Eingangsbelastung der Patient(inn)en und Ergebnisqualität der Rehabilitation nach Schlaganfall

Abstract

Rehabilitation after stroke has to face specific problems when treating patients with more or less severe disabilities in cognition and communication. Correspondingly, stroke rehabilitation takes a special position within the larger field of rehabilitation, and relatively little is known outside the neurological scientific community about the status of patients at admission, the case mix in the centres and the short- and medium-term effects of rehabilitation. The present study describes in some detail the initial status in unselected samples of consecutive patients (n = 768) from three neurological rehabilitation centres. The description shows a very inconsistent picture in all centres, ranging from patients with no neurological deficits to patients needing intensive care. Across the centres, we found remarkable differences in case mix. In order to measure the effects of rehabilitation after stroke, an instrument was developed that combines a physicians' questionnaire aiming at an assessment of the severely disabled cases with a patients' questionnaire for the less severe cases for which the physicians' questionnaire would show "ceiling effects" so that improvements could no longer be depicted. The application of the instrument showed that about 50 % of the sample were not capable of answering the patients' questionnaire. For the patients with neurological deficits, the functional parameters of the physicians' questionnaire showed significant improvements at discharge that can be interpreted as "strong" effects (effect sizes 1.0-1.3). For the patients with less severe deficits (and usually in later stages of the rehabilitation process), the patients' questionnaire showed "strong" improvements on the somatic and psychosocial scales both at discharge and 6 months later. On the functional scales, however, only small improvements were found. Finally, predictors could be identified that explain a large amount of the variance for length of stay (R(2) =.42) as well as for the effects of rehabilitation (R(2) =.74). When comparing effects across rehabilitation units with differences of case mix, these predictors should be statistically controlled in order to assure fair comparisons.