+ Reply to Thread
Results 1 to 4 of 4

Thread: Dealing with multiple confounding variables

  1. #1
    Points: 33, Level: 1
    Level completed: 66%, Points required for next Level: 17

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Dealing with multiple confounding variables




    for example, we have 3 confounding random variables, x1, x2 and x3 with 3 different variances. If we had to treat each variable alone, we would have used odd ratio or relative risk ( depending on what kind of study we are using whether retrospective or prospective) to determine the risk of the disease given the variable. What if we have 3 variables? how can we combine all of them to the get the odd ratio of diagnosing the disease? Do we have to use logistic regression analysis with the logit as a linear combination function of multiple variables? Also what does it have to do with our specific patient who comes with a combination of x1, x2 and x3, do we have to plug those values into the logistic analysis in order to get the probability of having the disease?

    To complicate the issue ( although it is not neccessarily at this stage), what if we conduct a metaanalysis with different studies are conducted at different places, how would we combine those data into a the result?

  2. #2
    Omega Contributor
    Points: 38,253, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,989
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Dealing with multiple confounding variables

    Three confounder, all related to the same variable or different variables? How do you know they are confounders without running a multiple regression yet?
    Stop cowardice, ban guns!

  3. #3
    Points: 33, Level: 1
    Level completed: 66%, Points required for next Level: 17

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Dealing with multiple confounding variables

    I consider them independent variables. I may omit the term confounding for now.

    Now suppose my study is retrospective one, case-control, which is suppose to get the odd of the exposure given the disease. Then how to get the probabilty of the disease given the variables in hand? Because this is what I care about, namely the probability of the disease given the exposure not the opposite. How can I run the logistic regression in the retrospective study and get a meaningful inference about the probability of the disease given the exposure?

  4. #4
    Omega Contributor
    Points: 38,253, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,989
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Dealing with multiple confounding variables


    Yes, you would want to run logistic regression. It would produce the log odds which can easily convert to probability.
    Stop cowardice, ban guns!

  5. The Following User Says Thank You to hlsmith For This Useful Post:

    PeterFlom (05-24-2015)

+ 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