+ Reply to Thread
Results 1 to 4 of 4

Thread: COX models and continuous variables

  1. #1
    Points: 2,465, Level: 30
    Level completed: 10%, Points required for next Level: 135

    Posts
    12
    Thanks
    1
    Thanked 0 Times in 0 Posts

    COX models and continuous variables




    Hello everyone.
    I have a simple and basic question about cox model. When I introduce a continuous variable in the model, the hazard ratio assessed for this variable relates the increase (decrease) of the risk after an increase of one unit of the continuous variable. How to extrapolate for more than one unit (for example 10 units) increase . Is that only the HR multiplied by 10 or is there a specific formula for this extrapolation ?
    For example, a scientific question would be what is the hazard ratio of cordiovacular disease for an increase of 10 mmhg of blood pressure.
    Thanks for your help!!

  2. #2
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Location
    Konstanz
    Posts
    372
    Thanks
    3
    Thanked 146 Times in 123 Posts

    Re: COX models and continuous variables

    The easiest solution is to just prepare your data such that it has the unit of your choice. So in your case just divide your continuous variable by 10 before estimating your Cox model. Alternatively, the hazard ratio for a 10 unit increase is the original hazard ratio to the power 10.

  3. The Following 2 Users Say Thank You to maartenbuis For This Useful Post:

    didye (07-30-2014), hlsmith (07-30-2014)

  4. #3
    Points: 2,465, Level: 30
    Level completed: 10%, Points required for next Level: 135

    Posts
    12
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Smile Re: COX models and continuous variables

    Thank tou so much

  5. #4
    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: COX models and continuous variables


    If I remember correctly, I believe you exponentiate model terms to get the relative values, so maartenbuis suggestion is good approach. If using a software package, you could also request these estimates for different unit values.
    Stop cowardice, ban guns!

+ 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