Shawn Baker (on his instagram), Chris Kresser (on Joe Rogan), and the youtube channel "What I've Learned" amongst others, have all stated, in various forms, that:
The main problem is that these exact characters will turn around and post, for instance, fake pictures of vegan bones or use the Epic-Oxford study and claim that numerous epidemiological studies show that vegans have bone problems. So then which is it? Are they good or bad? It would sure seem as though "It's good when it supports my case, but in the vast majority of cases I look silly so... not good!" But you can't have it both ways and cherry-pick the epidemiological studies you don't like.
Recall that the carnist is required to show that eating animal products is necessary. Failing to do so means that one is special pleading the rights of specific species of animals away for no demonstrable reason. It would be indistinguishable from any possible vague "oh it's necessary, trust me bro" argument. In short, the carnist carries the burden of proof. We recognize this as justification for unethical actions in any other context in any other context. "Why did you steal that car?" "oh it was necessary. It just was. Trust me.". You would be almost obligated at that point to say that they need to show that it's necessary; they can't just assert it. And "well, you can't prove it wasn't necessary" isn't a defense. You carry the burden of proof in such cases because it is an unfalsifiable position.
Thus, any argument that goes "I don't know and you don't know either" is an admission of defeat. Because you're saying your statement, for which you have the burden of proof, is unprovable. It is to state "Animals are necessary, but I can't prove it because we can't trust any of the science, so you'll just have to trust me instead".
Not all epidemiology is created equal. There are multiple levels of epidemiology and a systematic review of the literature is not magically equivalent to some random one-off case study on a single subject. So to lump it all together and dismiss everything is ridiculous. Good evidence is good, bad evidence is bad, and scientific consensus is the consensus.