Duplicates or singletons? An analysis of the need for replication in immunoassay and a computer program to calculate the distribution of outliers, error rate and the precision profile from assay duplicates.

The nature, frequency, statistics and methods of detection of both Gaussian and catastrophic errors in immunoassay are described. The method of sequential analysis is applied to the rates of catastrophic errors and is used to calculate the number of results which must be examined in order to achieve a stated confidence that the error rate is acceptable so that a rational decision on use of duplicate or singleton assay can be made. A simple computer program is described, which allows computation of the precision profile and an analysis of errors using accumulated data from duplicate assays. These methods have been applied to an enzyme amplification immunometric assay for thyroid-stimulating hormone (TSH) and an analysis is made of the errors detected in a sample of 1419 patient results.