Measurements in laboratory medicine have a degree of uncertainty; this uncertainty is often called “error” and refers to imprecisions and inaccuracies in measurement. This measurement error refers to the difference between the true value of the measured sample and the measured value. One of the types of error is systematic error, also called bias, because these errors errors are reproducible and skew the results consistently in the same direction. A common approach to identify systematic error is to use control samples with a method comparison approach. An alternative is use of statistical methods that analyze actual patient values either as an “Average of Normals” or a “Moving Patient Averages.” Fundamental questions should be decided before a quality control method is used: how are weights assigned to the results? Is preference given to more recent samples or to the older samples? How sensitive should the model be? In this chapter, we will expand the fundamental notion of systematic error and explain why it is difficult to identify and measure and current statistical methods that are used to detect systematic error or bias.
Part of the book: Quality Control in Laboratory