How do practitioners interpret systematic reviews?

A study published in BMC Medicine this week by Lai and colleagues sought to determine how well hospital staff and medical students were able to interpret systematic reviews (SRs), and also to what extent they were influenced by their prior beliefs and the direction (i.e. beneficial or not) of the study results.

SRs are reviews that are focussed on a specific research question that aim to synthesize data from current evidence into a single source of reference. They are widely accepted as the most reliable source of clinical evidence, especially regarding information on potential therapies.  Therefore, they are an attractive source of information for evidence based medicine (EBM) practitioners, such as clinicians, who often require ‘quick answers’ when deciding whether or not to implement a particular practice or therapy.

However, although being relatively quick to read, SRs also require that the reader to be able to interpret the data within the paper as a whole. As the conclusions are often the most readable part of the paper, there is the possibility that clinicians with weaker EBM backgrounds will read and thus be influenced primarily by this section and so will follow through with these messages in their practice without due regard to the caveats of the analyses presented in the Results section.

This study highlights the importance of appropriate interpretation of systematic reviews, which can have a ‘knock-on’ effect on treatments carried out by EBM practitioners. A linked commentary for BMC Medicine by Andrea Tricco and David Moher emphasizes this point, and underlines important  methodological issues relating to the conduct and reporting of SRs. Interpretation of SRs is bidirectional. As well as authors taking responsibility for the correct design and reporting of SRs, readers themselves should be able to interpret them correctly. To this end, Tricco et al also provide useful guidance to EBM practitioners on SR interpretations. 

View the latest posts on the On Medicine homepage

Comments