Combining within study effect sizes for multiple measures of the

Forum (Archive)

This forum is kept largely for historic reasons and for our latest changes announcements. (It was focused around the older EPPI Reviewer version 4.)

There are many informative posts and answers to common questions, but you may find our videos and other resources more informative if you are an EPPI Reviewer WEB user.

Click here to search the forum. If you do have questions or require support, please email eppisupport@ucl.ac.uk.

<< Back to main Help page

Home

Using EPPI-Revi...

Questions about...

Combining within study effect sizes for multiple measures of the same outcome

30/10/2017 18:02

Sally Abbott

Joined: 27/06/2017

Posts: 4

Combining within study effect sizes for multiple measures of the same outcome

Is there a way to combine the effect sizes of several measures of the same outcome within a study?

i.e. different measures of assessing level of physical activity from a questionnaire - some studies report kcal/week and mins/week. Different measures are used between studies.

I have run meta-analyses individually for individual studies with several outcome measures, which has given me the mean effect size for that study. But how do I then run a meta-analysis to pool the effect size across studies?

Thanks

31/10/2017 16:13

Zak Ghouze

Joined: 20/03/2017

Posts: 165

Re: Combining within study effect sizes for multiple measures of the same outcome

Dear Sally,

I’m not sure if I completely understood your question, so I forwarded it to one of my colleagues who is more familiar with different meta-analysis methods. They were a bit confused as well, and were unsure what was meant by “I have run meta-analyses individually for individual studies with several outcome measures, which has given me the mean effect size for that study.“

They did say “Usually, if there are different measures of the same construct they should pick the most appropriate to combine within models. However, if they want to combine data from multiple outcomes in the same trial they need to be thinking about multivariate meta-analysis – it’s not currently possible but coming soon to ER5!”

Do let us know further details if you want more assistance or information on the matter!

Kind regards,

Zak Ghouze

Page 1 of 1

Home

Using EPPI-Revi...

Questions about...

Combining within study effect sizes for multiple measures of the same outcome