Effective evaluation of novel low-cost CD4 monitoring assays

Indian generic pharmaceutical manufacturers have dramatically reduced the price of antiretroviral therapy thus improving access, but monitoring costs of the patients on therapy remain high. Few low-cost monitoring assays are available and validation against gold standard is exceedingly important before implementation. Validity is defined as the degree to which the results of the new technique correspond to the true state of the phenomena being measured by the gold standard. Majority of publications report validation of low-cost CD4 assay using correlation coefficient, which is the best measure of association, and a few publications report intra-class correlation coefficient, which is a measure of reliability. However, Bland-Altman analysis that directly quantifies the difference seen between the novel low-cost CD4 assay and the gold standard is the most ideal approach. Absolute CD4 counts estimated for samples collected from 110 HIV+ and 118 HIV- individuals using novel EasyCD4 assay and the gold standard FACSCount assay were validated using correlation coefficient, intra-class correlation coefficient and Bland Altman analysis. The results from these different analyses have been compared and the significance of each has been explained. Hence, to validate a low-cost assay Bland-Altman analysis and plot is recommended. (c) 2006 Elsevier B.V. All rights reserved.