ABSTRACT
In clinical trials, it is important to identify the primary response variables that will be used to address the scientific and/or medical questions of interest. The response variables, also known as the clinical endpoints, are usually chosen during the planning stage of the study protocol in order to achieve the study objectives. Once the response variables are chosen, the possible outcomes of treatment are defined and the corresponding information would be used to assess the efficacy and safety of the test treatment under study (Paul 2000). For evaluation of the efficacy and safety of a test treatment, a typical approach is to test whether there is a statistically significant difference between the test treatment and an active control or a placebo control. If a statistically significant difference is observed, the trial is powered to have a high probability of correctly detecting a clinically meaningful difference if such a difference truly exists (Chow and Liu 2004). As a result, in practice, a prestudy power analysis for sample size estimation is usually performed to ensure that the trail with the intended sample size will have a desired power, say 80%, for addressing the scientific/medical questions of interest. Note that a prestudy power analysis for sample size estimation is often performed based on some required information such as the clinically meaningful difference of interest, variability associated with the observed response, significance level, and the desired power provided by the investigator (Chow, Shao, and Wang 2008).
