[The Barthel Index in geriatrics. A context analysis for the Hamburg Classification Manual].

The Barthel Index (BI) is firmly established as an assessment instrument in geriatrics. It is a proven, clear and easy-to-use instrument for the recording of basic daily functions. However it is increasingly finding new applications beyond its original, clinically orientated use. It has been applied as a quality parameter, as an instrument for the management of service delivery provision and as an instrument to record treatment efforts relevant to care or costs. This study considers the basic suitability prerequisites of the Barthel Index for these applications.With the Hamburg Classification Manual for the BI, German geriatrics has made a contribution to the standardized operationalization of the items and to the standardized evaluation of aids. An evaluation was performed on a total of 5262 Barthel classifications at seven geriatric hospitals according to existing classification practice by comparison with 5483 classifications after introduction of the Hamburg Classification Manual. No significant differences were found in Barthel Index total scores either on admission or on discharge. With respect to further applications it is essential to be aware that the usability of the total score (including its changes over the course of time) is limited because of the ordinal scaling of the BI. Studies have been carried out which show how important this is, although they have so far received little attention. As a grading criterion the BI takes account not only of the "functional status" but also of the "extent of support effort". This can lead to positive changes in one focus -- especially with the use of aids -- without associated improvements in the other focus. Whether the BI in this form is meaningful for a specific application must be tested separately for each individual context. There is no justification for assuming that the BI has general validity irrespective of application. The results of a systematic literature survey on the testing quality criteria of the BI indicate an astonishingly generous approach to the question of the validity of the BI. Contrary to widespread opinion, cognitive-psychological components do influence the classification result of the BI. As an addition to the Hamburg Classification Manual we recommend that the extent of this influencing factor should also be more clearly operationalized and that "stimulation required" for the carrying out of an activity be included in the evaluation equivalent to the factor "supervision required" already introduced by Barthel and Mahoney. The BI has shown itself to be an efficient but nonetheless multidimensional global parameter in clinical practice, whose meaningfulness is on the level of the individual item and whose validity, particularly as total score and course parameter, needs to be proven for each new application. The Hamburg Classification Manual, as a standardized and consensus-based operationalization of the BI, provides an important basis for this.