Analytical Variation in Immunoassays and its Importance for Medical Decision Making

Abstract

The analytical variation of immunoassays is at present significantly larger than that of other methods in clinical chemistry. The main sources of this large variation are variation between methods, lot to lot variation, and the large imprecision between-runs and within-runs when the methods are not performed on automated instruments. The variation can be substantially reduced by standardization, by ensuring the comparability between lots, and by automation of the analytical procedure. The extent to which the variation should be reduced is determined by the needs of medical decision making. It is demonstrated using the ana-lytes TSH and CA 15-3 as examples that quality requirements have to be individually specified for each analyte. Such guidelines should be established by teams of international experts. They should include suggestions for reference methods, guidelines on the extent of comparability between routine methods, the maximum allowable deviations from Jot to lot, and the minimum requirements for between-run precision. Such guidelines could form the basis for improvements which are directed towards the medical requirements.