predictive performance (AUC = 0.837), followed by aTRISS (AUC = 0.836), TRISS (AUC = 0.823), ISS (AUC = 0.807), NISS (AUC = 0.805), GCS (AUC = 0.735), RTS (AUC = 0.698), PHI (AUC = 0.662), and AIS (AUC = 0.651). In the present study, we concluded the superiority of the TRISS and its two derived counterparts, aTRISS and NTRISS, compared to other scoring systems, to efficiently discerning individuals who possess a heightened susceptibility to unfavorable consequences. The significance of these findings underscores the necessity of incorporating these metrics into the realm of clinical practice.

Asunto(s)