Requirements for reproducibility, trueness and error of measurement in internal quality control schemes.

Among the various models proposed to meet requirements for metrological performance in clinical laboratory measurements, that most widely used is intra- and inter-biological variation. However, this model has weaknesses: a) in some cases, intra-individual biological variation is so small that requirements based on such data may not be met by routine methods; b) there is great variation among the intra- and inter-individual biological variations themselves. We therefore propose integration between "state-of-the-art" methods to minimize imprecision, and also suggest that requirements calculated on the basis of biological variations should be used in setting requirements for minimizing imprecision. Furthermore, we describe a new scale for setting requirements from biological variation data and suggest that systematic error should be corrected whenever it is statistically significant.