The Feingold diet is a food elimination program developed by Ben F. Feingold, MD to treat hyperactivity. It eliminates a number of artificial colors and artificial flavors, aspartame, three petroleum-based preservatives, and (at least initially) certain salicylates. There has been much debate about the efficacy of this program. The mainstream medical establishment dismissed it as an "outmoded approach" lacking evidence and efficacy,[1] however some medical practitioners, as well as many people living with ADHD and parents of children with ADHD, claim that it is effective in the management of ADHD as well as a number of other behavioral, physical and neurological conditions including salicylate sensitivity. The debate has continued for more than 30 years, involving not only consumers and physicians, but scientists, politicians, and the pharmaceutical and food industries.
Contents |
The Feingold Program eliminates three groups of synthetic food additives and one class of synthetic sweeteners:
The above-listed colorings and preservatives are made from petroleum.[2][3] The word "synthetic" is used instead of "artificial" because not all artificial colorings are eliminated by the program. For example, titanium dioxide and coloring made from iron oxides are acceptable, posing no problem for most people. Only FD&C and D&C colorings are eliminated. There are thousands of synthetic flavorings, from a variety of sources, most of which are not specified in ingredient lists. Due to the "de minimis" principle,[4] safety and neurotoxicity studies are not required for these chemicals. Aspartame and its related chemicals have recently been eliminated from the Feingold Program because of evidence that they may be harmful to the nervous system.[5]
During the initial weeks of the program, certain foods containing salicylates are removed and may later be reintroduced and tested for tolerance, one at a time. Most of the problematic salicylate-rich foods are common temperate-zone fruits, as well as a few vegetables, spices, and one tree nut. During this early period, foods like pears, cashews and bananas are used instead of foods like apples, almonds and grapes.
Contrary to popular misconception, soft drinks, chocolate and sugar have never been eliminated on the Feingold Program, although moderation is encouraged when consuming such items. Families can often continue to eat the types of food to which they are accustomed, including desserts. It is a matter of picking brands free of the unwanted additives. Most of the acceptable foods are easily available at supermarkets.
Dr. Feingold was a pediatrician and allergist, and was considered a pioneer in the fields of allergy and immunology.[6] He served as Chief of Pediatrics, Cedars of Lebanon Hospital, Los Angeles, CA; later he established a number of allergy centers for Kaiser Permanente of Northern California, and served as Chief of Allergy at the Kaiser Permanente Medical Center in San Francisco.
Since the 1940s, researchers worldwide had discussed cross-reactions of aspirin (a common salicylate) and Tartrazine (FD&C Yellow #5).[7] Dr. Stephen Lockey[8] at the Mayo Clinic and later Dr. Feingold at Kaiser, found that eliminating both salicylates and synthetic food additives from patients' diets not only eliminated allergic-type reactions, but also induced behavioral changes in some of their patients.
The first clear case for Dr. Feingold was an adult patient referred to him for treatment of her severe hives in 1965.[9] Typical treatments had not worked for her. Dr. Feingold placed her on a low-salicylate diet with no synthetic coloring or flavoring. Soon her hives were gone, and the patient was happy.
Ten days later, however, her psychiatrist called Dr. Feingold to ask, "What did you do to my patient?" She had been receiving treatment for a personality disorder for years, but in less than two weeks on the diet, her behavior had noticeably improved. Both doctors were puzzled. Dr. Feingold asked his staff to watch for other patients who did not respond to standard treatments. He suggested the diet regimen to them, and sometimes it worked. As more reports of behavioral improvement came in, he began to use the diet for people - especially children - with behavioral problems as well as allergy, and eventually found the diet often worked for children with behavioral problems even without allergy symptoms.
He named the diet the K-P Diet for Kaiser-Permanente (and he liked the pun of "K-P" as Kitchen Police).[10] Later, as this diet became more well-known for helping hyperactive and learning disabled children, the media dubbed it the "Feingold diet."
In his early work with children (and adults) who suffered from what was then called "minimal brain dysfunction" (MBD), Hyperkinesis, or Hyperkinesis-Learning Disability (H-LD), Dr. Feingold found that response was variable, depending on the age of the child and the presence or absence of a history suggestive of neurological damage. In 1976, he reported that in five separate programs, totalling 360 children managed with the K-P diet, 30% to 50% of them showed favorable responses and could be removed from medication.[11][12]
In 1977, Dr. Arnold Brenner published a study he had begun with the intention of disproving Dr. Feingold's claims. Of 32 children who had been under various medication treatments for years with poor success, and who were now put on the K-P diet, 11 (about 33%) were "markedly improved." Surprised, Dr. Brenner wrote that "the startling changes in patients who had been followed for years with other forms of therapy suggest strongly that this improvement was genuine."[13] Although 33% does not appear to be a high rate of success, these were children who had already tried everything else - they were what is called "medication failures."
In the beginning, Dr. Feingold did not eliminate the preservatives BHA or BHT. By 1979, he reported having treated over 600 children with the diet, and that his success rate had risen to 60-70% once he had begun eliminating the preservatives as well.[14] At that time TBHQ and aspartame, which are eliminated today, did not yet exist.
In late 1972, Dr. Feingold was invited to present his findings at the annual conference of the American Medical Association in June 1973, which he did.[15][16] Before that time, he had been treating children and adults with allergy symptoms as well as behavioral symptoms, but by 1972 he had arranged for a group of 25 children with a primary complaint of behavioral problems to be managed by the K-P diet. Dramatic improvement in behavior was repeated in this group as well. Dr. Feingold was again invited to participate in the June 1974 AMA conference, where he reported on 169 children in five separate samples that had experienced a success rate of 30 to 50 percent, depending on the age of the child and the sample. (Note, this was also before the diet eliminated preservatives.)
This led to the first controlled double-blind crossover study, at the University of Pittsburgh, which was funded by the National Institute of Education. It was directed by Dr. C. Keith Connors, with Dr. Feingold as consultant, and the favorable results were published in the August 1976 issue of Pediatrics, the official publication of the American Academy of Pediatrics.[17]
Meanwhile, in 1974, the food industry became involved via its industry-supported organization called the "Nutrition Foundation of New York."[16] This is an organization whose members included Dow Chemical, Coca Cola, and several companies who make, use, and distribute the food additives removed from the K-P diet.[18] In December, 1974, the Nutrition Foundation issued a "proposal for the study of dietary relationship to hyperkenesis" which promised, in its final paragraph, that "no publicity will be given to the findings until the committee has approved the report for release." A two-week long conference, by invitation only, was arranged in January, 1975, in Glen Cove, Long Island. There, they created what they called the National Advisory Committee, which was not government-related in any way. One week later - and certainly long before any of the planned studies had been done, let alone published - this committee widely published its preliminary report concluding that "no controlled studies have demonstrated that hyperkinesis is related to the ingestion of food additives."[16]
This conclusion is still quoted today, more than 30 years later.[19][20] Adding to the confusion is the assumption that the National Advisory Committee is a governmental agency rather than an arm of industry. Over the next few years, the Nutrition Foundation funded and designed several small studies, each concluding that the diet produced little effect.[21][22] Although the studies were conducted at universities, which would normally have lent them credibility, Congressman Ben Rosenthal of New York complained, "Despite the compelling need for experts who can examine the food industry with a critical eye, nutrition and food science communities have fallen under the $200 billion industry's influence. At our more prominent universities, eminent nutritionists have traded their independence for food companies' favors."[16]
A review of the early studies published by M.A. Lipton in 1983[23] concluded that possibly 2% of children respond adversely to food additives, and that "even the 2% are questionable." Dr. Lipton further asserted that there "is no need for high-priority research or for changes in public policy regarding the use and labeling of foods containing additives." It should be noted that Dr. Lipton chaired the National Advisory Committee set up by the food industry's Nutrition Foundation discussed above.[16] However, when toxicologist Bernard Weiss[24] and autism expert Bernard Rimland[22] analyzed these same studies, they found that they actually did support the positive effects of the Feingold diet.
Meanwhile, the media had begun calling the K-P Diet the "Feingold Diet," but because of the confusion with weight-loss diets, and because more than just food is involved in the management of ADHD suggested by the Feingold Association, the "Feingold Diet" was renamed the "Feingold Program."
Many studies show that 70% or more of hyperactive children respond positively to the removal of synthetic additives, especially when salicylates or allergens are removed.[25][26] There is controversy, however, over what happens when researchers take children whose behavior has improved on a diet that eliminates several thousand additives, and then challenge them with one or a few additives, usually synthetic colors.
Especially in the early studies, if such a challenge did not produce a change in behavior, researchers often concluded that the diet had not directly caused the initial improvement in behavior. Rather, the assumption was that the improvement had been due to a placebo effect.
There are other possible reasons for the failure of a challenge to evoke a response, however. For example, the amount of additive used as a challenge might have been too small to cause an effect.[22] Rowe & Rowe in 1994 found a dose-related effect; the higher the amount of coloring he used for his double-blind challenge, the stronger (and longer) the reaction of the children.[25] The following chart lists the amount of coloring used in various studies along with the rate of response:
| Name of Researcher
and Year Published |
Amount of Food Dye
Challenge Used in Study |
Percent of Children w/Behavioral
Reactions to Food Dye Challenge |
|---|---|---|
| Levy 1978[27] | 4 mg in cookies w/1 mg each | 0% of 8 children
1 child was dropped from study when behavior deteriorated on challenge |
| Levy 1978[27] | 5 mg in cookies w/1 mg each
testing done day AFTER challenge |
0% of 12 children |
| Wilson 1989[28] | 17 mg | 5% of 19 children |
| Weiss 1980[29] | 35.26 mg | 9% of 22 children (not ADHD) |
| Williams 1978[30] | 26 mg | 11% of 26 children |
| Goyette 1978 (a)[31] | 26 mg | 19% of 16 children showed visual tracking problem;
Behavior of all the children was worse after eating food dye, but not significantly so |
| Goyette 1978 (b)[31] | 26 mg, using younger children
This is a higher dose in a younger child |
100% of 8 children were impaired by food dye |
| Rowe 1988[32] | 50 mg | 25% of 8 children |
| Rowe 1994[25] | 50 mg | 64.7% of 34 children |
| Boris 1994[26] | 100 mg or 5 g other provoking food | 81% of 16 children |
| Swanson 1980[33] | 100 & 150 mg | 85% of 20 children |
| Pollock 1989[34] | 125 mg | 89.5% of 19 children |
| Egger 1985[35] | 150 mg | 79% of 34 children |
Considering that in 1976 an FDA scientist estimated that children may be consuming up to 315 mg food dye per day,[36] all the above studies appear to be overly conservative in their choice of challenge amounts. In addition, the effect of an additive might only be seen in synergy with other additives or foods,[5] or the additive used for the challenge may simply not be among those causing the original effect.[36]
As with many new developments, the first reports of improvement of behavior via diet were anecdotal. This was followed by clinical trials and eventually by larger, double-blind placebo-controlled studies.
In 1976, a double-blind crossover diet trial found that both parents and teachers saw fewer hyperkinetic symptoms on the K-P diet as compared to the pretreatment baseline.[37] A 1978 double-blind crossover study using cookies with 13 mg food dye each combined with either medication or placebo found, "The results of this study offer data that a diet free of artificial flavors and colors results in a reduction of symptoms in some hyperactive children."[38] In 1980, forty children were put on a diet free of artificial food dyes and other additives for five days. They then performed the usual double-blind placebo-controlled test but used 100 mg or 150 mg of the food dye mix. They found that the food dyes impaired the performance of the 20 hyperactive children on paired-associate learning tests. The dyes did not hurt the performance of the 20 non-hyperactive children. The study states: "Our data suggest that a large dose of food dye blend decreases attention span in hyperactive children as reflected by performance on the learning test."[39]
Seven small studies adding up to a total of 190 children were published starting in the late 1970s. Some of them were elaborate double-blind diet studies using a Feingold-type diet. In some of these studies, the children were taken off their medication, while in others, they continued on stimulant medications including artificial colorants, in spite of the interference that this would create with the diet.
One of the studies in 1978,[40] for example, used 36 children between 6 and 12, and 10 children between 3 and 5. The teachers of the school-aged children did not record any improvement, but 63% of the mothers reported improved behavior, as well as 100% of the mothers of the preschoolers; however, since the improvement was reported by the parents of the children rather than teachers, and locomotor activity tests were unaffected, it was reported that there was "no diet effect."
In 1980 the Nutrition Foundation set up a review team to review studies related to the Feingold diet.[41] They published a report that stated that there was no response at all to the diet. In 1983, the review team's co-chairman and a colleague reviewed a variety of studies and concluded that no more than 2% of children respond adversely to dye additives.[23]
An influential comparative diet study was conducted in 1987 by Gross et al.[42] This was a study of 39 children, of whom 18 were hyperactive, and the balance had other learning disorders. Of those 18, all but one were on behavior-modifying medications during the entire study. The researchers provided a Feingold-type diet for a single week that was, by their own description, unpalatable. They particularly noted that the children missed mustard and ketchup; however, mustard is not eliminated by the Feingold diet, and no reason was given for its exclusion.
This diet week was followed by an additive-rich diet the next week. Although the study reported that the camp director and all teachers felt that the children were noisier and more active during the second, additive-rich week, they discounted these observations in favor of filmed 4-minute sequences made during meals. These films were intended to measure reaction to additives in the meals in spite of the fact that any such reaction would not be expected to occur for some time after eating.
During the course of the study, three children were dropped: one who was not on stimulant medication, whose behavior became worse during the second week; one who refused to behave altogether; and one whose dose of Cylert became "inadequate" and whose behavior worsened when additives were allowed during the second week.
They concluded that the "Feingold diet has no beneficial effect on most children with learning disorders" and moreover that the diet was "distasteful to the typical American child."
A number of studies conducted since 1980 using diets similar to the Feingold Program report greater than 70% of children responding positively to the diets. Others that eliminated synthetic colors and flavors, but included salicylates still reported greater than 50% positive response.[25][26][32][33] [35][43][44][45][46]
In the biggest such study ever performed, published in 1986, the performance of over a million children in 803 New York City public schools was studied for seven years. The children's average standardized test scores rose 8.1% when levels of sucrose (normal table sugar) were restricted to 11 percent along with the removal of two synthetic food colors; when the remaining food colors and all artificial flavors were removed the next academic year, performance rose another 3.8%; when no further changes were made the following academic year, test performance also remained stable; finally, when the petroleum-based preservatives BHT and BHA were removed from the menu in the next academic year, performance improved another 3.7% for an overall improvement of 15.7% in mean national percentile rankings (from 39.2% to 54.9%).[47] Although it appears that improvement increases as the diet approaches the guidelines of the Feingold Program, the researchers suggest that by removing sugar and additives - thereby removing empty calories and processed foods - malnutrition is reduced. It is not clear what portion of the effects can be contributed to limiting sugar intake or the foods containing BHT and BHA (the related preservative TBHQ did not exist at that time) and what portion can be attributed to the removal of the artificial colorants themselves.
A most important and often overlooked detail in this study is that all the children did not improve equally. There was a dramatic decline in learning disabled and repeat-failure children. In 1979, 12.4% of the million children were performing two or more grades below their proper level. By the end of the study in 1983, the percent of children two or more grades below proper level had dropped to 4.9%. Moreover, before the dietary changes, the more school food that was consumed, the worse the children did academically. After the changes, however, the more school food the children ate, the better they did academically.[48]
In 1997, an association between brain electrical activity and intake of provoking foods was shown in children with food-induced ADHD. (Picture) [49] Another study showed that an oligoantigenic diet can work as well as Ritalin for conduct-disordered children.[50] Other research demonstrated the positive effect of treating young criminals with dietary intervention and correction of mineral imbalances,[51][52] and that toddlers show both significant reductions in hyperactive behaviour when additives are removed from their diet, as well as increased hyperactivity when exposed to a very small (20 mg) amount of food coloring and a benzoate preservative. This effect was observed by parents whether or not the child was hyperactive or atopic.[53]
A 2007 British study at the University of Southampton[54] has pointed to food additives as a health hazard for all children, whether they have ADHD or not. The study concluded that artificial colors or a sodium benzoate preservative (or both) in the diet result in increased hyperactivity in 3-year-old and 8/9-year-old children in the general population. In response to this study and a massive media and grass-roots campaign, the major supermarket chains in the UK have removed additives from their house brands. Several American-based candy companies have done the same with the candies they sell in the UK.
In response to that study, which had been financed by Britain’s Food Standards Agency and published online by the British medical journal The Lancet, the American Academy of Pediatrics concluded that a low-additive diet is a valid intervention for children with ADHD. It released the following statement in the February 2008 issue of its publication, AAP Grand Rounds:
“Although quite complicated, this was a carefully conducted study in which the investigators went to great lengths to eliminate bias and to rigorously measure outcomes. The results are hard to follow and somewhat inconsistent. For many of the assessments there were small but statistically significant differences of measured behaviors in children who consumed the food additives compared with those who did not. In each case increased hyperactive behaviors were associated with consuming the additives. For those comparisons in which no statistically significant differences were found, there was a trend for more hyperactive behaviors associated with the food additive drink in virtually every assessment. Thus, the overall findings of the study are clear and require that even we skeptics, who have long doubted parental claims of the effects of various foods on the behavior of their children, admit we might have been wrong.”
Anne Swain, of the Allergy Unit at Royal Prince Alfred Hospital in Sydney Australia, measured amounts (but not type) of salicylate in 333 foods in 1985,[55] and has done other research based on the Feingold Diet.[56]
Over the years, a number of criticisms of the Feingold Program have been presented. Many of these center on the difficulty in avoiding synthetic additives, especially in processed or fast food or while eating out, or with social or emotional side-effects the diet may cause. Others center on the range of symptoms claimed to be improved by the Feingold Program.
Some critics say that the Feingold Program requires a significant change in family lifestyle and eating patterns because families are limited to a narrow selection of foods, and that such foods are often expensive, and must be prepared "from scratch", greatly increasing the amount of time and effort a family must put into preparing a meal.[19] However, the Feingold Association publishes a Foodlist & Shopping Guide that provides thousands of acceptable brand-name items to choose from, including many convenience foods available in local supermarkets. The Foodlist books are organized by region (seven regions in the U.S., one encompassing Canada). They also provide a Fast Food & Restaurant Guide as well as other help.
Like any change in diet, the Feingold Program does require that patients make changes in the food that they eat. However, these changes do not usually require significant changes in the types or cost of food a family may choose or the way a family chooses to prepare them. Such choices can be more difficult when little is known about the exact ingredients used in a product, such as at a restaurant or when purchasing food from a vending machine. This requires that a family identify restaurants or products that are not likely to create a problem. Questionable choices can also be avoided by bringing appropriate food when necessary, such as bringing a lunch to school. Parents are encouraged to keep treats available at home and school, so that the children never need feel deprived or left out.[57]
Nutritionally, the Feingold Program is little different from a normal diet. While some fruits and a few vegetables are eliminated in the first weeks of the Program, they are replaced by others. Often, some or all of these items can be returned to the diet, once the level of tolerance is determined. Studies have found that children on the Feingold Program actually ate better than those eating a "usual" diet, and were more likely to achieve the Recommended Dietary Allowance (RDA) of various nutrients.[58][59]
Other critics express concerns about social or emotional effects of putting children on a specific diet. These include that their self esteem may be undermined by implanting notions that they are unhealthy and fragile, or that children may experience situations in which the children's eating behavior or "fear of chemicals" are regarded as peculiar by other children.
No clinical evidence supports these speculations, although the attitude of the parents may be important. The issues that a child on the Feingold Program faces are very similar to the issues that a child with an allergy to a common product such as eggs or peanuts must deal with, or a medical condition such as diabetes. Compared to these situations, the Feingold Program could be seen as relatively liberal, as it includes sugary foods, junk foods and even fast food.
Some feel that it is absurd to think that one intervention could improve symptoms as diverse as asthma, allergies, bedwetting, chronic ear infections, headaches, and insomnia all at once. Critics point to the fact that effectiveness against a wide range of unrelated symptoms is frequently a hallmark of treatments that work via the placebo effect.
The diet was originally designed as a diagnostic elimination diet to improve food-related asthma and allergic reactions and was only later found to also be effective in treating behavioral issues. In addition, many children with ADHD suffer from multiple comorbid symptoms. It has been found that there is a profile of the child most likely to benefit from the diet. [60] The child may not have all of these symptoms: some may have few symptoms and others seem to have all of them. While the underlying physiological reason is not understood, when a patient eliminates the additives to which they are sensitive, many or even most of the symptoms contained within the profile are improved. Research supports dietary intervention for each of the symptoms in turn.[5][25][44][61][62][63][64][65][66][67][68][69][70][71]
Dr. Feingold wrote a book directed to parents, entitled Why Your Child is Hyperactive,[72] as well as The Feingold Cookbook,[73] written in collaboration with his wife, Helene. In addition, more up-to-date books have also been published.
As parents began using this diet for their children, many saw dramatic success and formed grass roots support groups. When they gathered in 1976 to form a non-profit national organization, they chose the name "Feingold Association" to honor Dr. Feingold. As time passed, due to the increasing number of double-income families, fewer mothers were available to run these local "kitchen table" support groups, and today the Feingold Association of the United States provides member support services. Recently, some support has been added for Canadian members, and there is some information on the website suitable for people in other countries, as well.
The Feingold Association provides information and support for those starting the Program. Members can purchase comprehensive materials[74] including a book listing thousands of brand name foods that have been researched by the Association and are free of the eliminated additives. Newsletters, updates, and phone and email support are also provided. Acceptable products — food, toiletries, cleaning supplies — are included in the Foodlist and Shopping Guide, the Mail Order Guide, the Supplements Guide and the Fast Food Guide.
A good introduction to the Feingold Program, as well as 400 pages of compiled wisdom from over 30 years of working with families using the diet, is provided by the book Why Can't My Child Behave?[75]