Bland Altman diagrams are widely used to assess the concordance between two different instruments or two measurement techniques. Bland Altman diagrams identify systematic differences between measurements (i.e. solid distortions) and potential outliers. The mean difference is the estimated distortion, and the SD of differences measures random variations around this mean. If the mean value of the difference is significantly 0 based on a test of 1 sample t, this indicates the existence of a solid distortion. If there is a consistent distortion, it can be adjusted by subtracting the mean difference from the new method. It is customary to calculate 95% of match limits for each comparison (mean difference ± 1.96 standard deviation of the difference), which tells us to what extent measurements with two methods were more likely for most individuals. If the differences in the mean ± SD 1.96 are not clinically important, the two methods can be used interchangeably. If this option is selected, the measurement reports are presented instead of the differences (MedCalc performs the calculations with log-transformed data in the background). This option is also useful when the variability of differences increases, when the size of the measurement increases. However, the program issues a warning if either of the two techniques contains null values.
The original Bland Altman graph (Bland & Altman, 1986) shows the differences between the two methods compared to the averages of the two methods. Bland and Altman point out that two methods for measuring the same parameter (or property) should have a good correlation when a group of samples is selected, so the property to be determined varies greatly. A high correlation for two different methods designed to measure the same property could therefore in itself only be a sign that a widespread sample has been chosen. A high correlation does not necessarily mean that there is a good agreement between the two methods. The limit of the agreements, expressed in relation to the standard deviation of the differences. If md is the mean value of the differences and sd the standard deviation of these differences, then the conformity limits that are displayed are md +/- sd_limit * sd. The default value of 1.96 produces 95% confidence intervals for the mean values of the differences. If sd_limit = 0, no limit value is pissed and the ylimitation limit of the diagram is by default at 3 standard deviations on both sides of the mean. Horizontal lines are drawn at the mean difference and compliance limits, defined as an average difference plus and minus 1.96 times the standard deviation of the differences.
You can select the following variants of the Bland Altman plot (see Bland &Altman, 1995; Bland & Altman, 1999; Krouwer, 2008): The Bland Altman diagram can also be used to assess the repeatability of a method by comparing repeated measurements to a single method on a number of subjects. . . .