👋 In today‘s #JournalClub📚, we read & discussed the article ‘There is no such thing as a validated prediction model’ by #BenVanCalster, @ESteyerberg, @laure_wynants, & @MaartenvSmeden.

👉For more, read

#PredictionModel #valdation #diagnosis #prognosis

doi.org/10.1186/s12916…

@BenVanCalster @GSCollins How can we stop this sort of nonsense? A multi-institutional "validation" of a breast cancer model that looks at discrimination only. not a whiff of calibration to be found. https://ascopubs.org/doi/pdf/10.1200/JCO.21.01337

@BenVanCalster and I looking for papers in the literature showing that doctors appropriately use percentage risks in decision making, and don't need these to be converted into yes / no classifiers. here is one https://pubmed.ncbi.nlm.nih.gov/26839521/ Any others?

Medical prediction models typically have binary endpoint & give risk of cancer/death/MI etc. Any examples of prediction models with continuous endpoint? eg.amount pain relief from surgery. @f2harrell @mike__kattan @BenVanCalster @ESteyerberg @MaartenvSmeden @GSCollins