Judging from this comment, the choice to remove fluoride seems to be a way to avoid additional FDA testing (including animal studies).
I find it amazing that only basic testing is needed for cosmetic products and something one is easily ingesting (slightly) on a day to day basis, but yet... fluoride requires animal testing?
I would think at this point, a simple chemical composition testing the fluorine levels would be appropriate enough. They obviously don't find the caffeine a threat enough to warrant such things in the product.
In the case of toothpaste, we've already tested fluoride. I'm going to guess we know the safe limits and the recommended amounts in the stuff. So as long as there isn't any new sorts of drug in the toothpaste and one is using pretty standard safe sort of ingredients, there isn't really a need for some of the testing - animal or human testing. Tests to verify the percentage of fluoride in the toothpaste? Absolutely. Getting the fluoride from verified sources? Sure thing. And so on.
Cosmetics have relatively few standards comparatively [1]. In a way, it seems to be enough. It likely isn't an issue with common ingredients - same situation as fluoride. But on the other hand, if a company uses a new ingredient, I'd rather have some sort of official tests for safety. I think this is probably especially important with things used around the eyes and lips - and I'm not sure some of the warning labels are quite enough. For example, some of the Halloween makeup will have warnings not to use certain colors around the eyes (usually black or red). Of course, this is in small print on the back of the package, and not on the container holding the makeup. Yet if you look at the package, it often has that very color near the eyes.
[1]http://www.fda.gov/Cosmetics/GuidanceRegulation/LawsRegulati...
This is probably exactly what the testing is for. Consider a drug that is (fluoride + X), where both fluoride and X are "already tested and safe". Are we sure that for all values for X, (fluoride + X) is safe also?
Of course, it might also be prudent to get certified to handle fluoride as well as manufacture toothpaste with it and other such things.
But then again, caffeine is a drug occasionally as well. Excedrine has to list it on their FDA required product information: No-Doz (caffeine pills) has the same sort of label. Granted, it might be more benign than fluoride - yet the FDA sees no problem with it being in toothpaste so long as he doesn't have the fluoride.
I don't know all the details, mostly because I've never wanted to manufacture toothpaste. But it seems there should be a simple way of doing this stuff with simple tests in most cases, and that the information should be easily accessible.