HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text Full Text (PDF) All Versions of this Article: clinchem.2007.096529v1 54/1/17    most recent Submit an electronic Letter to the Editor about this paper Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Cook, N. R. Search for Related Content PubMed PubMed Citation Articles by Cook, N. R.

Statistical Evaluation of Prognostic versus Diagnostic Models: Beyond the ROC Curve

From the Division of Preventive Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, and Department of Epidemiology, Harvard School of Public Health, Boston, MA.

^aAddress correspondence to the author at: ncook{at}rics.bwh.harvard.edu

Abstract

Background: Diagnostic and prognostic or predictive models serve different purposes. Whereas diagnostic models are usually used for classification, prognostic models incorporate the dimension of time, adding a stochastic element.

Content: The ROC curve is typically used to evaluate clinical utility for both diagnostic and prognostic models. This curve assesses how well a test or model discriminates, or separates individuals into two classes, such as diseased and nondiseased. A strong risk predictor, such as lipids for cardiovascular disease, may have limited impact on the area under the curve, called the AUC or c-statistic, even if it alters predicted values. Calibration, measuring whether predicted probabilities agree with observed proportions, is another component of model accuracy important to assess. Reclassification can directly compare the clinical impact of two models by determining how many individuals would be reclassified into clinically relevant risk strata. For example, adding high-sensitivity C-reactive protein and family history to prediction models for cardiovascular disease using traditional risk factors moves approximately 30% of those at intermediate risk levels, such as 5%–10% or 10%–20% 10-year risk, into higher or lower risk categories, despite little change in the c-statistic. A calibration statistic can asses how well the new predicted values agree with those observed in the cross-classified data.

Summary: Although it is useful for classification, evaluation of prognostic models should not rely solely on the ROC curve, but should assess both discrimination and calibration. Risk reclassification can aid in comparing the clinical impact of two models on risk for the individual, as well as the population.

The following articles in journals at HighWire Press have cited this article:

				S. Masson, R. Latini, I. S. Anand, S. Barlera, L. Angelici, T. Vago, G. Tognoni, J. N. Cohn, and for the Val-HeFT Investigators Prognostic Value of Changes in N-Terminal Pro-Brain Natriuretic Peptide in Val-HeFT (Valsartan Heart Failure Trial) J. Am. Coll. Cardiol., September 16, 2008; 52(12): 997 - 1003. [Abstract] [Full Text] [PDF]

				O. M. Gutierrez, H. Tamez, I. Bhan, J. Zazra, M. Tonelli, M. Wolf, J. L. Januzzi, Y. Chang, and R. Thadhani N-terminal Pro-B-Type Natriuretic Peptide (NT-proBNP) Concentrations in Hemodialysis Patients: Prognostic Value of Baseline and Follow-up Measurements Clin. Chem., August 1, 2008; 54(8): 1339 - 1348. [Abstract] [Full Text] [PDF]

				T. Kurth and G. Logroscino The Metabolic Syndrome: More Than the Sum of Its Components? Stroke, April 1, 2008; 39(4): 1068 - 1069. [Full Text] [PDF]