[The ICF in use: testing of a new concept in four clinics]

The International Classification of Functioning, Disability and Health (ICF) consists of more than 1 400 items. Due to its size and complexity, using the ICF is hardly feasible in practice. Moreover, the severity of problems is given only on a five-point scale (qualifiers), which seems too insensitive for measurement of changes. Thus, in cooperation with four rehabilitation hospitals, Jacobi, Urban & Kaluscha have identified 148 relevant items for musculoskeletal diseases. For easier usage we assigned mnemonic three letter codes to these items, e. g. WAL for the item "Walking" (ICF d450). Then the physician assigns those few items which he and/or the patient consider most important. The severity of the problem is judged on a scale from 0 (no restriction) to 100 (maximum restriction) for each item. By drawing a bar on the scale the physician gives an interval instead of just a single point (fuzzy judgement). The centre of the interval represents the severity of the problem and the width represents the uncertainty of judgement, e. g. when there is no gold standard for measurement or when the problem varies with time. As both the assignment of descriptors from an item pool and the fuzzy judgements are a new approach to medical documentation, we carried out this pilot study to examine acceptance and feasibility of the approach. In the pilot study, 264 in-patients in four rehabilitation clinics have been included. 138 (94%) of the 148 descriptors in the item pool have been used. Altogether there were 1,764 single judgements. On average six descriptors per patient have been assigned, giving an efficient and precise individual documentation. This is an advantage of the documentation principle called combination of terms compared to the principle of classification. In contrast to the ICF core sets patients may get individual sets of descriptors as appropriate. On discharge, the judgements were about 25 points better compared to admission. During treatment the fluctuations of disorders decreased and the certainty of the judgements increased, resulting in reduction of interval width of 2.8 points. The feedback of the rehabilitation clinics involved showed good acceptance and feasibility. Our data can easily be mapped into the ICF. In addition to ICF core sets, our approach could facilitate the use of the ICF in practical settings.