Comparison of clinical data based on limits of agreement

Two different medical measurement methods, which usually do not produce exactly the same results, are used to analyse the serum levels of folic acid in a blood sample. We assess the (dis)agreement of the available data in order to replace the old method (the reference method, which involves a lot of human intervention) with the new one (which uses mostly machines), without causing problems in clinical interpretation. The 95% limits of agreement are estimated, before and after a logarithmic transformation, and an appropriate use of regression and a nonparametric approach are also considered. The application of these different statistical techniques is very useful and easily interpreted by medical researchers, but the results obtained do not provide confidence that the new method can be used in place of the old one for clinical purposes.