Substantial discussion has emerged on measures to quantify the clinical usefulness of a prediction model. This issue was discussed at a symposium in Cleveland, Sept 29, 2008. A related paper in Epidemiology (Jan 2010) aimed to present a perspective on some traditional and modern performance measures (R code here). It was accompanied by a vivid discussion in an Editorial and response from the authors.
Michael Pencina worked on extensions of the NRI, including a weighted version, and a proposal for survival time data, published in Stat Med, 2011. We also made a simple proposal for NRI in survival data (a letter in Ann Int Med 2010). Other papers are on the evaluation of diagnostic and prognostic markers EJCI 2011, and on the evaluation of models and markers (Spanish journal 2011).
Data and R code are available for didactic purposes:
For recent work on marker evaluations, including the Net Reclassification Improvement (NRI),
see Research on markers