Harborview Assessment for Risk of Mortality: An Improved Measure of Injury Severity on the Basis of ICD-9-CM

Abstract

Background: There have been sev-

The impact of time of admission on major complications and mortality in patients undergoing emergency trauma surgery

“…Our study suggests that for patients requiring emergency trauma surgery, major complica- tions and mortality are not associated with whether patients presented during the night or day, on a weekend versus a weekday, month of presentation, or year of presentation. In our analysis, sex was not a significant predictor of outcome following trauma care and this is consistent with other studies (West et al 2000, Brennan et al 2002, Croce et al 2002, Rappold et al 2002. The narrow confidence intervals surrounding many of our odds ratios provide greater confidence in these estimates of effect.…”

Diagnosis based injury severity scaling: investigation of a method using Australian and New Zealand hospitalisations

“…The recently developed severity measure HARM (Harborview Assessment for Risk of Mortality) included some comorbidities. 24 One approach would be to calculate SRRs for each chapter of the noninjury diagnoses.…”

The trauma severity model: An ensemble machine learning approach to risk prediction

“…The potential impact of trauma injuries on one's quality of life has inspired several studies on how we can improve the quality of trauma care, and consequently improve trauma patient outcomes. However, several of these studies require that we take a trauma patient's injury severity (risk of mortality) into account, and there is no consensus as to which risk prediction model is most appropriate for use [2][3][4][5][6][7][8][9][10][11][12][13][14].…”

“…One reason for this is that several risk prediction models have been developed from small data sets [2-4, 6, 8, 10-14], which implies that these models may not represent the population appropriately [15]. Another reason is that even if a model was developed from a large data set, nearly every model proposed thus far has been a linear model [2][3][4][5][6][7][8][9], and linear models may be insufficient for capturing the potentially complex relationships that exist within trauma data [16][17][18]. theory, which are leading to substantial changes in both the availability of large data sets and the ability to perform intensive analyses on such data sets, our objective is to develop a risk prediction model from machine learning algorithms and compare its predictive performance to the performance of other established risk prediction models.…”

“…theory, which are leading to substantial changes in both the availability of large data sets and the ability to perform intensive analyses on such data sets, our objective is to develop a risk prediction model from machine learning algorithms and compare its predictive performance to the performance of other established risk prediction models. We achieve this by comparing three leading risk prediction models -the Bayesian Logistic Injury Severity Score (BLISS) [7], the Harborview Assessment for Risk of Mortality (HARM) [8], and the Trauma Mortality Prediction Model (TMPM-ICD9) [9] -to a new machine learning model for risk prediction. This machine learning model is the Trauma Severity Model (TSM).…”