A survival analysis has been conducted based on secondary data. How can we calculate power of the study to compare mean survival among two groups of patients?
The data that I have is the mean survival expected among test and control group patients.
The South West Oncology Group (SWOG) provides web-based tools for calculating sample size and power for various designs. Here is the link for a 2-sample survival analysis power/sample-size calculator:
http://www.swogstat.org/stat/public/survival/two_survival.htm
This sounds like a t-test comparison of two groups. The power can easily be calculated in R using for example the pwr.t.test() function in the pwr library. There are also different options if you work with proportions, single sided or two sided t-tests etc..
No. This is not like a t-test comparison. A major difference is that survival data are subject to censoring. You cannot compare means for censored data using a t-test. Special handling is required. Log-rank tests and other similar tests are commonly used.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress