Related Experiment Videos
Improving the Glasgow Coma Scale score: motor score alone is a better predictor
C Healey1, Turner M Osler, Frederick B Rogers
1Department of Surgery, University of Vermont, College of Medicine, Burlington 05401, USA.
The Journal of Trauma
|April 23, 2003
Summary
The motor component of the Glasgow Coma Scale (GCS) better predicts patient outcomes than the full GCS. This motor assessment is statistically sound and should replace the GCS in trauma outcome models.
Area of Science:
- Trauma research
- Neurological assessment
- Outcome prediction modeling
Background:
- The Glasgow Coma Scale (GCS) is widely used for head trauma assessment and in physiological outcome models.
- However, its rigorous examination as a direct predictor of patient outcomes is lacking.
Purpose of the Study:
- To compare the predictive power and calibration of the GCS against its individual components.
- To evaluate the GCS's utility in large-scale trauma data analysis.
Main Methods:
- Utilized the National Trauma Data Bank (N = 204,181) for analysis.
- Compared the predictive performance (pseudo R2, ROC) and calibration of the GCS and its components (motor, verbal, eye).
Main Results:
- The GCS score is a sum of 120 combinations, leading to varied mortality predictions for the same score.
- The motor (m) component alone demonstrates a linear relationship with survival and retains nearly all predictive power of the full GCS.
- The GCS score shows a non-linear relationship with survival and results in poorly calibrated logistic models.
Conclusions:
- The motor component of the GCS is a superior predictor of outcome compared to the full GCS.
- It is measurable in intubated patients and statistically more robust.
- The motor component should replace the GCS in outcome prediction models, with appropriate mathematical transformation for non-linear log odds of survival.