Dec 31 2012

Nutrigenomics – Personalized Pseudoscience

I wrote last week about the problem of stem-cell quackery throughout the world, mostly in poorly regulated countries but with the purpose of attracting international customers. Stem cells are real, and the science of developing medical applications of stem cells is both real and promising, but these stem cell clinics are making claims that are years or decades ahead of the science. They are capitalizing on stem cell hype as a marketing ploy to those who are more desperate than scientifically savvy.

I was asked to comment on yet another example of the same phenomenon – nutrigenomics. That’s a very impressive-sounding name, just like a real science, but as always the devil is in the details. The claim is that by analyzing one’s genes a personalized regimen of specific nutrients can be developed to help their gene’s function at optimal efficiency. One website that promises, “Genetics Based Integrative Medicine” contain this statement:

Nutrigenomics seeks to unravel these medical mysteries by providing personalized genetics-based treatment. Even so, it will take decades to confirm what we already understand; that replacing specific nutrients and/or chemicals in existing pathways allows more efficient gene expression, particularly with genetic vulnerabilities and mutations.

The money-quote is the phrase, ” it will take decades to confirm what we already understand.” This is the essence of pseudoscience – using science to confirm what one already “knows.” This has it backwards, of course. Science is not use to “confirm” but to determine if a hypothesis is true or not.

As with the stem-cell quackery, this idea (of studying one’s genes in order to personalize therapy) is not itself pseudoscience. It is, in fact, an area of legitimate research. We already use genetic analysis to diagnose certain diseases, and to target chemotherapy. We are beginning to identify specific genes that affect how different individuals metabolize and respond to specific drugs. While genes are not necessarily destiny, our genes do exert a powerful influence over our health. They are already an important part of science-based medicine.

Further, as genetic analysis become more rapid and cost effective, there is the increasing potential that it can be used as part of a routine screening health evaluation in order to identify susceptibilities, target preventive treatments, adjust behaviors to target risks, and guide therapy. However, similar to stem cell treatments, our current knowledge base with respect to genetic predispositions is still in its infancy. What is well established is already incorporated into mainstream medical practice. The rest is a matter for research, not current practice.

This creates an opportunity for exploitation, however – using current cutting edge research to make clinical claims that are years or decades premature by pretending to have knowledge that simply does not exist. This type of medical pseudoscience is also a manifestation of one common tactic among dubious practitioners – basing clinical claims on pre-clinical scientific research. This is especially insidious and difficult for the non-expert to properly evaluate (which makes for effective pseudoscientific marketing).

A November 2012 review of the field of nutrigenomics states:

It has, however, proven very challenging to define an individual’s responsiveness to complex diets based on common genetic variations. In addition, there is a limited understanding of what constitutes an optimal response because we lack key health biomarkers and signatures.

In other words – the relationship between nutrition and personal health is complex and we do not yet have the scientific knowledge to apply to practice.  The article concludes that, while promising, the field is not yet ready for the marketplace.

There is a great deal of basic science research going on (test tube and petri dish type research), which asks questions about how the body works and how it is affected by all conceivable factors. There is also translational or clinical research which seeks to apply this knowledge to specific medical interventions. Biology is horrifically complex, however, and so it is extremely difficult to extrapolate from basic science knowledge to net clinical effects. Most guesses that derive from basic science research turn out to be wrong. Sometimes our guesses are even the opposite of what we predicted – interventions cause harm rather than improvement.  The only way to know is to conduct careful rigorous clinical research to measure the actual effects (good and bad) of a specific intervention in a specific population.

Because of the vast reservoir of published basic-science research, however, it is possible to find studies that seem to support almost any conceivable intervention you wish. To the public this can make any intervention seem as if it is science-based and legitimate, even when the treatment is nothing but deception.

The nutrigenomics website, for example, claims to treat the following conditions:

Welcome to Genetics Based Integrative Medicine (GBIM), a telemedicine practice dedicated to the education, treatment, and recovery of those with autism spectrum disorders, ADD/ADHD, and PANDAS as well as highly complex & disabling disorders affecting adults such as CFS/ME/FM, Multiple Sclerosis, ALS, Parkinson’s and mitochondrial dysfunction.

There is no compelling evidence for any nutritional treatment for the above diseases, let alone for personalized nutritional treatment based on specific genetic types. How the practitioners of GBIM came by the knowledge they are claiming to have is a mystery. As with the stem cell treatments I discussed previously, such clinical claims, if legitimate, would have a paper trail of hundreds of published studies in the literature. Further, if such studies existed such practice would be standard of care, not isolated to one or a few special clinics.

I would add “nutrigenomics” to the list of red flags for dangerous quackery.  It is a shame because, like stem cells, it is a legitimate field of research, and the current quackery is likely to taint the reputation of what in the future might be a promising approach.

7 responses so far

7 thoughts on “Nutrigenomics – Personalized Pseudoscience”

  1. tmac57 says:

    I’m not sure what their angle is here.Is it the testing,supplements,consultations…or what? The genetic test is about $500,but I have no idea if that is fair for that type of test.
    They seem to be just in that sweet spot of plausible sounding (if you are not a skeptic) sciencey sounding stuff,with a bit of toxins and supplements thrown in.These kind of sites are so common now that I wonder if they aren’t cannibalizing each other since they all have different answers for the same thing.

  2. Yes nutrigenomics is a legitimate field of research, has been for years ( The stuff you highlight about autism etc. is not legitimate though, there is no validity to the claims made and the quote “it will take decades to confirm what we already understand” is amazing, it just doesn’t make any sense – stupid!

    But you also need to be careful about implying that there is not any use at all, now, for nutrigenomics and personalising nutrition based on genotype – especially as you only cite one review. Yes nutrition is complex, individual SNPs have small overall effects on long term disease risk but… individual nutrients also have only small effects on overall disease risk, so what’s the point of any nutritional advice??

    BTW – in the review you cite in the final paragraph it says:

    In conclusion, most impact on health and disease prevention
    could be achieved in the short term if personalised
    nutrition focused on helping individuals to make healthy
    dietary choices from existing natural products. This should,
    for the moment, be based on population dietary advice but
    where possible, also take into account environmental
    factors and individual genotypes and metabotypes, where
    scientifically justified.

    So what you say is not correct (“In other words – the relationship between nutrition and personal health is complex and we do not yet have the scientific knowledge to apply to practice. The article concludes that, while promising, the field is not yet ready for the marketplace.”)

    The full article itself does not say that we don’t have the knowledge and does not say it is not ready – the abstract mentions commercial viability related to several things, that’s different

    Using it to cure autism is not scientifically justified – but there are some companies, a few serious companies, that are using it, in a less dramatic way, but in a way that is scientifically justified.

    The latter is one aspect we are working on making more transparent in the EU project

    see also

  3. Kieth – I disagree with your interpretation of the current state of affairs, but I am willing to change my mind if you can provide published studies that demonstrate that specific genetic types have different clinical outcomes from targeted nutritional therapy.

  4. There are quite a few validated gene x diet interactions. The best known I suppose are

    MTHFR x folic acid x homocysteine

    CYP1A2 x caffeine x heart disease risk

    GSTs x cruciferous x DNA damage and some cancers

    SOD2 x antioxidants x breast & prostate cancer

    The main question though is utility – is it useful to have low homocysteine? Does the CYP1A2 x caffeine data mean that heart attacks can be prevented? These are difficult questions to answer, they never will be with standard RCTs (eg see here:

    But can the genetic information be used to supplement the standard guidelines (none of which have be, can be or will be “proven” by RCT)? That’s where the debate lies – and as with standard nutrition, there are many points of view

  5. Factoidjunkie says:

    I agree with Steven on the current status. It does seem that claims are way ahead of evidence at this stage of investigation. The greatest difficulty it seems to me about individual nutrition is the high number of variables and the dynamics of health. Enhancing a system of this complexity will require a refinement in monitoring that does not yet exist.

  6. Aardwark says:

    I also agree with Dr Novella, but I would like to highlight what I feel could be a very important distinction. Namely, there is pseudoscience and there is pseudoscience, if one might say so.

    What I mean is that it is important to distinguish between two things. Firstly, there is ‘pure’ pseudoscience (or shall I say PSEUDOscience, with strong emphasis on the ‘pseudo’ part) such as, for instance, homeopathy, or the ‘four humors’ type of diagnostic/treatment systems so delightfully dissected on SBM by David Gorski. To this category fall all attempts at giving medical advice based on principles that are – as a matter of principle – not correct, in that they do not correspond to any reality, or at the very least, there is no methodologically acceptably gathered data to support the conclusion that they do.

    Secondly, there is the type of pseudoscience that this post, as well as the previous one on stem cell treatments, deals with: ‘bogus’ (or, in a more PC speech, unsubstantiated) claims concerning one or more particular diagnostic/therapeutic applications of what otherwise may be principles that are themselves actually – again as a matter of principle – validated by lots of scientific research. I propose to call this ‘type II’ pseudoscience or pseudoSCIENCE, since the emphasis lies on the ‘science’ bit.

    This is, in a way, like reading a document in, say, German or Chinese to people who do not speak those languages and replacing the contents with whatever suits one to invent.

    My motivation to strongly insist upon the above distinction lies in the fact that, in the latter case, smashing the ‘pseudo’ part should be somehow (and this is, admittedly, very difficult) guarded from having a negative bearing on the ‘science’ part – for German and Chinese are, after all, real languages, and there is nothing wrong with the idea of reading and translating documents. Just as there is absolutely nothing wrong with those of nutrigenomics and stem cell therapy.

    Therefore I propose that, while debunking pseudoscience, we should all be very careful to distinguish whether we are dealing with something that is incorrect in principle (‘type I’ pseudoscience) or a piece that merely abuses real scientific knowledge to promote dishonest (or at least illusory) practice – ‘type II’ pseudoscience. In the two cases, the counter-tactics has to be quite different. In the former case, it is most important to promote scientific literacy (i.e. to point to the error of principle involved), while in the latter one must start from the ‘common ground’ (such as noting that nutrigenomics may, in fact, work, but not against autism right now and not in the way it is being claimed).

    Another vital point to make is that ‘type II’ pseudoscience may infect and threaten to discredit any part of science-based medicine. For instance, I would certainly place the frightfully common practice of prescribing antibiotics (i.e. antibacterial drugs) for obviously viral infections in this category as well, since it is certainly no more ‘scientifically based’ than the examples here discussed.

  7. tmac57 says:

    Aardwark- I like your language analogy,but I would like to extend it a bit.
    Nutrigenomics and stem cell therapy are applications of a ‘language’ that has neither been completely deciphered,nor is fluently spoken or understood by anyone yet.
    Scientists are still working out the ‘syntax’, ‘morphology’,’phonology’,and ‘semantics’, while the pseudoscientists pretend that they not only have the dictionary,but also encyclopedias and novels written in those yet un-deciphered languages.

Leave a Reply