[KempyKolibri]

Even that wouldn't be sufficient. Look at the heterogeneity in dairy - the same amount of SFA in yoghurt has a markedly different impact on LDL-c compared to butter, likely because of both the calcium content of the yoghurt _and_ the actual molecular structure (non-churned dairy has intact milk fat globule membranes).

I actually think we need to go the other way and look at foods as foods where we have the data, rather than individual components. Most recent dietary guidelines are more "x% of your plate should be vegetables" than "you should consume x% of your energy as cereal fibre", at least in their headline advice.

[bilsbie]

I agree but I think both approaches are needed.

If this device simply found most bad stuff (when above safe limits) we’d be in a way better position. Eg. Arsenic, lead, pesticides, etc.

* edited to add “above safe limits” since folks seem to be strawmanning my point. In case it really wasn’t clear.

[KempyKolibri]

This is what regulation already does (quite effectively too, at least over here in the UK). We already know that harmful substances aren’t likely to be present in our foods thanks to regulatory checks.

Then we’d be left with checks for substances at levels lower than regulations are concerned with, but I’m not sure why we’d care about that.

Fish has mercury present in it, but increased consumption seems to be associated with positive health outcomes. If the device said “danger, mercury”, what are we replacing it with? Red meat? Sausage? The current evidence would suggest that would be a retrograde step.

[mrguyorama]

So you never eat rice or apples right?

They always contain arsenic. They always have