Article Text

Download PDFPDF
  1. Kate Jewitt,
  2. Richard Hughes
  1. Cochrane Neuromuscular Disease Group, Department of Neuroimmunology, Guy’s, King’s and St Thomas’ School of Medicine, Guy’s Campus, London, UK
  1. Correspondence to:
 Ms Kate Jewitt, Cochrane Neuromuscular Disease Group, Guy’s, King’s and St Thomas’ School of Medicine, Department of Neuroimmunology, Hodgkin Building, Guy’s Campus, London SE1 1UL, UK; 

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Doctors face an impossible task in appraising, synthesising, and regularly revisiting the evidence, even in a relatively small field such as neuromuscular disease. Randomised controlled trials are the most powerful tool for assessing and comparing the efficacy of different interventions.1 With the possible exceptions of interventions with very large or very small effect sizes,2,3 they provide the best research design to address questions of therapeutic efficacy. Properly designed randomised controlled trials minimise selection bias by ensuring the homogeneity of the comparison groups from the very start of the trial. Interventions for neuromuscular disease need testing in randomised trials as much as interventions in any other field. However, most neuromuscular diseases are uncommon so that collecting sufficient patients to answer even simple questions about interventions with moderate effects is difficult.


Conventional reviews of treatment risk bias from failure to identify all trials, inclusion of flawed trials, the personal opinion of the author and, often, lack of peer review. The Cochrane Collaboration, which was established in 1992, …

View Full Text