ASSESSING REPRODUCIBILITY BY THE WITHIN-SUBJECT COEFFICIENT OF VARIATION WITH RANDOM EFFECTS MODELS

In this paper we consider the use of within-subject coefficient of var iation (WCV) for assessing the reproducibility or reliability of a mea surement. Application to assessing reproducibility of biochemical mark ers for measuring bone turnover is described and the comparison with i ntraclass correlation is discussed. Both maximum likelihood and moment confidence intervals of WCV are obtained through their corresponding asymptotic distributions. Normal and log-normal cases are considered. In general, WCV is preferred when the measurement scale bears intrinsi c meaning and is not subject to arbitrary shifting. The intraclass cor relation may be preferred when a fixed population of subjects can be w ell identified.