Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Forest plot of Hazard ratios

    Is there a way to create a forest plot of Hazard ratios by various subgroups? I am doing a Cox Hazard regression model and would like to include something that looks like the attached. I did not have any luck with "fintplot" as some had suggested in older posts. Thanks for any suggestions.
    Attached Files

  • #2
    Christine, try looking at http://www.statalist.org/forums/foru...p-interactions for a start.

    Comment


    • #3
      Thank you for all your hard work with these commands. I came across your posts previously and made some progress using your examples as a guide but am stuck again. I am trying to create a plot of hazard ratio for outcome= death based on a number of categorical variables. There is no interaction or treatment, just the single outcome variable for which I want to compare HRs between a variety of subgroups. When I try to use:
      ipdover, over(binvar1 binvar2 binvar3 binvar4) forest(nonull): stcox
      I get the error message "explist error". I'm certain the syntax is the problem, but cannot find a fix for it. Thank you!

      Comment


      • #4
        Christine,

        Unfortunately at present ipdover is not designed for analyses where an outcome is compared across subgroups with reference to a base category; it assumes that a treatment variable is present. This extra functionality has been requested by numerous users and I intend to program it over the coming months, time permitting.

        The current issue of the Stata Journal has an article on ipdmetan, and a minor (mainly bug-fixing) update will be available via SSC in the coming days. I intend to start a thread on Statalist when this happens, and in that thread I will post some example code which should achieve your desired result, albeit with a bit of fiddling.

        Thanks,

        David (author of ipdmetan and ipdover).

        Comment


        • #5
          See http://www.statalist.org/forums/foru...es#post1301387

          Comment

          Working...
          X