With this sampling scheme, the expected sample size will be around 275,000 women.
This design has an expected sample size of 6286%6% of the fixed sample value) and an expected study length of 3.1 years (77% of the fixed sample value).
Design parameters are chosen to minimize expected sample size subject to frequentist error constraints.
Through simulation studies, we evaluate the sensitivity of Type I error, power, expected sample size, and trial duration to different design characteristics.
In addition, the study parameters were determined by comparing the assessment indexes such as total sample size, expected sample size and the test power at the first stage.
The profiles of p-specific expected sample size, mean squared error and risk under different approaches motivate us to introduce the concept of "admissible sample size (design)".
If the effect size is smaller than expected, sample size recalculation would be a reasonable choice in the face of an underpowered study.
The conventional optimal design, for given type 1 and type 2 error rates, is the one which minimises the expected sample size under the null hypothesis.
We dropped this assumption and computed the expected power and expected sample size to evaluate the performance of the procedure under the omniscient model.
Extending the ideas of step-up and step-down methods for multiple comparisons to sequential designs, the new techniques improve over the Bonferroni and closed testing methods proposed earlier by a substantial reduction of the expected sample size.
We conclude that it is relevant to consider optimal designs under a range of hypotheses about the true response rate, and that allowing early stopping for efficacy is always advantageous in terms of expected sample size.
