The approach taken to evaluate performance for FXIII assays complies with specifications detailed in ISO 13528 Statistical Methods for use in proficiency testing by interlaboratory comparisons and is as follows:
The consensus median is taken as the central reference point or "target value".
A z score is calculated using the following formula:
[result – mean] / SD
The mean is a “robust” mean value, with outliers excluded; the SD is a “robust” SD determined after exclusion of outliers.
A result identical to the mean will give a Z score of 0. ISO13528 defines Z score above 3 or below -3 as an “action signal”. A Z score above 2 and below 3 (or below -2 and above -3) is a “warning signal”. It is suggested in this ISO guide that one action signal, or two successive warning signals require investigation.
Assessment is against peer group (method-specific) medians, when n =>10. If the assay is performed by 10 but <20 centres, performance is assessed but centres should note that z-scores can be unreliable with small numbers of samples (n<20) as in this case. If <10 centres return results for a method, performance will not be assessed.
|
|
|