Predictive Likelihood Research Articles

The most likely common cause

The common cause principle for two random variables A and B is examined in the case of causal insufficiency, when their common cause C is known to exist, but only the joint probability of A and B is observed. As a result, C cannot be uniquely identified (the latent confounder problem). We show that the generalized maximum likelihood method can be applied to this situation and allows identification of C that is consistent with the common cause principle. It closely relates to the maximum entropy principle. Investigation of the two binary symmetric variables reveals a non-analytic behavior of conditional probabilities reminiscent of a second-order phase transition. This occurs during the transition from correlation to anti-correlation in the observed probability distribution. The relation between the generalized likelihood approach and alternative methods, such as predictive likelihood and minimum common entropy, is discussed. The consideration of the common cause for three observed variables (and one hidden cause) uncovers causal structures that defy representation through directed acyclic graphs with the Markov condition.

Influence of Corrosion on Lifespan of Reinforced Concrete Structures: A Comprehensive Review

Carbonation-induced corrosion is a major concern for reinforced cement concrete (RCC) structures, impacting their long-term durability and structural integrity. This review synthesizes the findings on the deterioration mechanisms in concrete structures, focusing on carbonation, chloride-induced corrosion, and time-dependent deterioration. The analysis includes discussions on predictive likelihood methods for estimating bridge reliability, the impact of environmental factors on carbonation, and standardized testing methods for assessing concrete durability. The review highlights the importance of understanding material characteristics and environmental conditions in designing durable concrete structures, emphasizing the processes of carbonation, its impact on rebar corrosion, and strategies for mitigation. Sheltered concrete carbonation resistance in metropolitan tropical climates is 10-20% lower than open exposure. SCM concretes exhibit equivalent or greater long-term carbonation resistance to OPC concretes, as evidenced by the increase in carbonation depth ( Δ xd) at ages greater than 5 years. The paper concludes with recommendations for integrating advanced modeling techniques and empirical studies to develop robust maintenance strategies and improve concrete mix designs for enhanced durability.

Awareness and Knowledge of Congenital Cytomegalovirus as an Agent of Hearing Loss: A Descriptive Evidence from Nigeria

Background Neonatal hearing loss has become a serious public health concern over the last two decades, largely because of the increasing number of reported cases associated with prelingual causes. Objective This study provided a descriptive report on the awareness and knowledge of congenital cytomegalovirus (cCMV) as a causative agent of hearing loss in the Nigerian population. Methods The study adopted a multistage sampling procedure. Data generation for analysis was achieved using a structured research questionnaire. A questionnaire on the issue of cCMV and hearing disability was sent to a total of 324 relevant stakeholders, and 324 completed questionnaires were received in return. The data was carefully coded and analyzed using descriptive and inferential statistical procedures. Results Experience sharing among professional colleagues, particularly those in the healthcare sector, remains the major source of information about cCMV in Nigeria. A significant association was found between the respondents’ professional work experience, organizational type, highest educational qualifications, and awareness and knowledge of cCMV. No predictive likelihood could be established between gender differences (male or female), organizational type (private or public), and awareness or knowledge of cCMV. Conclusion Information from and experience sharing among professional colleagues, family and friends and through the internet were major sources of information about cCMV in Nigeria. Work experience, organizational type, and educational qualifications determine the level of awareness and knowledge about cCMV in Nigeria.

A Robust Interacting Multi-Model Multi-Bernoulli Mixture Filter for Maneuvering Multitarget Tracking under Glint Noise.

In practical radar systems, changes in the target aspect toward the radar will result in glint noise disturbances in the measurement data. The glint noise has heavy-tailed characteristics and cannot be perfectly modeled by the Gaussian distribution widely used in conventional tracking algorithms. In this article, we investigate the challenging problem of tracking a time-varying number of maneuvering targets in the context of glint noise with unknown statistics. By assuming a set of models for the possible motion modes of each single-target hypothesis and leveraging the multivariate Laplace distribution to model measurement noise, we propose a robust interacting multi-model multi-Bernoulli mixture filter based on the variational Bayesian method. Within this filter, the unknown noise statistics are adaptively learned while filtering and the predictive likelihood is approximately calculated by means of the variational lower bound. The effectiveness and superiority of our proposed filter is verified via computer simulations.

Applying XGBoost and SHAP to Open Source Data to Identify Key Drivers and Predict Likelihood of Wolf Pair Presence.

Wolves have returned to Germany since 2000. Numbers have grown to 209 territorial pairs in 2021. XGBoost machine learning, combined with SHAP analysis is applied to predict German wolf pair presence in 2022 for 10 × 10 km grid cells. Model input consisted of 38 variables from open sources, covering the period 2000 to 2021. The XGBoost model predicted well, with 0.91 as the AUC. SHAP analysis ranked the variables: distance to the closest neighboring wolf pair was the main driver for a grid cell to become occupied by a wolf pair. The clustering tendency of related wolves seems to be an important explanatory factor here. Second was the percentage of wooded area. The next eight variables related to wolf presence in the preceding year, except at fifth, eighth and tenth position in the total order: human density (square root) in the grid, percentage arable land and road density respectively. Other variables including the occurrence of wild prey were the weakest predictors. The SHAP analysis also provided crucial added value in identifying a variable that had threshold values where its contribution to the prediction changed from positive to negative or vice versa. For instance, low density of people increased the probability of wolf pair presence, whereas a high density decreased this probability. Cumulative lift techniques showed that the model performed almost four times better than random prediction. The combination of XGBoost, SHAP and cumulative lift techniques is new in wolf management and conservation, allowing for the focusing of educational and financial resources.

Investigation of breast cancer molecular subtype in a multi-ethnic population using MRI.

Accurate subtyping of breast cancer is crucial for its diagnosis, management, and prognostication. This study aimed to determine the association of magnetic resonance imaging (MRI) breast features with the molecular subtype and aggressiveness of breast cancer in a multi-ethnic population. Treatment-naive patients with invasive breast carcinoma were included in this retrospective study. Breast MRI features were recorded based on the American College of Radiology-Breast Imaging Reporting and Data System (ACR-BIRADS) criteria, with tumour size, and apparent diffusion coefficient value (ADC). The statistical association was tested with Pearson Chi-Square Test of Independence for categorical data or the Kruskal-Wallis/ Mann Whitney U test for numerical data between the MRI features and molecular subtype, receptor status, tumour grade, lymphovascular infiltration (LVI) and axillary lymph node (ALN). Multinomial logistic regression was used to test the predictive likelihood of the significant features. The breast cancer subtypes were determined via immunohistochemistry (IHC) and dual-color dual-hapten in-situ hybridization (D-DISH). The expression statuses of ER, PR, and HER-2, LVI, and ALN were obtained from the histopathology report. The ER / PR / HER-2 was evaluated according to the American Society of Clinical Oncology / College of American Pathologists. The study included 194 patients; 41.8% (n = 81) Chinese, 40.7% (n = 79) Malay, and 17.5% (n = 34) Indian, involving 71.6%(n = 139) luminal-like, 12.9%(n = 25) HER-2 enriched, and 15.5%(n = 30) Triple-negative breast cancer (TNBC). TNBC was associated with rim enhancement (p = 0.002) and peritumoral oedema (p = 0.004). HER-2 enriched tumour was associated with larger tumour size (p = 0.041). Luminal-like cancer was associated with irregular shape (p = 0.005) with circumscribed margin (p = 0.003). Other associations were ER-negative tumour with circumscribed margin (p = 0.002) and PR-negative with round shape (p = 0.001). Tumour sizes were larger in ER-negative (p = 0.044) and PR-negative (p = 0.022). Rim enhancement was significantly associated with higher grade (p = 0.001), and moderate peritumoral oedema with positive axillary lymph node (p = 0.002). Certain MRI features can be applied to differentiate breast cancer molecular subtypes, receptor status and aggressiveness, even in a multi-ethnic population.

A Hybrid Decision Tree and Deep Learning Approach Combining Medical Imaging and Electronic Medical Records to Predict Intubation Among Hospitalized Patients With COVID-19: Algorithm Development and Validation.

Early prediction of the need for invasive mechanical ventilation (IMV) in patients hospitalized with COVID-19 symptoms can help in the allocation of resources appropriately and improve patient outcomes by appropriately monitoring and treating patients at the greatest risk of respiratory failure. To help with the complexity of deciding whether a patient needs IMV, machine learning algorithms may help bring more prognostic value in a timely and systematic manner. Chest radiographs (CXRs) and electronic medical records (EMRs), typically obtained early in patients admitted with COVID-19, are the keys to deciding whether they need IMV. We aimed to evaluate the use of a machine learning model to predict the need for intubation within 24 hours by using a combination of CXR and EMR data in an end-to-end automated pipeline. We included historical data from 2481 hospitalizations at The Mount Sinai Hospital in New York City. CXRs were first resized, rescaled, and normalized. Then lungs were segmented from the CXRs by using a U-Net algorithm. After splitting them into a training and a test set, the training set images were augmented. The augmented images were used to train an image classifier to predict the probability of intubation with a prediction window of 24 hours by retraining a pretrained DenseNet model by using transfer learning, 10-fold cross-validation, and grid search. Then, in the final fusion model, we trained a random forest algorithm via 10-fold cross-validation by combining the probability score from the image classifier with 41 longitudinal variables in the EMR. Variables in the EMR included clinical and laboratory data routinely collected in the inpatient setting. The final fusion model gave a prediction likelihood for the need of intubation within 24 hours as well. At a prediction probability threshold of 0.5, the fusion model provided 78.9% (95% CI 59%-96%) sensitivity, 83% (95% CI 76%-89%) specificity, 0.509 (95% CI 0.34-0.67) F1-score, 0.874 (95% CI 0.80-0.94) area under the receiver operating characteristic curve (AUROC), and 0.497 (95% CI 0.32-0.65) area under the precision recall curve (AUPRC) on the holdout set. Compared to the image classifier alone, which had an AUROC of 0.577 (95% CI 0.44-0.73) and an AUPRC of 0.206 (95% CI 0.08-0.38), the fusion model showed significant improvement (P<.001). The most important predictor variables were respiratory rate, C-reactive protein, oxygen saturation, and lactate dehydrogenase. The imaging probability score ranked 15th in overall feature importance. We show that, when linked with EMR data, an automated deep learning image classifier improved performance in identifying hospitalized patients with severe COVID-19 at risk for intubation. With additional prospective and external validation, such a model may assist risk assessment and optimize clinical decision-making in choosing the best care plan during the critical stages of COVID-19.

Multi-step estimators and shrinkage effect in time series models

Many modern statistical models are used for both insight and prediction when applied to data. When models are used for prediction one should optimise parameters through a prediction error loss function. Estimation methods based on multiple steps ahead forecast errors have been shown to lead to more robust and less biased estimates of parameters. However, a plausible explanation of why this is the case is lacking. In this paper, we provide this explanation, showing that the main benefit of these estimators is in a shrinkage effect, happening in univariate models naturally. However, this can introduce a series of limitations, due to overly aggressive shrinkage. We discuss the predictive likelihoods related to the multistep estimators and demonstrate what their usage implies to time series models. To overcome the limitations of the existing multiple steps estimators, we propose the Geometric Trace Mean Squared Error, demonstrating its advantages. We conduct a simulation experiment showing how the estimators behave with different sample sizes and forecast horizons. Finally, we carry out an empirical evaluation on real data, demonstrating the performance and advantages of the estimators. Given that the underlying process to be modelled is often unknown, we conclude that the shrinkage achieved by the GTMSE is a competitive alternative to conventional ones.

Using Quantum hybrid join prediction model to predict future trajectory motion during transit train movement

Predicting future train motion from the yard to the main track via an autonomous driving train is an essential task in a terminal railway station. Existing models fail to address the multiple agents that cross the track as the train transits. Still, there is available research in this area, such as how to predict scene-compliant trajectories across multiple agents jointly. The number of agents significantly increases prediction space as the number of challenges grows exponentially. This study focuses on joint motion prediction between an interacting agent and a transit train. The complex joint prediction problem is divided into marginal sub-prediction issues. In the joint prediction space where the marginal prediction model and conditional prediction model are processed, our Quantum Hybrid Joint Prediction (QHJP) model effectively classifies both influencers and reactors. The proposed model combined the interacting agent features to improve prediction likelihood at the joint relative motion. We test and analyse our model's effectiveness as subjected to the acquired database from publicly real-world surveillance camera records from terminal railway junctions. We recreate a simulation view of an authentic, interactive agent scenario and compare the evaluation result with the well-known prediction benchmark Waymo Open Motion (WOM) dataset for experimental purposes.

A Dynamic Binary Probit Model with Time-Varying Parameters and Shrinkage Prior*

This article studies a time series binary probit model in which the underlying latent variable depends on its lag and exogenous regressors. The regression coefficients for the latent variable are allowed to vary over time to capture possible model instability. Bayesian shrinkage priors are applied to automatically differentiate fixed and truly time-varying coefficients and thus avoid unnecessary model complexity. I develop an MCMC algorithm for model estimation that exploits parameter blocking to boost sampling efficiency. An efficient Monte Carlo approximation based on the Kalman filter is developed to improve the numerical stability for computing the predictive likelihood of the binary outcome. Benefits of the proposed model are illustrated in a simulation study and an application to forecast economic recessions.

Unobserved components model estimates of credit cycles: Tests and predictions

This paper estimates unobserved components (UC) models with real and financial trends and business and credit cycles to assess different measures of the credit cycle used by policymakers. The permanent components of the real and financial sectors are a Beveridge–Nelson and local linear trend, respectively. The business and credit cycles evolve jointly as a second-order vector autoregression. Bootstrap methods are applied to UC model estimates retrieved from classical optimization of the predictive likelihood of the Kalman filter. Results indicate the slope of the financial trend better predicts the credit to GDP ratio in the United States than the estimated business and credit cycles and the Basel gap. This suggests policymakers should consider permanent shocks to the financial sector when gauging the state of financial stability.

Machine learning model identifies aggressive acute pancreatitis within 48 h of admission: a large retrospective study

BackgroundAcute pancreatitis (AP) with critical illness is linked to increased morbidity and mortality. Current risk scores to identify high-risk AP patients have certain limitations.ObjectiveTo develop and validate a machine learning tool within 48 h after admission for predicting which patients with AP will develop critical illness based on ubiquitously available clinical, laboratory, and radiologic variables.Methods5460 AP patients were enrolled. Clinical, laboratory, and imaging variables were collected within 48 h after hospital admission. Least Absolute Shrinkage Selection Operator with bootstrap method was employed to select the most informative variables. Five different machine learning models were constructed to predictive likelihood of critical illness, and the optimal model (APCU) was selected. External cohort was used to validate APCU. APCU and other risk scores were compared using multivariate analysis. Models were evaluated by area under the curve (AUC). The decision curve analysis was employed to evaluate the standardized net benefit.ResultsXgboost was constructed and selected as APCU, involving age, comorbid disease, mental status, pulmonary infiltrates, procalcitonin (PCT), neutrophil percentage (Neu%), ALT/AST, ratio of albumin and globulin, cholinesterase, Urea, Glu, AST and serum total cholesterol. The APCU performed excellently in discriminating AP risk in internal cohort (AUC = 0.95) and external cohort (AUC = 0.873). The APCU was significant for biliogenic AP (OR = 4.25 [2.08–8.72], P < 0.001), alcoholic AP (OR = 3.60 [1.67–7.72], P = 0.001), hyperlipidemic AP (OR = 2.63 [1.28–5.37], P = 0.008) and tumor AP (OR = 4.57 [2.14–9.72], P < 0.001). APCU yielded the highest clinical net benefit, comparatively.ConclusionMachine learning tool based on ubiquitously available clinical variables accurately predicts the development of AP, optimizing the management of AP.

The electrophysiological correlates of word pre-activation during associative word learning

Human beings continuously make use of learned associations to generate predictions about future occurrences in the environment. Such memory-related predictive processes provide a scaffold for learning in that mental representations of foreseeable events can be adjusted or strengthened based on a specific outcome. Learning the meaning of novel words through picture-word associations constitutes a prime example of associative learning because pictures preceding words can trigger word prediction through the pre-activation of the related mnemonic representations. In the present electroencephalography (EEG) study, we used event-related potentials (ERPs) to compare neural indices of word pre-activation between a word learning condition with maximal prediction likelihood and a non-learning control condition with low prediction. Results revealed that prediction-related N400 amplitudes in response to pictures decreased over time at central electrodes as a function of word learning, whereas late positive component (LPC) amplitudes increased. Notably, N400 but not LPC changes were also predictive of word learning performance, suggesting that the N400 component constitutes a sensitive marker of word pre-activation during associative word learning.

Predicting the Risk of Diabetic Foot Ulcers From Diabetics With Dysmetabolism: A Retrospective Clinical Trial.

BackgroundDiabetic foot ulcer (DFU) in patients with type 2 diabetes mellitus (T2D) often leads to amputation. Early intervention to prevent DFU is urgently necessary. So far, there have been no studies on predictive models associated with DFU risk factors. Our study aimed to quantify the predictive risk value of DFU, promote health education, and further develop behavioral interventions to reduce the incidence of DFU.MethodsData from 973 consecutive patients with T2D was collected from two hospitals. Patients from the Guangxi Medical University First Affiliated Hospital formed the training cohort (n = 853), and those from the Wuming Hospital of Guangxi Medical University formed the validation cohort (n = 120). Independent variable grouping analysis and multivariate logistic regression analysis were used to determine the risk factors of DFUs. The prediction model was established according to the related risk factors. In addition, the accuracy of the model was evaluated by specificity, sensitivity, predictive value, and predictive likelihood ratio.ResultsIn total, 369 of the 853 patients (43.3%) and 60 of the 120 (50.0%) were diagnosed with DFUs in the two hospitals. The factors associated with DFU were old age, male gender, lower body mass index (BMI), longer duration of diabetes, history of foot disease, cardiac insufficiency, no use of oral hypoglycemic agent (OHA), high white blood cell count, high platelet count, low hemoglobin level, low lymphocyte absolute value, and high postprandial blood glucose. After incorporating these 12 factors, the nomogram drawn achieved good concordance indexes of 0.89 [95% confidence interval (CI): 0.87 to 0.91] in the training cohort and 0.84 (95% CI: 0.77 to 0.91) in the validation cohort in predicting DFUs and had well-fitted calibration curves. Patients who had a nomogram score of ≥180 were considered to have a low risk of DFU, whereas those having ≥180 were at high risk.ConclusionsA nomogram was constructed by combining 12 identified risk factors of DFU. These 12 risk factors are easily available in hospitalized patients, so the prediction of DFU in hospitalized patients with T2D has potential clinical significance. The model provides a reliable prediction of the risk of DFU in patients with T2D.

Accelerated parallel non-conjugate sampling for Bayesian non-parametric models

Inference of latent feature models in the Bayesian nonparametric setting is generally difficult, especially in high dimensional settings, because it usually requires proposing features from some prior distribution. In special cases, where the integration is tractable, we can sample new feature assignments according to a predictive likelihood. We present a novel method to accelerate the mixing of latent variable model inference by proposing feature locations based on the data, as opposed to the prior. First, we introduce an accelerated feature proposal mechanism that we show is a valid MCMC algorithm for posterior inference. Next, we propose an approximate inference strategy to perform accelerated inference in parallel. A two-stage algorithm that combines the two approaches provides a computationally attractive method that can quickly reach local convergence to the posterior distribution of our model, while allowing us to exploit parallelization.

Bayesian Context Trees: Modelling and Exact Inference for Discrete Time Series

AbstractWe develop a new Bayesian modelling framework for the class of higher-order, variable-memory Markov chains, and introduce an associated collection of methodological tools for exact inference with discrete time series. We show that a version of the context tree weighting alg-orithm can compute the prior predictive likelihood exa-ctly (averaged over both models and parameters), and two related algorithms are introduced, which identify the a posteriori most likely models and compute their exact posterior probabilities. All three algorithms are deterministic and have linear-time complexity. A family of variable-dimension Markov chain Monte Carlo samplers is also provided, facilitating further exploration of the posterior. The performance of the proposed methods in model selection, Markov order estimation and prediction is illustrated through simulation experiments and real-world applications with data from finance, genetics, neuroscience and animal communication. The associated algorithms are implemented in the R package BCT.

Posterior Summaries of Grocery Retail Topic Models: Evaluation, Interpretability and Credibility

AbstractUnderstanding the shopping motivations behind market baskets has significant commercial value for the grocery retail industry. The analysis of shopping transactions demands techniques that can cope with the volume and dimensionality of grocery transactional data while delivering interpretable outcomes. Latent Dirichlet allocation (LDA) allows processing grocery transactions and the discovering of customer behaviours. Interpretations of topic models typically exploit individual samples overlooking the uncertainty of single topics. Moreover, training LDA multiple times show topics with large uncertainty, that is, topics (dis)appear in some but not all posterior samples, concurring with various authors in the field. In response, we introduce a clustering methodology that post-processes posterior LDA draws to summarise topic distributions represented as recurrent topics. Our approach identifies clusters of topics that belong to different samples and provides associated measures of uncertainty for each group. Our proposed methodology allows the identification of an unconstrained number of customer behaviours presented as recurrent topics. We also establish a more holistic framework for model evaluation, which assesses topic models based not only on their predictive likelihood but also on quality aspects such as coherence and distinctiveness of single topics and credibility of a set of topics. Using the outcomes of a tailored survey, we set thresholds that aid in interpreting quality aspects in grocery retail data. We demonstrate that selecting recurrent topics not only improves predictive likelihood but also outperforms interpretability and credibility. We illustrate our methods with an example from a large British supermarket chain.

Diagnostic Utility of Cerebrospinal Fluid Procalcitonin in Neonatal Meningitis.

We aimed to study the diagnostic utility of cerebrospinal fluid (CSF) procalcitonin (PCT) in neonates with meningitis. All the neonates with sepsis who qualified for lumbar puncture were prospectively evaluated. The neonates were classified into Meningitis and No meningitis group based on predefined criteria. CSF PCT was estimated in these neonates along with cytological and biochemical parameters. A total of 113 neonates were included in the study with 29 in the meningitis group and 84 in the no meningitis group. The median PCT levels were higher in babies with meningitis as compared to those without meningitis [0.194 (0.034-0.534) in meningitis group vs. 0.012 (0.012-0.012) ng/ml in no meningitis group, p < 0.001]. The area under curve for CSF PCT was 0.867 (0.77-0.95) and at a cut-off level of 0.120 ng/ml CSF PCT had a sensitivity of 83%, specificity of 84% and positive and negative predictive likelihood ratios of 5.35 and 0.20, respectively for the diagnosis of meningitis. CSF PCT has a good diagnostic accuracy similar to other parameters in the diagnosis of neonatal meningitis and can be considered as an additional diagnostic marker particularly when CSF culture is negative and cytochemical analysis is inconclusive.

The Brief Environmental Exposure and Sensitivity Inventory (BREESI): an international validation study

BackgroundChemical intolerance (CI) is a condition that may result in multisystem symptoms triggered by low levels of exposure to xenobiotics such as chemical inhalants, foods, and/or drugs. The population prevalence of self-reported chemical intolerance is estimated to be between 4 and 25% across several countries. Clinicians and researchers require a brief, practical screening tool for identifying chemical intolerance.ObjectivesWe investigated the validity of a three-item screening questionnaire for CI, the Brief Environmental Exposure and Sensitivity Inventory (BREESI). The internationally validated, and widely used 50-item Quick Environmental Exposure and Sensitivity Inventory (QEESI) was used as the reference standard.MethodsFive thousand individuals (n = 1000 in each of five countries: the US, Japan, Italy, Mexico, and India) responded to both the QEESI and the BREESI using an online research survey platform. We determined the statistical performance metrics for the BREESI, comparing the number of items chosen on the BREESI with QEESI scores for chemical intolerance. Logistic regression was used to determine the likelihood of chemical intolerance based on endorsing 0, 1, 2, or 3 items on the BREESI. We report the BREESI's sensitivity and specificity, positive and negative predictive values, and positive and negative likelihood ratios.ResultsCompared to the QEESI reference standard, the BREESI had excellent sensitivity, specificity, positive and negative likelihood ratios, and positive and negative predictive values for chemical intolerance in all countries except than in Japan, the negative predictive value was poor. Notwithstanding, logistic regression curves show that in all of the countries, for each one-unit increase in the number of BREESI items, there is a 4- to 5-fold increase in the odds of CI.DiscussionThis study confirms the results of two recently published validation papers in the US. The BREESI performs well as a screening tool for chemical intolerance. It is a practical screening tool for researchers, clinicians, and epidemiologists seeking to understand and address this important and prevalent condition.

Trauma triage criteria as predictors of severe injury - a Swedish multicenter cohort study

BackgroundAdequate performance of trauma team activation (TTA) criteria is important in order to accurately triage trauma patients. The Swedish National Trauma Triage Criteria (SNTTC) consists of 29 criteria that trigger either a Trauma Alert, the highest level of TTA, or a Trauma Response. This study aimed to evaluate the SNTTC and its accuracy in predicting a severely injured patient in a multicenter setting.MethodsA cohort study in Sweden involving six trauma receiving hospitals. Data was collected from the Swedish Trauma Registry. Some 626 patients were analyzed with regard to the specific criteria used to initiate the TTA, injury severity with New Injury Severity Score (NISS) and emergency interventions. Sensitivity, specificity, positive predictive value (PPV) and positive likelihood ratio (LR+) of the criteria were calculated, as well as undertriage and overtriage.ResultsAll 29 criteria of SNTTC had a sensitivity > 80% for identifying a severely injured patient. The 16 Trauma Alert Criteria had a lower sensitivity of 62.6% but higher LR+ (3.5 vs all criteria 1.4), specificity (82.3 vs 39.1%) and PPV (55.4 vs 37.6%) and the highest accuracy (AUC 0.724). When using only the six physiological criteria, sensitivity (44.8%) and accuracy (AUC 0.690) decreased while LR+ (6.7), specificity (93.3%) and PPV (70.2%) improved.ConclusionSNTTC is efficient in identifying severely injured patients. The current set of criteria exhibits the best sensitivity compared to other examined combinations and no additional criterion was found to improve the protocol enough to promote a change.