+ Reply to Thread
Results 1 to 3 of 3

Thread: power calculation for two groups (parallel)

  1. #1
    Points: 10, Level: 1
    Level completed: 19%, Points required for next Level: 40

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Lightbulb power calculation for two groups (parallel)




    Hi
    I have got 400 and 600 samples in control and test groups respectively. And in each of these two groups, i have an effect variable that occurs at the rate of 30%. this is a retrospective data analysis.

    I want to investigate whether the effect variable impairs the patient outcome (response rate as well as median survival) in control and treatment groups and if so, whether such effect is any different in treatment and control groups.

    Any ideas which variables i need for this calculation ? What formula can i use to calculate the power for this study. thanks

  2. #2
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: power calculation for two groups (parallel)

    So you have two groups (test and control) with 600 and 400 people respectively in them. You have an exposure variable which is equal in proportion for both groups and you want to see if it explains median survival (as a binary variable) and response rate (please better define this).


    So it sound like you will be running a survival analysis, correct?
    Stop cowardice, ban guns!

  3. #3
    Points: 10, Level: 1
    Level completed: 19%, Points required for next Level: 40

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: power calculation for two groups (parallel)


    Quote Originally Posted by hlsmith View Post
    So you have two groups (test and control) with 600 and 400 people respectively in them. You have an exposure variable which is equal in proportion for both groups and you want to see if it explains median survival (as a binary variable) and response rate (please better define this).


    So it sound like you will be running a survival analysis, correct?
    Thanks, hlsmith.
    Median survival for each group can be computed by number of years of disease free survival for each case. So it is a continuous variable but can be dichotomized in case it helps analysis.

    Response rate is percentage of patient responded in each group, which is based on dichotomous variable for each patient- responded or not.

    I want to see whether the effect variable (which is discrete) in control and/or test groups can influence survival.

    I want to know if the number of cases are adequate based on power calculation.

+ 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