Jump to ContentJump to Main Navigation

Clinical Chemistry and Laboratory Medicine (CCLM)

Published in Association with the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM)

Editor-in-Chief: Plebani, Mario

Ed. by Gillery, Philippe / Lackner, Karl J. / Lippi, Giuseppe / Melichar, Bohuslav / Schlattmann, Peter / Tate, Jillian R. / Tsongalis, Gregory J.

12 Issues per year

IMPACT FACTOR 2013: 2.955
Rank 5 out of 29 in category Medical Laboratory Technology in the 2013 Thomson Reuters Journal Citation Report/Science Edition

SCImago Journal Rank (SJR): 0.860
Source Normalized Impact per Paper (SNIP): 1.046



Statistical methods for assessment of added usefulness of new biomarkers

Michael J. Pencina1, 4, 5 / Ralph B. D'Agostino2, 4 / Ramachandran S. Vasan3, 4

1Department of Biostatistics, Boston University, Boston, MA, USA

2Department of Mathematics and Statistics, Boston University, Boston, MA, USA

3School of Medicine, Boston University, Boston, MA, USA

4NHLBI's Framingham Heart Study, Framingham, MA, USA

5Harvard Clinical Research Institute, Boston, MA, USA

Corresponding authors: Michael J. Pencina, PhD, Department of Biostatistics, Boston University, Framingham Heart Study, Harvard Clinical Research Institute, 111 Cummington St., Boston, MA 02215, USA Phone: +1 617-358-3386, Fax: +1 617-353-4767, Ramachandran S. Vasan, MD, DM, FACC, Boston University School of Medicine, Framingham Heart Study, 73 Mount Wayte Avenue, Suite 2, Framingham, MA 01702-5803, USA Phone: +1 508-935-3450,

Citation Information: Clinical Chemistry and Laboratory Medicine. Volume 48, Issue 12, Pages 1703–1711, ISSN (Online) 1437-4331, ISSN (Print) 1434-6621, DOI: 10.1515/CCLM.2010.340, August 2010

Publication History

Published Online:


The discovery and development of new biomarkers continues to be an exciting and promising field. Improvement in prediction of risk of developing disease is one of the key motivations in these pursuits. Appropriate statistical measures are necessary for drawing meaningful conclusions about the clinical usefulness of these new markers. In this review, we present several novel metrics proposed to serve this purpose. We use reclassification tables constructed on the basis of clinically meaningful disease risk categories to discuss the concepts of calibration, risk separation, risk discrimination, and risk classification accuracy. We discuss the notion that the net reclassification improvement (NRI) is a simple yet informative way to summarize information contained in risk reclassification tables. In the absence of meaningful risk categories, we suggest a ‘category-less’ version of the NRI and integrated discrimination improvement as metrics to summarize the incremental value of new biomarkers. We also suggest that predictiveness curves be preferred to receiver operating characteristic curves as visual descriptors of a statistical model's ability to separate predicted probabilities of disease events. Reporting of standard metrics, including measures of relative risk and the c statistic, is still recommended. These concepts are illustrated with a risk prediction example using data from the Framingham Heart Study.

Clin Chem Lab Med 2010;48:1703–11.

Keywords: calibration; discrimination; integrated discrimination improvement; net reclassification improvement; reclassification; risk prediction

Comments (0)

Please log in or register to comment.