There are so many incidental factors that it's very hard to form a practical conclusion from such studies. I find the only way to approach this is similar to a neural network with a very slow learning rate: each data point nudges my understanding at a weight determined by how credible, objective, and applicable to me (because everyone is different) the study seems.

I see no other way than to ultimately treat this as a soft science rather than a hard one, focusing on qualitative takeaways, and sometimes taking away a "consideration" rather than a conclusion, e.g. "there's a chance eggs might ultimately be bad, best not overdo it; rice contains arsenic, it's probably best to get rice produced in low-arsenic places, and follow a soaking/rinsing procedure before eating it..."