Abstract:
The TAR System
and Other Patient Classification Systems for Rehabilitation

Abstract (English)

Patient classification systems for acute inpatient treatment have been introduced in many countries. They can be used to describe the patient case mix and to reimburse condition based flat rates. In the face of these activities, the question now arises as to how to transpose these approaches into the rehabilitation setting.

In this study we will, on the one hand, present the research we have done in the context of the Swiss TAR project concerning neurological rehabilitation. On the other hand we will have a look at the developments in other countries, especially in Germany (RBG and RMK), in France (PMSI SSR), in Australia (CRAFT and AN-SNAP) and in the USA (FRG and IRF PPS).

A 12-week survey in four facilities with 274 patients formed the basis of the TAR classifications. In order to categorise the treatment weeks we used the CART algorithm («Classification and Regression Trees»). From among the proposals obtained by this method, we then – in discussions with experts – selected those categorisations that lent themselves well to interpretation from a clinical point of view.

The first classification proposed in the TAR study groups the daily nursing expenses according to the patient condition as assessed on a weekl­y bases. The latter was quantified by means of the FIM instrument. Six nursing cost groups were defined on the basis of the cognitive and the practical-motor FIM values. They explain 65 % of the daily nursing time.

The second classification proposed in the TAR study groups the daily therapy expenses. Because we did not achieve to extract patient characteristics which explain them significantly, we divided the therapy expenses up into for therapy intensity levels according to weighted therapy hours.

The six nursing cost groups can be combined with the for therapy cost groups into 24 TAR groups.

The nursing classification presented is practicable because it can be interpreted relatively well by clinicans. It still has to be validated in other neurological rehabilitation facilities. Its transferability into other rehabilitation domains remains to be tested. More research needs to be conducted on the therapy classification. Furthermore, treatment goals should be taken into consideration as additional criteria.