Bridging Medical Research and Clinical Work by Evaluating Predictive Ability

Clinical practice guidelines are developed to aid clinical decision making of healthcare practitioners and are derived from best available evidence including clinical prediction rules (CPRs). Unfortunately,there is little evidence that current CPRs can be used to accurately predict the intervention outcomes for individual patients. Clinicians must, therefore, understand when and how a CPR should be used for personalized prediction. This viewpoint aims to empower clinicians with respect to personalized prediction. We describe how to evaluate the predictive ability of a validated CPR, and the important role of existing data within clinics, hospitals, and registries for such predictive evaluation.