Corrected Supplemental Data For

Corrected Supplemental Data for:

Standardization of High Sensitivity Immunoassays for Measurement of C-Reactive Protein. Phase II: Evaluation of Three Approaches for Evaluating Commutability of a Secondary Reference Material

Mary M. Kimberly[1]*, Samuel P. Caudill1, Hubert W. Vesper1, Elizabeth A. Monsell1, W. Greg Miller[2], Robert Rej[3], Nader Rifai[4], Francesco Dati[5], Gary L. Myers1

The original published version of this article (Clin Chem 2009;55:342-350) contains errors in the Results section. Our evaluation of within-run precision for the assays was performed incorrectly. Methods 2, 3, and 23 should not have been omitted; thus, 18 methods should have been evaluated further for commutability and harmonization. We have performed all of the calculations and analyses for the 18 methods. All of the figures and tables in both the main paper and the Data Supplement were affected. The correct tables and figures appear in this corrected on-line Data Supplement. These errors did not alter the conclusions.

Normalized Residuals for Weighted Least Squares with Polynomial Regression

The following table replaces Table 1 in the publication:

Weighted least squares normalized residuals for undiluted PRM (1:10 dilution of CRM470)a

a Normalized residuals ≥2 are in bold.
The following Table replaces the original Supplemental Data Table 4:

Normalized Residuals for 25% Dilution of PRM (0.25 fraction diluent plus 0.75 fraction PRM)

The following Table replaces the original Supplemental Data Table 5:

Normalized Residuals for 50% Dilution of PRM (0.50 fraction diluent plus 0.50 fraction PRM):

The following Table replaces the original Supplemental Data Table 6:

Normalized Residuals for 75% Dilution of PRM (0.75 fraction diluent plus 0.25 fraction PRM):

The following Figure replaces Figure 1 in the manuscript:

Correspondence Analysis (CA) results obtained by 18 analytical assays (□, open squares identified by numbers) and 40 clinical samples (∆, open triangles) as active elements; the PRM samples (○, open circles with numbers) are inactive elements.

PRM, the 15% dilution, the 50 % dilution, and the 75% dilution are represented by the open circles with 1, 2, 3, and 4, respectively. The area defined by the dotted boundary is the 95% tolerance ellipse for the methods; solid boundary is the 95% tolerance ellipse for the samples.

The following Figure replaces Figure 2 in the manuscript:

Correspondence Analysis (CA) results obtained by 17 analytical assays (□, open squares identified by numbers) and 40 clinical samples (∆, open triangles) as active elements; the PRM samples (○, open circles with numbers) are inactive elements.

PRM, the 15% dilution, the 50 % dilution, and the 75% dilution are represented by the open circles with 1, 2, 3, and 4, respectively. The area defined by the dotted boundary is the 95% tolerance ellipse for the methods; solid boundary is the 95% tolerance ellipse for the samples.

The following Figure replaces Figure 3 in the manuscript:

Total CV, before and after harmonization with PRM.

Solid line (■, solid squares), total CV before harmonization; dotted line (▲, solid triangles), total CV after harmonization.

The following Figure replaces Figure 4 in the manuscript:

Among-method CV, before and after harmonization with PRM.

Solid line (■, solid squares), among-method CV before harmonization; dotted line (▲, solid triangles), among-method CV after harmonization.

This figure replaces the plot in the original Supplemental Data file.

Plot of difference between adjusted (harmonized) and measured values for methods. Individual runs are plotted separately; thus there are 3 curves for each method.

[1] Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30341-3717.

[2] Medical College of Virginia Hospitals at Virginia Commonwealth University, Richmond, VA 23298.

[3] Wadsworth Center for Laboratories and Research, New York State Department of Health, Albany, NY 12201

* Address correspondence to this author at: Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Mailstop F25, 4770 Buford Hwy NE, Atlanta, GA 30341-3717; email ; fax (770) 488-7030.

[4] Departments of Laboratory Medicine and Pathology, Children’s Hospital and Harvard Medical School, Boston, MA 02115.

[5] IVD Consulting, Marburg, Germany.