Approaches for calculating power for case-cohort studies

Abstract

An approach for determining the power of a case-cohort study for a single binary exposure variable and a low failure rate was recently proposed by Cai and Zeng (2004, Biometrics 60, 1015-1024). In this article, we show that computing power for a case-cohort study using a standard case-control method yields nearly identical levels of power. An advantage of the case-control approach is that existing sample size software can be used for the calculations. We also propose an additional formula for computing the power of a case-cohort study for the situation when the event is not rare.