Sample sizes and power computation for clinical intervention trials

The purpose of this article was to describe basic concepts of sample size and power estimation for planning nursing intervention trials and interpreting their results. Simple mathematical calculations, using the formulas presented here, can be used to estimate the number of subjects required to conduct a study with a designated effect size and level of power. These methods are of great importance, since most funding agencies require sample size and power estimations before a grant is awarded. In general, studies with power lower than .7 or .8 need careful consideration before they are implemented. In these situations, it may be wise to consider various alternatives for obtaining study subjects or deleting treatment groups for investigations involving more than two groups. The formulas presented here can also be useful in estimating the power of published research findings. Through a quick calculation, the consumer of nursing research can critically evaluate the meaning of a negative trial and draw appropriate conclusions for future research and practice.