Growing pains of meta-analysisBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7059.702 (Published 21 September 1996) Cite this as: BMJ 1996;313:702
- Ida Sim, Postdoctoral fellow,
- Mark A Hlatky, Professor
- Department of Health Research and Policy, Department of Medicine, Stanford University School of Medicine, Stanford, CA 94305-5092, USA
- Supported in part by Grant HS 08362 from the Agency for Health Care Policy and Research, Rockville, MD.
Advances in methodology will not remove the need for well designed trials
Meta-analysis is a statistical technique for pooling clinical trials of similar design. It achieves higher statistical power for detecting differences between treatments and obtaining more precise estimates of outcome. Meta-analysis quickly chalked up some notable successes after its introduction to medicine. Meta-analysis of several small studies suggested thrombolytic treatment would reduce mortality from acute myocardial infarction by 20%,1 a prediction confirmed by subsequent large randomised trials.2 The value of the meta-analysis in providing a quantitative overview of data from many studies soon became apparent. Optimists suggested that cumulative meta-analysis of small trials would identify effective treatments sooner,3 …