Use of Diet for ADHD -- Myths and Facts

The American Academy of Pediatrics (AAP) recommends in its Policy Statement RE0049 that physicians should educate themselves about alternative and complementary therapies. We agree. The AAP recommends four websites for this purpose, two of which were authored by Stephen Barrett, MD, a retired psychiatrist in Allentown, PA. We believe that unbiased information cannot be found on these sites, and doctors depending on them are misinformed.

See all four websites recommended by the AAP, and what they say about the Feingold Program here.

After reading this page, please visit the rest of our web site.

Food additives, and the products that rely upon them, are big business. In addition, the drugs used to treat symptoms that are triggered by these additives, are also extremely profitable.

Ever since Dr. Feingold first described the harmful effects of these chemicals, this work has been the target of well-funded and intensive distribution of misinformation from the food and chemical lobbies, which continues to this day.

The following order is chronological - not in order of importance - following the order in which the statements occur in Barrett's article called "The Feingold Diet: Dubious Benefits, Subtle Risks".
Barrett claims that the Feingold diet requires a "change in family lifestyle and eating patterns." This is not true. Families do not need to change their lifestyle and eating patterns. They can buy any of the thousands of acceptable brand name foods and eat when and how they like.
Barrett says that the diet requires "special foods." We are not aware that foods made without artificial colorings and flavorings are "special." They used to be considered just plain food. Barrett apparently believes that foods of such dubious quality that they require strong chemicals to look and taste edible should be the norm.
Barrett attempts to frighten parents by misrepresenting a sentence from the out-of-print and out-of-date Feingold Cookbook, claiming that a "single bite" may cause a lengthy "failure" of the diet. First, it is important to know that going "off-diet" will usually cause a reaction -- just like forgetting the pill will cause a problem for the child on stimulant medication. It is not a "failure" of the treatment. It is also important to realize that once the child is well established on the diet, they are usually much less sensitive and can often begin to tolerate minor excursions "off-diet" without any dramatic problem. Indeed, in experiments where children doing well on the diet were given small amounts of colorings, most of them remained well. Instead of realizing that this meant the diet was working so well that the children were stable, this was misinterpreted to mean that the colorings were harmless at all doses.
Barrett says that "carefully designed experiments" do not support use of the diet. Even the older, poorly designed studies (designed and paid for by the companies that make and use the additives) showed a positive effect from the Feingold diet. Newer studies with better designs have shown even greater success.

Tobacco companies have funded "carefully designed" studies to show that smoking doesn't cause lung cancer. One should not accept these studies without questioning them.

Barrett says any improvement "appears related to changes in family dynamics, such as paying more attention to the children" All mainstream professionals now agree that ADHD is not the result of poor parenting, and improved parenting or extra attention will not cure it. No scientific studies support Barrett's opinion.
Barrett claims that the "salicylates" eliminated by the Feingold Association (FAUS) do not match some other list of salicylates organized by weight or quantity. Plants contain many kinds of salicylate, which are mildly toxic chemicals used as protection against insects. There is ethyl salicylate, methyl salicylate, octyl salicylate, acetyl salicylate, sodium salicylate, etc. - and their toxicity per amount is not necessarily the same for all of them. (Think of the difference between ethyl alcohol and methyl alcohol - ethyl alcohol is in beer and wine, while even small amounts of methyl alcohol, also called "wood alcohol," can kill you.) The list used for the Feingold Program is based on clinical response. It works. FAUS hopes to see some research on the different types of salicylate one day.
Barrett quotes results from seven small studies funded and designed by the Nutrition Foundation way back in the 1970's. At an average of only 27 children per study, he adds them up to sound better, and claims this is proof the diet does not work. These studies are almost 30 years old. In A careful re-examination of these studies in 1982 by Bernard Weiss, PhD, Professor of Toxicology at University of Rochester School of Medicine, under a grant from the National Institute of Mental Health, shows that even these studies actually did show support for the diet-behavior connection. They cannot be considered "proof" that the Feingold diet does not work. The now-defunct Nutrition Foundation was the major food industry lobby, representing the interests of companies like Dow Chemical and General Mills. See listing of members at the end of this article.
Barrett says that the Rowe 1988 study "reported no effect." This study is #6 in the references and is used to support his statement that "Some researchers have reported little or no adverse effect during challenge experiments " This statement is false. The 1988 Rowe study reported that 72.7% of children improved on the diet.
Barrett also uses the Gross 1987 study to support his contention that scientists "reported no effect" of the diet. It's true that the authors of this study reported no effect of the diet on the children. But the study was so badly designed, no conclusion can be drawn from it. Of the 39 children in the study, only 18 had ADHD. Of those, 17 took stimulant medicine (with artificial coloring) the entire time. The 18th boy went home, so the number of children with ADHD who could actually be tested with the diet was: zero.

There are more problems with the study design, which you may see below:

Barrett says it is "clear that the percentage of children who may become hyperactive in response to food additives is, at best, very small." Nine studies using Feingold-type diets report the following percentages of children with ADHD responding to dietary intervention:
§ Egger 1985 = 81.6%
§ Swanson 1980 = 85%
§ Rowe 1988 = 72.7%
§ Egger 1989 = 80%
§ Kaplan 1989 = 50%+
§ Egger 1992 = 76%
§ Carter 1993 = 75.6%
§ Rowe 1994 = 75%
§ Boris 1994 = 73%
Barrett claims that sugar does not cause ADHD. FAUS agrees. The Feingold Program does not eliminate sugar. Why do these so-called experts continue to insist that we do? It is the additives frequently found in such foods that we eliminate.
Barrett thinks it is absurd to think that children and adults may be sensitive to environmental chemicals. From other articles on his site, Barrett clearly believes that any chemical not strong enough to kill you outright cannot make you ill or produce any neurological impairment. The Feingold Program focuses on several groups of food additives and on salicylates. While we occasionally carry articles in our newsletter about environmental chemicals, their elimination is not required on the Program.
Barrett says the diet teaches children that their behavior is related to what they eat rather than what they feel. The Feingold Program is a simple test to determine if certain foods or food additives are affecting a person. If there is a change in a child's behavior after he eats blue applesauce or green ketchup, it's likely the child will notice it himself. There is no need to teach a child how he feels.
Barrett says the diet makes children feel unhealthy and fragile. No research supports his opinion. Actually, the opposite is true. Once they go on the Feingold diet, many children stop having ear infections, or asthma, or stomachaches, or eczema or chronic sinus infections, etc. Children who literally were fragile can begin to feel healthy and strong.
Barrett says the diet makes the children look peculiar to other children. A typical lunch on the Feingold diet might be a peanut butter and jelly sandwich or a turkey sandwich, or a cheese sandwich with mayo and lettuce; the beverage might be whole milk bought at school, or lemonade in a juice box, and maybe Swiss Miss Chocolate Pudding for dessert. Would you consider a child eating such a lunch "peculiar?"
Barrett says the children's "fear of chemicals" would make them look peculiar to others. Children who understand that synthetic food additives are made from crude oil don't fear them; they think they're gross.
Barrett claims the Feingold Program deprives the children of "appropriate professional help." He apparently means they are being deprived of receiving stimulant drugs. When symptoms improve on the Feingold Program, then the Feingold Program is an appropriate treatment. Children who still need drugs can use them. Children who still need behavior modification or psychotherapy or classroom accommodations can still get them. They are deprived of nothing. The diet can be used along with any other required treatment.
Barrett claims the Feingold website "displays abstracts of scientific articles with portions highlighted out-of-context to falsely suggest that fluoridation is hazardous." On the information page about Fluoride, the articles are linked to their abstracts in MedLine. The reader can see the entire abstract - and where possible, the entire article. Is Barrett saying that the MedLine abstracts are "out of context?" It is well known by scientists that the type of fluoride used in our water supply increases uptake of lead, which has neurological effects, and that fluoride itself has neurological effects. For a minority of our members, this is a significant factor. Fluoride as used in our water supply has never been proven safe and effective for children. Fluoridation of water is an ongoing controversy, but Barrett does not want you to know it.
Barrett reports that Dr. Sinaiko, one of the Feingold Association's medical advisors, was placed on probation by the Medical Board of California. He links to the judge's findings. This is true; however, Barrett neglects to tell you the rest of the story - that many organizations, including the California Medical Association and the Union of American Physicians & Dentists, have objected to this ruling. The case is in appeal. See more at
The statements made by the Barrett websites, such as, are either not documented at all, or not adequately documented. The author misquotes the research and the facts, and relies on very old or poorly done studies while ignoring newer and better ones. But since this misinformation is presented to doctors in lieu of appropriate education, one can understand why a child's pediatrician fails to support or understand a trial or continued use of the Feingold Program.

Dietary elimination is not the same as megavitamins or other supplementation, and should not be lumped together with them any more than one would lump together stimulants and opiates as treatments for ADHD. They have different evidence bases, which should be separately evaluated.

Here is the research evidence in peer review literature, including 37 double-blind, controlled studies supporting the use of diet for ADHD, and more studies supporting use of diet for asthma and other symptoms that these same children often have.

Who is the Nutrition Foundation?

This is an organization made up of food industry companies, food additive companies, pharmaceutical companies, universities who presumably receive research funds from the Nutrition Foundation, etc.

In the 1970's, its members included executives of the companies listed below. You may well wonder what connection some of them have with nutrition. This was the group whose mission was to decide on the credibility of the Feingold diet. You may think there is a conflict of interest on the part of many on this list. You will be right.

  • The Nestle Company
  • The Coca Cola Company
  • Revere Sugar Corporation
  • Mrs. Paul's Kitchens, Inc.
  • Oscar Mayer & Company
  • Josia Macy Jr. Foundation
  • Hoffmann-LaRoche, Inc. - makers of pharmaceuticals/drugs
  • Mayo Clinic
  • Cornell University
  • Standard Brands, Inc.
  • Dept. of Chemistry, U of Arizona
  • The Commonwealth Fund
  • Uncle Ben's Foods
  • Hedrick & Lane - attorneys
  • Miles Laboratories, Inc.
  • Campbell Soup Company
  • Fritzsche-D & O, Inc. - makers of artificial flavors
  • Stange Company - makers of artificial colors & flavors
  • Beatrice Foods Company
  • Wm Underwood Company
  • Universal Foods Corporation
  • Gerber Products Company
  • Florasynth Inc. - makers of artificial flavorings
  • California Canners & Growers
  • Massachusetts Institute of Technology
  • Swift & Company
  • Griffith Laboratories - makers of nitrites
  • Rockefeller Foundation
  • Vanderbilt University
  • H.J. Heinz company
  • Henningsen Foods, Inc. - makers of dehydrated Eggs & Meat
  • American Home Products Corp
  • Dow Chemical Company
  • Monell Chemical Senses Center - developers of artificial flavorings
  • General Mills, Inc.
  • Carnation Company
  • Kellogg Company
  • General Foods Corporation
  • Tulane University
  • Kraft, Inc.
  • Philadelphia Macaroni Company
  • Borden, Inc.
  • U of California, San Diego
  • Welch Foods, Inc.
  • Life Savers, Inc.
  • Swanson Center for Nutrition
  • H. Kohnstamm & Co. - makers of artificial colors & flavors
  • Amstar Corporation - sugar company
  • Nabisco, Inc.
  • Shaklee Corporation
  • Pet Incorporated
  • Rockefeller University
  • M.J. Murdock Charitable Trust
  • The Quaker Oats Company
  • Milwaukee School of Engineering
  • Apnomoto Company, Inc. - makers of MSG
  • Thomas J. Lipton Inc.
  • PFW, Inc. - makers of artificial flavorings
  • CPC North America - makers of corn syrup
  • ICI Americas inc. - makers of dyes, petrochemicals, pesticides
  • Proctor & Gamble Co.
  • McCormick Co.
  • Agricultural Experiment Station, Michigan State U

Gross, M. D., Tofanelli, R. A., Butzirus, S. M., Snodgrass, E. W. (1987).

The effect of diets rich in and free from additives on the behavior of children with hyperkinetic and learning disorders. Journal of American Academy of Child and Adolescent Psychiatry, 26, 53-55.

  • Gross used an unpalatable (his own words) diet which is certainly not a requirement of the Feingold Program.

  • Gross allowed no condiments, although many are permitted by the Feingold Program.

  • Gross told the children "the good stuff didn't come yet," which made them oppositional.

  • Gross apparently ignored additives in soap, toothpaste, medicine, cleaning supplies, etc., so the children were certainly exposed to artificial colorings, flavorings, and petroleum-based fragrances.

  • Since the children continued taking their stimulant medication (with artificial coloring), and their hyperactivity was already theoretically under control, there would be little ability to compare with any other treatment.

  • The actual diet lasted only ONE WEEK. It usually takes a week in the best environment for any improvement to begin to appear.

  • The justification for a one-week diet was that when children eat items forbidden by the diet, parents report a reaction within minutes to hours. Apparently, the researchers were not aware that while one may get a reaction in minutes to hours, one first has to have an actual response to the diet, which takes several days to several weeks, depending on age, circumstances, errors, etc.

  • The children were filmed during meals, whereas any effect from what they ate would have taken place after the meal.

  • The people analyzing the 4-minute film were supposed to be blind to whether the children were on the diet during the particular segment. However, it is likely that the food was visible on the table. Gross himself (certainly not a blind reviewer) also reviewed the taped segments, and his results were averaged with the others. Moreover, the person doing the filming had control over what appeared in the film, and the researchers do not say the photographer was blind to the diet order.

  • The purpose of this study was to test a diet for hyperactive children. The behavioral ratings of all the children were averaged. Since half of the children were not hyperactive to begin with, this certainly appears to have been calculated to wipe out any possible significant difference.

  • Considering the careful design of this study, any actual response to the diet week would be astonishing.

  • In spite of the many pitfalls listed above, however, one hyperactive boy did become "more boisterous during the second ("additive-rich") week, and was sent home on the 12th day." The authors did not conclude that the additives bothered him, but that his dosage of Cylert was inadequate.