| Abstract: |
Context.—The goal of the College of American Pathologists Accuracy-Based Proficiency Testing Program is to promote the quality, standardization, and harmonization of clinical laboratory results through proficiency testing specimens that are free from matrix effects, have target values that are traceable to reference methods, and that probe the limitations of current methods. Objective.—To summarize the first 6 years of the Accuracy-Based Vitamin D Survey and highlight key insights from the data generated as it relates to assay performance. Design.—Accuracy-based challenges were created by using pooled human serum samples. Certain samples were derived from participants in an institutional review board–approved protocol in which vitamin D–deficient participants were treated with ergocalciferol (vitamin D2). Reference targets for the survey were set by the Centers for Disease Control and Prevention using isotope-dilution liquid chromatography–tandem mass spectrometry. Each method was compared with the reference method procedure over the course of the program (n 1⁄4 43 proficiency testing samples). Results.—Linear regression versus the reference method procedure revealed proportional biases across the methods, ranging from 0.0% to 16.7%. Pearson correlation coefficients (r2) ranged from 0.902 to 0.996. Results were influenced by the concentration of 25-hydroxyvitamin D2 as well as the C-3 epimer of 25-hydroxyvitamin D3. During the 6 years, 2 manufacturers altered their assays to match the reference method procedure more closely. Conclusions.—There is considerable bias, both proportional bias and sample-specific matrix effects, affecting many assays. This ongoing accuracy-based proficiency testing program for vitamin D will provide the data needed for laboratories and manufacturers to improve their assays and thereby patient care. © 2019 College of American Pathologists. All rights reserved. |
