A study of the conditions of measurement required to evaluate bias in analytical results illustrated by the use of data from a multi-round, blind-duplicated, proficiency test

Abstract

Measurement uncertainty estimated under repeatability conditions in batch chemical analysis using calibrated instruments may be considered to be composed of contributions from two major effects: (i) precision of the analysis that encompasses sufficient variability of the measurement and (ii) the assessment of trueness by quantifying and if necessary correcting for bias. This paper considers under what conditions of measurement to assess bias, and from the results of a six-round blind-duplicated interlaboratory proficiency program for creatinine in urine shows that bias is present in each individual run with components from that batch and from the laboratory over the rounds of the program. We conclude that bias should be determined in each batch run under repeatability conditions. Measurement of laboratory bias alone is not sufficient to account for effects in each batch run.