| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mini-Review |
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:
|
|
 |
|
  M. J. Price Bedside Evaluation of Thienopyridine Antiplatelet Therapy Circulation, May 19, 2009; 119(19): 2625 - 2632. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Q. Cabell, L. M. Justice, T. A. Zucker, and C. R. Kilday Validity of Teacher Report for Assessing the Emergent Literacy Skills of At-Risk Preschoolers Lang Speech Hear Serv Sch, April 1, 2009; 40(2): 161 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K. Sato, T. Hayashi, N. Harita, T. Yoneda, Y. Nakamura, G. Endo, and H. Kambe Combined Measurement of Fasting Plasma Glucose and A1C Is Effective for the Prediction of Type 2 Diabetes: The Kansai Healthcare Study Diabetes Care, April 1, 2009; 32(4): 644 - 646. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.M. V. Narayan, M. B. Weber, J. Gulcher, K. Stefansson, V. Lyssenko, P. Nilsson, and L. Groop Clinical risk factors, DNA variants, and the development of type 2 diabetes. N. Engl. J. Med., March 26, 2009; 360(13): 1360 - 1360. [Full Text] [PDF]
|
|
|
|
 |
|
 K. V. Patel, L. Ferrucci, W. B. Ershler, D. L. Longo, and J. M. Guralnik Red Blood Cell Distribution Width and the Risk of Death in Middle-aged and Older Adults Arch Intern Med, March 9, 2009; 169(5): 515 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Paynter, D. I. Chasman, J. E. Buring, D. Shiffman, N. R. Cook, and P. M. Ridker Cardiovascular Disease Risk Prediction With and Without Knowledge of Genetic Variation at Chromosome 9p21.3 Ann Intern Med, January 20, 2009; 150(2): 65 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Soreide Receiver-operating characteristic curve analysis in diagnostic, prognostic and predictive biomarker research J. Clin. Pathol., January 1, 2009; 62(1): 1 - 5. [Full Text] [PDF]
|
|
|
|
|
|
P. M Ridker, N. P. Paynter, N. Rifai, J. M. Gaziano, and N. R. Cook C-Reactive Protein and Parental History Improve Global Cardiovascular Risk Prediction: The Reynolds Risk Score for Men Circulation, November 25, 2008; 118(22): 2243 - 2251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Janes, M. S. Pepe, and W. Gu Assessing the Value of Risk Predictions by Using Risk Stratification Tables Ann Intern Med, November 18, 2008; 149(10): 751 - 760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
