Dear all,

For some time, we have been working with Stata metaprop in order to analyse our results from our review on two surgical techniques (for now A and B).

Through the metaprop command (Metaprop: a Stata command to perform meta-analysis of binomial data, by Nyaga et al.) we were able to perform a meta-analysis of our data.

However, where the 'normal' Stata program for meta-analysis has a test for group differences ( Qb with corresponding p value) to compare overall effect sizes of the two groups.

Metaprop gives a significance test with hypothesis effect size = 0 (ES = 0).

This results in tree Z values and their p values, one for each group and one overall outcome.

Eg. outcome A : z= 7.68, p=0.00

outcome B: z= 8.77, p=0.00

Overall: z =9.79, p=0.00

Is there a way of telling or calculating the group difference between the effect sizes of group A versus group B?

Thanks in advance,

Regards,

Iris Hochstenbach