Estimation of uncertainty in measurement: interest of short-term Bayesian model as a complement to the conventional approach

Estimating uncertainty in measurement in patient results is a thorny question that continues to stir controversy in the med lab community. Various well established approaches are advocated, which we shall detail below. These are all based on a large set of a-posteriori data that the laboratory may in fact not have, especially at the point of setting up new analysis methods. The objective of the current study was to present a uncertainty in measurement estimation approach that will be useful when setting up new routine functioning methods, alongside conventional approaches that may be applied afterward.