I did not mean to imply the science was not ready to do this testing. Food labs in manufacturing facilities do this testing every day. This is not new technology.
I'm suggesting that this MAY not necessarily be so.
There exist standard protocols in critical batch testing in any industry. They are
known to be matrix-limited, and to require validation (basically, big data) in any new application.
I fear that the science IS NOT actually ready, but that there are people who
think that it is.
Foods are not a uniform matrix, and food allergen proteins will partition differently in different foods by virtue of their denaturation, temperature-related characteristics, and chemical behavior within the matrix. That creates enormous sampling problems that the pharmaceutical industry basically doesn't have to manage in manufacturing.
Oh, sure, "protein" testing works pretty well.
But that isn't useful here-- because pretty much any food is going to come up positive... so then you get into specificity assays, ALL of which have limitations, and different limitations/intereferents for different assays.
So even occasional line testing is still a job for someone with significant training and expertise if it's going to be accurate enough to mean anything, and you'll need to do it regularly (every 10 hours of run time on a line, for example) BEFORE you can say what it means.
All for information that John Q. Public will neither appreciate nor (mostly) understand.
That kind of cost is NOT something that industry is going to absorb without being forced to do so. They really don't have any reason to.
I mean, ask yourself this--
what is the difference between
20 mg
and
20 mg + or - 15 mg
?
Is one of those safer than the other? Is one of them more informative than the other? Would you be more willing to buy one over the other? Why?
My answer is that I would NEVER buy the former because it indicates that the company doesn't understand analytical statistics, and that the latter indicates to me that the assay is running near the limit of determination, and that it's possible that the assay is inaccurate. I wouldn't buy either one of them for that reason. What I might like better is 25 mg + or - 2 mg.