Perhaps you could download the entire meta study that is linked next to the abstract and go through it?
No, I am not refereeing a paper because some commenter links it in a web forum. Why would you think that’s even close to what anyone should do in this environment?