[Evaluation of added value of diagnostic tests]

In the evaluation of diagnostic tests emphasis is placed on the evaluation of a particular diagnostic test in order to estimate it's sensitivity, specificity, or area under the 'receiver operating characteristic' curve (so called test research). This kind of test research is only useful in situations in which a diagnosis is set by one test (i.e. screening) and in the initial phases of test development for efficiency reasons. But, test research is not suitable to evaluate the clinical or added value of a test. The value of a diagnostic test can be evaluated on its value to discriminate between the presence and absence of a particular disease, to guide therapeutic decisions and to improve the prognosis of the patient. For evaluation of the added value of a diagnostic test within the diagnostic process, the study design should take into account the patient group in which the test will be applied in practice, the target disease of the test and the prior probability (prevalence) of the target disease in that population, the subsequent diagnostic stages and the reference test. In analysing the data it is calculated for each step of the diagnostic process how the test result changes the probability of presence or absence of the target disease.