+ Reply to Thread
Results 1 to 4 of 4

Thread: Median Survival CI - SPSS vs Stata

  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

    Median Survival CI - SPSS vs Stata




    I've been generating some Kaplan Meier survial curves and the related stats. SPSS and Stata give me the same Median survival but very different 95% confidence intervals. Could someone explain the disparity for me?

    Group 1
    SPSS: Median 60, SE 5.5, CI 49.2-70.8
    Stata: Median 60, SE 5.5, CI 51-94

    Group 2
    SPSS: Median 39, SE 5.8, CI 27.6-50.4
    Stata: Median 39, SE 5.8, CI 27-78

    Group 3
    SPSS: Median 116, SE 14.6, CI 87.3-144.7
    Stata: Median 116, SE 14.6, CI 66-136

    Many thanks!

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Median Survival CI - SPSS vs Stata

    Hmm, I haven't run survival analyses in either of those programs before. The SPSS calculation seems to be the traditional +/- 1.96(SE), while the STATA seems to not be symmetrical around the median. Perhaps meaning it is using an exponentiate'd value maybe.


    Can you provide the output as well. I was initially thinking it could be one is using confidence interval and the other is using prediction interval, but I don't think the asymmetry would exist.
    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: Median Survival CI - SPSS vs Stata

    Stata outputs this:

    . stci if MRI_in_30d==0,by(First_MRI_SEP)

    failure _d: RIP == 1
    analysis time _t: FUpDurWeeks

    Code: 
                 |    no. of 
    First_MRI_~P |  subjects         50%     Std. Err.     [95% Conf. Interval]
    -------------+-------------------------------------------------------------
             Equ |        33          60      5.507446           51         94
            Prog |        15          39      5.796551           27         78
         Sta+Reg |        14         116      14.61789           66        136
    -------------+-------------------------------------------------------------
           total |        62          71      11.10534           57         94

    Not sure if that's what you were after....


    I did some more digging and found this:
    http://www.stata.com/manuals13/ststci.pdf

    The methodology for working out the CI is over my head but I think it's safe to say that it's a more realistic method than just +/-1.96 SE?

  4. #4
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Median Survival CI - SPSS vs Stata


    It has been some time since I ran a survival model, it kind of seemed like the link was for K-M based values and it grabbed the interval value closest to the respective percentile. Sorry I can't be of more help off the top of my head.
    Stop cowardice, ban guns!

+ Reply to 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