# Agreement Between Methods Of Measurement With Multiple Observations Per Individual

The literature has proposed various methods of continuous data compliance assessment, of which the match coefficient [3, 4] and compliance limits [5] are most frequently used. The probability of coverage [6], the overall difference index [6, 7] and the coefficient of certain agree methods [8, 9] were also described. All five methods can be calculated on linear mixed-effect models. With an emphasis on practical application and interpretation, this study aims to show how these five approaches can be applied to the same problem of agreement and to highlight the strengths and weaknesses of each method so that researchers can decide which methods they wish to use in their own studies. Comments on the contractual indices have already been presented in the literature of Barnhart et al. (2007) [2], Obuchowski et al. (2015) [10], Barnhart et al. (2016) [11] and Barnhart (2018) [12]. with these last three titles, including examples of real-life for comparison between thought indices. However, the examples presented come almost exclusively from the fields of quantitative imaging and nuclear laboratory research.

In this article, we expand the methodological work already done in the field of analysis of unbalanced data grouped in applied clinical research, particularly in the field of respiratory frequency measurement in copd patients. In addition, we focus on the linear mixed effect implementation of the methods and not on the broader approach used in the above documents. Especially for the limitations of the agreement, this application of the method is not taken into account in previous audits. This is because mixed effects modeling is increasingly being used in clinical research and has advantages over fixed-effect methods (e.g. B variance analysis (ANOVA) for several reasons described in Brown (2015). In particular, (i) missing or unbalanced data are less problematic for analysis and (ii) conclusions can be drawn on the basis of a larger patient population [13]. We also focus on problems agreeing with repeated observations, as these are recommended when evaluating the agreement [14]. Finally, to assist practitioners of the methods of agreement, we also provided the R code necessary to implement the methods in a file of complementary materials. In true value is constant in each object model (see Bland -Altman, 2007), there is only one marker for each pattern in the diagram.

The size of the marker is relative to the number of observations for the subject. The number of marks corresponds to the number of patterns. Bland and Altman first proposed the limits of the Agreement Method (LoA) more than 30 years ago in their 1986 paper [5] as an alternative to correlation-based methods, which they thought did not accurately characterize the agreement [19].