The one-sample log-rank test is a type of statistical test that can be used when comparing survival distributions between a study sample that is going to be collected prospectively, and a standard population or historical control. This test can be employed in clinical trials and epidemiological studies. To make sure the planned study reaches a desired statistical power, researchers need to calculate the sample size of the study before they start to collect data. However, the previous methods developed for one-sample log-rank test tend to underestimate the required sample size. Therefore, a new method is proposed by Wu, 2015, using the exact variance under the alternative hypothesis to derive the sample size for one-sample log-rank test. Simulation study has been done (Wu, 2015) to show that the sample sizes calculated based on the new method has better performance compared to the two previous methods. The new method gives empirical type I error and empirical power that are closer to the nominal type I error and power, compared to previous methods.
The goal of this package is to provide functions that can be used by clinical and epidemiological researchers to calculate sample size, power, and perform simulation study to check empirical type I error and power in the one-sample log-rank test. The statistical methods used in this package and the examples mentioned in this vignettes are based on methods and examples described in Wu, 2015.
References Wu, J. R. (2015). Sample size calculation for the one-sample log-rank test. Pharmaceutical Statistics, 14, 26-33. https://doi.org/10.1002/pst.1654