Predictors of Outcome in Traumatic Brain Injury: New Insight Using Receiver Operating Curve Indices and Bayesian Network Analysis

Abstract

BackgroundTraumatic brain injury remains a global health problem. Understanding the relative importance of outcome predictors helps optimize our treatment strategies by informing assessment protocols, clinical decisions and trial designs. In this study we establish importance ranking for outcome predictors based on receiver operating indices to identify key predictors of outcome and create simple predictive models. We then explore the associations between key outcome predictors using Bayesian networks to gain further insight into predictor importance.MethodsWe analyzed the corticosteroid randomization after significant head injury (CRASH) trial database of 10008 patients and included patients for whom demographics, injury characteristics, computer tomography (CT) findings and Glasgow Outcome Scale (GCS) were recorded (total of 13 predictors, which would be available to clinicians within a few hours following the injury in 6945 patients). Predictions of clinical outcome (death or severe disability at 6 months) were performed using logistic regression models with 5-fold cross validation. Predictive performance was measured using standardized partial area (pAUC) under the receiver operating curve (ROC) and we used Delong test for comparisons. Variable importance ranking was based on pAUC targeted at specificity (pAUCSP) and sensitivity (pAUCSE) intervals of 90–100%. Probabilistic associations were depicted using Bayesian networks.ResultsComplete AUC analysis showed very good predictive power (AUC = 0.8237, 95% CI: 0.8138–0.8336) for the complete model. Specificity focused importance ranking highlighted age, pupillary, motor responses, obliteration of basal cisterns/3rd ventricle and midline shift. Interestingly when targeting model sensitivity, the highest-ranking variables were age, severe extracranial injury, verbal response, hematoma on CT and motor response. Simplified models, which included only these key predictors, had similar performance (pAUCSP = 0.6523, 95% CI: 0.6402–0.6641 and pAUCSE = 0.6332, 95% CI: 0.62–0.6477) compared to the complete models (pAUCSP = 0.6664, 95% CI: 0.6543–0.679, pAUCSE = 0.6436, 95% CI: 0.6289–0.6585, de Long p value 0.1165 and 0.3448 respectively). Bayesian networks showed the predictors that did not feature in the simplified models were associated with those that did.ConclusionWe demonstrate that importance based variable selection allows simplified predictive models to be created while maintaining prediction accuracy. Variable selection targeting specificity confirmed key components of clinical assessment in TBI whereas sensitivity based ranking suggested extracranial injury as one of the important predictors. These results help refine our approach to head injury assessment, decision-making and outcome prediction targeted at model sensitivity and specificity. Bayesian networks proved to be a comprehensive tool for depicting probabilistic associations for key predictors giving insight into why the simplified model has maintained accuracy.

Data driven analysis reveals shared transcriptome response and immune cell composition across aetiologies of critical illness

“…We also tested internal validation on the trauma dataset using k-fold cross-validation. Model performance was assessed using area under the receiver operator curve (33,34).…”

Logic Synthesis as an Efficient Means of Minimal Model Discovery from Multivariable Medical Datasets

“…We analyzed the dataset of Corticosteroid Randomization after Significant Head Injury (CRASH) trial [47] using an implementation of ESPRESSO-II in R [48]. Study variables include demographics, injury characteristics and computed tomography (CT) findings which are clinically important predictors of TBI outcome [48] [49] [50]. Outcome measure is death or severe disability versus moderate disability or good recovery at 6 months.…”

“…Outcome measure is death or severe disability versus moderate disability or good recovery at 6 months. For a full description of the dataset we refer the readers to [50]. Table 5.…”

“Multimorbidity states with high sepsis-related deaths: a data-driven analysis in critical care”

“…Critical care is one of the most data-intense health care environments (17) yielding a broad range of high dimensional data (18).This lends itself well to advanced analytics such as machine learning (19,20), which allows detection of complex, clinically relevant patterns. Latent class analysis has become increasingly utilized in the discovery of clinically relevant patient subclasses (14,15,21).…”