Hello,

If by "dimensions" you're referring to the different treatment groups (arms) within each study, then a network meta-analysis would normally be a good approach. However, since none of the studies compare the same intervention, a standard network meta-analysis isn't feasible.

You may do a proportional meta-analysis of each arm and combine the results in one paper if the arms are closely related.

PS. Assuming you referred to study arm.

Khawaja Abdul Rehman Thanks for your feedback.

This is a multivariate meta-analysis. You would need to do something like a network meta-analysis. You can do this in Stata

Prof James Hurley

Thanks for your comment. If possible, please share any resources or articles that I can use.