Precision-based sample size calculation for one survival curve

data analysis
, ,
1

I am designing a cohort study whose main objective is estimation of the survival probability for the group.

Many packages are available to estimate the sample size for a single mean or a single proportion based on the margin of error (or CI width), yet, to my knowledge there is none on single-arm time-to-event outcomes. Any guidance?

2

Here’s a start: Regression Modeling Strategies - 20  Cox Proportional Hazards Regression Model

3

The calculator: Sample size calculation for one sample non-parametric survival (survival proportion) | Kengo Nagashima - Keio University
I would appreciate if you could explain how to calculate ‘Null survival probability’ and ‘Alternative survival probability’ :slightly_smiling_face:

1 Like
4

Thank you for the link, but this seems to still do a “one sample test” comparing a sample’s survival probability to a fixed probability (null hypothesis testing), which is different from what I am looking for (precision).