Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: sample size calculation for survival data


From   Paul Seed <paul.seed@kcl.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: sample size calculation for survival data
Date   Wed, 07 Mar 2007 15:02:43 +0000

Paolo Di Stefano <paola.di.stefano@medtronic.com>
asked:

Does anybody know how to calculate sample size in the following
situation with Stata?


H0: s1=s2 (survival at 2 years) H1: one side (logrank non paramentric
test for exponential data)
Size ratio =1
S1=77,5%, s2=83,125%
Alfa=5%
Beta=20%
Lost follow-up 10% per year
Accrual time =2 years
Total study time =4 years (2 enrollment+2 follow-up)

As Svend Juul pointed out, the standard power calculations is
given in Stata by :

        sampsi .83125 .775 , onesided power(.8)

To correct for 10% LTFU per year, the answer should be divided by 0.9^2

        di (r(N_1) + r(N_2) )/.9/.9

This assumes that a simple chi-sq test will be used, rather than
a non parametric logrank test.  However, you don't seem to
have any information on when the events might occur.

Using or ignoring the time to events has minimal
impact on the power, so this approach should be
fairly safe.


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index