ABSTRACT
This chapter introduces the concepts of confounding and interaction effects in clinical trials. It discusses the controversial issues between the use of a one-sided test and a two-sided test in clinical research. The chapter summarizes the difference in sample size calculation between a crossover design and a parallel design. It also discusses the concepts of group sequential boundaries and alpha spending function in subgroup/interim analyses in clinical trials. The chapter provides a discussion regarding some practical issues that have impact on sample size calculation in clinical trials. These issues include unequal treatment allocation in randomization, sample size adjustment for dropouts or covariates, the effect of mixed-up treatment codes during the conduct of clinical trials and the loss in power for treatment and/or center imbalance. The issues also include the issue of multiplicity in multiple primary endpoints and/or multiple comparisons, multiple-stage design for early stopping and sample size calculation based on rare or extremely low incidence rate in safety assessment.
