[Skip to Content]
[Skip to Content Landing]
January 4, 1995

Reporting of Power and Sample Size in Randomized Controlled Trials

Author Affiliations

Valley Care Medical System Pleasanton, Calif

JAMA. 1995;273(1):22. doi:10.1001/jama.1995.03520250036017

To the Editor.  —Mr Moher and colleagues1 have done a great service to the medical community by reviewing a large sample of articles from important journals for sample size adequacy. They noted that most trials with negative results did not have sample sizes large enough to detect a 25% or a 50% relative difference and that few trials discussed whether the observed differences were clinically important.Pressure from managed care organizations to save money can take the form of providing fewer services, with the rationale that there appears to be no outcome difference between a more costly and a less costly path of diagnosis or treatment.For example, it has been suggested that biochemical profiles be reduced from 23 to three tests,2 that three stool specimens be pooled for a single examination,3 and that laboratories should not search for Yersinia organisms in patients with diarrhea.4These

First Page Preview View Large
First page PDF preview
First page PDF preview