Trauma And Injury Severity Score Model Research Articles

Prediction of Outcome Based on Trauma and Injury Severity Score, IMPACT and CRASH Prognostic Models in Moderate-to-Severe Traumatic Brain Injury in the Elderly.

Objectives:This study aimed to evaluate the trauma and injury severity score (TRISS), IMPACT (international mission for prognosis and analysis of clinical trials), and CRASH (corticosteroid randomization after significant head injury) prognostic models for prediction of outcome after moderate-to-severe traumatic brain injury (TBI) in the elderly following road traffic accident.Design:This was a prospective observational study.Materials and Methods:This was a prospective observational study on 104 elderly trauma patients who were admitted to tertiary care hospital, over a consecutive period of 18 months from December 2016 to May 2018. On the day of admission, data were collected from each patient to compute the TRISS, IMPACT, and CRASH and outcome evaluation was prospectively done at discharge, 14th day, and 6-month follow-up.Results:This study included 104 TBI patients with a mean age of 66.75 years and with a mortality rate of 32% and 45%, respectively, at discharge and at the end of 6 months. The predictive accuracies of the TRISS, CRASH (computed tomography), and IMPACT (core, extended, laboratory) were calculated using receiver operator characteristic (ROC) curves for the prediction of mortality. Best cutoff point for predicting mortality in elderly TBI patients using TRISS system was a score of ≤88 (sensitivity 94%, specificity of 80%, and area under ROC curve 0.95), similarly cutoff point under the CRASH at 14 days was score of >35 (100%, 80%, 0.958); for CRASH at 6 months, best cutoff point was at >84 (88%, 88%, 0.959); for IMPACT (core), it was >38 (88%, 93%, 0.976); for IMPACT (extended), it was >27 (91%, 89%, 0.968); and for IMPACT (lab), it was >41 (82%, 100%, 0.954). There were statistical differences among TRISS, CRASH (at 14 days and 6 months), and IMPACT (core, extended, lab) in terms of area under the ROC curve (P < 0.0001).Conclusion:IMPACT (core, extended) models were the strongest predictors of mortality in moderate-to-severe TBI when compared with the TRISS, CRASH, and IMPACT (lab) models.

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

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. 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. 1840 patients were included, of whom 1236/1840 (67%) were male. Median age was 59years 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. 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.

Validation of age-specific survival prediction in pediatric patients with blunt trauma using trauma and injury severity score methodology: a ten-year Nationwide observational study

BackgroundIn-hospital mortality in trauma patients has decreased recently owing to improved trauma injury prevention systems. However, no study has evaluated the validity of the Trauma and Injury Severity Score (TRISS) in pediatric patients by a detailed classification of patients’ age and injury severity in Japan. This retrospective nationwide study evaluated the validity of TRISS in predicting survival in Japanese pediatric patients with blunt trauma by age and injury severity.MethodsData were obtained from the Japan Trauma Data Bank during 2009–2018. The outcomes were as follows: (1) patients’ characteristics and mortality by age groups (neonates/infants aged 0 years, preschool children aged 1–5 years, schoolchildren aged 6–11 years, and adolescents aged 12–18 years), (2) validity of survival probability (Ps) assessed using the TRISS methodology by the four age groups and six Ps-interval groups (0.00–0.25, 0.26–0.50, 0.51–0.75, 0.76–0.90, 0.91–0.95, and 0.96–1.00), and (3) the observed/expected survivor ratio by age- and Ps-interval groups. The validity of TRISS was evaluated by the predictive ability of the TRISS method using the receiver operating characteristic (ROC) curves that present the sensitivity, specificity, positive predictive value, negative predictive value, accuracy, area under the receiver operator characteristic curve (AUC) of TRISS.ResultsIn all the age categories considered, the AUC for TRISS demonstrated high performance (0.935, 0.981, 0.979, and 0.977). The AUC for TRISS was 0.865, 0.585, 0.614, 0.585, 0.591, and 0.600 in Ps-interval groups (0.96–1.00), (0.91–0.95), (0.76. − 0.90), (0.51–0.75), (0.26–0.50), and (0.00–0.25), respectively. In all the age categories considered, the observed survivors among patients with Ps interval (0.00–0.25) were 1.5 times or more than the expected survivors calculated using the TRISS method.ConclusionsThe TRISS methodology appears to predict survival accurately in Japanese pediatric patients with blunt trauma; however, there were several problems in adopting the TRISS methodology for younger blunt trauma patients with higher injury severity. In the next step, it may be necessary to develop a simple, high-quality prediction model that is more suitable for pediatric trauma patients than the current TRISS model.

Factors affecting mortality in major trauma patients in Trinidad and Tobago – A view from the developing world

IntroductionThere are few data on major trauma in the developing world. This study investigated the characteristics and outcomes of seriously injured patients in Trinidad and Tobago, using Trauma and Injury Severity Score (TRISS) methodology. We also aimed to assess the predictive accuracy of the TRISS model in patients in Trinidad and Tobago.MethodsRetrospective data from major trauma patients attending the Emergency Department of a tertiary hospital in Trinidad between 2010 and 2014 were analysed. Patients ≥18 years having an Injury Severity Score &gt;15 were included. The impact of age, gender, comorbidities, mechanisms and patterns of injury on mortality was investigated. Using TRISS methodology, predicted mortality was calculated and compared to actual mortality.ResultsOf 323 patients analysed, 284 were male and 24 were aged ≥65 years. The commonest injury mechanisms in younger people were motor vehicle accidents (34.1%) and stabbings (30.8%) compared to falls (66.7%) and motor vehicle accidents (20.8%) in people aged ≥65 years. The commonest areas injured were the chest in younger patients (81.9%) and the head and neck in patients aged ≥65 years (58.3%). Women’s mortality rates were similar to men (RR 1.8; 95% CI 0.7–4.9). Mortality was higher with age ≥65 years (RR 7.0; 95% CI 3.1–15.9), blunt trauma (RR 7.6; 95% CI 1.8–32.4) and Charlson Comorbidity Index of 1 or more (RR 3.2; 95% CI 1.3–8.0). The TRISS model performed well at lower ISS scores and was excellent at predicting survival (discrimination statistic 0.94).ConclusionMultiple factors influence mortality in major trauma patients in Trinidad and Tobago, including age, co-morbidities and injury mechanism. TRISS methodology accurately predicted survival in this population but was better at predicting mortality in patients with lower Injury Severity Score.

External validation of the TRISS, CRASH, and IMPACT prognostic models in severe traumatic brain injury in Japan.

In Japan, a range of patients with traumatic brain injury (TBI) has been recorded in a nationwide database (Japan Neurotrauma Data Bank; JNTDB). This study aimed to externally validate three international prediction models using JNTDB data: Trauma and Injury Severity Score (TRISS), Corticosteroid Randomization After Significant Head Injury (CRASH), and International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT). We also aimed to validate the applicability of these models in the Japanese population. Of 1,091 patients registered in the JNTDB from July 2009 to June 2011, we analyzed data for 635 patients. We examined factors associated with mortality in-hospital and unfavorable outcomes 6 months after TBI by applying the TRISS, CRASH, and IMPACT models. We also conducted an external validation of these models based on these data. The patients’ mean age was 60.1 ±21.1 years, and 342 were alive at the time of discharge (53.9%). Univariate analysis revealed eight major risk factors for mortality in-hospital: age, Glasgow Coma Scale (GCS), Injury Severity Score (ISS), systolic blood pressure, heart rate, mydriasis, acute epidural hematoma (AEDH), and traumatic subarachnoid hemorrhage. A similar analysis identified five risk factors for unfavorable outcomes at 6 months: age, GCS, ISS, mydriasis, and AEDH. For mortality in-hospital, the TRISS had a satisfactory area under the curve value (0.75). For unfavorable outcomes at 6 months, the CRASH (basic and computed tomography) and IMPACT (core and core extended) models had satisfactory area under the curve values (0.86, 0.86, 0.81, and 0.85, respectively). The TRISS, CRASH, and IMPACT models were suitable for application to the JNTDB population, indicating these models had high value in Japanese patients with neurotrauma.

Machine Learning Models of Survival Prediction in Trauma Patients.

Background: We aimed to build a model using machine learning for the prediction of survival in trauma patients and compared these model predictions to those predicted by the most commonly used algorithm, the Trauma and Injury Severity Score (TRISS). Methods: Enrolled hospitalized trauma patients from 2009 to 2016 were divided into a training dataset (70% of the original data set) for generation of a plausible model under supervised classification, and a test dataset (30% of the original data set) to test the performance of the model. The training and test datasets comprised 13,208 (12,871 survival and 337 mortality) and 5603 (5473 survival and 130 mortality) patients, respectively. With the provision of additional information such as pre-existing comorbidity status or laboratory data, logistic regression (LR), support vector machine (SVM), and neural network (NN) (with the Stuttgart Neural Network Simulator (RSNNS)) were used to build models of survival prediction and compared to the predictive performance of TRISS. Predictive performance was evaluated by accuracy, sensitivity, and specificity, as well as by area under the curve (AUC) measures of receiver operating characteristic curves. Results: In the validation dataset, NN and the TRISS presented the highest score (82.0%) for balanced accuracy, followed by SVM (75.2%) and LR (71.8%) models. In the test dataset, NN had the highest balanced accuracy (75.1%), followed by the TRISS (70.2%), SVM (70.6%), and LR (68.9%) models. All four models (LR, SVM, NN, and TRISS) exhibited a high accuracy of more than 97.5% and a sensitivity of more than 98.6%. However, NN exhibited the highest specificity (51.5%), followed by the TRISS (41.5%), SVM (40.8%), and LR (38.5%) models. Conclusions: These four models (LR, SVM, NN, and TRISS) exhibited a similar high accuracy and sensitivity in predicting the survival of the trauma patients. In the test dataset, the NN model had the highest balanced accuracy and predictive specificity.

Refining the Trauma and Injury Severity Score (TRISS) to Measure the Performance of the UK Combat Casualty Care System.

The Trauma and Injury Severity Score (TRISS) methodology is used in both the UK and US Military trauma registries. The method relies on dividing casualties according to mechanism, penetrating or blunt, and uses different weighting coefficients accordingly. The UK Military Joint Theatre Trauma Registry uses the original coefficients devised in 1987, whereas the US military registry uses updated civilian coefficients, but it is not clear how either registry analyzes explosive casualties according to the TRISS methodology. This study aims to use the UK Military Joint Theatre Trauma Registry to calculate new TRISS coefficients for contemporary battlefield casualties injured by either gunshot or explosive mechanisms. The secondary aim of this study is to apply the revised TRISS coefficients to examine the survival trends of UK casualties from recent military conflicts. The Joint Theatre Trauma Registry was searched for all UK casualties injured or killed in Iraq and Afghanistan by explosive or gunshot mechanisms between January 1, 2003 and December 31, 2014. Details of these casualties including injuries and vital signs were reviewed. Logistic regression analysis was performed to devise new TRISS coefficients; these were then used to examine survival over the 12 yr of the study. Comparing the predictions from the gunshot TRISS model to the observed outcomes, it demonstrates a sensitivity of 98.1% and a specificity of 96.8% and an overall accuracy of 97.8%. With respect to the explosive TRISS model, there is a sensitivity of 98.6%, a specificity of 97.4%, and an overall accuracy of 98.4%. When this updated and mechanism-specific TRISS methodology was used to measure changes in survival over the study period, survival following these injuries improved until 2012 when performance was maintained for the last 2 yr of the study. This study for the first time refines the TRISS methodology with coefficients appropriate for use within combat casualty care systems. This improved methodology reveals that UK combat casualty care performance appears to have improved until 2012 when this standard was maintained.

Comparison of risk-adjusted survival in two Scandinavian Level-I trauma centres.

BackgroundAssessment of trauma-system performance is important for improving the care of injured patients. The aim of the study was to compare risk-adjusted survival in two Scandinavian Level-I trauma centres.MethodsThis was an observational, retrospective study of prospectively-collected trauma registry data for patients >14 years from Karolinska University Hospital – Solna (KUH), Sweden, and Oslo University Hospital – Ullevål (OUH), Norway, from 2009-2011. Probability of survival (Ps) was calculated according to the Trauma and Injury Severity Score (TRISS) method. Risk-adjusted survival per patient was calculated by assigning every patient a value corresponding to gained or lost fractional life: Each survivor contributed a reward of 1-Ps and each death a penalty of -Ps. The sum of penalties and rewards, corresponding to the difference between expected and actual mortality, was compared between the centres. We present the data as excess survivors per 100 trauma patients.ResultsThere were 4485 admissions at KUH and 3591 at OUH. The proportion of severely injured patients was higher at OUH compared with KUH (Injury Severity Score [ISS] >15: 33.9 % vs. 21.1 %, p <0.001). OUH had a larger proportion of patients >65 years (16.0 % vs. 13.4 %, p <0.001) and greater comorbidity (ASA-PS ≥3: 14.6 % vs. 6.9 %, p <0.001) compared with KUH. The frequency of helicopter transport and presence of prehospital physicians was higher at OUH compared with KUH (27.6 % vs. 15.5 % and 30.5 % vs. 3.7 %, both p <0.001). Secondary admissions were 5.2-fold more common at OUH compared with KUH (p <0.001). There were no differences in 30-day mortality for severely injured patients (ISS >15). Risk-adjusted survival rate was higher at OUH than at KUH for primary (0.59 vs. 0.51) but lower for secondary (1.41 vs. 2.85) admissions (both p <0.001).ConclusionAdjustments for age as a continuous variable and comorbidity should be made when comparing risk-adjusted survival between hospitals, but this is not possible with the TRISS model. A survival prediction model that takes this into account may be a better choice for Scandinavian trauma populations. The current study could not rule out the influence of the system differences between the centres on risk-adjusted survival.

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...

The Trauma and Injury Severity Score (TRISS) revised

BackgroundThe Trauma and Injury Severity Score (TRISS) remains the most commonly used tool for benchmarking trauma fatality outcome. Recently, it was demonstrated that the predictive power of TRISS could be substantially improved by re-classifying the component variables and treating the variable categories nominally. This study aims to develop revised TRISS models using re-classified variables, to assess these models’ predictive performances against existing TRISS models, and to identify and recommend a preferred TRISS model. Materials and methodsRevised TRISS models for blunt and penetrating injury mechanism were developed on an adult (aged ≥15 years) sample from the National Trauma Data Bank National Sample Project (NSP), using 5-category variable classifications and weighted logistic regression. Their predictive performances were then assessed against existing TRISS models on the unweighted NSP, National Trauma Data Bank (NTDB), and New Zealand Database (NZDB) samples using area under the Receiver Operating Characteristic curve (AUC) and Bayesian Information Criterion (BIC) statistics. ResultsThe weighted NSP sample included 1,124,001 adults with blunt or penetrating injury mechanism events and known discharge status, of whom 1,061,709 (94.5%) survived to discharge. Complete information for all TRISS variables was available for 896,212 (79.7%). Revised TRISS models that included main-effects and two-factor interaction terms had superior AUC and BIC statistics to main-effects models and existing TRISS models for patients with complete data in NSP, NTDB and NZDB samples. Predictive performance decreased as the number of variables with missing values included within revised TRISS models increased, but model performances generally remained superior to existing TRISS models. DiscussionRevised TRISS models had importantly improved predictive capacities over existing TRISS models. Additionally, they were easily computed, utilised only those variables already collected for existing TRISS models, and could be applied and produce meaningful survival probabilities when one or more of the predictor variables contained missing values. The preferred revised TRISS model included main-effects and two-factor interaction terms and allowed for missing values in all predictor variables. A strong case exists for replacing existing TRISS models in trauma scoring systems benchmarking software with this preferred revised TRISS model.

Trauma and Injury Severity Score (TRISS): Is it time for variable re-categorisations and re-characterisations?

BackgroundDespite its limitations, the Trauma and Injury Severity Score (TRISS) continues to be the most commonly used tool for benchmarking trauma outcome. Since its inception, considerable energy has been devoted to improving TRISS. However, there has been no investigation into the classification or characterisation of the TRISS variables. Using a major nationally representative database, this study aims to explore the adequacy of the existing TRISS model by investigating variable re-categorisations and alternative characterisations in a logistic model used to predict survival in adults after traumatic injury. Materials and methodsData were obtained from the National Trauma Data Bank National Sample Project (NSP). Each variable in the TRISS model was related to discharge status and various categorisations considered using weighted logistic regression. Categorisations were treated nominally, using a series of indicator variables. For each variable and classification level, the best category combination was ascertained using the Bayesian Information Criterion (BIC). All best 5-category classified TRISS variables were combined, as were all best 10-category classified TRISS variables, and their predictive performance assessed against two conventionally defined TRISS models on the unweighted NSP sample using area under the Receiver Operating Characteristic curve (AUC) and BIC statistics. ResultsOverall, the weighted sample included 1,124,001 adults with injury events and known discharge status, of whom 1,061,709 (94.5%) were alive at discharge. When separately related to discharge status, each re-classified TRISS variable yielded a superior BIC statistic to its original specification. When investigating predictive performance, complete information was available for 167,239 (79.9%) adults with blunt and 20,643 (82.3%) adults with penetrating injury mechanisms. AUC and BIC estimates for the re-classified TRISS models were superior to the conventionally defined TRISS models. While having better predictive precision, the complexity associated with the best 10-category model resulted in the best 5-category model being preferred for penetrating mechanism injuries and being negligibly inferior for blunt mechanism injuries. DiscussionSubstantial improvements in the predictive power of TRISS were demonstrated by re-classifying the component variables and treating the variable categories nominally. However, before a new TRISS model with updated coefficients can be published, variable interactions and the effect of missing data needs thorough statistical evaluation.

Predicting trauma patient mortality: ICD [or ICD‐10‐AM] versus AIS based approaches

The International Classification of Diseases Injury Severity Score (ICISS) has been proposed as an International Classification of Diseases (ICD)-10-based alternative to mortality prediction tools that use Abbreviated Injury Scale (AIS) data, including the Trauma and Injury Severity Score (TRISS). To date, studies have not examined the performance of ICISS using Australian trauma registry data. This study aimed to compare the performance of ICISS with other mortality prediction tools in an Australian trauma registry. This was a retrospective review of prospectively collected data from the Victorian State Trauma Registry. A training dataset was created for model development and a validation dataset for evaluation. The multiplicative ICISS model was compared with a worst injury ICISS approach, Victorian TRISS (V-TRISS, using local coefficients), maximum AIS severity and a multivariable model including ICD-10-AM codes as predictors. Models were investigated for discrimination (C-statistic) and calibration (Hosmer-Lemeshow statistic). The multivariable approach had the highest level of discrimination (C-statistic 0.90) and calibration (H-L 7.65, P= 0.468). Worst injury ICISS, V-TRISS and maximum AIS had similar performance. The multiplicative ICISS produced the lowest level of discrimination (C-statistic 0.80) and poorest calibration (H-L 50.23, P < 0.001). The performance of ICISS may be affected by the data used to develop estimates, the ICD version employed, the methods for deriving estimates and the inclusion of covariates. In this analysis, a multivariable approach using ICD-10-AM codes was the best-performing method. A multivariable ICISS approach may therefore be a useful alternative to AIS-based methods and may have comparable predictive performance to locally derived TRISS models.

Análisis de la mortalidad en los pacientes politraumatizados según la metodología TRISS en un hospital traumatológico de referencia

A growing number of patients with multiple injuries are being treated. Injury severity scales can be used to assess outcomes objectively. This study aimed to assess our hospital's cases on the basis of the Trauma and Injury Severity Score (TRISS) and compare outcomes to those reported in the Major Trauma Outcome Study, as well as to determine preventable mortality and analyze causes of death and associated factors. Data were extracted from the records of patients admitted with multiple injuries in 2005 and were used to calculate the Revised Trauma Score (RTS), the Injury Severity Score (ISS), and the TRISS or probability of survival. Hospital mortality was also calculated. A TRISS between 25 and 50 was considered to indicate a preventable avoidable death; a TRISS over 50 indicated a preventable death. Logistic regression analysis was used to identify factors associated with mortality. We studied the cases of 198 patients with a mean (SD) age of 43.9 (19) years. Ninety-three percent had suffered blunt trauma. The mean ISS, the prehospital RTS, and the TRISS were 16.9 (11.2), 10.8 (2.5), and 0.95 (0.2), respectively. Twenty-five patients died. Fifteen deaths were classified as preventable or potentially preventable. Factors related to exitus were head injury and age (odds ratios, 4.6 and 4.0, respectively). The rate of preventable death in our hospital was higher than expected. Mortality was strongly associated with head injury and age. The TRISS model can identify preventable deaths objectively.

Base Deficit-Based Predictive Modeling of Outcome in Trauma Patients Admitted to Intensive Care Units in Dutch Trauma Centers

Worldwide, the base deficit is available as an objective indicator of acid base status. We used the base deficit as a measure of physiologic derangement in a Trauma and Injury Severity Score (TRISS)-like model as a predictor for outcome in trauma patients. We prospectively recorded data of 349 consecutive trauma patients admitted to the intensive care unit and calculated Revised Trauma Score, Injury Severity Score and Abbreviated Injury Scale, and TRISS and correlated them with the simultaneously determined base deficit value. The delta base deficit is introduced, which is the absolute difference of the base deficit from its normal range (-2 to 2). A statistical model analogous to the TRISS model was designed in which the physiologic disturbance reflected by the Revised Trauma Score was replaced by the delta base deficit [Base Excess Injury Severity Scale (BISS) model]. Calculating the area under the curve (AUC) of the respective receiver operating characteristic curve compared these two models. Finally, the BISS model was validated in a patient group from another tertiary referral hospital in which similar data were recorded prospectively. We demonstrated a significant correlation between the delta base deficit and the calculated trauma scoring systems. Moreover, the delta base deficit is significantly correlated with mortality. The BISS performed better than the TRISS did when evaluated by the AUC of the receiver operating characteristic curves (AUC 0.806 vs. 0.803, respectively). Validation in an independent prospectively compiled dataset from another referral center showed comparable and even better results (AUC 0.891 vs. 0.885, respectively). The performance of our proposed BISS model was superior to that of the TRISS model in the populations under investigation. Nevertheless, given the ease of assessment and the objective value of the base deficit, it may be considered as a good method to predict outcome and evaluate care of trauma patients. Whether this can be translated to trauma patients in general needs further investigation.

Incorporating Recent Advances To Make the TRISS Approach Universally Available

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. 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). 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. 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.

Comparison of Current Injury Scales for Survival Chance Estimation: An Evaluation Comparing the Predictive Performance of the ISS, NISS, and AP Scores in a Dutch Local Trauma Registration

Prediction of survival chances for trauma patients is a basic requirement for evaluation of trauma care. The current methods are the Trauma and Injury Severity Score (TRISS) and A Severity Characterization of Trauma (ASCOT). Scales for scoring injury severity are part of these methods. This study compared three injury scales, the Injury Severity Score (ISS), the New ISS (NISS), and the Anatomic Profile (AP), in three otherwise identical predictive models. Records of the Rotterdam Trauma Center were analyzed using logistic regression. The variables used in the models were age (as a linear variable), the corrected Revised Trauma Score (RTS), a denominator for blunt or penetrating trauma, and one of the three injury scales. The original TRISS and ASCOT models also were evaluated. The resulting models were compared in terms of their discriminative power, as indicated by the receiver-operator characteristic (ROC), and calibration (Hosmer-Lemeshow [HL]) for the entire range of injury severity. For this study, 1,102 patients, with an average ISS of 15, met the inclusion criteria. The TRISS and ASCOT models, using original coefficients, showed excellent discriminative power (ROC, 0.94 and 0.96, respectively), but insufficient fits (HL, p = 0.001 and p = 0.03, respectively). The three fitted models also had excellent discriminative abilities (ROC, 0.95, 0.97, and 0.96, respectively). The custom ISS model was unable to fit the entire range of survival chances sufficiently (p = 0.01). Models using the NISS and AP scales provided adequate fits to the actual survival chances of the population (HL, 0.32 and 0.12, respectively). The AP and NISS scores particularly both managed to outperform the ISS score in correctly predicting survival chances among a Dutch trauma population. Trauma registries stratifying injuries by the ISS score should evaluate the use of the NISS and AP scores.

Improving the TRISS methodology by restructuring age categories and adding comorbidities.

The Trauma and Injury Severity Score (TRISS) methodology was developed to predict the probability of survival after trauma. Despite many criticisms, this methodology remains in common use. The purpose of this study was to show that improving the stratification for age and adding an adjustment for comorbidity significantly increases the predictive accuracy of the TRISS model. The trauma registry and the hospital administrative database of a regional trauma center were used to identify all blunt trauma patients older than 14 years of age admitted with International Classification of Diseases, Ninth Revision codes 800 to 959 from April 1993 to March 2001. Each individual medical record was then reviewed to ascertain the Revised Trauma Score, the Injury Severity Score, the age of the patients, and the presence of eight comorbidities. The outcome variable was the status at discharge: alive or dead. The study population was divided into two subsamples of equal size using a random sampling method. Logistic regression was used to develop models on the first subsample; a second subsample was used for cross-validation of the models. The original TRISS and three TRISS-derived models were created using different categorizations of Revised Trauma Score, Injury Severity Score, and age. A new model labeled TRISSCOM was created that included an additional term for the presence of comorbidity. There were 5,672 blunt trauma patients, 2,836 in each group. For original TRISS, the Hosmer-Lemeshow statistic (HL) was 179.1 and the area under the receiver operating characteristic (AUROC) curve was 0.873. Sensitivity and specificity were 99.0% and 27.8%, respectively. For the best modified TRISS model, the HL statistic was 20.35, the AUROC curve was 0.902, the sensitivity was 99.0%, and the specificity was 27.8%. For TRISSCOM, the HL statistic was 14.95 and the AUROC curve was 0.918. Sensitivity and specificity were 99.0% and 29.7%, respectively. The difference between the two models almost reached statistical significance (p = 0.086). When TRISSCOM was applied to the cross-validation group, the HL statistic was 10.48 and the AUROC curve was 0.914. The sensitivity was 98.6% and the specificity was 34.9%. TRISSCOM can predict survival more accurately than models that do not include comorbidity. A better categorization of age and the inclusion of comorbid conditions in the logistic model significantly improves the predictive performance of TRISS.

Validation of Trauma and Injury Severity Score in Blunt Trauma Patients by Using a Canadian Trauma Registry

To compare outcomes in blunt trauma by using Trauma and Injury Severity Score (TRISS) models derived from the Major Trauma Outcome Study (MTOS) and the Ontario Trauma Registry (OTR) as well as to evaluate the role of the Revised Trauma Score within the TRISS model. Consecutive blunt trauma cases from 11 Level I trauma centers over a 4-year period were identified from the OTR. Coefficients of the Revised Trauma Score were modified using the Ontario data and this score was tested by using the Hosmer-Lemeshow Goodness of Fit Test. Two Ontario-specific TRISS models were developed with revised coefficients. The first used the standard Revised Trauma Score and the second used the Revised Trauma Score with regenerated coefficients. The accuracy of mortality predictions for all models were compared by using a Hosmer-Lemeshow Goodness of Fit procedure. Additionally, each TRISS models performance characteristics and receiver operating characteristic (ROC) curves were used to evaluate their discriminative capabilities. A total of 5,436 cases were incorporated in the analysis. Patients with all component TRISS variables had a significantly lower mortality compared to all blunt trauma patients (7.0% vs. 15.5%,p < 0.01). Use of the Revised Trauma Score led to the exclusion of 40% of cases because of absent data necessary to compute the score. The Hosmer-Lemeshow Goodness of Fit statistic for the Revised Trauma Score was 79.45 (p = 0.0001). The Hosmer-Lemeshow Goodness of Fit Statistic ranged from 11.42, p = 0.175 and 13.1, p = 0.125 for the Ontario TRISS models compared to 25.62, p < 0.005 for the MTOS TRISS model. Sensitivity of all three TRISS models ranged from 98% to 99% with specificity ranging from 24% to 35%. ROC curves were identical for all three TRISS models. TRISS demonstrated satisfactory performance in a Canadian blunt trauma population. Although revision of coefficients led to a better fit on the Hosmer-Lemeshow statistic, ROC curves demonstrated virtually identical performance of the MTOS and Ontario-based TRISS models. The poor performance of the Revised Trauma Score and the observation that its use led to the exclusion of 40% of cases with a higher mortality raises concerns regarding its use in the TRISS model.

Validation of TRISS and ASCOT using a non-MTOS trauma registry.

To validate the Trauma and Injury Severity Score (TRISS) and A Severity Characterization of Trauma (ASCOT) models for patients with blunt injuries using an independent trauma registry, and to develop new TRISS and ASCOT models for types of patients with blunt injuries and examine their fit. Retrospective analysis of clinical data from the Institute for Trauma and Emergency Care (ITEC). Statistical models were developed using TRISS and ASCOT variables applied to ITEC data for patients with blunt injuries. These models were compared to Major Trauma Outcome Study (MTOS) models with regard to the resulting coefficients and hospital quality assessments. Also, separate models were developed for different groups of blunt injuries, and these models were compared with one another and tested for adequacy of fit. ASCOT performed acceptably well when new coefficients were derived using ITEC data, but TRISS did not. Although the models developed from MTOS and from ITEC coefficients generally yielded similar hospital quality assessments, there were some notable exceptions. Some TRISS and ASCOT variables were not significantly related to survival for some subgroups of blunt injuries, and neither the TRISS nor the ASCOT model was an adequate predictor of survival for patients suffering from low falls. New TRISS and ASCOT coefficients should be derived if survival for patients with blunt injuries is to be predicted accurately in independent trauma registries. Also, it may be wise to consider developing separate models for subgroups of patients, particularly if hospitals in the registry have different mixes of patient types.