ABSTRACT
Estimation of sample size and power analysis for time-to-event analysis in randomized controlled trials poses higher complexity and considerations as compared to the routine sample size and power calculations for continuous or binary outcomes. In addition to the usual considerations for Type I and Type II error probabilities, expected treatment effect, losses-to-follow-up, and event probability distribution, factors such as duration of recruitment, total study duration, staggered entry distribution, treatment non-compliance, competing risks, and possibly non-proportional hazards are important components of sample size and power analysis for randomized controlled trials with time-to-event outcomes. This chapter focuses on strategies and considerations for different situations likely to be encountered in randomized controlled trials that generate time-to-event outcomes. This chapter will also touch on recent novel developments in the area of time-to-event analysis.
