Sample size is obviously a key factor in determining the validity and generality of trial outcomes. It needs to be determined carefully to ensure that the research time, effort and support costs invested in any clinical trial are not wasted. Ideally, clinical trials should be large enough to detect reliably the smallest possible differences in the primary outcome with treatment that are considered clinically worthwhile. My review found that it was common for studies to be “underpowered”, failing to detect even large treatment effects because of inadequate sample size suggesting that resources may have been wasted for want of a slightly larger sample. Some ethics committees may object to recruiting patients into a study that does not have a large enough sample size for the trial to deliver meaningful information. Despite the importance of appropriate sample size, only 12 studies reported the use of a sample size calculation.
It is a widely accepted rule of thumb that trials with 30 or less participants per treatment arm are unsuited to conventional statistical analysis. In fact trials with less than 15 participants per treatment arm, while useful for generating hypotheses for further research, are not at all reliable for making conclusive statements. In my review, 78 studies used 30 participants or less per treatment arm . In fact 17 studies used 10 or less participants per treatment arm, making any kind of analysis futile (although this did not stop the investigators from conducting statistical analyses), whereas only 42 studies used more than 30 participants per treatment arm.