|
Abstract:
|
For most of the oncology products, the purpose of QTc studies is to assess the effect on QTc interval, usually no hypothesis has been involved. In this situation, what people normally do is to calculate the precision level, which is the half-width of the 1-sided 95% confidence interval of (1) the time-matched changes from baseline if no placebo involved, or (2) the time-matched changes from placebo after baseline adjustment. In this presentation, I try to answer the following questions: (1) What are the margin of error one will encounter for different combinations of sample size and the standard deviation? (2) In order to have the same margin of error as a typical TQT, how many more subjects needed for the oncology ECG monitoring? (3) If increasing sample size is realistically impossible, how much will be the margin of error increased?
|
