DietBB Seminar Series: From systematic reviews and network meta-analyses

December 19, 2017

The DietBB seminar on 4 December 2017 was dedicated to the methods of review articles

"Every systematic review is a review, but not every review is systematic." With these words, Dr. Sabrina Schlesinger, head of the Junior Research Group Systematic Reviews at the German Diabetes Center in D├╝sseldorf, opened the seminar to around 50 participants.

Systematic reviews and meta-analyses are important tools of science. In evidence-based medicine, studies are evaluated by type and quality in order to make statements on the risk factors and benefits of therapies. On this rating scale systematic reviews and meta-analyses are at the top. A systematic review combines published literature on a specific topic and follows a protocol that defines the process of literature research and the selection of literature included in the review. This form of review can be complemented by a meta-analysis. Therefore, data from the individual studies are statistically evaluated together and so, for example, the average effect of a risk factor or a therapy can be estimated. A combination of the data increases statistical power, heterogeneity between studies and subgroups can be analysed, and the results can be used to identify research gaps. However, it should be noted that each meta-analysis depends on the quality of the original studies included. In addition, the results of the meta-analysis can be influenced by publication bias and incorrect application of statistical methods", said Schlesinger.

New developments in the field of meta-analyses include the network meta-analyses, which can be used to compare different interventions with each other. While pairwise meta-analyses investigate, e.g. the effect of a particular diet on the risk of a disease, a network meta-analysis can be used to compare the effect of different diets or interventions on the risk and to identify the intervention with the strongest effect", explains Schlesinger. In the field of nutritional epidemiology, in which observational studies are available for most questions, they are not yet widespread. Other recent developments include the Umbrella Reviews, which provide a summarized overview of published meta-analyses on a particular field of research. The aim is to obtain and critically evaluate an overview of the existing evidence on a certain research field. 


Text: TA6, Dr. Maike Gutmann, DGE