Seleccionar página

The intraclassical correlation coefficient (ICC) is a number that normally has a value between 0 and 1. It is a well-known statistical tool used, for example, in medical, psychological, biological and genetic research. These are correlations within a category of data (for example. B correlations in repeated weight measurements) and not correlations between two different categories of data (e.g. B the correlation between weight and length). There are different versions of CCI; The choice of the right version depends on the experimental situation. Consider a situation in which we would like to evaluate the adequacy between hemoglobin measurements (in g/dl) with a hemoglobinometer on the hospital bed and the formal photometric laboratory technique in ten people [Table 3]. The Bland Altman diagram for this data shows the difference between the two methods for each person [Figure 1]. The mean difference between the values is 1.07 g/dl (with a standard deviation of 0.36 g/dL) and the 95% match limits are 0.35 to 1.79. This means that the hemoglobin level measured by a given person`s photometry can vary from 0.35 g/dl greater than 1.79 g/dl measured by photometry (this is the case for 95% of people; for 5% of individuals, variations could be outside these limits).

This obviously means that the two techniques cannot be used as substitutes. It is important that there is no single criterion for acceptable compliance limits; This is a clinical decision that depends on the variables to be measured. The system of equation Eq (10) is overdetermined and has five equations, but only three unknowns (σr, σc and σv); Two of the equations can be derived from the others. The solution eq (10) for the σ:s we receive (for example) the expressions (11) Eq (11) provides us with a statistical estimate of the variances σr2, σc2 and σv2. With Eq (11) in Eq (9), we obtain (12), i.e. the known ICC sampling formula icc (A,1) [6], which is therefore a statistical estimate of the intraclass population coefficient ρ2A. The ICC formula (A,1) gives an estimate of the reliability of the method if one wants an absolute correspondence between different measurements [6]. . . .