Many studies, especially in medical research areas, do not allow direct (significant) conclusions to be drawn due to insufficient data volumes. One way out is the combined evaluation of several studies (with identical research questions). This is done, for example, with the help of meta-analyses, which evaluate studies in combination and lead to significant, usable results. However, in this approach there are often challenges regarding too much heterogeneity due to the joint evaluation of partly very different studies.

Therefore, on the one hand, we investigate approaches to reduce the existing heterogeneity of a meta-analysis. Besides the combination of other methods of data analysis (such as end-point-oriented cluster analyses with the actual meta-analysis), we also consider different methods in the area of underlying data transformation.

At the same time, we are working on indicators for the assessment of existing heterogeneity, especially in combined methods such as end-point-oriented cluster-based meta-analysis.