The classification of patients for phase B, C and D of neurological rehabilitation follows the suggestions of the Bundesarbeitsgemeinschaft für Rehabilitation (BAR) based on defined clinical criteria and on neurological rehabilitation assessment. The focus of this study is to define the intervals of the complete FIM(tm)-index, intervals covered empirically as well as by evaluations of physicians, that permit utmost accuracy in assigning patients to phase B, C and D of neurological rehabilitation. Therefore, data records of 3686 patients from 4 neurological rehabilitation centres were evaluated. The patients' functional autonomy was classified by FIM(tm) on admission, in intervals of 14 days and at discharge, at the same time all patients in addition were assigned to phases B, C or D by the rehabilitation centre physicians. Statistical analysis of a total of 11,247 links of the phase classifications and FIM-indexes at 6 measurement points showed that correct correlation to phase B, C and D could be obtained on average in 79 to 89 % of the cases, based on the assumption that 18 - 36 points of the FIM-index assign to phase B, 37 - 90 points to phase C and 91 - 126 points to phase D. Discrimination between phases B and C could be obtained accurately in an average of 84 %, discrimination between phases C and D in an average of 89 %. Conformance of the FIM-intervals with TAR-based groups of care efforts compared to the evaluation by physicians indicate that the FIM(tm) represents the need for care with greater validity. If assignment to phases B, C and D would have been done on the basis of the FIM-index instead of evaluation by a physician, 8,9 % fewer patients would on admission have been classified for phase C but, instead, 4,5 % more patients for phase B and 4,4 % for phase D. In case of using the FIM-intervals for classification, 12,1 % more phase B patients could have changed to phases C or D. Of the phase C cases, 17,7 % could have been transferred to phase D before discharge. The number of phase D patients would have remained unchanged comparing admission and discharge. FIM-orientated classification for the phases of neurological rehabilitation offers considerable advantages: operationalized criteria, possibility of statistical evaluation, objectiveness, reliability, validity, reproducibility of the decisions, sensitivity to change, prognostic sensitivity, and suitability as an instrument for internal and external quality assurance.
Read full abstract