Letter to the editor
Should scoring rules be based on odds ratios or regression coefficients?

https://doi.org/10.1016/S0895-4356(02)00453-5Get rights and content

Section snippets

Acknowledgements

We gratefully acknowledge the research support by the Netherlands Organization for Scientific Research (NWO: nr. R96–203).

Cited by (207)

  • Predicting mortality in moderate-severe TBI patients without early withdrawal of life-sustaining treatments including ICU complications: The MYSTIC-score

    2022, Journal of Critical Care
    Citation Excerpt :

    We assessed model discrimination using the C-statistic (area-under-the-receiver-operating-characteristics-curve [AUC-ROC]) and examined calibration with both the Hosmer-Lemeshow-goodness-of-fit (HL) test and calibration plots. Next, we developed the MYSTIC-Score based on the regression coefficients from the final logistic regression model [22]. Continuous parameters for age, Marshall CT score, and motor GCS were categorized into clinically meaningful groups to facilitate easier scoring at the bedside [23].

View all citing articles on Scopus
View full text