Last September I gave a two-part talk for Sean Croxton’s Real Food Summit over at Underground Wellness called “Weston Price on Primitive Wisdom,” which covered Price’s scientific approach, perspective, and conclusions. The talk was free for a few days, but listening to it now requires purchasing the entire conference for $99, a great deal well worth the money considering the price includes 27 talks and tons of free bonuses. Since many people can’t afford this, however, and since I got a lot of great feedback on the talk, I wanted to turn it into a series of blog posts that will cover the most important points.
It can be easy for us to mistake Price’s conclusion — that “primitive wisdom” protected many groups following ancestral diets and lifestyles from disease and provided them with vibrant health — for his scientific approach. We may suppose, for example, that Price simply observed these groups and their good health and attributed this health to their diets, potentially overlooking all kinds of confounding factors. We may suppose he went out looking for “primitive” people because he somehow knew they were “wise.” This would would be a disservice to Price’s work and would make it seem more like soft observation than science. So in this post I’d like to take a deeper look at Price’s study design.
Before documenting the devastating consequences of the nutritional transition from traditional to modern diets around the world, Price had conducted 25 years of animal research on the causes and consequences of tooth decay as the director of the American Dental Association’s Research Institute, with a team of sixty scientists working under him and an advisory board of eighteen of the leading scientists in a variety of disciplines, including Victor Vaughn, then president of the American Medical Association, and Charles Mayo, a founder of the Mayo Clinic. One of Price’s impressions from that research experience was that susceptibility to tooth decay resulted from the absence of protective factors, but he didn’t know what they were. The research he later published in Nutrition and Physical Degeneration (NAPD) had as its purpose the identification of those protective factors.
Price understood the need for scientific controls, and Americans without tooth decay at that time were a rarity. He also wanted to establish “standards of excellence” to see just what sort of freedom from physical degeneration was possible. Emphasizing the need for scientific controls, he set out not simply to observe foreign people free of tooth decay, but rather to observe people free of tooth decay who could be closely matched for heredity, latitude, altitude and climate with other people suffering from tooth decay. Thus, he looked specifically for groups of homogeneous genetic stock on the cusp of modernization to serve as a series of what we could call “natural experiments.”
Price described his search in this way (NAPD, p. 472):
A comprehensive study of modern degeneration will require the use of controls in order that standards of excellence may be established. Not finding adequate controls in our affected groups, it became necessary to look elsewhere in Nature’s great biological laboratory, which has been in operation throughout the history of life.
Price sought out “primitive racial stocks,” by which I believe he meant genetically homogeneous groups that were “primitive” both in the sense that they hadn’t yet modernized and also in that they hadn’t mixed with other races, since a common theory at the time was that dental deformities were caused by race mixing. Wherever he could, he studied these groups on the cusp of modernization so that some subgroups could be found who had remained “isolated” from modernizing forces and others could be found who had already modernized or who were in the process of modernizing. Price performed dental examinations, took thousands of pictures, collected dietary and other ethnographic information, and collected samples of food, soil, and saliva for laboratory analysis. He then used this data to make comparisons between the isolated and modernized subgroups of each “primitive racial stock.” By comparing the isolated Swiss to the modernized Swiss, and the isolated Gaelics to the modernized Gaelics, and so on, Price made comparisons wherein both isolated and modernized subgroups of each racial stock had similar genetics, a similar cultural history, and lived in similar climates and at similar altitudes and latitudes.
Presumably, within each community we could find a random distribution of potentially confounding risk factors. Differences in diet and lifestyle between individuals would be small in these groups, but probably would still exist. Differences in genetics and other hereditary factors would undoubtedly exist. For these reasons, some individuals would be at a higher risk for tooth decay or other degenerative diseases than others. Although the distributions are unlikely to be identical from, for example, one Swiss village to the next, they are likely to be somewhat similar.
In order to make a strong inference about cause and effect when we observe a difference between two groups, we want the distribution of confounding factors to be as close to identical as possible between the two groups. The only reliable way to do this is to randomly distribute participants into each group. Even randomization, however, is no guarantee. As I’ve discussed previously, the LA Veterans Administration Hospital Study randomly allocated just under 850 men to two different groups to study the effect of dietary fat composition on heart disease, yet one group wound up with far more moderate and heavy smokers than the other. The number of potentially confounding factors probably greatly exceeds the number we can currently imagine, so there were probably other confounding factors that were unevenly distributed in that study and in any other study of similar size.
Randomized controlled trials have other limitations as well: they are usually far too short to understand the long-term effects of anything, and they are always conducted in a dramatically different context than the one we would call “real life.” At a minimum, people are being monitored and told what to do, which puts them in a different category than people acting spontaneously, they are likely undergoing blood draws, which in and of itself affects a person’s biology (most obviously by removing iron and red blood cells), and in extreme cases they may be locked in metabolic wards under close surveillance with even their physical activity controlled.
The strength of observational studies is that the context is much closer to the day-to-day reality of most people, while their weakness is their lack of control for factors that confound our ability to infer cause-and-effect relationships. The closer we observe a system, and the more rigorously we control it, the more we distort it. And, the more we know how that distorted system works. Thus, any piece of evidence falls somewhere on the following spectrum:
Yet the trade-off isn’t a zero-sum game. Scientists usually work with highly distorted yet very knowable experimental systems, but always search for ways to maximize the relevance of these systems so that their amenability to being studied isn’t all for naught.
Price’s study design targeted the best of both worlds. It was observational, and firmly rooted in real-life context, free of experimental controls. At the same time, he exploited an opportunity that was not entirely but largely unique to the brief window of time open to him, wherein so many different groups were on the cusp of modernization all over the globe. This was the opportunity to make repeated observations in many different geographical, cultural, and genetic contexts the world over, wherein he could compare isolated and modernized groups with as similar distributions of confounding factors as possible. While this is not, overall, the most effective way to remove factors that confound inferences of cause and effect, it is the most effective way to do so in an entirely natural and spontaneous context. Few other studies could get the best of both worlds in this way, and probably none has ever done so on such large a scale.
In this sense, Price’s study design was to observe the closest thing to a randomized controlled trial that nature has to offer.
As most readers of this blog already know, Price found that the most consistent change that occurred with modernization was the replacement of traditional diets with the “displacing foods of modern commerce.” These foods, according to Price, included white sugar, white flour, white rice, syrups, jams, canned goods, and vegetable oils. Price verified in the laboratory that the nutrient contents of soil and individual foods were also much higher among the “primitive” groups, and that their saliva had different properties that appeared to protect against tooth decay. In every instance, modernization was associated with a proliferation of tooth decay in the first generation and dental deformities in the second. Price also provided less rigorous but nevertheless fascinating evidence suggesting that with tooth decay and dental deformities came an onslaught of other degenerative diseases. Attributing the disease to the dietary changes isn’t completely free of confounding, but the finding was robust across so many cultures and genetic stocks in so many different latitudes, altitudes, and even continents, that the risk of confounding is low.
Price didn’t end there. He tested his nutritional theory of tooth decay by reversing tooth decay among his patients with a nutrient-dense therapeutic diet, and conducted animal studies showing the effect of nutrition on tooth decay. He also made copious use of the nutritional literature available at the time to explain these findings.
Before we delve into that aspect of Price’s work, however, I’d like to spend one more post on the importance of making comparisons within each “primitive racial stock” rather than making comparisons between wildly different groups. It can be tempting for us, for example, to compare the Inuit to the Swiss and make inferences about whether one or the other diet is superior. In the next post, I’ll take a look at specific adaptations to vitamin D metabolism that appear to exist in the Inuit to show why we have to resist this temptation and stick to Price’s primary study design when interpreting his work.
Read more about the author, Chris Masterjohn, PhD, here.