How did they calculate the RDAs, anyway?

One thing led to another on the internet and I find myself wondering exactly how we came up with the Recommended Dietary Allowances (RDAs) and Daily Recommended Intakes (DRIs). There are all sorts of resources for finding out what the numbers are, but I’m curious about what kind of testing led to the original recommendations in the 1940s.

The guy who made the connection between pellagra and diet did so - in part - by depriving test subjects of certain foods until they got sick.

(1) How does a nutrition scientist go from “less than this amount of Vitamin Y will make you sick” to “that amount of Vitamin Y is sufficient/optimum”?

(2) How do you ethically test these values today?

There’s a description in Chemistry, by Pauling and Pauling, on how it was done to determine what amino acids were essential. That is the citation I am offering. You can check it for the results, and a good description of the deviations from the mean of a relatively small sample size. Such research has probably been done again and again, since then, and there may be more references.

In the case cited above, to summarize, a bunch of college age men were put on diets with less and less of one amino acid, and their urine was monitored for amino acid excretion. As they were fed less and less, output decreased, then suddenly shot up, at that point, we know, their bodies were breaking down thier existing tissues to get the amino acids it needed. From this we know how much is too low to go, as it were.

Basically, all such knowledge is aquired this way. A healthy, young, adult, male is offered a chance to advance human knowledge (and earn themselves a bit of cash) if they will deliberately, in a calulated manner, starve themselves for a short time. Once we’ve developed a mean sort of baseline from this sort of research, more studies may be done, with women, and older people. And from the deviations we try to extrapolate.

Nothing in what I’ve said gives the specificity of values that you asked for in question number one. And this is true. Fact is, the RDA are not meant to be absolutes for all people, in all circumstances.

Thanks, Arkcon! (I’m now curious about how one gets a meaningful mean and standard deviation off relatively few data points, but that’s another question.)

Sidetrack: Why is this all in the USDA’s department rather than the FDA? Historical reasons?

It’s all detailed in Recommended Dietary Allowances: 10th Edition (Dietary Reference Intakes) by Subcommittee on the Tenth Edition of the Recommended Dietary Allowances, Food and Nutrition Board, Commission on Life Sciences, and National Research Council.

Fascinating stuff, but not easily summarizable. If you’re really interested there’s a copy on sale for a penny (plus shipping).