Additional Information for Diet Detectives

Diet in ADHD – the position in 2001

The first connection between diet and hyperactivity, now called attention deficit hyperactivity disorder [ADHD], was made in 1974, by an allergist, Dr Ben Feingold. His hypothesis was that "diet caused hyperactivity". Researchers in the late 1970's investigated this. Their research showed that diet did not cause hyperactivity. This solved the public health problem of whether additive colours and flavours should be banned from the food supply. However, this research did also show that a small number of children reacted. Although Feingold's hypothesis had appeared simple to test, the issue was far more complex than had been appreciated. With hindsight we can note that the researchers had used what are now known to be low doses of dye. More importantly, they did not consider additive flavours, which are used in ten times the dose of colours. In some studies the colour was hidden in chocolate biscuits. We now know that chocolate causes reactions in around 75% of those who react to dyes.

Later research conducted in the 1980's and 1990's had a broader approach and better methodology than the initial work. Several teams expanded earlier work with increased attention to diet content, and to symptoms other than hyperactivity. They excluded more additives – all preservatives, added natural colours and flavours, as well as more salicylates, amines and monosodium glutamate. They also excluded more smells and perfumes, and some whole foods such as milk and wheat. They investigated a broader range of behaviours including irritability and sleep problems. One team developed a questionnaire especially for diet related research. Most of these studies used double blind placebo controlled methodology. They reported that there is a statistically significant diet behaviour connection in some ADHD children.

So the position on the role of diet in ADHD is that dietary factors have been shown to have a role in some children who are susceptible.

Which children are susceptible?

Firstly, it is important that other medical causes for the behaviours have been excluded, and that the behaviours of concern are outside the normal range.

The research found that where a child or close relative has or had allergic-type symptoms, a positive outcome to diet is much more likely. Symptoms present in the family found to correlate with a positive diet response are eczema, migraine and irritable bowel syndrome. Non-allergic children may still be diet responders. However their diet is usually less complicated.

Another excellent indicator of positive diet response was if parents could give an example of any diet reaction at some time in the child's life.

Food sensitive people are generally "super tasters" and "super smellers". They are likely to have strong food likes and dislikes. Many are fussy eaters. Just as ADHD children have developmental delays in speech or coordination, the eating problems can be seen as "eating delay".

There are more boys than girls diagnosed with ADHD, and there are more ADHD boys who respond to diet: 85% boys to 15% girls. If you are a boy who is restless, disruptive and have trouble with reading and spelling you may well be diagnosed with ADHD. If you are a girl who is scatty, talkative, and have trouble with maths, you are – well – a girl! There are probably many girls with ADD who are undiagnosed.

What does diet change?

Early research focussed on changes in hyperactivity, but later workers examined the diet effect on many other problems and found some surprising results. The diet was reported to change many problems, with the most change being in irritability. Other problems that decreased were poor concentration, impulsivity, unreasonableness, restlessness, argumentativeness, tantrums, uncontrollability, aggression, sleep problems and excitability. In one study of non-ADHD children using the Connors rating scale, the overall scores were improved on the diet. This showed that ADHD behaviours, as well as changing in those with an ADHD diagnosis (as has been well-documented in many studies), also changed in those without a diagnosis of ADHD. Diet also affected reading (less guessing), language (more mature), and coordination (less spills and bumps into furniture). It now can be said that diet aggravates the underlying tendency in susceptible children. Note that all behaviours are somewhere on a continuum, and in diet responders they shift towards or into the normal range.

Changes often relate to mood. Gradually different groups emerged. One was the "happy-high and silly" children. They are likeable but wear people down! The second were the "touchy" people: they were "outward directed". They say things like "It's not my fault, he started it", "You are always asking me to do too much", and "The teacher is always picking on me!" The third group are the supersensitive ones. They are "inward directed" and say things like "I'm no good, I'll never have any friends", "I am not good looking" and "I won't be able to do it". This group are more often girls. Parents report that all of these moods decrease if the children respond to diet. When they eat foods to which they have reactions, silly children become more impulsive and talkative, touchy children become angry, and worriers become anxious. The effect whereby alcohol can cause specific reactions such as aggression, silliness or depression in certain people is a useful example of another food component effecting people. The idea that I am raising here is that in susceptible people, the diet effect may be a more general effect on mood, regardless of the original diagnosis (eg ADHD).

As well as ADHD behaviours changing, non-ADHD, physical symptoms also change with diet. I call these "allergic-type" symptoms as they include symptoms usually classes as "allergic" and other symptoms that often also occur in atopic families (families that have allergic symptoms). The symptoms often reported to improve with diet include eczema, headaches and migraine, tummy aches and irritable bowel syndrome, looseness, diarrhoea and constipation, halitosis [bad breath], earaches, hay fever, sinusitis, wheezing, night terrors, bedwetting, and limb pains. Susceptible individuals can vary in which part of the body they have symptoms. This is called "target organ sensitivity". A food sensitive family may present with one child with ADHD, another with eczema, an infant with diarrhoea, and a parent with migraine.

Since diet "aggravates the underlying tendencies" it can be the " icing on the cake" for those who have worked hard on managing behaviour and fine tuning medication. It can also be very useful if it reduces the severity of symptoms. Alternatively, it can be at its most important when the child seems unable to respond to any other intervention.

Sometimes I am asked if diet changes the same symptoms as stimulant medication (eg Ritalin). This cannot be discussed in detail here. A useful simplification is that when using diet parents more often say the child is "easier to live with" and on medication they more often say, "school work has improved".

What diet factors are important?

Several diet components need to be excluded or minimised. These include: Additive colour and flavours, most preservatives; Natural chemicals including salicylates, (eg in spice and tomato sauce); amines (eg in chocolate and matured cheese), and natural and added monosodium glutamate; Whole foods commonly implicated in allergy, such as milk or wheat, especially if known to produce physical symptoms in the ADHD child or a close relative; Inhalants commonly implicated in allergy, such as pollen, particularly if seen as suspect by a close relation; Smells – perfumes, paint, petrol, cigarettes, perfumed plants, bubble bath, glues; Contact dye on the skin – finger or face paint, playdough that leaves colour on the skin; Infections – viral, bacterial, parasitic or fungal; Significant stress – for the child or in the family. Temperature change or particular seasons; Insect bites.

All these factors add up to become the "total body load". Individuals vary in which factors they are sensitive to, and which are more important. Rarely is anyone sensitive to all. However, in the initial trial diet, it is worthwhile to minimise all possible suspect factors. Some excluded foods and additives produce only mild reactions. Others, such as red cordial, Coca Cola, chocolate or tomato sauce are often reported to produce clear reactions. Dose and build-up effects can occur.

A common factor in substances excluded is flavour – artificial flavour, spicy and acid fruit flavour, tea and mint flavours, chocolate and aged food flavour and, of course, flavour enhancers such as MSG. If you add perfumes and smells you can see that aromatic compounds are a problem. Artificial colour also has an aromatic chemical structure.

Food sensitivity is not related to the nutritional quality of the diet. Food sensitivity is not a nutritional deficiency. Poorly nourished children are not more susceptible. Nutritional aspects to consider are ensuring sufficient total energy for thin active children, sufficient protein and iron, vitamins (especially vitamin C as fruit intake is lowered), and calcium if milk has to be limited or excluded.

Diet therapy using the "Diet detective method"

The only way to see if diet has a role is to conduct a diet trial. The initial diet needs to exclude or minimise all the diet and environmental factors mentioned above. It is important to attend to all of them at the same time. Attention to additives without also reducing natural chemicals or smells may mean a diet effect present is not seen. It is often wise to use the help of a dietitian, preferably one with experience in diet investigation. The diet can be managed incorporating the breadth of factors mentioned above, but with different levels of strictness depending on age, motivation and severity of symptoms. See my book, "Are you food sensitive?" for details. Attention to nutrition is also important. Withdrawal or worsening of symptoms often occurs in the first week. The reduction of the "total body load" needs to continue for four weeks to provide a baseline.

The initial trial diet may produce some improvement, but it is the challenging by reintroduction of excluded foods and additives that shows if the worsening of behaviours and ADHD symptoms is due to diet.

Through challenges and trials of increasingly risky food, the diet can and should be gradually expanded. Individuals do differ in what they tolerate. Generally, tolerance improves with age and maturity.

Diet interacts with behaviour

When diet therapy is begun, the child can learn he is being a "diet detective" to "see if he can handle himself better". So when challenges are being conducted and a suspect food is being reintroduced, the child can make an effort to show he "can handle himself". If he becomes touchy, feels irritable but does not get irritable, he is showing that he can tolerate the food. I use the "Three pain rule" whereby if three people are seen as "a pain", (with comments such as "It's his fault, he started it!" "The teacher is always picking on me!" or "You make me do more that my brother!"), when this level of touchiness was not present the week before, then the child is reacting. An adult expressed this by saying "When I have a reaction, not only do I feel annoyed with people but I feel I have good reason!"


Diet is both more and less important than initially envisaged. It is more important in that it changes more than hyperactivity. It is less important as it does not affect all children. It is not all-powerful or even useful for every person. However for those who are susceptible it can definitely decrease allergic-type symptoms, mood problems, and it can be an important treatment for ADHD.

Joan Breakey Presentation at the ADHD in the Third Millennium – Perspectives for Australia Conference Sydney March 2001. The abstract is available in the Children's Hospital Education Research Institute Proceedings of the conference.

The role of diet in mood.

Talk given to the Institute of Australasian Psychiatrists. Nov, 2000

Joan Breakey Dietitian / Nutritionist.

Diet and mood

We are all familiar with people feeling like picking a fight, dancing on the table, or crying into a drink as being due to alcohol. What I propose is that there is an equally wide range of mood changes which result from a different group of substances in food - in a particular, that is, susceptible, subgroup of the population. Furthermore, just as it is possible for alcohol use to interact with depression or any other psychiatric disorder, it is also possible for food sensitivity to interact with these disorders.

There is a gradually accumulating body of evidence that some factors in food can affect mood. I will provide key papers from the literature and describe the phenomenon as it occurs in my dietetic practice, in which I specialise in the investigation of suspected food sensitivity.

Let me give you a description of my food sensitive patients. They usually present with ADHD, ADD, and allergic-type (atopic) symptoms. What I mean by "allergic-type symptoms" are symptoms that are usually seen as allergic; such as eczema, hay fever, sinusitis, rashes, diarrhoea, and asthma, and also symptoms such as headaches, migraine, tummy aches, irritable bowel syndrome, mouth ulcers and car sickness. These allergic-type symptoms often occur in families susceptible to food-sensitivity.

Where did the idea that diet could affect behaviour originate?

The idea that diet can affect behaviour has been around for hundreds of years, but it was only relatively recently that such wisdom was subjected to scientific study. Early in this century there were reports of behavioural, or neuropathic, changes due to food allergy [Shannon Duke and Rowe]. Schneider in 1945 implicated allergy in childhood hyperkinesis. In the 1950's, Speer included behaviour changes in children at the hypoactive end of the spectrum, in an allergic tension-fatigue syndrome. In 1970, Kittler reported improvement in Minimal Brain Dysfunction in children when allergens such as milk and chocolate were excluded.

The work that caused the most controversy was that of Feingold, who reported in 1973 that when his patients were placed on a low-additive, low-salicylate diet, the parents reported that they "became docile, better adjusted to the home environment ...". He went on to hypothesise that additives and salicylates caused hyperactivity. He also reported that diet reduced aggression.

The early research, up until 1985, refuted his hypothesis. However, later research, with improved methodologies, showed that there was a susceptible group of patients who did react. The detail of this research is presented in my 1997 review paper in the Australian Journal Paediatrics and Child Health.

A report on adverse reactions to additives by Lessof in 1987 in the Journal of the Royal College of Physicians of London cautioned about the possibility of food phobia occurring. Lessof's position was that individuals could develop phobias about certain foods, which may account for perceived adverse reactions. Certainly, psychiatrists deal with phobias regularly. However, in diet therapy I rarely see them. In fact, my patients are usually keen to try reintroducing suspect foods as much as possible, to see what they can tolerate. Where caution should be highlighted, I believe, and this is still rare, is in anorexics using suspected food sensitivity as a reason for food exclusion.

Why is mood worthy of being singled out for attention?

Mood was not singled out for attention initially. There are different strands to the research - that on ADHD, that on allergy and some on depression.

1. Research on allergy.

Those researching allergy reported changes in physical allergic symptoms. Where irritability was associated and it resolved, its resolution was assumed to have been due to the mitigation of the distress of the physical symptoms. The direct question of whether allergic exposure can provoke psychological symptoms in adults was addressed in a double blind study of sublingually administered allergenic extracts by King (1981). He notes that one usual explanation of psychological symptoms is that such reactions are psychosomatic, the second is that they are an effect of the somatic symptoms, and the third, confirmed in his study, is that allergic exposure may be directly responsible for both psychological and somatic symptoms. In fact, people can have symptoms related to mood, without, or in addition to, physical allergic symptoms.

2. Research on ADHD

When early workers researched Feingold's hypothesis from 1975 to 1985, hyperactivity was emphasised. Later research groups developed their work over time, researching a variety of behavioural and physical symptoms, as well as a broader view of the diet. Mood as a separate factor was first reported by Rowe of Melbourne in 1988. As Rowe and Rowe did not feel that the rating scales used for medication were as applicable to diet they developed the Rowe Behavioural Rating Scale [The RBRI] for diet research. They emphasised irritability, restlessness, settling and sleep problems as part of symptoms associated with additive colour challenges.

Kaplan's team in Canada [1989] used the Conners scale and added extra issues noted by parents. These included whininess, sleep and physical symptoms. They reported in 1989 that a decrease in night awakenings correlated with a decrease in ADHD behaviours on diet, but that physical symptoms did not, showing that difficult behaviour did not correlate with physical discomfort. On the other hand, Bock and Atkins [1990] did report irritability in children in whom abdominal and cutaneous reactions developed. They had selected atopic subjects and had not used any separate behavioural rating scales.

In my early research of 1978, I reported that symptoms of anxiety and tearfulness in the presenting ADHD child, and in siblings, as well as aggression and poor socialisation, improved in addition to hyperactivity. In the follow up study of five hundred families [1991] I, like others, reported that diet changed other factors as well as hyperactivity. I had categorised areas of improvement as; behaviour, learning, activity, social [having few friends, not being invited to birthday parties], sleep problems and physical symptoms. Rather than the "all or nothing" reaction described by Feingold, I found that diet was, "aggravating the underlying predisposition in susceptible children".

My most detailed study, following up 120 children, was in the early 90's. I used the RBRI, and found that the symptoms most improved on the diet were irritable, touchy, cranky, and reported that these children were hyper-reactive as much as hyperactive.

The following is an overview of the types of mood-based reactions that, over the years of my research, parents have described to me personally, and coded on questionnaires such as the RBRI. In food-sensitive children, diet has had a significant effect on all these symptoms.

Many children are reported as impatient, even more are irritable, touchy or cranky. Parents often say, "living with him is like walking on eggshells, one minute he's okay, and the next his mood has changed completely!" Many are described as having a chip on their shoulder - sometimes a log! The questionnaire does not include the word angry, but children are often comfortable with using it. When asked how they feel on the diet, some children will say, "I don't feel angry any more". I have also heard several variations on a teenager's clear statement that, "whenever I have red cordial I feel angry and just want to be by myself!" "He's so agro", even "cruel" are words used when coding aggression. I have discovered that when parents say their child goes "hypo" after a diet reaction, it is a mixture of hyperactive and aggressive!

Most patients are reported as excitable and this does not change with diet. However, the diet responders are reported to "come down faster" after any stimulus.

"Arguing about everything!" is a frequent comment about children over the age of seven and getting past the tantrum stage. Sometimes parents comment that, "he acts as if he wants to pick a fight". Parents will unhesitatingly pick the extreme end of the scale when responding to such questions, and sometimes even ask if they can code, "off the page"!

When we come to the question on being happy the parents often stop to think for some time. They comment that the child is not really unhappy. They recognise that it is unusual for them to be so emphatic about how irritable the child is, and yet they would not code him or her as unhappy. However, a subgroup are described as unhappy or 'cries often'. In a very small number, "I want to kill myself" was said often before diet and parents report it is said no longer in the diet responders.

Many do relate well to others, some do except at home, and even those who relate well may also be exasperating! Parents will say, "I love him but sometimes I could murder him!"

Easily frustrated is often noted, as is difficult to reason with. Parents generally use phrases like "off with the fairies", or, "on another planet", and they might comment he is "off his face" when having a bad reaction. Aimlessness is connected to being bored, with the message from the child being that more interesting stimulus is desired. It may also reflect the level of irritability.

Preschoolers are reported to have frequent tantrums – "they never grew out of the terrible two's!" They do decrease in the diet responders. Parents often say, "Joan, he runs just as fast, but I can get through to him now, and he is easier to live with".

To return to other work on ADHD being published by the 1990's: Egger and Carter and their London team reported in 1991 that, "symptoms showing change are not the attention deficit that is considered the core of the ADHD but rather irritability" and that "children had become more manageable and more amenable to reasoning rather than less active or better able to concentrate". This sounds rather like Feingold's initial report, doesn't it? In a double blind study in Melbourne in 1994, Rowe reported that behavioural changes in irritability, restlessness and sleep disturbance are associated with the ingestion of tartrazine in some children.

3. Research on depression.

The third strand to the research was not related to ADHD, but to atopic [having allergic symptoms] families. A child psychiatrist who visited Queensland in 1999, Mariannne Wamboldt, has researched and published on the role of atopy in depression. She reports that there is a shared genetic risk for atopic and depressive symptoms. This is interesting particularly because my research has indicated that there is a correlation between those coming from atopic families, and those who are likely to be food-sensitive. Wambodlt has completed the chain of thought suggested by the above research. By linking atopy and depression, we can see that atopic symptoms are a common factor that links mood and diet. Indeed, for many years earlier I had been asking my patients to fill out a "family sensitivity history" of atopic symptoms and suspect causes. It is not unusual for depression to be mentioned as a family problem.

How does mood come up during the diet therapy process?

In my work I am aware that I teach, particularly children, to be aware of and monitor their mood. Diet therapy for suspected food sensitivity in children begins by running a trial individual elimination diet, "to see if the child can handle himself better". After the four-week trial, various categories of foods are reintroduced as challenges to see what changes. I call this "being a diet detective". I am happy for children to prove they can handle themselves rather than having to stop eating the gradually increasing challenge foods. I am happy to teach children that if they feel cranky, but don't get cranky, then they can still have the food [providing their sleep, physical symptoms and ADHD symptoms are also not returning]. I often tell the story of the bright early teen who said, "you mean 'fake it?". Big brother could manage that, but little brother could not. He insisted he wasn't reacting. He said things like, "it was big brother's fault, he started it", "the teacher is always mean to me!", and, "you are always asking me to do more than my brother". I use this in what I call "the Three-Pain Rule". If in the space of a few hours, the child complains that three people, such as the teacher, another child and his mother, are "a pain", when none were the week before, then he or she has to stop the trial food, and "maybe trial the food again next year when he or she may feel less touchy. An adult expressed the level of touchiness well by saying that before diet, and after challenges she, "not only felt annoyed with people, but felt she had good reason!" In saying this I am reminded that it is the improvements in this mood component, and even more important, the behaviours resulting from the touchy mood, that are particularly important to the parents.

As my patient group becomes broader, different groups with mood effects have emerged. In those who respond to diet, these problem areas all decrease. They may include:-

1 Happy high and silly - likeable, flighty, distractible, often talkative children.

2 Touchy people - Outward directed. If all is not well it is because someone else "started it!", "It is all their fault", "They are always picking on me"

3 Supersensitive folk - inward directed. They are more likely to say, "I'm no good, I'll never have any friends, I'll never be able to do this, I'm not good looking."

If we look at all of these groups we have a fair representation of the general population. It is possible that the diet has a general effect on mood (regardless of the diagnostic group), as well as a specific effect on certain symptoms of atopic people. Indeed, it is possible to postulate that the ADHD presentation in the research groups of dietary responders, may be secondary to a primary mood variation affected by food in susceptible people.


Breakey J. Review article The role of diet and behaviour in childhood. Journal of Paediatrics and Child Health 1997: Vol 33 No 3 190-194.

King DS. Can allergic exposure provoke psychological symptoms? A double-blind test. Biological Psychiatry 1981;16:3-7.

Wamboldt MZ. Multiple atopic disorders, adult depression may be linked. Am J Genet 2000:96: 146-49

Breakey J. Are you food sensitive? How to investigate your own diet. 1998 CE Breakey (Medical) Pty Ltd. Brisbane Australia.

Breakey J, Connell HM, Reilly C. The role of food additives and chemicals in behavioural, learning, activity and sleep problems in children. In: Branen AR, Davidson PM, Salminen S, eds. Food Additives New York:Marcel Dekker, in press.

Link to: | Order "Are you Food Sensitive?" | Book Reviews | Back to Home |

This site was last updated on: 7 October 2001