Abstract
Combined errors are the commonest systematic errors in laboratory results and occur with most tests studied. In these errors, results are reduced (or, less commonly, increased) by a factor, and this effect is compensated by the addition (or subtraction) of a constant amount. The effect is that results are in error in opposite directions at high and low levels. Inconsistency is predominant and is due mainly to imprecision, although for some tests (iron, cholesterol, calcium, and triglycerides among those studied) other factors such as non-specificity are significant. An interlaboratory survey based upon external method assessment using linear regression analysis provided objective information about analytical error in laboratories which is not usually obtained, while at the same time meeting the usual functions of surveys in the quality audit of performance.