There are a multitude of methods for assessing continuous coherence in the literature, which vary in complexity and in their underlying assumptions. In this article, we looked at five different methods for analyzing the same problem of agreement that concerns aggregated and unbalanced data; including some of those that are known and commonly used in the literature, and others that understand recent advances in contract research. The main constraints of this work are the search for ad hoc literature in a single database (MEDLINE/PubMed) and the fact that only one critic has verified titles and abstracts, read complete texts, extracted data and synthesized qualitative evidence [24]. This may have led to the ignore of other lists of report posts that were important than those identified. However, the investigations are, in our view, reasonable and complete. Another limitation is that the analysis of the example treated [25] by a Bland-Altman analysis oversimplified the structure of the underlying data, since a physician was compared to a number of radiologists. As a result, BA-LoAs were slightly underestimated compared to the repeatability coefficients derived from Section 3.2.3. On the other hand, the limitations of the agreement and TDI methods have the advantage of being based on the initial unit of measurement and are compared to a clinically acceptable difference [43]. In the reviews of Barnhart et al. [11] and Barnhart [12], the authors point out that it is possible for the LoA to have 95% of the differences in the clinically acceptable difference, but not to reach an agreement (for example, if one of the limits is outside the CAD). This can happen in the event of distorted data or another error in accepting normality.

We agree that this may be a problem in the search for loA interpretation and that it is particularly important to review the assumptions in the implementation of loA. However, we believe that the ability of the methodology (and in particular the Bland-Altman plot) to highlight relative average distortions, patterns in the data and therefore sources of differences of opinion, is valuable; and that the simple calculation of a TDI or CP synthesis index can mask this detail.