METAANALYSIS IN THE ASSESSMENT OF THERAPY FOR STROKE PREVENTION

The meta-analysis statistical method enables the pooling of data from a large number of similar studies so that highly significant conclusions can be drawn because of the increased patient numbers. This method should never compare patients in one trial directly with those in another because the patients, treatments, duration of treatment, quality of follow-up and end point definitions may be different. Only like can be compared with like within one trial. It must not be assumed that the sizes of risk reductions in different trials will be similar. When these caveats are ignored, inappropriate conclusions may be drawn. Data are presented challenging an assumption based on meta-analysis that 325 mg a day or less of aspirin is not only safer than 975 mg a day or more, but is equally beneficial for patients with transient ischemic attacks or stroke. The evidence to date suggests that although the higher dose may be more effective than the lower, this can only be determined by a properly designed, large, prospective study.