The thing about toxicity though - not all poisons have any effect whatsoever in low doses. Some do have effects that don't go to zero at low doses - radiation for example - because it does not dilute below 1 particle track through nucleus, which is harmful. The carcinogens in general, as they don't dilute below causing 1 mutation to dna in 1 cell, which has probability of harm. But organic compounds which are not carcinogenic - not as in natural but as in organic chemistry - and especially those from plants, they may well be entirely harmless at the doses below those that actually result in toxicity. Of course there may be situations where they are 'the last straw', but in general not everything is harmful all the way down.
It's not quite that the toxicity effects don't go to zero at low doses for things like radiation — the effects of low radiation exposure may indeed be zero. The toxicity may be zero at even relatively high doses. Rather, the value of radiation's dosage curve at the low end is stochastic, where for chemical toxins the value of the dosage curve on the entire domain is the severity of toxicity.
Nutrition is a case where we have to try to make the best possible use of the data we have no matter how terrible, because we have to eat something now to sustain us while we plan and conduct more experiments.
I want to apply Bayes theorem to make rational health decisions from relatively weak data. I am generally wondering how one can synthesize historical human experiences with incomplete scientific data, in order to make risk-adverse and healthy decisions about human nutrition given limited research.
Example question/hypothesis: Does gluten cause health problems (ie exhibit chronic toxicity) in non-coeliac humans? Is there enough evidence to suggest that avoiding gluten might be a prudent risk-adverse decision for non-coeliacs?
We have some (mostly in vitro) scientific data suggesting that gluten may cause health problems in non-coeliac humans (such as these articles http://evolvify.com/the-case-against-gluten-medical-journal-references/). Let's say for the sake of arguing, that I can somehow convert these studies into a non-unity likelihood ratio for gluten toxicity in humans (although suggestions are welcome here too).
However, we also have prior information that a population of humans has been consuming gluten containing foods for at least 10,000 years, without any blatantly obvious toxic effects. Is there some way to convert this observation (and observations like this) into a prior probability distribution?