Conflicting conclusions from two systematic reviews of epidural steroid injections for sciatica: which evidence should general practitioners heed?

Systematic reviews and meta-analyses are becoming increasingly important in informing clinical practice and commissioning. Two systematic reviews of a treatment for low back pain and sciatica using epidural steroid injections, published in the same year, arrived at conflicting conclusions. Only one was reported in a digest for evidence-based medicine. This paper aims to find the reasons for the discordance between the reviews, and draw conclusions for users of reviews. Using comparative analysis of two published systematic reviews and their source material, it was found that the two reviews had the same overall aims and met the criteria for review methods. They differed in their choice of methods, including the judgement of quality of studies for inclusion and for summing-up evidence. Estimation of summary odds ratios in one review led to stronger conclusions about effectiveness. In conclusion, the choice of methods for systematic review may alter views about the current state of evidence. Users should be aware that systematic reviews include an element of judgement, whatever method is used.