metan for mean differences?

[Talia Sherrard]

Hello,

I am conducting a systematic review and the majority of studies present their results as means and SDs of groups, with a p-value generated from chi squared/t test/ANOVA.

Would it be possible to conduct a meta-analysis of these effect estimates in stata? As far as I can tell it looks like metan is geared towards meta analysis from regression outputs.

If there are any systematic reviewers, or anyone who has experience in this area who have time to advise it would really be appreciated, many thanks

[David Fisher]

Dear Talia,

Any of metan, or the more recent user-written alterative admetan, or the built-in Stata 16 routine meta, will quite happily handle means and SDs by group.

For example, with metan the syntax would be:

metan n1 mean1 sd1 n0 mean0 sd0

where 1=treatment group and 0=control group.

Regards,

David.

[Talia Sherrard]

Hi David,

Many thanks for your quick reply! Please could I just confirm, if I have a dataset in the following format;

case mean, case SD, control mean, control SD
25, 0.2, 24, 0,2
22, 01, 21, 0,1
23, 0.2, 22, 0.2
..etc

What information is stata requiring in the n1 and n0 groups?

[Leonardo Guizzetti]

Those would be the sample size in each group.

[Talia Sherrard]

Those would be the sample size in each group.

Thank you Leonardo, so obvious in retrospect!