Which test do I use?


First time poster; apologies if the question is misplaced.

I'm attempting to do some basic meta-analysis. I have a group of studies looking at the impact of different surgical resections on risk of a particular complication - biliary leak. For example, if I have a group of 10 studies looking at resection of the pancreatic head, 12 studies looking at resection of the pancreatic tail, and 4 studies looking at resection of whole pancreas. Each study has a different sample size and proportion developing bile leak. I have been able to calculate a pooled effect size with random effects model to state what the pooled risk is for each type of surgery for the complication. How can I then compare between the three different types of surgery to prove whether the difference in pooled risk is statistically significant? I have done some reading around relative risk - but does not seem to completely apply to my situation as no studies have a direct comparison of different surgeries within their cohort.

Thank you in advance if anyone is able to assist.