TRISS methodology: an inappropriate tool for comparing outcomes between trauma centers

Does the Probability of Survival Calculated by the Trauma and Injury Severity Score Method Accurately Reflect the Severity of Neurotrauma Patients Admitted to Regional Trauma Centers in Korea?

BACKGROUND

Assessing and improving the quality of trauma care is crucial in modern trauma systems and centers. In Korea, evaluations of regional trauma centers are conducted annually to assess and improve trauma management quality. This includes using the Trauma and Injury Severity Score (TRISS) method to calculate the W-score and mortality Observed-to-Expected ratio (O:E ratio), which are used to evaluate the quality of care. We analyzed the potential for overestimation of the probability of survival using TRISS method for patients with neurotrauma, as well as the potential for errors when evaluating and comparing regional trauma centers.

METHODS

We included patients who visited the regional trauma center between 2019 and 2021 and compared their probability of survival of the TRISS method, W-score, mortality O:E ratio, and misclassification rates. The patient groups were further subdivided into smaller subgroups based on age, Glasgow Coma Scale (GCS), and Injury Severity Score, and comparisons were made between the neurotrauma and non-neurotrauma groups within each subgroup.

RESULTS

A total of 4,045 patients were enrolled in the study, with 1,639 of them having neurotrauma. The neurotrauma patient group had a W-score of -0.68 and a mortality O:E ratio of 1.044. The misclassification rate was found to be 13.3%, and patients with a GCS of 8 or less had a higher misclassification rate of 37.4%.

CONCLUSION

The limitations of using the TRISS method for predicting outcomes in patients with severe neurotrauma are exposed in this study. The TRISS methodology demonstrated a high misclassification rate of approximately 40% in subgroups of patients with GCS less than 9, indicating that it may be less reliable in predicting outcomes for severely injured patients with low GCS. Clinicians and researchers should be cautious when using the TRISS method and consider alternative methods to evaluate patient outcomes and compare the quality of care provided by different trauma centers.

Potential of Hematologic Parameters in Predicting Mortality of Patients with Traumatic Brain Injury

Traumatic brain injury (TBI) occurs frequently, and acute TBI requiring surgical treatment is closely related to patient survival. Models for predicting the prognosis of patients with TBI do not consider various factors of patient status; therefore, it is difficult to predict the prognosis more accurately. In this study, we created a model that can predict the survival of patients with TBI by adding hematologic parameters along with existing non-hematologic parameters. The best-fitting model was created using the Akaike information criterion (AIC), and hematologic factors including preoperative hematocrit, preoperative C-reactive protein (CRP), postoperative white blood cell (WBC) count, and postoperative hemoglobin were selected to predict the prognosis. Among several prediction models, the model that included age, Glasgow Coma Scale, Injury Severity Score, preoperative hematocrit, preoperative CRP, postoperative WBC count, postoperative hemoglobin, and postoperative CRP showed the highest area under the curve and the lowest corrected AIC for a finite sample size. Our study showed a new prediction model for mortality in patients with TBI using non-hematologic and hematologic parameters. This prediction model could be useful for the management of patients with TBI.

National trauma system establishment based on implementation of regional trauma centers improves outcomes of trauma care: A follow-up observational study in South Korea

Although South Korea is a high-income country, its trauma system is comparable to low- and middle-income countries with high preventable trauma death rates of more than 30%. Since 2012, South Korea has established a national trauma system based on the implementation of regional trauma centers and improvement of the transfer system; this study aimed to evaluate its effectiveness. We compared the national preventable trauma death rates, transfer patterns, and outcomes between 2015 and 2017. The review of preventable trauma deaths was conducted by multiple panels, and a severity-adjusted logistic regression model was created to identify factors influencing the preventable trauma death rate. We also compared the number of trauma patients transferred to emergency medical institutions and mortality in models adjusted with injury severity scores. The preventable trauma death rate decreased from 2015 to 2017 (30.5% vs. 19.9%, p < 0.001). In the severity-adjusted model, the preventable trauma death risk had a lower odds ratio (0.68, 95% confidence interval: 0.53-0.87, p = 0.002) in 2017 than in 2015. Regional trauma centers received 1.6 times more severe cases in 2017 (according to the International Classification of Diseases Injury Severity Score [ICISS]; 23.1% vs. 36.5%). In the extended ICISS model, the overall trauma mortality decreased significantly from 2.1% (1008/47 806) to 1.9% (1062/55 057) (p = 0.041). The establishment of the national trauma system was associated with significant improvements in the performance and outcomes of trauma care. This was mainly because of the implementation of regional trauma centers and because more severe patients were transferred to regional trauma centers. This study might be a good model for low- and middle-income countries, which lack a trauma system.

Artificial intelligence to predict in-hospital mortality using novel anatomical injury score

The aim of the study is to develop artificial intelligence (AI) algorithm based on a deep learning model to predict mortality using abbreviate injury score (AIS). The performance of the conventional anatomic injury severity score (ISS) system in predicting in-hospital mortality is still limited. AIS data of 42,933 patients registered in the Korean trauma data bank from four Korean regional trauma centers were enrolled. After excluding patients who were younger than 19 years old and those who died within six hours from arrival, we included 37,762 patients, of which 36,493 (96.6%) survived and 1269 (3.4%) deceased. To enhance the AI model performance, we reduced the AIS codes to 46 input values by organizing them according to the organ location (Region-46). The total AIS and six categories of the anatomic region in the ISS system (Region-6) were used to compare the input features. The AI models were compared with the conventional ISS and new ISS (NISS) systems. We evaluated the performance pertaining to the 12 combinations of the features and models. The highest accuracy (85.05%) corresponded to Region-46 with DNN, followed by that of Region-6 with DNN (83.62%), AIS with DNN (81.27%), ISS-16 (80.50%), NISS-16 (79.18%), NISS-25 (77.09%), and ISS-25 (70.82%). The highest AUROC (0.9084) corresponded to Region-46 with DNN, followed by that of Region-6 with DNN (0.9013), AIS with DNN (0.8819), ISS (0.8709), and NISS (0.8681). The proposed deep learning scheme with feature combination exhibited high accuracy metrics such as the balanced accuracy and AUROC than the conventional ISS and NISS systems. We expect that our trial would be a cornerstone of more complex combination model.

RISC II is superior to TRISS in predicting 30-day mortality in blunt major trauma patients in Hong Kong

PURPOSE

Hong Kong (HK) trauma registries have been using the Trauma and Injury Severity Score (TRISS) for audit and benchmarking since their introduction in 2000. We compare the mortality prediction model using TRISS and Revised Injury Severity Classification, version II (RISC II) for trauma centre patients in HK.

METHODS

This was a retrospective cohort study with all five trauma centres in HK. Adult trauma patients with Injury Severity Score (ISS) > 15 suffering from blunt injuries from January 2013 to December 2015 were included. TRISS models using the US and local coefficients were compared with the RISC II model. The primary outcome was 30-day mortality and the area under the receiver operating characteristic curve (AUC) for tested models.

RESULTS

1840 patients were included, of whom 1236/1840 (67%) were male. Median age was 59 years and median ISS was 25. Low falls were the most common mechanism of injury. The 30-day mortality was 23%. RISC II yielded a superior AUC of 0.896, compared with the TRISS models (MTOS: 0.848; PATOS: 0.839; HK: 0.858). Prespecified subgroup analyses showed that all the models performed worse for age ≥ 70, ASA ≥ III, and low falls. RISC II had a higher AUC compared with the TRISS models in all subgroups, although not statistically significant.

CONCLUSION

RISC II was superior to TRISS in predicting the 30-day mortality for Hong Kong adult blunt major trauma patients. RISC II may be useful when performing future audit or benchmarking exercises for trauma in Hong Kong.

A Leg to Stand on: Trauma Center Designation and Association with Rate of Limb Salvage in Patients Suffering Severe Lower Extremity Injury

BACKGROUND

Mangled extremities are one of the most difficult injuries for trauma surgeons to manage. We compare limb salvage rates for a limb-threatening lower extremity injuries managed at Level I vs Level II trauma centers (TCs).

STUDY DESIGN

We identified all adult patients with a limb-threatening injury who underwent primary amputation or limb salvage (LS) using the American College of Surgeons (ACS) Trauma Quality Improvement Program database at ACS Level I vs II TCs between 2007 and 2017. A limb-threatening injury was defined as an open tibial fracture with concurrent arterial injury (Gustilo type IIIc). Multivariable analysis and propensity score matching were performed to minimize confounding by indication.

RESULTS

There were 712 records for analysis; 391 (54.9%) LS performed and 321 (45.1%) underwent amputation. The rate of LS was statistically higher among patients treated at Level I TCs vs those treated at Level II TCs (47.4% vs 34.8%; p = 0.01). Patients with penetrating injuries (13% vs 9.5%; p = 0.046) and tibial/peroneal artery injury (72.9% vs 50.4%; p < 0.001), as opposed to popliteal artery injury (30.8% vs 58.8%; p < 0.001), were more likely to have LS. The risk-adjusted odds of LS was 3.13 times higher at Level I TCs vs Level II TCs (95% CI, 1.59 to 6.34; p = 0.001). Limb salvage rates were significantly higher at Level I TCs compared with Level II TCs (53.0% vs 34.8%; p = 0.004), even after propensity matching.

CONCLUSIONS

In patients with a mangled extremity, limb salvage rates are 50% higher at Level I TCs compared with Level II TCs, independent of case mix and injury severity.

High-Performance Trauma Centers in a Single-State Trauma System : Big Saves or Marginal Gains?

BACKGROUND

Trauma centers with low observed:expected (O:E) mortality ratios are considered high performers; however, it is unknown whether improvements in this ratio are due to a small number of unexpected survivors with high mortality risk (big saves) or a larger number of unexpected survivors with moderate mortality risk (marginal gains). We hypothesized that the highest-performing centers achieve that status via larger numbers of unexpected survivors with moderate mortality risk.

METHODS

We calculated O:E ratios for trauma centers in Pennsylvania for 2016 using a risk-adjusted mortality model. We identified high and low performers as centers whose 95% CIs did not cross 1. We visualized differences between these centers by plotting patient-level observed and expected mortality; we then examined differences in a subset of patients with a predicted mortality of ≥10% using the chi-squared test.

RESULTS

One high performer and 1 low performer were identified. The high performer managed a population with more blunt injuries (97.2% vs 93.6%, P < .001) and a higher median Injury Severity Score (14 vs 11, P < .001). There was no difference in survival between these centers in patients with an expected mortality of <10% (98.0% vs 96.7%, P = .11) or ≥70% (23.5% vs 10.8%, P = .22), but there was a difference in the subset with an expected mortality of ≥10% (77.5% vs 43.1%, P < .001).

CONCLUSIONS

Though patients with very low predicted mortality do equally well in high-performing and low-performing centers, the fact that performance seems determined by outcomes of patients with moderate predicted mortality favors a "marginal gains" theory.

Validation of the Norwegian survival prediction model in trauma (NORMIT) in Swedish trauma populations

BACKGROUND

Trauma survival prediction models can be used for quality assessment in trauma populations. The Norwegian survival prediction model in trauma (NORMIT) has been updated recently and validated internally (NORMIT 2). The aim of this observational study was to compare the accuracy of NORMIT 1 and 2 in two Swedish trauma populations.

METHODS

Adult patients registered in the national trauma registry during 2014-2016 were eligible for inclusion. The study populations comprised the total national trauma (NT) population, and a subpopulation of patients admitted to a single level I trauma centre (TC). The primary outcome was 30-day mortality. Model validation included receiver operating characteristic (ROC) curve analysis and GiViTI calibration belts. The calibration was also assessed in subgroups of severely injured patients (New Injury Severity Score (NISS) over 15).

RESULTS

A total of 26 504 patients were included. Some 18·7 per cent of patients in the NT population and 2·6 per cent in the TC subpopulation were excluded owing to missing data, leaving 21 554 and 3972 respectively for analysis. NORMIT 1 and 2 showed excellent ability to distinguish between survivors and non-survivors in both populations, but poor agreement between predicted and observed outcome in the NT population with overestimation of survival, including in the subgroup with NISS over 15. In the TC subpopulation, NORMIT 1 underestimated survival irrespective of injury severity, but NORMIT 2 showed good calibration both in the total subpopulation and the subgroup with NISS over 15.

CONCLUSION

NORMIT 2 is well suited to predict survival in a Swedish trauma centre population, irrespective of injury severity. Both NORMIT 1 and 2 performed poorly in a more heterogeneous national population of injured patients.

Variability in management of blunt liver trauma and contribution of level of American College of Surgeons Committee on Trauma verification status on mortality

BACKGROUND

Patients who sustain blunt liver trauma and are treated at an American College of Surgeons Committee on Trauma-verified Level I trauma center have an overall lower risk of mortality compared with patients admitted to a level II trauma center. However, elements contributing to these differences are unknown. We hypothesize that practice variation exists between trauma centers in management of blunt liver injury. Our objective is to identify practice variations and their effect on clinical outcomes.

METHODS

Data from a statewide collaborative quality initiative for trauma were used. The data set contains information from 29 American College of Surgeons Committee on Trauma verified Levels I and II trauma centers from 2011 to 2016. Propensity score matching was used to create cohorts of patients treated at Levels I or II trauma centers. The 1:1 matched cohorts were used to compare in-hospital mortality, management strategy, complications, intensive care unit (ICU) and hospital length of stay, and failure to rescue.

RESULTS

Four hundred fifty-four patients with grade 3 or higher blunt liver injury were included. Patients treated at level II trauma centers had higher in-hospital mortality than those treated at Level I trauma centers (15.4% vs 8.8%, p = 0.03). Level II trauma centers used angiography less compared with Level I centers (p = 0.007) and admitted significantly fewer patients to the ICU (p = 0.002). The ICU status was associated with reduced mortality (7.2% vs 23.9%, p < 0.001). Despite a lower rate of overall complications, Level II trauma centers were more likely to fail in rescuing their patients (p = 0.045).

CONCLUSION

Admission with a high-grade liver injury to a Level II trauma center is associated with increased in-hospital mortality. Level II trauma centers were less likely to use angiography or admit high-grade liver injuries to the ICU. This variation in practice may lead to the inability to rescue critically ill patients. Future research should investigate contributors to underutilization of resources for patients with high-grade liver injuries.

LEVEL OF EVIDENCE

Care management, level IV.

Mortality prediction in patients with isolated moderate and severe traumatic brain injury using machine learning models

BACKGROUND

The purpose of this study was to build a model of machine learning (ML) for the prediction of mortality in patients with isolated moderate and severe traumatic brain injury (TBI).

METHODS

Hospitalized adult patients registered in the Trauma Registry System between January 2009 and December 2015 were enrolled in this study. Only patients with an Abbreviated Injury Scale (AIS) score ≥ 3 points related to head injuries were included in this study. A total of 1734 (1564 survival and 170 non-survival) and 325 (293 survival and 32 non-survival) patients were included in the training and test sets, respectively.

RESULTS

Using demographics and injury characteristics, as well as patient laboratory data, predictive tools (e.g., logistic regression [LR], support vector machine [SVM], decision tree [DT], naive Bayes [NB], and artificial neural networks [ANN]) were used to determine the mortality of individual patients. The predictive performance was evaluated by accuracy, sensitivity, and specificity, as well as by area under the curve (AUC) measures of receiver operator characteristic curves. In the training set, all five ML models had a specificity of more than 90% and all ML models (except the NB) achieved an accuracy of more than 90%. Among them, the ANN had the highest sensitivity (80.59%) in mortality prediction. Regarding performance, the ANN had the highest AUC (0.968), followed by the LR (0.942), SVM (0.935), NB (0.908), and DT (0.872). In the test set, the ANN had the highest sensitivity (84.38%) in mortality prediction, followed by the SVM (65.63%), LR (59.38%), NB (59.38%), and DT (43.75%).

CONCLUSIONS

The ANN model provided the best prediction of mortality for patients with isolated moderate and severe TBI.

Derivation and validation of different machine-learning models in mortality prediction of trauma in motorcycle riders: a cross-sectional retrospective study in southern Taiwan

OBJECTIVES

This study aimed to build and test the models of machine learning (ML) to predict the mortality of hospitalised motorcycle riders.

SETTING

The study was conducted in a level-1 trauma centre in southern Taiwan.

PARTICIPANTS

Motorcycle riders who were hospitalised between January 2009 and December 2015 were classified into a training set (n=6306) and test set (n=946). Using the demographic information, injury characteristics and laboratory data of patients, logistic regression (LR), support vector machine (SVM) and decision tree (DT) analyses were performed to determine the mortality of individual motorcycle riders, under different conditions, using all samples or reduced samples, as well as all variables or selected features in the algorithm.

PRIMARY AND SECONDARY OUTCOME MEASURES

The predictive performance of the model was evaluated based on accuracy, sensitivity, specificity and geometric mean, and an analysis of the area under the receiver operating characteristic curves of the two different models was carried out.

RESULTS

In the training set, both LR and SVM had a significantly higher area under the receiver operating characteristic curve (AUC) than DT. No significant difference was observed in the AUC of LR and SVM, regardless of whether all samples or reduced samples and whether all variables or selected features were used. In the test set, the performance of the SVM model for all samples with selected features was better than that of all other models, with an accuracy of 98.73%, sensitivity of 86.96%, specificity of 99.02%, geometric mean of 92.79% and AUC of 0.9517, in mortality prediction.

CONCLUSION

ML can provide a feasible level of accuracy in predicting the mortality of motorcycle riders. Integration of the ML model, particularly the SVM algorithm in the trauma system, may help identify high-risk patients and, therefore, guide appropriate interventions by the clinical staff.

Assessing Field Triage Decisions and the International Classification Injury Severity Score (ICISS) at Predicting Outcomes of Trauma Patients

Florida considers the International Classification Injury Severity Score (ICISS) from hospital discharges within a geographic region in the apportionment of trauma centers (TCs). Patients with an ICISS <0.85 are considered to require triage to a TC, yet many are triaged to an emergency department (ED). We assess outcomes of those with an ICISS <0.85 by the actual triage decision of emergency medical services (EMS). From October 2011 to October 2013, 39,021 consecutive admissions with injury ICD-9 codes were analyzed. ICISS was calculated from the product of the survival risk ratios for a patient's three worst injuries. Outcomes were compared between patients with ICISS <0.85 either triaged to the ED or its separate, neighboring, free-standing TC at a large urban hospital. A total of 32,191 (83%) patients were triaged to the ED by EMS and 6,827 (17%) were triaged to the TC. Of these, 2544 had an ICISS <0.85, with 2145 (84%) being triaged to the TC and 399 (16%) to the ED. In these patients, those taken to the TC more often required admission, and those taken to the ED had better outcomes. When the confounders influencing triage to an ED or a TC are eliminated, those triaged by EMS to the ED rather than the TC had better overall outcomes. EMS providers better identified patients at risk for mortality than did the retrospective application of ICISS. ICISS <0.85 does not identify the absolute need for TC as EMS providers were able to appropriately triage a large portion of this population to the ED.

Demographic Patterns and Outcomes of Patients in Level I Trauma Centers in Three International Trauma Systems

Introduction

Trauma systems were developed to improve the care for the injured. The designation and elements comprising these systems vary across countries. In this study, we have compared the demographic patterns and patient outcomes of Level I trauma centers in three international trauma systems.

Methods

International multicenter prospective trauma registry-based study, performed in the University Medical Center Utrecht (UMCU), Utrecht, the Netherlands, John Hunter Hospital (JHH), Newcastle, Australia, and Harborview Medical Center (HMC), Seattle, the United States. Inclusion: patients ≥18 years, admitted in 2012, registered in the institutional trauma registry.

Results

In UMCU, JHH, and HMC, respectively, 955, 1146, and 4049 patients met the inclusion criteria of which 300, 412, and 1375 patients with Injury Severity Score (ISS) > 15. Mean ISS was higher in JHH (13.5; p < 0.001) and HMC (13.4; p < 0.001) compared to UMCU (11.7). Unadjusted mortality: UMCU = 6.5 %, JHH = 3.6 %, and HMC = 4.8 %. Adjusted odds of death: JHH = 0.498 [95 % confidence interval (CI) 0.303–0.818] and HMC = 0.473 (95 % CI 0.325–0.690) compared to UMCU. HMC compared to JHH was 1.002 (95 % CI 0.664–1.514). Odds of death patients ISS > 15: JHH = 0.507 (95 % CI 0.300–0.857) and HMC = 0.451 (95 % CI 0.297–0.683) compared to UMCU. HMC = 0.931 (95 % CI 0.608–1.425) compared to JHH. TRISS analysis: UMCU: Ws = 0.787, Z = 1.31, M = 0.87; JHH, Ws = 3.583, Z = 6.7, M = 0.89; HMC, Ws = 3.902, Z = 14.6, M = 0.84.

Conclusion

This study demonstrated substantial differences across centers in patient characteristics and mortality, mainly of neurological cause. Future research must investigate whether the outcome differences remain with nonfatal and long-term outcomes. Furthermore, we must focus on the development of a more valid method to compare systems.

The Applicability of Trauma and Injury Severity Score for a Blunt Trauma Population in Korea and a Proposal of New Models Using Score Predictors

PURPOSE

The purpose of this study was to verify the utility of existing Trauma and Injury Severity Score (TRISS) coefficients and to propose a new prediction model with a new set of TRISS coefficients or predictors.

MATERIALS AND METHODS

Of the blunt adult trauma patients who were admitted to our hospital in 2014, those eligible for Korea Trauma Data Bank entry were selected to collect the TRISS predictors. The study data were input into the TRISS formula to obtain "probability of survival" values, which were examined for consistency with actual patient survival status. For TRISS coefficients, Major Trauma Outcome Study-derived values revised in 1995 and National Trauma Data Bank-derived and National Sample Project-derived coefficients revised in 2009 were used. Additionally, using a logistic regression method, a new set of coefficients was derived from our medical center's database. Areas under the receiver operating characteristic (ROC) curve (AUC) for each prediction ability were obtained, and a pairwise comparison of ROC curves was performed.

RESULTS

In the statistical analysis, the AUCs (0.879-0.899) for predicting outcomes were lower than those of other countries. However, by adjusting the TRISS score using a continuous variable rather than a code for age, we were able to achieve higher AUCs [0.913 (95% confidence interval, 0.899 to 0.926)].

CONCLUSION

These results support further studies that will allow a more accurate prediction of prognosis for trauma patients. Furthermore, Korean TRISS coefficients or a new prediction model suited for Korea needs to be developed using a sufficiently sized sample.

External validation of the Norwegian survival prediction model in trauma after major trauma in Southern Finland

BACKGROUND

The Norwegian Survival Prediction Model in Trauma (NORMIT) is a newly developed outcome prediction model for patients with trauma. We aimed to compare the novel NORMIT to the more commonly used Trauma and Injury Severity Score (TRISS) in Finnish trauma patients.

METHODS

We performed a retrospective open-cohort study, using the trauma registry of Helsinki university hospital's trauma unit, including severely injured patients (new injury severity score > 15) admitted from 2007 to 2011. We used 30-day in-hospital mortality as the primary outcome, and discharge functional outcome as a secondary outcome of interest. Model performance was evaluated by comparing discrimination (by area under the receiver operating characteristic curve [AUC]), using a re-sample bootstrap technique, and by assessing calibration (GiViTI belt).

RESULTS

We identified 1111 patients fulfilling the study inclusion criteria. Overall mortality was 13% (n = 147). NORMIT showed slightly better discrimination for mortality prediction (AUC = 0.83, 95% confidence interval [CI] = 0.80-0.86 vs. AUC = 0.79, 95% CI = 0.75-0.83, P = 0.004) and functional outcome prediction (AUC = 0.78, 95% CI = 0.76-0.82 vs. AUC = 0.75, 95% CI = 0.72-0.78, P < 0.001) than TRISS. Calibration testing revealed poor calibration for both NORMIT and TRISS (P < 0.001), by giving too pessimistic predictions (predicted survival significantly lower than actual survival).

CONCLUSION

NORMIT and TRISS showed good discrimination, but poor calibration, in this mixed cohort of severely injured trauma patients from Southern Finland. We found NORMIT to be a feasible alternative to TRISS for trauma patient outcome prediction, but trauma prediction models with improved calibration are needed.

Short-term outcome and differences between rural and urban trauma patients treated by mobile intensive care units in Northern Finland: a retrospective analysis

Background

Emergency medical services are an important part of trauma care, but data comparing urban and rural areas is needed. We compared 30-day mortality and length of intensive care unit (ICU) stay for trauma patients injured in rural and urban municipalities and collected basic data on trauma care in Northern Finland.

Methods

We examined data from all trauma patients treated by the Finnish Helicopter Emergency Medical Services in 2012 and 2013. Only patients surviving to hospital were included in the analysis but all pre-hospital deaths were recorded. All data was retrieved from the national Helicopter Emergency Medical Services database, medical records, and the Finnish Causes of Death Registry. Patients were defined as urban or rural depending on the type of municipality where the injury occurred.

Results

A total of 472 patients were included. Age and Injury Severity Score did not differ between rural and urban patients. The pre-hospital time intervals and distances to trauma centers were longer for rural patients and a larger proportion of urban patients had intentional injuries (23.5 % vs. 9.3 %, P <0.001). The 30-day mortality for severely injured patients (Injury Severity Score >15) was 23.9 % in urban and 13.3 % in rural municipalities. In the multivariate regression analysis the odds ratio (OR) for 30-day mortality was 2.8 (95 % confidence interval 1.0 to 7.9, P = 0.05) in urban municipalities. There was no difference in the length of ICU stay or scores. Twenty patients died on scene or during transportation and 56 missions were aborted because of pre-hospital death.

Conclusions

The severely injured urban trauma patients had a trend toward higher 30-day mortality compared with patients injured in rural areas but the length of ICU stay was similar. However, more pre-hospital deaths occurred in rural municipalities. The time before mobile ICU arrival appears to be critical for trauma patients’ survival, especially in rural areas.

Using population-based critical care data to evaluate trauma outcomes

BACKGROUND

The analysis of mortality is an integral part of the evaluation of trauma care. When specific data are not available, general prediction models can be used to adjust for case mix. The aim of this study was to evaluate the feasibility of conducting a population-based analysis of trends in trauma mortality, using critical care audit data, and to investigate whether such data could provide a benchmark for the assessment of service reconfiguration.

METHODS

Retrospective cohort study of adult trauma patients, requiring admission to a critical care unit in Scotland, 2002-2011, using nationally collected data. Results are presented as standardised mortality ratios of observed mortality divided by APACHE II predicted mortality. Tests for trends in numbers and ratios over time were performed using linear regression.

FINDINGS

4503 patients were identified. There was a significant increase in the number of trauma patients admitted per year (p = 0.011). The median predicted probability of in-hospital death was 7% (interquartile range 1-13%), against an actual mortality was 11.6%. There was no significant change in the standardised mortality ratios of trauma patients (p = 0.1224).

CONCLUSIONS

This study demonstrated the feasibility of utilising critical care unit audit data for analysing outcomes from trauma care. It also showed the potential of such an approach to establish a baseline against which to compare the impact of future service reconfiguration. In contrast to healthcare systems with regionalised trauma care, there appears to have been little change in the mortality of trauma patients requiring critical care unit admission in Scotland.

Norwegian survival prediction model in trauma: modelling effects of anatomic injury, acute physiology, age, and co-morbidity

INTRODUCTION

Anatomic injury, physiological derangement, age, and injury mechanism are well-founded predictors of trauma outcome. We aimed to develop and validate the first Scandinavian survival prediction model for trauma.

METHODS

Eligible were patients admitted to Oslo University Hospital Ullevål within 24 h after injury with Injury Severity Score ≥ 10, proximal penetrating injuries or received by a trauma team. The derivation dataset comprised 5363 patients (August 2000 to July 2006); the validation dataset comprised 2517 patients (August 2006 to July 2008). Exclusion because of missing data was < 1%. Outcome was 30-day mortality. Logistic regression analysis incorporated fractional polynomial modelling and interaction effects. Model validation included a calibration plot, Hosmer-Lemeshow test and receiver operating characteristic (ROC) curves.

RESULTS

The new survival prediction model included the anatomic New Injury Severity Score (NISS), Triage Revised Trauma Score (T-RTS, comprising Glascow Coma Scale score, respiratory rate, and systolic blood pressure), age, pre-injury co-morbidity scored according to the American Society of Anesthesiologists Physical Status Classification System (ASA-PS), and an interaction term. Fractional polynomial analysis supported treating NISS and T-RTS as linear functions and age as cubic. Model discrimination between survivors and non-survivors was excellent. Area (95% confidence interval) under the ROC curve was 0.966 (0.959-0.972) in the derivation and 0.946 (0.930-0.962) in the validation dataset. Overall, low mortality and skewed survival probability distribution invalidated model calibration using the Hosmer-Lemeshow test.

CONCLUSIONS

The Norwegian survival prediction model in trauma (NORMIT) is a promising alternative to existing prediction models. External validation of the model in other trauma populations is warranted.

Identification of dynamic prehospital changes with continuous vital signs acquisition

OBJECTIVE

In most trauma registries, prehospital trauma data are often missing or unreliable because of the difficult dual task consigned to prehospital providers of recording vital signs and simultaneously resuscitating patients. The purpose of this study was to test the hypothesis that the analysis of continuous vital signs acquired automatically, without prehospital provider input, improves vital signs data quality, captures more extreme values that might be missed with conventional human data recording, and changes Trauma Injury Severity Scores compared with retrospectively compiled prehospital trauma registry data.

METHODS

A statewide vital signs collection network in 6 medevac helicopters was deployed for prehospital vital signs acquisition using a locally built vital signs data recorder (VSDR) to capture continuous vital signs from the patient monitor onto a memory card. VSDR vital signs data were assessed by 3 raters, and intraclass correlation coefficients were calculated to test interrater reliability. Agreement between VSDR and trauma registry data was compared with the methods of Altman and Bland including corresponding calculations for precision and bias.

RESULTS

Automated prehospital continuous VSDR data were collected in 177 patients. There was good agreement between the first recorded vital signs from the VSDR and the trauma registry value. Significant differences were observed between the highest and lowest heart rate, systolic blood pressure, and pulse oximeter from the VSDR and the trauma registry data (P< .001). Trauma Injury Severity Scores changed in 12 patients (7%) when using data from the VSDR.

CONCLUSION

Real-time continuous vital signs monitoring and data acquisition can identify dynamic prehospital changes, which may be missed compared with vital signs recorded manually during distinct prehospital intervals. In the future, the use of automated vital signs trending may improve the quality of data reported for inclusion in trauma registries. These data may be used to develop improved triage algorithms aimed at optimizing resource use and enhancing patient outcomes.

Comparative effectiveness of helicopter emergency medical services compared to ground emergency medical services

The use of helicopter emergency medical services (HEMS) for the transportation and treatment of trauma patients, while commonplace in most developed nations, remains controversial. The purported beneficial effects of HEMS compared to ground emergency medical services is likely to be some combination of speed, crew expertise, and the fact that HEMS is part of an organized trauma system. When the HEMS literature is assessed as a whole, considerable heterogeneity of effects and study methodologies preclude an accurate estimate of composite effect. However, when the outcome of mortality is studied using advanced multivariable regression techniques to control for multiple known confounders, an improved odds of survival has been repeatedly demonstrated. Future HEMS research must rely on robust observational study designs and assessments of a variety of patient outcomes. Questions about the role of speed, distance, and other potentially beneficial elements of HEMS remain.

Trauma survival prediction in Asian population: a modification of TRISS to improve accuracy

The probability of survival (PS) in blunt trauma as calculated by Trauma and Injury Severity Score (TRISS) has been an indispensable tool in trauma audit. The aim of this study is to explore the predictive performance of the latest updated TRISS model by investigating the Age variable recategorisations and application of local Injury Severity Score (ISS) and Revised Trauma Score (RTS) coefficients in a logistic model using a level I trauma centre database involving Asian population.

METHODS

Prospectively and consecutively collected 5684 trauma patients' data over a 10-year period at a regional level I trauma centre were reviewed. Four modified TRISS (mTRISS) models using Age coefficient from reclassifications of the Age variable according to their correlation with survival by logistic regression on the local dataset were acquired. RTS and ISS coefficients were derived from the local dataset and then applied to the mTRISS models. mTRISS models were compared with the existing Major Trauma Outcome Study (MTOS)-derived TRISS (eTRISS) model. Model 1=Age effect taken as linear; Model 2=Age classified into two groups (0-54, 55+); Model 3=Age classified into four groups (0-15, 16-54, 55-79, 80+) and Model 4=Age classified into two groups (0-69, 70+). Performance measures including sensitivity, specificity, accuracy and area under the Receiver Operating Characteristic (ROC) curve were used to assess the various models. The cross-validation procedure consisted of comparing the P(S) obtained from mTRISS Models 1 and 2 with the P(S) obtained from the MTOS derived from eTRISS.

RESULTS

A 5147 blunt trauma patients' dataset was reviewed. Model 1, where Age was taken as a scale variable, demonstrated a substantial improvement in the survival prediction with 91.6% accuracy in blunt injuries as compared with 89.2% in the MTOS-derived TRISS. The 95% CI for ROC derived from mTRISS Model 1 was (0.923, 0.940), when compared with the hypothesised ROC value 0.886 obtained from eTRISS, it clearly indicated a significant improvement in predicting survival at 5% level. Furthermore, ROCs have shown clearly the superiority of Model 1 over Model 2, and of Model 2 over MTOS-derived TRISS. The recategorisation of the Age variable (Models 3 and 4) also demonstrated improved performance, but their strength was not as intense as in Model 1. Overall, the results point to the adoption of Model 1 as the best model for PS. Cross-validation analysis has further assured the validity of these findings.

CONCLUSIONS

The present study has demonstrated that (1) having the Age variable being dichotomised (cut-off at 55 years) as in the eTRISS, but with the application of a local dataset-derived coefficients give better TRISS survival prediction in Asian blunt trauma patients; (2) improved performance are found with certain recategorisation of the Age variable and (3) the accuracy can further be enhanced if the Age effect is taken to be linear, with the application of local dataset-derived coefficients.

A structured autopsy-based audit of 370 firearm fatalities: Contribution to inform policy decisions and the probability of the injured arriving alive at a hospital and receiving definitive care

The objectives of this autopsy-based audit of firearm-related fatalities were to acquire data to inform policy decisions and to assess the probability of the injured arriving alive at a hospital and receiving definitive care.

EVALUATED VARIABLES

Demographics; co-morbidities; location and intention of the injury; toxicology; types of firearms; Abbreviated Injury Scale; Injury Severity Score (ISS); transfer means and time; and location of death.

RESULTS

Of a total of 370 fatalities, 85.7% were male. The median age was 38 (9-95) years. Suicides (47%) and assaults (45.1%) were the most common underlying intentions. The most seriously injured regions were the head (44.5%), thorax (25.7%), abdomen (10.7%), and spine (5.7%). Of the 370 total subjects, 4.9% had an ISS<16 and 59.5% had an ISS≤74; both groups were classified as potentially preventable deaths. The majority (84%) died at the scene, and only 9.8% left the emergency department alive for further treatment. Multivariate analyses documented that postmortem ISS is an independent factor that predicts the probability of the injured reaching a hospital alive and receiving definitive care. Individuals injured in greater Athens and those most seriously injured in the face, abdomen or spine had significantly greater chances of reaching a hospital alive and receiving definitive care, whereas those injured by a shotgun and the positive toxicology group were significantly less likely to. In conclusion, this study provides data to inform policy decisions, calls for a surveillance network and establishes a baseline for estimating the probability regarding the location of firearm-related deaths.

Comparing regression-adjusted mortality to standardized mortality ratios for trauma center profiling

Background:

Trauma center profiling is commonly performed with Standardized Mortality Ratios (SMRs). However, comparison of SMRs across trauma centers with different case mix can induce confounding leading to biased trauma center ranks. We hypothesized that Regression-Adjusted Mortality (RAM) estimates would provide a more valid measure of trauma center performance than SMRs.

Objective:

Compare trauma center ranks generated by RAM estimates to those generated by SMRs.

Materials and Methods:

The study was based on data from a provincial Trauma Registry (1999-2006; n = 88,235). SMRs were derived as the ratio of observed to expected deaths using: (1) the study population as an internal standard, (2) the US National Trauma Data Bank as an external standard. The expected death count was calculated as the sum of mortality probabilities for all patients treated in a hospital conditional on the injury severity score, the revised trauma score, and age. RAM estimates were obtained directly from a hierarchical logistic regression model.

Results:

Crude mortality was 5.4% and varied between 1.3% and 13.5% across the 59 trauma centers. When trauma center ranks from internal SMRs and RAM were compared, 49 out of 59 centers changed rank and six centers changed by more than five ranks. When trauma center ranks from external SMRs and RAM were compared, 55 centers changed rank and 17 changed by more than five ranks.

Conclusions:

The results of this study suggest that the use of SMRs to rank trauma centers in terms of mortality may be misleading. RAM estimates represent a potentially more valid method of trauma center profiling.

Pre-hospital trauma care: A comparison of two healthcare systems

INTRODUCTION

The management of trauma patients differs depending upon the healthcare system available.

AIM

To compare the pre-hospital management and outcome of polytrauma patients between two countries with differing approaches to pre-hospital management.

MATERIALS AND METHODS

The Scottish trauma and audit group (STAG) and the German trauma registry (GTR) databases were used to compare the management and outcome of trauma patients in Scotland and Germany. Severely injured patients (injury severity score (ISS) > 16) were analyzed for a 3 year period (2000 to 2002). Patient demographics, pre-hospital interventions, ISS, revised trauma score (RTS), time from scene of injury to arrival to the emergency department (ED), 120 day mortality and standardized mortality ratios using TRISS methodology were compared.

RESULTS

There were 227 patients identified from the STAG registry and 6878 patients from the GTR registry. There was a significant difference in ISS (24.9 vs. 29.8, P = 0.001, respectively). No significant difference was observed for the RTS (P = 0.2). There was a significantly higher rate of pre-hospital interventions in the German group (P < 0.001). The mean time from an injury to arrival to the ED (73 vs. 247 minutes, P = 0.001) was longer for the Scottish patients. There was no difference for an unadjusted mortality rate between the groups, but the standardized mortality ratio was significantly greater for the Scottish population (3.8 vs. 2.2, P = 0.036).

CONCLUSION

Despite variation in pre-hospital transfer times and interventions, no significant difference was demonstrated in RTS upon arrival, or for the unadjusted mortality rates.

Using the abbreviated injury severity and Glasgow Coma Scale scores to predict 2-week mortality after traumatic brain injury

BACKGROUND

Prediction of outcome after traumatic brain injury (TBI) remains elusive. We tested the use of a single hospital Glasgow Coma Scale (GCS) Score, GCS Motor Score, and the Head component of the Abbreviated Injury Scale (AIS) Score to predict 2-week cumulative mortality in a large cohort of TBI patients admitted to the eight U.S. Level I trauma centers in the TBI Clinical Trials Network.

METHODS

Data on 2,808 TBI patients were entered into a centralized database. These TBI patients were categorized as severe (GCS score, 3-8), moderate (9-12), or complicated mild (13-15 with positive computed tomography findings). Intubation and chemical paralysis were recorded. The cumulative incidence of mortality in the first 2 weeks after head injury was calculated using Kaplan-Meier survival analysis. Cox proportional hazards regression was used to estimate the magnitude of the risk for 2-week mortality.

RESULTS

Two-week cumulative mortality was independently predicted by GCS, GCS Motor Score, and Head AIS. GCS Severity Category and GCS Motor Score were stronger predictors of 2-week mortality than Head AIS. There was also an independent effect of age (<60 vs. ≥60) on mortality after controlling for both GCS and Head AIS Scores.

CONCLUSIONS

Anatomic and physiologic scales are useful in the prediction of mortality after TBI. We did not demonstrate any added benefit to combining the total GCS or GCS Motor Scores with the Head AIS Score in the short-term prediction of death after TBI.

Association of direct helicopter versus ground transport and in-hospital mortality in trauma patients: a propensity score analysis

OBJECTIVES

Helicopter emergency medical services (HEMS) transport of trauma patients has been used for decades. Its use, however, is still a subject of debate, including issues such as high costs, increasing numbers of crashes, and conflicting results regarding effectiveness in reducing mortality. The aim of this study was to examine whether mode of transport (HEMS vs. ground EMS) is independently associated with mortality among trauma patients transported directly from the scene of injury to definitive care.

METHODS

All trauma patients transported directly to a Level I or Level II trauma center by either air or ground EMS over a 4-year period were selected from the Oklahoma State Trauma Registry. Multivariable logistic regression was used to develop propensity scores based on variables measured at the scene of injury. The propensity scores represented the predicted probabilities of a patient being transported by HEMS given a specific set of characteristics and were used as a composite confounding variable in subsequent models of the association of mortality and mode of transport. Along with the propensity scores, Injury Severity Scores (ISS), initial Revised Trauma Score (RTS), and distance from the trauma center were included in a Cox proportional hazards model of the association of mode of transport and 24-hour and 2-week mortality.

RESULTS

Overall, the hazard ratio (HR) for 2-week mortality in patients transported by HEMS was 33% lower (HR = 0.67, 95% confidence interval [CI] = 0.54 to 0.84) than in patients transported by ground EMS from the scene of injury, after adjustment for the propensity score and other covariates. In subanalyses, the apparent association of a reduction in the hazard of early mortality among patients transported by HEMS was most evident for patients with an RTS based on injury scene vital signs of 3 to 7 (HR = 0.61, 95% CI = 0.46 to 0.82). The point estimate of the HR was similar (HR = 0.65 95% CI = 0.34 to 1.2) in the 75% of cases who had normal vital signs at the scene of injury, although it was no longer statistically significant because crude mortality was very low (1.7%) in this group. Among those with a RTS of 3 or less at the scene, crude mortality was 58%, and mode of transport was not associated with mortality (HR = 1.02, 95% CI = 0.68 to 1.6).

CONCLUSIONS

Helicopter EMS transport was associated with a decreased hazard of mortality among certain patients transported from the scene of injury directly to definitive care. Refinements in scene triage and transport guidelines are needed to more effectively select patients that may benefit from HEMS transport from those unlikely to benefit.

Improved prediction from revised injury severity classification (RISC) over trauma and injury severity score (TRISS) in an independent evaluation of major trauma patients

The usefulness of the Revised Injury Severity Classification (RISC) analysis was compared with that of the Trauma and Injury Severity Score (TRISS) for evaluating the quality of treatment of severely injured patients at the General Hospital Celje, Slovenia. Over a period of 2 years, data from a cohort of 155 patients treated at the General Hospital Celje were included in the Traumaregister Deutsche Gesellschaft für Unfallchirurgie. The structure of the patient group was compared with that in the registry, and TRISS and RISC analyses were performed. The M statistic (0.83) showed a good match of the distribution of probability of survival between groups. Evaluation of RISC (area under the curve [AUC] 0.94, Hosmer-Lemeshow test 3.5) demonstrated the efficacy of this method in the patient group. TRISS (AUC 0.89, Hosmer-Lemeshow test 21.1) was not a reliable instrument for analysis of treatment of major trauma patients. We believe that RISC should replace TRISS for evaluation of the treatment of major trauma patients.

The regionalization of pediatric health care

Regionalization of health care is a method of providing high-quality, cost-efficient health care to the largest number of patients. Within pediatric medicine, regionalization has been undertaken in 2 areas: neonatal intensive care and pediatric trauma care. The supporting literature for the regionalization of these areas demonstrates the range of studies within this field: studies of neonatal intensive care primarily compare different levels of hospitals, whereas studies of pediatric trauma care primarily compare the impact of institutionalizing a trauma system in a single geographic region. However, neither specialty has been completely regionalized, possibly because of methodologic deficiencies in the evidence base. Research with improved study designs, controlling for differences in illness severity between different hospitals; a systems approach to regionalization studies; and measurement of parental preferences will improve the understanding of the advantages and disadvantages of regionalizing pediatric medicine and will ultimately optimize the outcomes of children.

Survival benefit of transfer to tertiary trauma centers for major trauma patients initially presenting to nontertiary trauma centers

OBJECTIVES

Recent evidence suggests a measurable reduction in mortality for patients transferred from a nontertiary trauma center (Level III or IV) to a Level I trauma center, but not for those transferred to a Level II trauma center. Whether this can be generalized to a predominantly rural region with fewer tertiary trauma care resources is uncertain. This study sought to evaluate mortality differences for patients initially presenting to nontertiary trauma centers in a predominantly rural region depending on transfer status.

METHODS

This was a retrospective cohort study of patients initially presenting to 104 nontertiary trauma centers in Oklahoma and meeting the state's criteria for major trauma. Patients dying within 1 hour of emergency department (ED) arrival at the nontertiary trauma center were excluded. The exposure variable of interest was admission status, which was categorized as either transfer to a tertiary (Level I or II) trauma center within 24 hours or admission to a nontertiary trauma center. Propensity scores were used to minimize the selection bias inherent in the decision to admit or transfer a patient for higher-level care. Multiple logistic regression was used to generate three propensity score models: probability of transfer to either a Level I or II, Level I only, and Level II only. Propensity scores were then included as a covariate in multivariable Cox regression models assessing outcome differences between admitted and transferred patients. The outcome of interest was 30-day mortality, defined as death at either the nontertiary trauma center or the tertiary trauma center within 30 days of arrival at the initial Level III/IV center's ED.

RESULTS

A total of 6,229 patients met study criteria, of whom 2,669 (43%) were transferred to tertiary trauma centers. Of those transferred, 1,422 patients (53%) were transferred to a Level I trauma center. Crude mortality was lower for patients transferred to tertiary trauma centers compared to those remaining at nontertiary trauma facilities (hazard ratio [HR] = 0.59; 95% confidence interval [CI] = 0.48 to 0.72). After adjusting for the propensity to be transferred, Injury Severity Score (ISS), presence of severe head injury, and age, transfer to a tertiary trauma center was associated with a significantly lower 30-day mortality (HR = 0.38; 95% CI = 0.30 to 0.50) compared to admission and treatment at a nontertiary trauma center. The observed survival benefit was similar for patients transferred to a Level I trauma center (HR = 0.36; 95% CI = 0.20 to 0.4) and those transferred to a Level II center (HR = 0.45; 95% CI = 0.33 to 0.61).

CONCLUSIONS

This study suggests a survival benefit among patients initially presenting to nontertiary trauma centers who are subsequently transferred to tertiary trauma centers compared to those remaining in nontertiary trauma centers, even after adjusting for variables affecting the likelihood of transfer. Although this survival benefit was larger for patients treated at a Level I trauma center, Level II trauma centers in a region with few tertiary trauma resources demonstrated a measurable benefit as well.

M-study from an urban trauma center in Tokyo

BACKGROUND

The original Trauma and Injury Severity Score (TRISS) methodology from the Major Trauma Outcome Study (MTOS) is the most widely used outcome prediction model. The coefficients from the MTOS cohorts are still used in the Japan Trauma Data Bank for evaluating the quality of patient care. The purposes are to determine whether the database of this institution is well matched to the MTOS study and whether the original TRISS coefficients are accurate predictors of the patient outcome in Japan.

METHODS

The M-statistic score was calculated based on the trauma registry data from 2000 to 2003 in Teikyo University.

RESULTS

Eight hundred fifty-four cases were analyzed. The crude mortality rate was 10.5%. The mean Injury Severity Score was 15.8 ± 13.6. The mean Revised Trauma Score was 7.00 ± 1.4. The M-statistic score was 0.811.

CONCLUSION

The trauma populations in this study differed significantly from the MTOS. The Modified TRISS coefficients should be adapted for outcome assessment based on the location of the injured population. This is the first report of an M-study from Japan to be published in the English literature.

Are all trauma centers created equally? A statewide analysis

OBJECTIVES

Prior work has shown differences in mortality at different levels of trauma centers (TCs). There are limited data comparing mortality of equivalently verified TCs. This study sought to assess the potential differences in mortality as well as discharge destination (discharge to home vs. to a rehabilitation center or skilled nursing facility) across Level I TCs in the state of Ohio.

METHODS

This was a retrospective, multicenter, statewide analysis of a state trauma registry of American College of Surgeons (ACS)-verified Level I TCs from 2003 to 2006. All adult (>15 years) patients transferred from the scene to one of the 10 Level I TCs throughout the state were included (n = 16,849). Multivariable logistic regression models were developed to assess for differences in mortality, keeping each TC as a fixed-effect term and adjusting for patient demographics, injury severity, mechanism of injury, and emergency medical services and emergency department procedures. Outcomes included in-hospital mortality and discharge destination (home vs. rehabilitation center or skilled nursing facility). Adjusted odds ratios (ORs) for each TC were also calculated.

RESULTS

Considerable variability existed in unadjusted mortality between the centers, from 3.8% (95% confidence interval [CI] = 3.7% to 3.9%) to 24.2% (95% CI = 24.1% to 24.3%), despite similar patient characteristics and injury severity. Adjusted mortality had similar variability as well, ranging from an OR of 0.93 (95% CI = 0.47 to 1.84) to an OR of 6.02 (95% CI= 3.70 to 9.79). Similar results were seen with the secondary outcomes (discharge destination).

CONCLUSIONS

There is considerable variability in the mortality of injured patients at Level I TCs in the state of Ohio. The patient differences or care processes responsible for this variation should be explored.

Prognostic value of Sequential Organ Failure Assessment and Simplified Acute Physiology II Score compared with trauma scores in the outcome of multiple-trauma patients

BACKGROUND

Prospective data regarding the prognostic value of the Sequential Organ Failure Assessment (SOFA) score in comparison with the Simplified Acute Physiology Score (SAPS II) and trauma scores on the outcome of multiple-trauma patients are lacking.

METHODS

Single-center evaluation (n = 237, Injury Severity Score [ISS] >16; mean ISS = 29). Uni- and multivariate analysis of SAPS II, SOFA, revised trauma, polytrauma, and trauma and ISS scores (TRISS) was performed.

RESULTS

The 30-day mortality was 22.8% (n = 54). SOFA day 1 was significantly higher in nonsurvivors compared with survivors (P < .001) and correlated well with the length of intensive care unit stay (r = .50, P < .001). Logistic regression revealed SAPS II to have the best predictive value of 30-day mortality (area under the receiver operating characteristic = .86 +/- .03). The SOFA score significantly added prognostic information with regard to mortality to both SAPS II and TRISS.

CONCLUSIONS

The combination of critically ill and trauma scores may increase the accuracy of mortality prediction in multiple-trauma patients.

Quality of life in trauma victims six months after hospital discharge

OBJECTIVE

Trauma is the third most important cause of death in Brazil. However, its impact on survivors' quality of life has been scarcely studied in this country. This study aimed to assess trauma victims' quality of life, cared for in an emergency hospital unit, six months after discharge.

METHODS

A total of 35 patients from the emergency unit of a university hospital in the city of Ribeirão Preto, Southeastern Brazil, were included in this study, between 2005 and 2006. Patients were interviewed in their homes, six months after hospital discharge. The short version of the World Health Organization Quality of Life (WHOQOL-BREF) instrument was applied to assess the physical, psychological, social relationships, and environmental domains. Associations between domain scores and hospital stay, age, sex and Injury Severity Score variables were analyzed with linear regression models.

RESULTS

Significant reduction in quality of life was found in the group studied, when compared to samples of normal people in national and international studies, especially as regards the physical, psychological, and environmental domains. The social relationships domain revealed the highest mean scores, with 69.7 points, whereas the environmental domain received the lowest score (52.4 points), both on the percentage scale. Variables associated with the physical domain were hospital stay (p=0.02), age (p<0.01) and sex (p=0.03). The analysis did not show association with the variables studied for the remaining domains.

CONCLUSIONS

Trauma victims showed a reduction in quality of life scores. Even though the physical aspect was the most affected, there is evidence that the psychological and environmental domains remained far from the ideal conditions expected for the general population.

The trauma risk adjustment model: a new model for evaluating trauma care

BACKGROUND DATA

The trauma injury severity score (TRISS) has been used for over 20 years for retrospective risk assessment in trauma populations. The TRISS has serious limitations, which may compromise the validity of trauma care evaluations.

OBJECTIVE

To derive and validate a new mortality prediction model, the trauma risk adjustment model (TRAM), and to compare the performance of the TRAM to that of the TRISS in terms of predictive validity and risk adjustment.

METHODS

The Quebec Trauma Registry (1998-2005), based on the mandatory participation of 59 designated provincial trauma centers, was used to derive the model. The American National Trauma Data Bank (2000-2005), based on the voluntary participation of any US hospitals treating trauma, was used for the validation phase. Adult patients with blunt trauma respecting at least one of the following criteria were included: hospital stay >2 days, intensive care unit admission, death, or hospital transfer. Hospital mortality was modeled with logistic generalized additive models using cubic smoothing splines to accommodate nonlinear relations to mortality. Predictive validity was assessed with model discrimination and calibration. Risk adjustment was assessed using comparisons of risk-adjusted mortality between hospitals.

RESULTS

The TRAM generated an area under the receiving operator curve of 0.944 and a Hosmer-Lemeshow statistic of 42 in the derivation phase. In the validation phase, the TRAM demonstrated better model discrimination and calibration than the TRISS (area under the receiving operator curve = 0.942 and 0.928, P < 0.001; Hosmer-Lemeshow statistics = 127 and 256, respectively). Replacing the TRISS with the TRAM led to a mean change of 28% in hospital risk-adjusted odds ratios of mortality.

CONCLUSIONS

Our results suggest that adopting the TRAM could improve the validity of trauma care evaluations and trauma outcome research.

Evaluation of trauma care using TRISS method: the role of adjusted misclassification rate and adjusted w-statistic

Background

Major trauma is a leading cause of death worldwide. Evaluation of trauma care using Trauma Injury and Injury Severity Score (TRISS) method is focused in trauma outcome (deaths and survivors). For testing TRISS method TRISS misclassification rate is used. Calculating w-statistic, as a difference between observed and TRISS expected survivors, we compare our trauma care results with the TRISS standard.

Aim

The aim of this study is to analyze interaction between misclassification rate and w-statistic and to adjust these parameters to be closer to the truth.

Materials and methods

Analysis of components of TRISS misclassification rate and w-statistic and actual trauma outcome.

Results

The component of false negative (FN) (by TRISS method unexpected deaths) has two parts: preventable (Pd) and non-preventable (nonPd) trauma deaths. Pd represents inappropriate trauma care of an institution; otherwise nonpreventable trauma deaths represents errors in TRISS method. Removing patients with preventable trauma deaths we get an Adjusted misclassification rate: (FP + FN - Pd)/N or (b+c-Pd)/N. Substracting nonPd from FN value in w-statistic formula we get an Adjusted w-statistic: [FP-(FN - nonPd)]/N, respectively (FP-Pd)/N, or (b-Pd)/N).

Conclusion

Because adjusted formulas clean method from inappropriate trauma care, and clean trauma care from the methods error, TRISS adjusted misclassification rate and adjusted w-statistic gives more realistic results and may be used in researches of trauma outcome.

The Utstein template for uniform reporting of data following major trauma: a joint revision by SCANTEM, TARN, DGU-TR and RITG

BACKGROUND

In 1999, an Utstein Template for Uniform Reporting of Data following Major Trauma was published. Few papers have since been published based on that template, reflecting a lack of international consensus on its feasibility and use. The aim of the present revision was to further develop the Utstein Template, particularly with a major reduction in the number of core data variables and the addition of more precise definitions of data variables. In addition, we wanted to define a set of inclusion and exclusion criteria that will facilitate uniform comparison of trauma cases.

METHODS

Over a ten-month period, selected experts from major European trauma registries and organisations carried out an Utstein consensus process based on a modified nominal group technique.

RESULTS

The expert panel concluded that a New Injury Severity Score > 15 should be used as a single inclusion criterion, and five exclusion criteria were also selected. Thirty-five precisely defined core data variables were agreed upon, with further division into core data for Predictive models, System Characteristic Descriptors and for Process Mapping.

CONCLUSION

Through a structured consensus process, the Utstein Template for Uniform Reporting of Data following Major Trauma has been revised. This revision will enhance national and international comparisons of trauma systems, and will form the basis for improved prediction models in trauma care.

Evaluation of trauma care applying TRISS methodology in a Caribbean developing country

There have been conflicting reports regarding the applicability of Trauma Injury Severity Score (TRISS) methodology to evaluate trauma care in a developing country setting. The objective of this study was to apply TRISS methodology to evaluate trauma care in the public hospitals of a Caribbean developing country. A prospective, observational study was conducted in the three major general hospitals in Trinidad. Major trauma patients were included. Demographic data, waiting time in the Emergency Department, and nature of injury (blunt or penetrating) were noted. Revised Trauma Score, Injury Severity Score, and Glasgow Coma Scale were applied to all patients on admission. Hospital outcomes were noted. Predicted outcomes were calculated for adult patients using TRISS methodology. M, Z statistics and receiver operating characteristic (ROC) curve analysis were done. There were 326 trauma patients studied, of whom 279 adults were evaluated by the TRISS methodology. Men were more frequently involved in trauma than women; there was more blunt trauma than penetrating trauma. The M statistic was 0.98 and the overall Z statistic was 5.81. The ROC curve analysis showed TRISS to be a fair discriminator in the study case-mix with an area under the curve of 0.82 (95% confidence interval 0.69-0.96). There is a considerable disparity between predicted and observed outcomes when trauma patients are evaluated by the TRISS methodology in a developing country setting.

Pre-injury ASA physical status classification is an independent predictor of mortality after trauma

BACKGROUND

The ability of an organism to withstand trauma is determined by the injury per se and inherent properties of the organism at the time of injury. We analyzed whether pre-injury morbidity scored on a four-level ordinal scale according to the American Society of Anesthesiologists Physical Status (ASA-PS) classification system predicts mortality after trauma.

MATERIALS

From a total of 3,773 prospectively collected patients (years 2000-2004), 3,728 patients were included. Main outcome measure was mortality 30 days after injury. The effect of pre-injury ASA-PS on mortality was assessed using linear logistic regression analysis, controlling for Revised Trauma Score (RTS), Injury Severity Score (ISS), and age.

RESULTS

Mortality increased with increasing pre-injury ASA-PS, age, and ISS, and with decreasing RTS. Unadjusted mortality rates were 5.7% in ASA-PS 1, 12.3% in ASA-PS 2, and 26.4% in ASA-PS 3-4. This increasing mortality trend across pre-injury ASA-PS group was evident in nearly all categories of ISS, RTS, and age. Odds ratio for death was 1.76 (95% CI, 1.14-2.72) for pre-injury ASA-PS 2, and 2.25 (95% CI, 1.36-3.71) for ASA-PS 3-4 compared with for ASA-PS 1 and adjusted for ISS, RTS, and age. There were no interaction effects between pre-injury ASA-PS and the other variables.

CONCLUSIONS

Pre-injury ASA-PS score was an independent predictor of mortality after trauma, also after adjusting for the major variables in the traditional TRISS (Trauma and Injury Severity Score) formula. Including pre-injury ASA-PS score might improve the predictive power of a survival prediction model without complicating it.

Canadian benchmarks in trauma

BACKGROUND

Benchmarks are used in trauma care for program evaluation, quality improvement, and research. National outcome benchmarks relevant to the Canadian trauma population need to be defined for evaluation of trauma care in Canada. The purpose of this study was to derive survival probabilities associated with trauma diagnoses using International Classification of Diseases, Ninth Revision (ICD-9) codes.

METHODS

All patients admitted to an acute care hospital with nonpenetrating trauma and submitted to the National Trauma Registry of Canada between 1994 through 2000 inclusively were included in analyses. Both inclusive and exclusive survival risk ratios (SRRs) were calculated for groups of ICD-9 injury codes between 800 to 959.

RESULTS

For the study period, there were 1,003,905 and 803,776 eligible trauma patients used to calculate inclusive SRRs and exclusive SRRs, respectively. Survival probabilities for injuries are given according to ICD-9 codes.

CONCLUSION

This is the first study to define national survival benchmarks for the Canadian trauma population. These results can be used to assess survival of patients using the ICISS [(ICD-9) based Injury Severity Score (ISS)] methodology. With regular updates, these data can further be developed for continual trauma outcome assessment, quality improvement, and research into trauma care in Canada.

The importance of quality of survival as an outcome measure for an integrated trauma system

Risk-adjusted survival rates have been the principle mode of comparison between trauma systems. In mature trauma systems, it is possible that there will be further improvements in survival but these are likely to be small. In the future, the largest gains will come from quality of life and improved function of the survivors. The issues related to measuring quality of survival for trauma systems are reviewed, including feasibility, ethical considerations, risk adjustment of outcomes of survivors, and challenges for selection of instruments and administration. In addition, the preliminary experiences of measuring outcomes in survivors through the Victorian State Trauma Registry are discussed. Although function and quality of life have been identified as important factors to measure in trauma populations, a standardised protocol has not been established. The experience in Victoria suggests that monitoring of population-based outcomes in survivors is feasible and may create the basis for benchmarking the level of morbidity in survivors.

Incorporating Recent Advances To Make the TRISS Approach Universally Available

BACKGROUND

The Trauma and Injury Severity Score (TRISS), used to garner predictions of survival from the Injury Severity Score (ISS), the Revised Trauma Score (RTS, for physiologic reserve), and age is difficult for many trauma facilities to compute because it requires 8 to 10 variables and ISS depends on the specialized Abbreviated Injury Scale (AIS) scale rather than the International Classification of Diseases scale (ICD-9). It has been shown that metrics describing a patient's worst injury (WORSTSRR) are a powerful predictor of survival (regardless of coding type, AIS versus ICD-9) and that the Glasgow Coma Scale (GCS) motor component contains the majority of the information found in the full GCS score. This study hypothesized that the TRISS approach could be made more predictive and efficient with fewer variables by incorporating these advances.

METHODS

A total of 310,958 patients with nonmissing TRISS variables were subset from the National Trauma Data Bank (NTDB). Logistic regression was used to model mortality as a function of anatomic, physiologic and age variables. A traditional TRISS model was computed (with NTDB-derived coefficients) that uses ISS, RTS, age index, and mechanism to predict survival. Four smaller three- or four-variable models employed the ICD-9 WORSTSRR, the GCS motor component, and age (both continuously and dichotomously). Two of the four models also use mechanism. These models were compared using the concordance index (c-index, a measure of model discrimination) and the pseudo-R statistic (estimates proportion of variance explained).

RESULTS

Each experimental model (two models with 3 variables and two models with 4 variables) have superior discrimination and explain more variance than the traditional TRISS model that employs 8-10 variables.

CONCLUSIONS

Recent advances in anatomic and physiologic scoring markedly simplify TRISS-type models at no cost to prediction. This approach uses routinely available data, requires up to seven fewer terms, and predicts at least as well as the original TRISS. These findings could increase the availability of accurate trauma scoring tools to smaller trauma facilities.