Letters to the Editor RE: ‘‘PROBLEMS WITH RISK RECLASSIFICATION METHODS FOR EVALUATING PREDICTION MODELS’’

In her recent Journal article, Pepe (1) takes a strong stance against reclassification, particularly the reclassification calibra-tion test. Although I agree with several comments, I disagree with her dismissal of this test. Pepe’s mathematical assumptions are based on the aver-age predicted risk from the full model equaling the observed risk within reclassified categories. There are several reasons beyond sampling variation why this may not hold in prac-tice. The model may not provide a good fit within the ranges of interest. This is particularly true when interest focuses on those at intermediate or high risk within generally low-risk population-based cohorts. The extent of reclassification may also be too small and random in the higher, more meaning-ful categories. Cook and Paynter (2) performed a series of simulations