I recently read a commentary  by Dominic Lawson, in the Times of London Sunday edition, entitled “Organic Food is Just a Tax on the Gullible.He was referring to  a couple of systematic reviews (health benefits and nutritional content) that suggest there are no measurable health benefits an no meaningful nutritional superiority associated with organically produced food .  The studies and the commentary add some useful detail to my previous comments about nutrition myths, and they illustrate nicely the seldom-appreciated complexity of making sound decisions about producing food, whether for animals or humans.

Now I’ll admit to a certain sympathy with organic food production, particularly as it seems likely to be less harmful to the environment than conventional industrial agriculture. And I have no doubt that the economic incentives of the agriculture industry lead to just as much malfeasance as those in the pharmaceutical game, or really any other area of economic activity. I believe we now produce more food, of better quality, more efficiently in terms of calories per acre that ever before, and that is itself a good thing, at least for our health as individuals and as a species. However, this increased production comes at a cost, and the use of fossil fuels and pesticides has consequences we must be mindful of.

Nevertheless, I believe in applying the same standards of evidence to all scientific claims, including those I personally find appealing. And I have been skeptical in the past of the nutritional and health claims of organic food proponents, which seem to be based more of the naturalistic fallacy  than on any evidence I’m aware of. A reader here suggested I had missed this evidence: “there are increasing numbers of studies documenting that (certified) organic food is more nutritious than conventional (non-certified organic food). They find significantly more minerals, vitamins, antioxidants in most all foods researcher look at.” Unfortunately, the person never responded to my subsequent request for details on these studies.

The first of the reviews Lawson comments on examined the literature from 1958-2008 and found only 11 studies that addressed the question of health benefits from organically produced foods or specific food ingredients. Of these, only 3 studies met the a priori defined criteria for methodological quality. The authors concluded, ” because of the limited and highly variable data available, and concerns over the reliability of some reported findings, there is currently no evidence of a health benefit from consuming organic compared to conventionally produced foodstuffs.” Clearly, this study does not definitively show that there are no health benefits to organically produced foods. But it does show that there is no high quality evidence for such benefits in the literature, which leaves the burden of proof squarely on those who make claims for the health benefits of eating organic foods.

The second study reviewed the literature for the same period, and did slightly better in terms of finding high quality studies. Of over 3000 studies examined, 55 met the quality standards. Based on these, the authors concluded:

“No evidence of a difference in content of nutrients and other substances between organically and conventionally produced crops and livestock products was detected for the majority of nutrients assessed in this review suggesting that organically and conventionally produced crops and livestock products are broadly comparable in their nutrient content. The differences detected in content of nutrients and other substances between organically and conventionally produced crops and livestock products are biologically plausible and most likely relate to differences in crop or animal management, and soil quality. It should be noted that these conclusions relate to the evidence base currently available, which contains limitations in the design and in the comparability of studies. There is no good evidence that increased dietary intake, of the nutrients identified in this review to be present in larger amounts in organically than in conventionally produced crops and livestock products, would be of benefit to individuals consuming a normal varied diet, and it is therefore unlikely that these differences in nutrient content are relevant to consumer health.”

These two reviews have been criticized, not surprisingly, by organic food producers, primarily for their exclusion of studies which support the producers’ claims. Ben Goldacre has already responded to the substance of these criticisms. Clearly, these two reviews cannot not be the final word on the subject. They strongly suggest that when the scientific literature does not support the claims that organically produced food is healthier or more nutritious than conventionally produced foods, but they also show that the evidence is sparse and of poor quality, so we must keep an open mind on the subject. And, of course, none of this bears on the other potential benefits of organic agriculture, including decreased environmental harm and subjective benefits such as better tasting food. These are themselves issues which ought to be investigated in a rigorous way.

Finally, Lawson’s commentary made what I thought was a critical point, nicely illustrate by an anecdote that is humorous, rather than tragic, only because it ended happily. I’ll let him tell it:

“A few years ago my wife decided we should have an entirely organic vegetable garden. To this end she refused all man-made fertilisers [sic] and ordered a truckload of pigeon droppings. What could be more natural? Neither was there anything unnatural in the germs I inhaled through the spores of our organic manure, thereby contracting psittacosis. This developed into “atypical” pneumonia, which was of course resistant to all standard antibiotics. Had a hospital doctor not guessed the cause and put me on a drip with the appropriate drugs – ooh, chemicals! – I could have become a fatal casualty of the organic movement. Obviously my wife might have ordered cow manure rather than pigeon poo[p]; then I could have been felled by E coli instead.”

This is a powerful illustration of a concept promoters of CAM often ignore; every choice has an array of costs and benefits that have to be evaluated and balanced. Vaccines do sometimes cause harm, but to a far lesser degree than the harm of not using them appropriately. And there may be benefits to reducing the use of chemical fertilizers, but one cannot just blithely follow the naturalistic fallacy and assume that organic fertilizers are always better. This sort of cost benefit analysis is always difficult, tedious, and fraught with uncertainty, and it lacks the appeal of simplicity, but it is a crucial part of making the best decisions about medicine, agriculture, or any other complex endeavor.

I’ve previously written an analysis of one of the most popular veterinary nutrition myths, the idea that cats and dogs should eat raw food. I was recently asked by one of my colleagues to address a couple of other veterinary nutrition myths, and by a happy coincidence Dr. David Dzanis, a board-certified veterinary nutritionist and consultant, gave a brief talk on the subject at the American Veterinary Medical Association conference from which I just returned. While I hope to someday put together a more comprehensive and thoroughly referenced list of such myths, I thought I’d share a few tidbits I learned from Dr. Dzanis, and a few other sources, on the subject of pet nutrition.

“Good” vs “Bad” Ingredients

Popular mythology assigns food ingredients statues as “good” or “bad,” ill-defined categories which are seldom based on any scientific facts but which have a powerful impact on the purchasing decisions of the pet-owning public. Unfortunately, pet food manufacturers play along and exacerbate this mythologizing of certain food ingredients for marketing advantage. When the public begins to believe a common ingredient is harmful, manufacturers will sometimes begin proudly announcing their foods are free from the ingredient, and this will appear to be confirmation of the myth. The chemical preservative ethoxyquin is a perfect example. While there is no evidence this ingredient causes any harm, and it is clear that spoilage of “preservative free” food can pose a real health risk, the ingredient is now rarely used because uninformed and unscientific public opinion demanded its removal from pet foods. A similar sort of process lead to the removal of thimersal from human vaccines despite the clear science refuting the public concerns about this preservative.

Often the growth of a movement against a food ingredient  is a purely aesthetic issue with no scientific merit. Eating chicken feet and viscera, unborn calf fetuses, and so on sounds gross, so people assume these aren’t healthy pet food ingredients. But the aesthetic standards of cats who like to chew on dead lizards and dogs who eat socks, rubber, and poop are different from ours, so it’s a meaningless way to judge such food constituents.

Such food ingredients are also disdained because of the confusion in many people’s minds of nutrients and foods. People will claim that “sea salt” is somehow different from mined salt or that wheat proteins are somehow less nutritious than beef proteins. The fact is that, properly prepared, many things we would not consider appropriate as foods in their original state can provide vital nutrients of the same quality, or better, as more aesthetically pleasing sources.

It is particularly common these days for people to claim that grains are “bad,” and that wheat and corn in particular are harmful or “allergenic” for our pets. As Dr. Dzanis puts it, “Corn and wheat are often criticized as inferior to other grains such as rice. While rice is generally more digestible than corn, that also means the starches are reduced to sugars and absorbed much more quickly. This may not be desirable for animals with clinical problems related to blood glucose control. As far as potential allergenicity, historical use of rice in therapeutic diets as a “novel ingredient” for diagnosis or control of food allergies may have led to a false impression. In cases where the cause of a food allergy has been determined, the incidence of allergy to corn is equivalent to the incidence of allergy to rice.”

The popular suspicion of wheat is likely related, to some extent, to the incident in which a Chinese firm supplying many American pet food manufacturers with wheat gluten adulterated their product with melamine, leading to kidney failure, sometimes fatal, in pets who consumed the tainted food. While this is a tragic and infuriating example of venality and inadequate regulation in China, it has nothing to do with the appropriateness of wheat as a pet food ingredient. And while a few breeds, such as the Irish Setter, have genetic gluten sensitivity, in general wheat is a healthy and nutritious ingredient perfectly appropriate as an energy and protein source in pet food.

Corn has also gotten a sinister reputation from popular author Michael Pollan. While I enjoy, and agree with, the balance of Mr. Pollan’s writing, his indictment of the corn industry and the use of corn-derivatives in food products can easily encourage an irrational and hysterical assessment that corn is somehow poisonous and nutritionally vacuous, which is simply not true. Corn and wheat proteins are common allergens for dogs and cats not because they are especially “allergenic” but simply because they are common proteins in pet food. As they are replaced, under the misguided pressure of public mythology, with rice and soy and so on, these newer ingredients will become the predominant allergens in pets who develop hypersensitivities or true allergies to commercial foods.

Hidden Food Ingredients

Pet food manufacturers are often accused of slipping roadkill, old shoes, and all sorts of other bizarre and disagreeable things into commercial pet food. While these companies exist to make a profit and certainly do what they can do encourage the public to view their products favorably, even if this involves a little slight-of-hand when it comes to describing the content of their foods, the reality is that they are legally required to identify any ingredient they use, and there is no evidence that they ignore this requirement in any routine or egregious way. Which brings me to one off the most shocking, and farfetched, concerns about pet food:

Dead Pets in Pet Food

Soylent Green is….Rover? Probably not. Promoters of this story take a few facts and weave them into an unlikely, but shocking narrative. It is true that in some parts of the country, euthanized dogs and cats are disposed of by rendering, a process which breaks whole carcasses down into potentially useful constituents. This is usually done by commercial services, and at facilities, not associated with slaughter and rendering of the agricultural animals generally used as pet food ingredient sources. however, the practice of using rendered cattle as an ingredient in cattle feed, and the subsequent epidemic of Bovine Spongiform Encephalopathy (BSE, aka “Mad Cow Disease) illustrates that there are potential risks to such practices. However, from a purely economic point of view, using rendered pets as an ingredient in pet food makes no sense. Additionally, the Pet Food Institute, an industry lobby which represents the manufacturers of  ~98% of commercial pet foods, specifically prohibits rendered pet ingredients in their members’ products.

FDA studies in 1998-2000 did find miniscule quantities of pentobarbital (2-32 parts per billion) in many commercial pet foods. This is an anesthetic often used to euthanize animals, and the finding provided some ammunition for those who claim dead pets have made their way into pet foods.

However, the same studies did not find any cat or dog DNA in the proteins from any of the tested foods. While the source of the pentobarbital was not identified, the best guess is that it comes from small numbers of cattle or possibly horses euthanized and then rendered and used in pet foods. In any case, follow up studies estimated the minimum amount of pentobarbital which has any measurable physiological effect and found that even the smallest dog eating large amounts of the food with the most pentobarbital could not get to this dose. So while the FDA cannot guarantee, that no rendered pet material ever makes it into any pet food, it is highly unlikely, and there is no evidence that even if this were the source of the anesthetic detected that there is any health risk associated with such miniscule amounts of the contaminate.

People concerned about such “toxins,” whether in food or vaccines, often fail to understand the concept of dose-dependent toxicity. As I’ve pointed out before, water and oxygen can kill in sufficient doses, despite being vital for life. And even an anesthetic which, when given as an overdose can kill, can be harmless in quantities measured in parts per billion.

The Role of AAFCO (Association of American Feed Control Officials)

AAFCO is a private organization that establishes non-binding guidelines for the production of animal feeds, including pet foods. While it is not a government agency, only government officials can be members. Most are from the U.S. but there are members from the FDA/USDA equivalents in other countries. It is often labeled, by promoters of raw or home-cooked pet diets, as an organ or lackey of the pet food industry. Industry and private groups can attend AAFCO meetings and contribute advice or information to task forces or working groups, but they cannot vote. Such groups have included pet food manufacturers, but also veterinary groups such as the Academy of Veterinary Nutrition, the American Veterinary Medical Association, and the American Animal Hospital Association, as well as advocacy groups such as People for the Ethical Treatment of Animals, Defend Our Pets, and others.

AAFCO creates guidelines and models but has no regulatory authority. However, their ingredient definitions and nutritional standards are often adopted by the FDA in their regulation of pet foods.  A food can be certified as meeting AAFCO standards based on nutrient analysis done by the manufacturer. Or it can be certified as “feeding trial tested” based on trials conducted by the manufacturer. These often last 10weeks (for growth diets) to 6 months (for maintenance diets) and involve regular clinical and laboratory monitoring. Unfortunately, a food can also be certified as “feeding trial tested” without an actual feeding trial if it is ruled by FDA substantively similar (in the same “product family”) to a food that has undergone feeding trial testing. Clearly,, this is less than ideal but without the political will to fund government testing of all pet foods marketed, these standards at least ensure a minimum level of adequacy, and they are certainly preferable to the complete lack of standards that apply to most home-cooked diets or those marketed outside the official regulatory system. Which leads me to the last category of veterinary nutritional mythology:

“Natural” and “Organic” Foods

By legal definition, to be “natural” an ingredient must come from an animal, plant, or mined source and must be minimally processed (it can be heated, fermented, and so on). Most vitamins added to foods are synthetic and do not meet this requirement. However, to be legally marketed, “natural” foods must still conform to the nutritional standards established by AAFCO and regulated by the FDA. There is zero evidence, of course, that this designation has anything to do with the safety or nutritional value of a food/food ingredient. It’s really a marketing issue, not a scientific one. The popularity of this designation is an example of the “naturalistic fallacy,” the idea that “natural” is synonymous with “good” or “healthy.” Salmonella, hydatid cyst disease, poison oak, and gamma radiation are all “natural,” while most antibiotics, antiparasitics, anti-inflammatories, and cancer treatments aren’t, but if I am unlucky to run into these “natural” things I’d sure like to have the “unnatural” ones around .

“Organic” foods and food ingredients, similarly, have a specific legal definition, enforced by USDA not FDA, that has detailed requirements for how a food ingredient can be grown and processed. The best way to ensure something is truly “organic” is too look for the USDA Organic seal. There is no evidence that food produced by these standards is superior in terms of health and nutrition. There is, however, good evidence that organic production methods have less harmful impact on the environment than more common industrial methods, so there may be some value to the designation despite its unfortunate affiliation with the mythology of the naturalistic fallacy. 

References

Comprehensive proceedings from the 2009 Hill’s Symposium on Nutrition Myths and Truths, Facts and Fallacies. Cats and Carbohydrates- What are the Concerns.

Cowell CS, Stout NP, Brinkmann MF, et al. Making commercial pet foods (Ingredient myths and facts). In: Hand MS, Thatcher CD, Remillard RL, et al. P (eds). Small Animal Clinical Nutrition, 4th

Edition. Topeka, KS: Mark Morris Institute, 2000; p. 141.

Food and Drug Administration/Center for Veterinary Medicine: Report on the risk from pentobarbital in dog food [Online]. Available: http://www.fda.gov/cvm/FOI/DFreport.htm. 31 March 2009.