Definition and application of performance specifications for measurement uncertainty of 23 common laboratory tests: linking theory to daily practice

Laboratories should estimate and validate [using analytical performance specifications (APS)] the measurement uncertainty (MU) of performed tests. It is therefore essential to appropriately define APS for MU, but also to provide a perspective on suitability of the practical application of these APS. In this study, 23 commonly ordered measurands were allocated to the models defined during the 2014 EFLM Strategic Conference to derive APS for MU. Then, we checked if the performance of commercial measuring systems used in our laboratory may achieve them. Most measurands (serum alkaline phosphatase, aspartate aminotransferase, creatine kinase, γ-glutamyltransferase, lactate dehydrogenase, pancreatic amylase, total proteins, immunoglobulin G, A, M, magnesium, urate, and prostate-specific antigen, plasma homocysteine, and blood red and white cells) were allocated to the biological variation (BV) model and desirable APS were defined accordingly (2.65%, 4.75%, 7.25%, 4.45%, 2.60%, 3.15%, 1.30%, 2.20%, 2.50%, 2.95%, 1.44%, 4.16%, 3.40%, 3.52%, 1.55%, and 5.65%, respectively). Desirable APS for serum total cholesterol (3.00%) and urine albumin (9.00%) were derived using outcome-based model. Lacking outcome-based information, serum albumin, high-density lipoprotein cholesterol, triglycerides, and blood platelets were temporarily reallocated to BV model, the corresponding desirable APS being 1.25%, 2.84%, 9.90%, and 4.85%, respectively. A mix between the two previous models was employed for serum digoxin, with a 6.00% desirable APS. In daily practice by using our laboratory systems, 16 tests fulfilled desirable and five minimum APS, while two (serum albumin and plasma homocysteine) exceeded goals, needing improvements.