A quality control program to evaluate accuracy and precision of clinical chemistry determinations.

ABSTRACT

An external quality control program is described which, apart from documenting the routine performance of clinical chemistry laboratories with respect to accuracy and precision, also indicates the control efficiency of intra-laboratory quality control procedures. Participants are supplied with a large lot of stable control serum to be used in their intra-laboratory quality control program during a prolonged period of time. The statistical analysis of the submitted data consists of an analysis of variance scheme in order to separate systematic from random errors. The systematic laboratory errors are thereafter graded according to a scoring system independent of the standard deviation. The control efficiency of intra-laboratory quality control procedures is derived from a modified version of the measurability and controllability rules of Van der Grinten (1968) Stat. Neerl. 22, 43--63). Application of these rules in clinical chemistry laboratories is a first step towards a dynamic system of process deviation control. Merits of such an approach are discussed. Principles of the analysis of variance technique are dealt with. A graphical procedure to display its results is shown. Testing procedures to justify application of the analysis of variance technique are discussed and several features are illustrated.