Search

For efficient rehabilitation it is important to identify, as early as possible, the patients likely to be successfully returned to work after rehabilitation. The aim of this pilot study was to develop a statistical model for predicting this return as reliably as possible. The model uses only information readily available at the beginning of rehabilitation. A multiple regression analysis with backward elimination was used from a routine data base and identified 8 variables of prognostic value. The model offers a comfortable possibility to predict the probability of return to work of a patient on the basis of routinely registered data. The prognosis was found correct in 68% of those returning to work after rehabilitation (sensitivity) and in 80% of those who did not (specificity). Further work to improve the model for prognosis in rehabilitation research is considered reasonable.

Methodology of clinical studies is highly sophisticated in drug research. But clinical trials are also necessary to demonstrate efficacy and safety of rehabilitation treatment. The call for evidence based medicine has also reached rehabilitation. However, in rehabilitation medicine it is much more difficult to design and conduct clinical trials with a high methodological standard. Among the reasons are: A comparable control group is necessary because spontaneous healing and unspecific measures contribute to therapeutic success, too. But what could "placebo rehabilitation" look like? The masking of therapies (blinded studies) will hardly ever be possible. Therefore, it is more difficult to achieve the same treatment and observation for the treatment and control group. Treatments in rehabilitation take longer to become effective than a drug and maybe the success will disappear after some time. Therefore, long-term trials and follow-ups are necessary. Such studies are expensive, need a strong organisation, and drop-outs are unavoidable. An appropriate outcome variable does not always exist. "Return to work" is an important, reliable and valid variable, but it delivers only one bit of information per patient. As a consequence, smaller progress in rehabilitation can only be demonstrated with large sample sizes. Outcome variables based on time enable studies with reasonable sample sizes. Sometimes it is more difficult to obtain acceptance of randomisation in rehabilitation patients than in acute patients. Some rehabilitation hospitals have only recently begun to take an interest in controlled clinical trials, hence are not so experienced. Nevertheless, controlled clinical trials delivering convincing results are possible in rehabilitation medicine as well. But biometrical consultation is necessary e.g. for study design, study conduction and evaluation. Most important points are the methodology of the study design and its practicability. Especially in these topics rehabilitation physicians and biometrician have to cooperate.

The Rehabilitation Research Network of Ulm on "Bausteine der Reha" consists of several departments of the University of Ulm, and various rehabilitation hospitals and cooperates with a pension insurance institute (LVA Wuerttemberg). The aim of this interdisciplinary approach is to evaluate process and outcome of rehabilitation measures considering specific clinical and rehabilitative questions with main emphasis on epidemiological and economic aspects. The Rehabilitation Research Nétwork of Ulm is subdivided into clinical projects and cross-sectional subjects. The clinical studies deal with cardiovascular and neurological rehabilitation and rehabilitation in disorders of the musculoskeletal system. The cross-sectional topics contain special fields such as biometry, health economics, epidemiology and occupational rehabilitation. All participating institutions have access to a central database with uniform structure of the data.