Skip to main content

Languages

A practical taxonomy proposal for systematic reviews of therapeutic interventions

Date and Location

Session: 

P2.084

Date

Saturday 21 September 2013 - 10:30 - 12:00

Location

Presenting author and contact person

Presenting author

Ralf Bender

Contact person

Ralf Bender
Abstract text
Background: No consistently used terminology is currently available to classify systematic reviews of therapeutic interventions according to their aim, their methodological rigor and the conclusiveness of their results. Some researchers consider systematic reviews to be merely observational studies, as in contrast to confirmatory randomized controlled trials (RCTs), all data used for the evidence synthesis are already available. This assumption results in the paradox that a high-quality RCT is qualified for drawing confirmatory conclusions, but a high-quality systematic review containing the same RCT together with other high-quality RCTs, is only qualified for generating hypotheses. Objectives: To avoid the paradox described above, a new taxonomy for systematic reviews of therapeutic interventions is proposed, taking into account the research question as well as the impact of methodological and outcome issues on the value of the systematic reviews. Methods: Different uses of systematic reviews and common sources of variability regarding the methodological rigor of systematic reviews and the resulting implications are discussed. Criteria to assess whether the results of systematic reviews allow firm conclusions regarding the considered therapeutic intervention are summarized. Results: In a first step the proposed taxonomy discriminates between exploratory and analytic systematic reviews considering their aim and their methodological rigor. For analytic systematic reviews, a further discrimination between conclusive and inconclusive systematic reviews follows. Criteria for the second step include the risk of bias within and across the studies considered in the systematic review as well as the precision of estimated treatment effects. Conclusions: The proposed taxonomy provides a simple and practical way to identify the value of systematic reviews for decision making. Furthermore, the new taxonomy avoids the paradox that a systematic review is on the one hand considered to be the highest evidence level, but on the other hand regarded merely as an observational study.