DNA testing is a recurrent topic on this blog, driven mainly by the significant technology advances that makes getting your DNA sequenced and analyzed both simple (usually a mouth swab) and inexpensive. Very quickly, this science has moved from the bench, to the bedside, and now to online test kits sold directly to consumers. What’s been missing in this weakly-regulated space is evidence that the information and advice derived from these tests is both reliable (i.e., you consistently get the same results) and valid (i.e., that the actions recommended are backed by robust evidence). An area that seems to have exploded in popularity is “DNA-based” diet advice, where your DNA is used to infer recommendations about your diet and to even recommend specific supplements. The idea of “personalized nutrition” sounds enticing (aren’t we all special snowflakes?), but as in many areas of science that are evolving quickly, the hype evolves even faster, outpacing the evidence. I was recently asked about a few products that have all appeared on the market over the past few years, each with impressive websites and lots of sciencey terms.

3D-printed vitamins?

Population-level evidence is convincing that there is no clear rationale to supplement with vitamins in the absence of a deficiency or demonstrated medical need. But is possible that benefits could be realized based on customized formulations defined by perceived need? The company Nourished promotes customized “3D-printed” “stacks” of vitamins that, while not strictly based on your DNA, are based on questions you answer. For context, I’m healthy, I exercise regularly, and I eat a well-balanced diet. I enter some demographic information, answered a few questions (How often do you eat vegetables? How often do you travel internationally?) and it suggested I need the following in my daily supplement – which Nourished will custom manufacture and ship to me monthly.

Each product has its own rationale included, none of which included citations or evidence to demonstrate that supplementation was necessary or beneficial, given my circumstances. The strangest one to me was the suggestion I need 1600mg of Vitamin A per day, presumably because I reported that I wear glasses. Not only is there no rationale to take vitamin supplements if you have myopia, like me, there’s pretty much no need for anyone in a resource-rich country (like me, in Canada) to take vitamin A – deficiencies are that rare.

Another company (Pure Genomics) goes a step further, using your results from 23andMe or testing to offer up specific vitamin supplement recommendations. There is no published evidence to show that specific vitamin supplements are warranted based on this type of testing, but this doesn’t prevent an array of recommendations based on your results. And when a company offers supplements that claim to help “detoxification” then it’s a big red flag for the scientific rigor of the information and advice presented. Unless you have known organ failure, your kidney and liver are “detoxing” you just fine, and you don’t need any supplements to help them.

There was also a company that makes wine recommendations based on your DNA. That may have tested the limits of consumer credulity, and it looks like they’re no longer in operation. (Besides, I don’t need your DNA results to suggest that you’ll probably like Beaujolais).

What about eating aligned with your microbiome?

I admit that this one surprised me when I saw it, but there is apparently a market for “personalized nutrition” using “artificial intelligence” based on a stool sample. Viome claims that with an examination of your microbiome this information can be used to identify your “superfoods”, your foods to minimize, or enjoy, and what supplements you should take. Dr. Mark Crislip doesn’t blog here anymore but his posts on the microbiome are all in the archive and I strongly recommend them as background on this topic. My take on the topic is that the scientific understanding of the microbiome is progressing but still preliminary, particularly nutritional recommendations based on stool analyses. While Viome has raised a significant amount of funding, I couldn’t locate any published information to demonstrate that the Viome test is validated or that the recommendations that it produces result in meaningful, measurable benefits. The website Gut UK says the following:

If the types of bacteria shown in your test are similar to what is commonly seen in thousands of other people’s poo, then congratulations: your poo is just like other people’s poo. Some companies will send you a report showing how your poo compares to thousands of other people’s poo, and also showing the levels of different types of bacteria associated with some diseases. These results are interesting, but do not indicate a particular risk of developing that disease. We have not yet made strong links between the presence or absence of any types of bacteria with many of the “big” diseases such as cancer, heart disease or dementia. So, DON’T PANIC, for instance, if your report says you have low levels of a bacterium associated with preventing Type 2 diabetes. We are not at a stage where we can confidently say that we need to increase levels of that bacterium to reduce your risk of getting Type 2 diabetes. At the moment, we just do not have enough knowledge to safely and accurately recommend any specific interventions (such as particular changes in diet, and/or particular probiotics) to cause beneficial long-term changes to the composition of the bacteria in the gut.

There is also the question of how much value there is in doing a single test taken at a single time point. Many doctors measure markers of disease more than once, just in case the one reading they took was an odd one. Some of these poo tests only become useful when you look at the profile or fingerprint of the bacteria at different times.

Additionally, we do not really know what a ‘healthy’ level of some of these bacteria and chemicals should be in poo, and are only beginning to understand to what degree levels of microbes naturally fluctuate in the gut in a healthy person over time. Until we have done more research, the information given by these tests can be confusing, and often does not help doctors or nurses in guiding their advice to make you healthier or in treating any conditions that you may have. We do know that if you have a well-balanced diet with a good variety of different foods, this usually leads to a wide variety of different bacteria in your large bowel and poo. Many experts feel that this is probably a good thing.

Another good review, in Nature, notes the following:

Basing dietary advice on microbiome studies also assumes that it is possible to predict the health effects of different microbial communities — a goal that is far from being realized. Even well-studied mechanisms have unclear consequences for human health. Take, for example, the production of short-chain fatty acids (SCFAs) from the bacterial digestion of fibre. SCFAs act on multiple tissues and targets of interest in both bacterial and host cells. Furthermore, not all SCFAs are the same, and they have not been considered in the context of products of carbohydrate and amino-acid fermentation — let alone the diversity of other microbial metabolites that could enhance or counteract the effects of these specific compounds. Therefore, although it is tempting to recommend diets that boost SCFAs to prevent metabolic and other diseases, the broader effects on the microbiome and its complex interactions with the host are hard to predict.

There are significant gaps in our knowledge and it’s premature to think that we understand the relationship between the consumption of specific foods, the composition of our microbiome, and health outcomes. But that hasn’t stopped testing vendors from making that leap. Given the state of the evidence, the results can’t be considered reliable or reproducible. Perhaps not surprisingly, one individual tried Viome and a competitor, DayTwo, and found the results and recommendations largely contradictory:

Perusing my list of foods with “Excellent” scores, I noticed some troubling conflicts with the other app. Lentils, which had been a no-no according to Viome, received high marks from DayTwo. Ditto for Kombucha. My purported superfood of cranberry received low marks. Almonds got an almost perfect score (9.7) while Viome told me to minimize them. I found similarly contradictory advice for foods I regularly eat, including navel oranges, peanuts, pork, and beets.

It seems a bit premature for a company to state “Imagine living in a world where illness is optional” when all it has are lots of testimonials yet no convincing scientific evidence that their “DNA-based” recommendations actually deliver meaningful health benefits.

Marketing outpaces the science

Our ability to collect DNA data is progressing quickly, and in time we may have more evidence to inform decisions about nutrition and possibly even supplements. At this time, however, there is a big gap between what we know, and what, if anything, to do with that information. You don’t need a DNA test or a stool sample to know that eating a varied, healthy diet, minimizing highly processed foods, eliminating trans fats, and keeping alcohol consumption moderate are reasonable approaches to designing your diet. It’s worth noting again, as we have blogged about many times before, that the evidence for taking supplementary vitamins, in general, is neutral to negative. In the absence of a specific medical need (e.g., pregnancy) there a few circumstances where routine supplementation is necessary or warranted. There is no robust evidence to date to show that personalized, “DNA-based” or “microbiome-based” nutritional recommendations give useful, actionable nutrition advice that actually improve health outcomes.


Posted by Scott Gavura