|
Abstract:
|
Models for risk prediction are widely used in clinical practice to risk stratify and assign treatment strategies. The contribution to risk prediction of new biomarkers, including genetic factors, has largely been based on the area under the ROC curve, but this measure is insensitive to small to moderate effects. More useful clinically is how the predictive model can risk stratify and whether new predictors can reclassify individuals within these strata. Risk reclassification, or cross-stratification into clinical risk categories, can be used to assess the potential impact on treatment decisions for individual patients. Measures based on this reclassification, including those assessing both calibration and discrimination, will be presented with particular application to genetic studies.
|
