In reply
I agree with Chiolero that the terms personal and personalized have been misinterpreted, fueling unrealistic expectations of predictive accuracy.1,2 Chiolero suggests that the term stratified should be used instead because it captures more appropriately the effort to create relatively accurate predictions for discrete strata rather than individuals. The term stratified (eg, stratified medicine) has gained some traction recently. I agree that it is technically sound and to the point in most cases where some predictive tool is applied.