Double-blind experiments are a great way to separate correlation and causation. See my (@zenkat's) wife's story above. Every so often, she accidentally and unknowingly eats foods she is sensitive to, and exhibits symptoms shortly afterwards. We go back and check the label and ... yep, there's the dairy/soy/yeast extract.
Oh, and there's no financially motivated third-parties here. We've done all this testing, elimination, and observation on our own.
Oh, and there's no financially motivated third-parties here. We've done all this testing, elimination, and observation on our own.