Sample Size Estimation

Introduction: In general, sample size calculation is conducted through a pre-study power analysis. Its purpose is to select an appropriate sample size in achieving a desired power for correctly detection of a pre-specified clinical meaningful difference at a given level of significance. This study/tool will provide statistical procedures for determination of sample size required not only for testing equality, but also for testing non-inferiority/superiority, equivalence (similarity), some survival analysis and phase II clinical trials designs that are commonly employed at various phases of clinical development. As a handful tool, it is very useful for clinical scientists and biostatisticians in the pharmaceutical industry, regulatory agencies, academia, and other researchers.

This program helps users to determine sample sizes and confidence intervals for a wide range of statistical techniques including means, proportions, survival analysis, phase II clinical trials, epidemiological studies and some general cases).

DIRECTIONS:

• Choose one of techniques from the following table
• From the chosen table, please select one of specified case that you are interested
• A new window will pop up and please read the instructions on how to use the calaulation tools.
• If you don’t know which program to use, please feel free to check : User guidance, definition and terminology

Means	Proportions	Survival Analysis	Phase II Clinical Trials	Epidemiological Studies	Confidence Intervals	Others
1. One-Sample Design Test for Equality Test for Non-Inferiority / Superiority Test for Equivalence 2. Two-Sample Parallel Design Test for Equality Test for Non-Inferiority / Superiority Test for Equivalence 3. Confidence Interval – Bristol 4. Compare Two Proportions – Casagrande, Pike & Smith

Note:

Parallel Design

A parallel design is a complete randomized design in which each subject receives one and only one treatment in a random fashion. It does not provide independent estimates for the intra-subject variability for each treatment. As a result, the assessment of treatment effect is made based on the total variability, which includes the inter-subject variability and the intra-subject variability.

Crossover Design

A crossver design is a modified randomized block design in which each block receives more than one treatment at different dosing periods. Subjects are randomly assigned to receive a sequence of treatments, which contains all the treatments in the study. The major advantage of a crossover design is that it allows a within subject comparison between subject variability from the comparison.