1
$\begingroup$

I am conducting a systematic review and meta-analysis which is investigating health-related quality of life in a particular patient population. I am having difficulties with conducting the meta-analysis as the reviewed studies do not posses a control group. I have the mean, standard deviation and sample sizes for the included studies. The majority of the included studies are cross-sectional to provide an insight into their health-related quality of life. I have seen similar meta-analyses conducted in different populations, but have no idea how to approach this. I have access to Revman, R and Stata. Any advice would be greatly appreciated. (I am still quite new to these programmes so the simpler the better!)

Improve this question
$\endgroup$
9
  • 1
    $\begingroup$ You can compute an effect size and its standard error for each study so I am not sure what your problem is. What else apart from effect sizes and se do you think you need? $\endgroup$
    – mdewey
    Commented Jan 17, 2020 at 17:12
  • $\begingroup$ Hi mdewey thank you for your response. I am just very lost with how to approach this. Every example I have found online has been geared towards meta-analyses with both an experimental group and control group, so I don't even know where to begin. Would you have any recommended resources for a novice? I would really appreciate it. $\endgroup$
    – Ragy Tadrous
    Commented Jan 17, 2020 at 17:19
  • 1
    $\begingroup$ Meta analysis combines a bunch of estimates. You have estimates, and you can calculate a standard error / variance of these estimates? Typically these estimates are differences, but they don't need to be. $\endgroup$
    – Jeremy Miles
    Commented Jan 17, 2020 at 17:49
  • 1
    $\begingroup$ Analyzing this will be easier in R, using metafor (I think it's been a while since I've used Stata or Revman), because that makes an explicit two stage process: get estimates and variances, then pool those estimates. $\endgroup$
    – Jeremy Miles
    Commented Jan 17, 2020 at 17:50
  • 2
    $\begingroup$ After opening metafor try ?escalc to find how to generate effect sizes from what you have then ?rma.uni to find what to do with them. Questions about how to program it in R are best asked on the dedicated mailing list stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis/ where you need to register first before asking. $\endgroup$
    – mdewey
    Commented Jan 18, 2020 at 14:18

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.