Sadly, You Should Ignore Non-Peer Review Nutritional Opinions

I was genuinely disappointed today with what I thought were some reliable sources of nutritional information in the blogosphere. The tipping point came when I was reading an article from (self-titled) "Authority Nutrition" against the adoption of a vegan diet. The first argument was a deficiency argument and the claim was that:

...B12 deficiency is very common in vegans, one study showing that a whopping 92% of vegans are deficient in this critical nutrient (1).

 I admit that, in the past, I have had a poor track record of actually checking references. I guess I was just too trustworthy, but because of all my recent pub-med crawling, I thought I would click on the reference. And I was sad. The claim made by Authority Nutrition only resembled what the reference said and a clearly crucial fact had been omitted. What the reference actually says in the abstract (in other words, all I had to do was click on the reference, without even reading the article):

Among subjects who did not supplement their diets with vitamin B12 or multiple vitamin tablets, 92% of the vegans (total vegetarians), 64% of the lactovegetarians, 47% of the lacto-ovovegetarians and 20% of the semivegetarians had serum vitamin B12 levels < 200pg/ml (normal = 200–900 pg/ml). However, their complete blood count values did not deviate greatly from those found for nonvegetarians, even though some had been vegans or lactovegetarians for over 10 years.

It is surely quite a different figure when those bolded statements are made clear. I am far from a representative socialite, but among vegans that I know, all of them supplement with vitamin B12 in some form. I am not sure whether it is more charitable to impute ignorance, carelessness or willful manipulation of the reader to the writer of the article, but it does not matter.

The unfortunate truth is that nutritional "facts" from these sorts of sources, as it turns out, are almost always misleading at minimum and plainly false in the worst of cases. Even the ones that claim to be based on facts, from people qualified to give advice on the area, can be wrong or misled; the diet advice you can receive on a popular platform in a small readable chunk will almost invariably lack the nuance of the original research and will equally invariably over-generalise. Studies are often not even done on humans, or are done on particular segments of the population in particular controlled environments. There are studies done, in fact, that demonstrate how poorly the results of other studies generalise to other portions of the population.

Even more unfortunate is the fact that the nuance and precision employed in the scientific literature often makes reading that literature inaccessible to the layperson. It is certainly possible to pick up the technical know-how to read and research the literature by yourself - but it takes a lot of practice to even break into one small subfield of the literature. Trying to get a good grip on a single topic is an excellent exercise if only to understand the complexity of even simple nutritional issues and why the researchers can do this as a full time job (without any of the glamour of a popular nutrition personality). Even after becoming proficient, there are few truths in nutrition that manage to actually transcend the bounds of the demographic studied to the general population, and fewer still that can be conclusively tied to some reductionistic element of a food. For example, the blood sugar effects of different foods cannot be tied to a single macronutrient (this video explains the research on berries and blood sugar). This implies that reductionistic accounts of nutrition of the ubiquitous "this was shown to be good so eat these other foods containing the compound" fail to account for the richness and complexity of foods as whole substances. The ineffectiveness of supplemental antioxidants compared to food-derived antioxidants is further evidence of this.

I sadly have to admit to myself that, not only is most of my popular level nutrition knowledge probably vastly over-simplified to the point of uselessness, but I am also unable to learn by going to popular sources. They are too unreliable and too simple. If I want to know what to eat - and I do! - I have to do my own homework. I may well post some of my findings here, but if you believe anything I have said, you will have to check my references for yourself.