Announcement

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

  • Indirect comparison of two interventional groups, meta analysis

    Hi,
    I have been searching all over the internet for an answer to my question.
    I have 10 studies, five which look at intervention #1 and five which look at intervention #2.
    I want to analyze the mean difference reported pain level on the pain scale pre and post intervention for both interventions.
    I have a table set up where I have:
    Author Year Sample Size Intervention #1 or Intervention #2 Mean Pain before Standard Deviation pain before Mean Pain after Standard deviation pain after Mean Difference Standard deviation of the difference
    AAA 20?? 30 1 5 2 3 1 2 0.7
    BBB 20?? 45 2 7 3 4 2 3 1.5
    I have looked into the metaprop command, but that is for binary variables. The variable I am analyzing at is continuous.
    I looked into using metan (metan n1 m1 sd1 n2 m2 sd2) command, but that compares an experimental group vs control group in each study. Each study I have contains only an "experimental" group.
    Lastly, I need to generate a forest plot with my data.
    Any help would be appreciated

  • #2
    There's the "indirect" user-written command that could help: https://www.stata-journal.com/articl...article=st0325

    Comment

    Working...
    X