provaxx,
A meta-analysis is only as good as the studies that it is using. If the studies are bad, then the meta-anlysis is also bad. If you look at that paper, it includes the paper that was retracted due to fraudulent data and since that paper had the strongest effects, the meta-analysis would need to be completely re-run. Indeed, the authors of another meta-analysis published around the same time in Open Forum Infectious Diseases have already retracted their paper and plan to reanalyze the data excluding the fraudulent studies.
More recently the ivermectin arm of the “Together Trial” in Brazil (I think 1,300 patients) was halted due to it not working and the patients were reassigned to other arms of the study. This was a very large scale trial done by researchers at a number of universities in Canada and Brazil. Admittedly it was tested on already sick patients so it does not tell us anything about prophylactic use, but it still shows that is is definitely not some magic bullet. The study did find that Fluvoxamine – an antidepressant which is cheaper than ivermectin and considered to be safer too – did have a significant effect, though I would wait until another high quality large size study of it comes out before jumping on that bandwagon either (there are couple such studies on Fluvoxamine currently underway).