Biometric planning of clinical-orthopedic studies. The optimal test

Abstract

A determinant for the evidence of a clinical trial is the magnitude of the sample size. The proper sample size can be easily computed with the knowledge of alpha, power and effect size. Standard values for alpha and power in clinical trials are 5% and 90%, respectively. As a consequence, effect size is crucial for the sample size. The effect size has to be determined by the clinician according to medical considerations. Possible consequences of sample sizes that are either too small or too large are discussed with regard to the meaning of alpha, Power and effect size. Trials with improper small sample sizes have a high risk of false negative results, and may subsequently prevent the application of a possibly effective therapy. Trials with improper large sample sizes may result in statistically significant differences without any clinical relevance.