+ Reply to Thread
Results 1 to 5 of 5

Thread: Please help me selecting a test for my study

  1. #1
    Points: 23, Level: 1
    Level completed: 45%, Points required for next Level: 27

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Please help me selecting a test for my study




    Hi,

    I was wondering if you guys can provide some ideas on best test for my samples size, since I am not very stat savvy.

    My project consist of 3 components on decreasing hospital readmission of pneumonia using these drugs
    drug A, B, and C comparing with control group with none of these drugs from last year.

    The sample size is kinda different between them.

    drug A + drug B + drug C compare with control -->small sample size (20 patient)
    drug A Only compare with control -->small sample size (20 patient)
    drug B Only compare with control --->larger sample size (250patients)
    drug C Only compare with control ---->moderate sample size (100 patients)

    It would be great if you guys can help me with this complicated test of different groups.

    Thanks alot!!

  2. #2
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 639 Times in 601 Posts

    Re: Please help me selecting a test for my study

    You control group has n=10 or so?

    Looks like 4 2x2 tables, with Chi² test or Fisher's exact test, respectively.
    The comparisons with n=20 subjects will probably have very low statistical
    power.

    With kind regards

    K.

  3. #3
    Points: 23, Level: 1
    Level completed: 45%, Points required for next Level: 27

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Please help me selecting a test for my study

    Thank you for the kind response! I had all my data and encounter a problem with disproportional sample size. My control data has about 200 pneumonia admissions receiving standard drug with 30 Readmissions, but the study group has about 250 pneumonia admissions with 20 Readmissions, but the number of patients in the study group receiving drug A is very small (used in about 20 admissions only, and not all these patients used drug A, so that cut down to like 10 readmissions who used drug A). Is there a way for way to make a comparison between control and my study group to determine whether drug A has an effect on readmission?

    Thank you very much for the help!

  4. #4
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 639 Times in 601 Posts

    Re: Please help me selecting a test for my study

    Still, 2x2 tables, with Chi² test or Fisher's exact test. Or what else do you want to know?

    With 170/30 (controls) versus 10/10 (drug A), total sample size is 220, which is not dramatically bad.

    With kind regards

    K.

  5. #5
    Points: 23, Level: 1
    Level completed: 45%, Points required for next Level: 27

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Please help me selecting a test for my study


    Is this how I set up my table data for comparison for 2015 control group and 2016 study group?

    Pneumonia ↓→ 2015 (2016)
    # Readmission 21 (21)
    ED admission visit 158 ( 239)
    TOTAL # 179 ( 260)

    Percentage of 12% ( 8.00% )
    Readmission

    As you can see above, 2016 study group has a total of 260 admissions and an overall of 8% of readmission compare to control group (12%). However, only about half of that 8% received drug A. It is hard for me to see how much of drug A having an effect on readmission since only half of those readmissions used drug A. Or am I just thinking over my head...sorry for the confusion as I am a little confused myself. Thank you!
    Last edited by kite6k; 05-24-2016 at 09:14 AM.

+ 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