The rSIG for trauma: one size fits all？

We read with great interest the article by Frieler et al.1 The authors validated the accuracy of the rSIG (SBP/HRxGCS) in predicting the risk of mortality in trauma patients. The results of this study are positive. However, we have some concerns before rSIG is commonly used in the prehospital setting. First, the site of the trauma is a significant contributor to the large degree of heterogeneity in trauma; hence, anatomical variables should be considered in clinical prediction models.2 For example, there was a U-shaped relationship between SI values and mortality in patients with head injuries, which means that hypertension is also associated with increased mortality in patients with head injuries.3 Second, this study did not evaluate the rSIG using calibration, which is a methodological issue with many clinical prediction models.4 Calibration measures are essential to determine whether the projected probabilities actually...

Source: Emergency Medicine Journal - June 22, 2023 Category: Emergency Medicine Authors: Shi, Q., Mao, Z. Tags: PostScript Source Type: research