CROSS-OVER TRIALS WITH CENSORED-DATA

Cross-over trials assign two or more treatments sequentially to the sa me subject; groups of subjects receive different treatment sequences. Both parametric and non-parametric methods of inference are available for cross-over trials with complete data. In this paper we develop met hods for estimation and testing in cross-over trials with censored dat a, based partly on methods used for complete data. Our estimator is co nsistent for the true effects. Simulation results show that both of ou r proposed tests have approximately nominal size. We compare our proce dures to a method for cross-over designs based on Cox regression propo sed by France, Lewis and Ray. We demonstrate that our method of estima tion is superior to the Cox-based method, which has considerable bias. Both of the tests presented here have more power than the Cox-based t ests in all of the situations we investigated. The estimation and test procedures apply to other designs, such as parallel trials and repeat ed measures designs.