Dishing On Diets

Veterinary nutritionists conclude that only they can properly formulate a homemade diet.


Four veterinarians from the School of Veterinary Medicine at the University of California, Davis, published a study in the June issue of the Journal of the American Veterinary Medical Association, entitled “Evaluation of recipes of home-prepared maintenance diets for dogs.” I don’t think any WDJ readers will be surprised to hear that their findings were not exactly positive.


The authors have a vested interest in opposition to any homemade diets prepared without the help of a veterinary nutritionist. Disclosed in the study is the fact that three of the authors are part of the veterinary college’s Nutrition Support Services, and one of those is an owner of DVM Consulting, Inc., the company that manufactures Balance IT, a supplement that is used extensively by veterinary nutritionists to balance limited-ingredient homemade diets.

The study analyzed 200 recipes. Of those, 133 were obtained from 2 veterinary textbooks and 9 pet care books (two of which were also written by veterinarians), and the remaining 67 recipes came from 23 different websites. Of all the recipes analyzed, 129 were written by veterinarians, and 71 by nonvets.

Of the nine non-textbooks examined in the study, there was only one that I would recommend, and it was not written by a vet. (Full disclosure: Another of the books has my name on it, but I don’t defend it. The recipes are not mine, but I did write the introductory text, which includes information about supplements that was not part of the recipes and therefore likely not taken into account when they were analyzed.)

The authors of the study are clear about their goals and expectations, stating, “Current recommendations are that home-prepared diets are best evaluated and formulated by a veterinary nutritionist.” They go on to say, “We believed that most of the recipes would not meet requirements for essential nutrients and that recipes written by nonveterinarians would have a higher number of deficiencies than recipes written by veterinarians. We also expected all recipes to require that at least one assumption would be necessary for preparation of the diet and dietary analysis.”

They found what they were looking for. The study says, “Overall, most (190/200 [95%]) recipes resulted in at least one essential nutrient at concentrations that did not meet NRC [National Research Council] or AAFCO [Association of American Feed Control Officials] guidelines, and many (167 [83.5%]) recipes had multiple deficiencies.” They also found that, “Most (184 [92%]) recipes contained vague or incomplete instructions that necessitated one or more assumptions for the ingredients, method of preparation, or supplement-type products.”

There are many nits I could pick with this study, but bottom line, I think the authors have a point.

Most home-prepared dog food recipes (in books and online) are incomplete, many seriously so. Instructions are often so vague that it’s impossible to determine a recipe’s nutrient content. For example, recipes may include “ground beef” without specifying the percentage of fat. Similarly, many recipes use “chicken,” without denoting dark meat or light meat, or whether skin is removed. When grains are used, some recipes do not state whether the amount given is before or after cooking. These factors greatly affect the caloric and nutritional value of the recipes. And of those that tell you to add supplements, many just say, “add a complete and balanced vitamin-mineral supplement,” with no further guidance. Others may tell you to use a specific amount of bone meal, without recognizing that different brands can vary considerably in how much calcium and phosphorus they contain. And, in general, those are some of the better books – at least they’re telling you to add calcium and supplements. Many recipes don’t include either.

When I analyze home-prepared diet recipes at the request of dog owners, I’m not concerned that the diets meet NRC guidelines exactly, but they should be in the same ballpark. Nutrition is not an exact science. I was suspicious that the study could easily claim that almost no recipe met every single guideline exactly, but if the deficiency is slight, I don’t consider that a concern.

That was not the case for many recipes, however. Not only were a number of nutrients lacking in most recipes, but “Some deficiencies were so severe that nutrient concentrations did not reach 50% of the NRC RA [recommended allowance].”

For example, 61 percent of recipes were low in vitamin D, and 95 percent of those provided less than half the NRC recommended amount. Zinc, copper, choline, and EPA/DHA were also short in more than half the recipes. Of those that were deficient, 55 percent had less than half the RA of zinc, 43 percent had less than half the RA of choline, and 39 percent had less than half the RA of vitamin E. In other words, those recipes were significantly, not just a little, deficient in these nutrients.

That didn’t surprise me, as it matches what I’ve found when I’ve analyzed many homemade diet recipes. Let’s look at the nutrients that were most often not only short, but seriously inadequate, in the recipes that they analyzed, and compare them to NRC recommendations per 1,000 calories for adult dogs, which is the amount NRC assumes is needed for a dog weighing 35 pounds:

For a comprehensive list of nutrients, continue to page 2!

Vitamin D is primarily found in fish, so any recipe that does not include fish will be short on vitamin D unless a supplement is added. NRC recommends 136 IUs vitamin D per 1,000 calories. It would take about 1 ounce of oily fish to provide this much vitamin D. Some yogurt is also fortified with vitamin D.

Vitamin E was short in every recipe I’ve analyzed, unless supplements are added. NRC recommends just 7.5 IUs per 1,000 calories. It’s okay to give more, but limit amounts to no more than 2 IUs per pound of body weight daily.

Zinc was at least a little short in most of the recipes I’ve analyzed. Significant amounts of zinc are found in red meat, with lesser amounts in pork and poultry. Turkey has more zinc than chicken, and dark meat poultry has more zinc than light meat. Organ meats, particularly liver and heart, provide substantial amounts of zinc. Egg yolks are also a good source of zinc. Diets that rely primarily on chicken, or that do not include organs, are likely to be significantly short on zinc. NRC recommends 15 mg zinc per 1,000 calories.

Choline, a member of the B vitamin family, is often short in recipes I analyze. Eggs are one of the best sources of choline, with one large egg providing 126 mg. Liver, particularly beef liver, is very high in choline, with almost 100 mg per ounce. Heart and kidney are also high in choline. Diets that do not include eggs and organs will be short on choline. NRC recommends 425 mg choline per 1,000 calories. This nutrient is hard to supplement, as most B-complex vitamins and multivitamins provide little or none. Sources include lecithin granules with 217 mg/Tbsp and brewer’s yeast (not nutritional yeast) with 63 mg/Tbsp choline; however, these sources also add about 50 calories per tablespoon.

Copper is plentiful in beef liver, which has 2.7 mg copper per ounce. Chicken, turkey, and pork liver provide very little, so diets that do not include beef liver are always low in copper. I recommend feeding about 1 ounce of liver (at least half of which is beef liver) per pound of other meat. NRC recommends 1.5 mg copper per 1,000 calories.

EPA and DHA are omega-3 essential fatty acids found in fatty fish and fish oil. If you do not feed fish or supplement with fish oil, the diet you feed will be short on EPA and DHA. NRC recommends just 110 mg EPA and DHA combined per 1,000 calories, but I prefer to give 100 to 150 mg EPA and DHA combined per 10 pounds of body weight daily for healthy dogs, up to twice that much for those with a variety of health problems.

Other common deficiencies include:

Calcium: 35 percent of the recipes analyzed were short on calcium. This was likely due to the multitude of recipes that do not include a calcium supplement. All homemade diets require added calcium, in amounts greater than a multivitamin will provide, unless you feed raw meaty bones that are fully consumed. NRC recommends 1,000 mg calcium per 1,000 calories for adult dogs.

B vitamins: Vitamins B1 (thiamin), B2 (riboflavin), and B5 (pantothenate) were short in 14.5, 40.5, and 27 percent of recipes, respectively. Cobalamin (vitamin B12) was also short in many recipes, but since cobalamin deficiency has only been linked to poor absorption due to genetic abnormalities or small intestinal bacterial overgrowth, rather than to dietary deficiency, I’m not as concerned about it. Most recipes met the rest of B vitamin needs. Thiamin is found primarily in legumes (beans, lentils), fruits, vegetables, and grains, so diets that omit these foods are likely to be short. Nutritional yeast is very high in both thiamin and riboflavin (brewer’s yeast has quite a bit less). Riboflavin is also found in vegetables, as well as mushrooms, yogurt, and eggs. I was surprised that pantothenate was short in so many recipes, as I almost never see that. Pantothenate is found in the same foods as thiamin, as well as egg yolks and meat, especially liver. NRC recommends 0.6 mg thiamin, 1.3 mg riboflavin, and 4 mg pantothenate per 1,000 calories.

Linoleic acid (omega-6 essential fatty acid): Found primarily in poultry fat and plant oils, so diets that include little poultry, or that use only skinless breast, which has little fat, will be deficient in linoleic acid. NRC recommends 2,800 mg linoleic acid per 1,000 calories. That amount would be provided by ½ tablespoon corn oil, soybean oil, walnut oil, or hempseed oil. It would take 1 tablespoon of canola oil, 1.5 tablespoons of safflower oil, or 2 tablespoons of olive oil to meet NRC recommendations. Six ounces of chicken breast with skin or ground turkey with 8 percent fat, or 12 ounces of dark meat chicken with skin and separable fat removed, will provide this much linoleic acid (based on raw weights).

Selenium: Just over one-third of recipes were found to be short on selenium. Fish, meat, and eggs are good sources of selenium, though the amount of selenium in foods can vary. Most of the recipes I’ve analyzed contained close to the recommended amount of selenium, which is 88 mcg per 1,000 calories.

As an alternative to these food sources, giving the dog a human one-a-day type of multivitamin can help to make up for most deficiencies (not including calcium, essential fatty acids, or choline). You can give the full adult human dose to dogs weighing 40-50 pounds, half the human dose to dogs weighing 20-25 pounds, or one-quarter the human dose to dogs weighing 10-12 pounds. Larger dogs would get proportionately more. This approach won’t work for really small dogs, as the dosage would be too high. Also, iron is often high, and copper low, in these supplements.

For more analysis and final thoughts, continue to page 3!

The study addressed the issue of “balance over time” by analyzing three groups of seven recipes (all from the same source), but found that even if a variety of recipes were used, they were still deficient in most of the nutrients listed above (14 nutrients had inadequate concentrations in at least 50 recipes).

Even if you vary ingredients, homemade diets are likely to be lacking if you leave out important food groups such as fish, eggs, or liver, or if you rely too much on one protein source, such as chicken or beef. Some supplements, such as calcium and vitamin E, are always needed when you feed a homemade diet. Others will be needed if you omit any of the following food groups from the diet: red meat, poultry (including some fat), fish, eggs, liver, dairy, vegetables, and fruits.

I’ve often said that you don’t need a degree in nutrition to feed your dog a homemade diet; if we can feed ourselves and our children, we should be able to feed our pets as well. That said, dogs have unique nutritional requirements; they need more calcium for their weight than we do, for example. Also, many people get in a rut when feeding a homemade diet, and may feed the same limited recipe for long periods.

I know that it’s possible for dog owners to feed their dogs a complete and balanced home-prepared diet without consulting a veterinary nutritionist, but the authors of this study clearly believe that canine diet formulation is best left to experts. The study concludes, “Formulation of recipes for home-prepared diets requires expert input to minimize the risk of problems, and we recommend that recipes for home-prepared diets for dogs be obtained from or evaluated by board-certified veterinary nutritionists or veterinarians with advanced training in nutrition who are experienced and able to understand and address these concerns.”

While veterinary nutritionists are unlikely to formulate a home-prepared canine diet that is deficient in essential nutrients, they are very likely to formulate a diet that is heavily grain-based, relies on plant-sourced oils for dietary fat, and obtains most of its vitamins and minerals from synthetic, not food-based, sources. In other words, the diets they create are not ideal, either!

I’ve analyzed quite a few diets that were formulated by board-certified veterinary nutritionists, and nearly every one looked like the following and allowed no substitutions:

A small amount of one type of lean meat, e.g., skinless chicken breast, for protein and amino acids.

A large amount of one type of starchy carbohyhdrate, e.g., white rice, used to provide calories.

Corn or canola oil, providing calories and fat (particularly omega-6 fatty acids).

Balance IT, or a number of other supplements, used to meet most nutritional requirements.

There are several problems with this approach. I believe that higher-protein diets provide many benefits to the body, including the immune system, nervous system, skin, and coat. Diets with more meat and fewer carbs help to build lean muscle rather than being stored as fat. Plant oils are more likely than animal fats to become rancid, contain hydrogenated trans fats, and be genetically modified.

Every five years, the U.S. Department of Agriculture (USDA) and Department of Health and Human Services (HHS) jointly update and issue a document called the “Dietary Guidelines for Americans.” These guidelines make it clear that a human’s nutritional needs should be met primarily through diet rather than supplements. Whole foods offer three main benefits over dietary supplements: greater nutrition from the micronutrients they contain, dietary fiber that can help with digestive disorders, and protective substances, such as antioxidants and phytochemicals. These factors apply to dogs as much as they do to humans.

People who feed their dogs a homemade diet usually want to improve nutrition through the use of fresh foods, rather than feeding highly processed commercial diets that rely on a long list of synthetic nutrients in order to meet nutritional requirements. Recipes from veterinary nutritionists are the homemade equivalent of poor-quality kibble, not the varied, fresh food diet that owners would like to feed their dogs.

Why can’t veterinary nutritionists design recipes that meet most nutritional needs through the use of whole foods, rather than synthetic supplements? Why can’t they accept that most people and dogs prefer variety rather than always feeding exactly the same thing every day? And why do they insist on using skinless chicken breast with added plant oils, rather than feeding dark meat chicken (or breast with skin) that would meet omega-6 fatty acid requirements without having to add plant oils?

As long as veterinary nutritionists think that a diet should consist only of the barest amount of fresh food propped up with synthetic supplements, people are going to continue to turn elsewhere for diet advice for their pets. It’s a shame that so many of the books and online resources that purport to provide that advice fail so miserably. Ideally, nutritionists would learn how to create complete and balanced diets that rely on fresh foods rather than supplements to meet most nutritional needs, and recognize the benefits provided by high-protein, meat-based diets (rather than just meeting minimal protein requirements).

Mary Straus is the owner of She and her Norwich Terrier, Ella, live in the San Francisco Bay Area.

Previous articleDog haters: What can be done about them?
Next articleEssential Oil Tick Repellents: What Works and What Doesn’t
Mary Straus has been a regular contributor to Whole Dog Journal since 2006. Mary first became interested in dog training and behavior in the 1980s. In 1997, Mary attended a seminar on wolf behavior at Wolf Park in Indiana. There, she was introduced to clicker training for the first time, and began to consider the question of how we feed our dogs after watching the wolves eat whole deer carcasses. Mary maintains and operates her own site,, which offers information and research on canine nutrition and health. has been created to help make people more "aware" of how to make the best decisions for their dogs. It's designed for people who like to ask questions and understand the reasoning behind decisions, rather than just being told what to do.  Mary has spent years doing research for people whose dogs have health problems, or who just want to learn how to feed them a better diet. Over this time, she has learned a great deal about dog nutrition and health, including the role of diet, supplements and nutraceuticals.  In 2007, she was asked by The Ivy Group to contribute to The Healthy Dog Cookbook. She previously also wrote a column for Dog World.


  1. I have a 16 week Moyen poodle. i would like to feed her homemade dog food. I am in search of a good homemade dog food recipe book. What supplements – brand names please – should I use? . I have read Dr Becker’s book but the book is poorly laid out and does not have an index.
    Wendy Cook