Some meta-analyses do kinda have the vibe of collateralized debt obligations and credit default swaps, combining weak studies and claiming a strong result. (Case in point: some of the bogus studies of ivermectin etc for covid). Not all of them, but if you’re not working in a field (and good at evaluating methods etc) you probably should wait for someone in the field to evaluate meta-analyses in the field.