Sample Size

The size of the study should be considered early in the planning phase. In some instances, no formal sample size is ever calculated. Instead, the number of participants available to the investigators during some period of time determines the size of the study. Many clinical trials that do not carefully consider the sample size requirements turn out to lack the statistical power or ability to detect intervention effects of a magnitude that has clinical importance. In 1978, Freiman and colleagues [1] reviewed the power of 71 published randomized controlled clinical trials which failed to find significant differences between groups. “Sixty-seven of the trials had a greater than 10% risk of missing a true 25% therapeutic improvement, and with the same risk, 50 of the trials could have missed a 50% improvement.” The situation was not much improved in 1994, when a similar survey found only 16% of negative trials had 80% power for a 25% effect, and only 36% for a 50% effect [2]. In other instances, the sample size estimation may assume an unrealistically large intervention effect. Thus, the power for more realistic intervention effects will be low or less than desired. The danger in studies with low statistical power is that interventions that could be beneficial are discarded without adequate testing and may never be considered again. Certainly, many studies do contain appropriate sample size estimates, but in spite of many years of critical review many are still too small [3, 4].