[Cluster formation - an operationalization of rehabilitation success in rehabilitants with chronic low back pain].

: To investigate the impact of the way of allocation for rehabilitation on rehabilitation success in patients with chronic low back pain it was necessary to define appropriate criteria of success. A multidimensional model of success is presented in this article. First, the outcomes of rehabilitation are described and statistically tested. No statistically or clinically relevant changes between pre and post measurements were found. For further analysis of predictors and group comparisons it is necessary to define multidimensional success. For this purpose, cluster analyses are introduced as a possible model, resulting in three groups of patients with differing outcomes, i. e.: 20 % patients with declined pain symptoms, 40 % patients without any changes, 40 % patients with increased symptoms. No impact of the way of allocation for rehabilitation could be verified.