This module discusses methodology for evaluating biomarkers and risk prediction models, covering principles, concepts, metrics, and graphical tools.

We will discuss motivations for risk prediction in clinical medicine and public health, clarify the concept of “personal” risk, and consider concepts of risk model calibration and performance. Metrics and graphical tools will include ROC curves and AUC; calibration plots for risk prediction models; and net benefit and decision curves. The module will also discuss methods for comparing risk prediction models and, in particular, assessing the incremental value of a new biomarker when there are already established predictors. We will consider the utility of a biomarker for prognostic enrichment of a clinical trial. Throughout the module, we will highlight some common myths and mistakes to avoid.

There will be an opportunity for hands-on practice in R using packages such as rms, rmda, and BioPET. The software component of this module is small and knowledge of R is not required for this module.