Related Experiment Video
Updated: Mar 9, 2026
10:08
Pseudofracture: An Acute Peripheral Tissue Trauma Model
Published on: April 18, 2011
15.3K
Mortality prediction models in the general trauma population: A systematic review
Leonie de Munter1, Suzanne Polinder2, Koen W W Lansink3
1Department Trauma TopCare, Elisabeth-TweeSteden Hospital, Tilburg, The Netherlands.
Injury
|December 25, 2016
Summary
This review highlights the need for improved trauma mortality prediction models. Current models vary in quality and predictive power, emphasizing the need for better validation and variable selection in trauma research.
Area of Science:
- Trauma research
- Medical informatics
- Public health
Background:
- Trauma is a leading cause of death for individuals under 40.
- Existing trauma mortality prediction models lack comprehensive review.
- This review addresses existing models, their quality, and relevant predictors.
Purpose of the Study:
- To describe existing mortality prediction models for the general trauma population.
- To assess the methodological quality of these models.
- To identify the most relevant variables for predicting mortality in trauma patients.
Main Methods:
- A systematic online search was conducted in June 2015 across multiple databases.
- Included peer-reviewed articles developed, validated, or updated mortality prediction models.
- Focused on models for the general trauma population.
Main Results:
- 90 articles yielded 258 distinct mortality prediction models.
- The Trauma and Injury Severity Score (TRISS) was most common.
- Only 40% of models were externally validated, with variable performance and predictor handling issues.
Conclusions:
- Improved trauma mortality prediction models are needed.
- Models require adequate sample sizes, robust handling of missing data, and use of continuous variables.
- Future models should incorporate diverse predictors and consider outcomes beyond mortality, such as quality of life.
