[Adaptive testing in rehabilitation: a way to economic assessment of patients' attributes]

For diagnostics and outcome measurement in clinical rehabilitation a multitude of questionnaires is used. In order to gain comparability of the diagnostic findings, generally, the same information is gathered of all patients, regardless of their state of health or how severely ill they are, by using identical groups of items. In this kind of assessment it is, however, problematic that (a) the assessment instrument usually only allows for adequate and reliable diagnostics of patients who suffer from injuries or illnesses of medium severity, and (b) that an uneconomic data collection is required because of an extensive set of items which may in individual cases be uninformative and unnecessary. This article shows how data assessment can be adjusted to the individual handicaps of the patient by using adaptive or tailored testing. Thus it can be ensured that the testing becomes more economical and results more precise. Furthermore, a high acceptance by the patients participating in the data collection can be achieved, as the test items submitted fit the ability of the patients adequately. Yet, adaptive test systems require a complex and very careful psychometrical development of the assessment procedure in such a way that in spite of presenting different items to patients, diagnostic outcomes are comparable between patients and for different points in time (e. g., pre-post comparison). This article shows the steps of development that have to be accomplished in order to set up a psychometrically sound adaptive test procedure. Moreover, it shows how its usage in the area of medical rehabilitation can be profitable for handlers and patients.