+ Reply to Thread
Results 1 to 3 of 3

Thread: Calculating sample size for non-inferiority trial using Stata cii

  1. #1
    Points: 6, Level: 1
    Level completed: 11%, Points required for next Level: 44

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Exclamation Calculating sample size for non-inferiority trial using Stata cii




    Hi, can I use Stata command cii to calculate the sample size required if I want to test that fasting TG is not lower than non-fasting TG by 0.2mmol/l? Given mean=1.4, sd=0.5, 95% confidence level. What is the exact command?

    How do I calculate sample size if I want to test for equivalence of the two TGs with the same information?

    Thanks!

  2. #2
    Omega Contributor
    Points: 39,242, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,094
    Thanks
    405
    Thanked 1,202 Times in 1,163 Posts

    Re: Calculating sample size for non-inferiority trial using Stata cii

    Not a regular STATA user, but got following in SAS.




    Hmm, if blurry - it said about 5 per group given power .8, alpha .05, etc.
    Attached Images  
    Stop cowardice, ban guns!

  3. #3
    TS Contributor
    Points: 1,930, Level: 26
    Level completed: 30%, Points required for next Level: 70

    Posts
    258
    Thanks
    39
    Thanked 73 Times in 64 Posts

    Re: Calculating sample size for non-inferiority trial using Stata cii


    Quote Originally Posted by hlsmith View Post
    Not a regular STATA user, but got following in SAS.




    Hmm, if blurry - it said about 5 per group given power .8, alpha .05, etc.
    Also, remember that sample size calculations are absolutely the minimum number of observations if you used the correct assumptions and if there are no sources of bias or any other issues. This calculation is only "true" under perfect conditions. So, if 5 is the minimum per group, it is probably wise to get more subjects per group, unless there is something extremely prohibitive about that. This gives you room for normal issues that arise and also protects you to an extent from being under powered if you over estimate the effect size or underestimate the variability, for example.

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats