+ Reply to Thread
Results 1 to 4 of 4

Thread: test for cumulative analysis of outcomes over time in meta-analysis

  1. #1
    Points: 450, Level: 8
    Level completed: 50%, Points required for next Level: 50

    Posts
    11
    Thanks
    4
    Thanked 0 Times in 0 Posts

    test for cumulative analysis of outcomes over time in meta-analysis




    A systematic review and meta-analysis I am reading references having tested for chronological cumulative analysis for the effects of the outcomes (mortality and neuro outcomes). They report no significance but don't give any information of what tests were actually performed. They seem to have some degree of sensitivity analysis built in whereby they adjusted with adding and removing specific trial data.
    My question is this, what would have been the most appropriate test to use for this analysis? 6 trials were included, considered moderately heterogeneous and conducted over a 15 year time period
    Last edited by Roundtoit; 06-02-2017 at 05:47 PM.

  2. #2
    Omega Contributor
    Points: 39,007, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,068
    Thanks
    402
    Thanked 1,191 Times in 1,152 Posts

    Re: test for cumulative analysis of outcomes over time in meta-analysis

    In clinical trials there is interim data analysis. I have only seen it once in meta analyses, but the is a similar approach in where they control for repeated testing and false discovery. It was my impression that for the most part you do the typical MA methods but correct the significance value. Yes, there is a name for this approach, but it slips my mind.

    Does this sound like your scenario?
    Stop cowardice, ban guns!

  3. #3
    Points: 450, Level: 8
    Level completed: 50%, Points required for next Level: 50

    Posts
    11
    Thanks
    4
    Thanked 0 Times in 0 Posts

    Re: test for cumulative analysis of outcomes over time in meta-analysis

    Honestly not completely sure. I know in a single trial, survival analysis like Kaplan-Meier curves can be used or perhaps proportional hazards. I am not sure how that would work, or of it can work looking at multiple studies being pooled. I find it an interesting idea though as the intervention in question - hypothermia in post resuscitation pts, is somewhat unchanged but other advances in care over a 15 year period have certainly changed. How to test that survival has improved or worsened over time between studies independent of hypothermia use is what I would like to know.

  4. #4
    Omega Contributor
    Points: 39,007, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,068
    Thanks
    402
    Thanked 1,191 Times in 1,152 Posts

    Re: test for cumulative analysis of outcomes over time in meta-analysis


    If you are referencing a particular study you can post a pubmed link so we can see what you are writing about. I think I remember the study.
    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