Determining the sample size in a clinical trial

Extract Overview Sample size determination is an important part of planning for clinical trials. One of the key aspects of the protocol is sample size estimation. The goal is to ensure that a trial is large enough to detect reliably the smallest possible differences in the primary outcome, with treatment that is considered clinically worthwhile. It is possible for studies to be underpowered, failing to detect even large treatment effects because of inadequate sample size. Therefore, sample size must be planned carefully to ensure that the resources invested including patient participation, are not wasted. It may be considered unethical to recruit patients into a study that does not have a large enough sample size to deliver meaningful information. Elements of sample size calculation The minimum information required to calculate the sample size for a randomized controlled trial includes: ... Power The power of a study is its ability to detect a true difference in outcome between the control arm and the intervention arm. Sample size increases as power increases. The higher the power, the lower the chance of missing a real effect of treatments. Type II error is directly proportional to sample size.