Ask The Expert
July 7, 2016
There is a lot of criticism of systematic reviews, both in medical journals and social media. It is claimed that systematic reviews have a bias towards non-recommendation of an intervention, often in the face of one or two very well structured randomized clinical trials. How do we handle this anomaly in our clinical practice, especially when patients are often very knowledgeable due to internet searches?
From the Editor: The thoughtful question regarding the value of systemic reviews and meta-analyses gets a thoughtful answer from our respondents. Please read.
By Doctor Jonathan D. Campbell1,2, David Price2,3, Alison Chisholm2
Systematic reviews and meta-analyses synthesize the findings of multiple evidence sources and present them as one, consolidated result. Yet with simplification of data comes the potential to over-simplify complex topics.[i] The challenge for clinicians and well-read patients is to realize that they must first take a step back from the systematic review’s bottom line to consider also the consistency and generalizability of the results of the trials involved. Interpretation and judgment are required in this science.
Indeed, systematic reviews and meta-analyses can be a powerful tool when used sparingly and appropriately; synthesizing evidence from like studies that pertain to the specific question at hand. Yet the availability of the necessary software and relative ease with which they can be conducted can result in their mis or overuse.
LeLorier et al propose that a well-conducted meta-analysis should involve a priori determination of strict standards to ensure that the criteria used for the inclusion of patients, the administration of the principal treatment, and the ascertainment of outcome events are similar in all the trials selected.[ii] Any heterogeneity among studies included in a systematic review that is not accounted for in the analysis can contribute to a biased recommendation.
Note: Please read disclaimer. Ask the Expert is for licensed physicians only.