Development of thinking in food chemical intolerance – 1975 to 2005 Joan Breakey
Presentation to the Dietitians Association of Australia 2006
Many people think that the role of additives and other food chemicals in conditions such as hyperactivity is confusing. This is reasonable as the claims, the research, and information available have all changed over time. Clear proof of a link between diet and attention deficit disorder [ADHD] was a long time coming, and not neat and tidy, compared to other treatments where a particular drug is known to affect a particular neurological pathway. Attitudes, both of the public and of the scientific community take a long time to change, even if the connection between food components and behaviour is now proved.
Dietary investigation of adverse reactions is also confusing as it is not a usual diet dealing with the nutrients we are familiar with: proteins, carbohydrate, vitamins and minerals. It is in fact, closer to medicine in that it concerns itself with the pharmacology of food components. It deals with some non nutrient substances in food and their effects on people.
We are all familiar with alcohol and caffeine in food affecting people, but it was a new idea there may be other food components that could have a pharmacological effect. We do at some level think about spicy food as stimulating to the senses, and expect to get a lift from chocolate, or some new drink or confection that is “the new sensation!” We know that dose matters with alcohol and caffeine. A little can be great but too much is detrimental. We also know that the amount of alcohol or caffeine that produces unpleasant symptoms varies with individuals, and that drunk people may exhibit different behaviours from each other.
Where chocolate, spices and other stimulating food chemicals are concerned, the same principle applies. Small amounts are enjoyed and tolerated in the majority of people. There is a degree of self regulation as we all know people who limit the amount of spice they eat. However it is still a new idea that the large amounts that people are able to enjoy in modern society are a significant problem to some. Another new idea is that the effect is due to the cumulative effect of several different stimulating compounds. As well it is interesting that the suspect chemicals are a problem only to a susceptible segment of the population.
It is important to realise that the thinking about the role of diet has evolved with the passage of time with regard to the diet itself, the symptoms it changes, the susceptible group, and the biochemical mechanism involved. As a dietitian I was in the fortunate position of being able to study the effects of the new dietary intervention as a treatment, research its role in my patients as well a in the scientific literature. This has enabled me to show how the thinking developed and changed over the last thirty years. There have been some surprising outcomes.
Development of the thinking with regard to exclusion of dietary factors
The concept that diet could affect behaviour is not new. As early as 1922 behavioural or “neuropathic” changes due to food “allergy” implicating milk and chocolate, had been described, and reports continued throughout the century. The main controversy began in 1973 when an allergist named Ben Feingold said that diet caused hyperactivity. He implicated artificial colours and flavours, just two preservatives BHA and BHT, and similar natural substances called salicylates, present in most fruit especially tomatoes. He used data from an analysis done in 1922! Monosodium glutamate [MSG] was mentioned but not emphasised. His K-P diet was clearly defined and so appeared easy to test. However his exclusions raised questions about other additives and other foods thought to contain salicylates.
In 1974 Dr Joan Woodhill, an Australian dietitian researching the diet also excluded all essences, including vanilla and lemon oil, all essential oils, all herbs and spices, coloured stock cubes and soups. Additional fruits excluded were blackcurrants, redcurrants, grapefruit, lemons, capsicum, coconut, and melons. In 1976 she further excluded all stock cubes, chocolate [drinks were still allowed]. Non food items she also found needed exclusion were perfume and fly spray, petrol, and felt pen smells.
I began investigating diet in hyperactive children in 1975 using Joan Woodhill’s diet. In 1977 I reported reactions to milk, wheat, potatoes, bananas, zucchini and peanuts in some individuals, as well as coloured chicken feed via eggs. Around that time Clinical Ecology, the study of adverse reactions to all aspects of the environment e.g. gas and petrol fumes, had become established, broadening the environmental substances that were suspect.
In 1977 Dr Richard Mackarness, a British Psychiatrist, published a book called ‘Not all in the mind’ describing psychiatric symptoms due to intake of any grains. He advocated a “cave man diet” excluding them. In correspondence after that I had a letter from Feingold saying that it was additives not grains that produced improvement with exclusion, and another from Mackarness saying it the other way around!
I had a food company make up a “nature identical” mango flavoured, colour free soft drink. Unfortunately, and to their disappointment, families reported it producing adverse reactions. I also learned that flavours are added to foods in ten times the dose of colours! This had important implications for the initial research into the Feingold hypothesis which only used dye in challenges.
In 1980 Minchin bought out a book, ‘Food for thought’, implicating milk as the source of many childhood problems, and also in that year Schauss implicated sugar in his book ‘Diet Crime and Delinquency’. He also implicated poor nutrition and encouraged megavitamin therapy. Neither sugar or nutrition proved relevant.
Woodhill had shown that Feingold’s diet needed modification. So, from the earliest days, as a dietitian, I was asking if the diet was right. I instructed parents to regard it as they would any new treatment and give me feedback. We called diet investigation “Diet Detective Work”. Along with Dr Joan Woodhill and some NZ researchers, I found that there was individual variation in tolerance to the Feingold exclusions, as well as reports of reactions to chocolate and milk, corned beef, pineapple juice, monosodium glutamate [MSG], and anything that smelled strongly, from house paint to jasmine flowers. Diet therapy had to incorporate the concept of individual variation in tolerance to various food chemicals, such as amines, and to whole foods, such as milk, to which children may be allergic. With the new addition of allergy an awareness that inhalant allergies could be relevant made parent reports of increases in adverse reactions in pollen season more explicable. With the feedback from parents the diet sheets came to contain allowed and disallowed foods, as well as use occasionally and use very occasionally foods.
The diet began to look more neat in the 1980’s when Dr Ann Swain’s work on the salicylate analysis of Australian foods was published. New exclusions by Swain were all amines, [not just chocolate], all preservatives, MSG, and in children with gastrointestinal symptoms, milk, wheat and rye.
Some fruit, e.g golden delicious apples and lemons became allowed, and others e.g pineapple and peaches were excluded. However, while analysis should be used as a guide to tolerance, families reported reactions to some fruit listed as very low in salicylate, and tolerated others listed as higher.
Around the world various research teams continued research so that each added to the knowledge of the suspect chemicals and foods implicated. All based their research on the exclusion of suspect additives and added what they saw as relevant. The Canadian team also excluded whole foods which the family identified as causing reactions. In 1978 work in London implicated whole foods as much as additives with all the foods usually implicated in allergy shown to cause behavioural reactions. These included milk, eggs, peanuts, fish, shellfish, soy, and tree nuts. Additionally fruit was reintroduced to show individual tolerance. Reported tolerance did not reflect salicylate content.
Given this dynamic milieu it was wise to have had a Diet Detective approach. The thinking changed greatly over time, and any idea that “The diet” could be reduced to two sheets of paper was overly hopeful. I also developed the concept of the “total body load” as a way of incorporating all the factors, yet letting families know that there were individual differences in relevance to them. After years of diet detective investigation it became clear that one important feature of suspect foods was that they usually were the more aromatic foods. In fact the more flavoured the food the greater the likelihood of it causing reactions. Suspect foods include flavoured sweets and drinks, spices, teas, acidic fruit and juices, peppermint, chocolate, sauces, aged foods, strong cheeses, and MSG which is a flavour enhancer. Flavour itself should now be one of the suspect chemicals. This is consistent with the knowledge that non food smells, such as flowers and paint are also implicated. A review of the literature on amines supports their exclusion with the addition of unintentional amines: those which develop as any food smells “stale” or “off” to the sensitive person. Fortunately the diet allows unlimited fresh plain foods.
What are the factors implicated now in 2005? Suspect chemicals, foods and factors which contribute to “The total body load” include:-
Additive colours and flavours – both natural and artificial,
Natural and medicinal salicylates
Natural amines – both intentional as in aged foods, and unintentional as those smelt as “stale” or “off”
Natural and added MSG
Smells – flowers, perfumes, bubble bath, paints, petrol, chemicals,
Skin contact with dyes –e.g. finger or face paint, hair dye, and also salicylate rubs,
Whole foods commonly implicated in allergy – where they occur in the family,
Inhaled allergens – also where they occur in the family,
Temperature change / seasons
Infections – viral, bacterial and fungal
Note that addition of vitamins and minerals have not been shown to change tolerance.
This paper will not cover the dietary management. See Are you Food sensitive? and, if possible, the professional input of a dietitian should be sought. Overall it can be said that the initial elimination diet minimises all the above factors emphasising the known family sensitivities and incorporating information on whole foods and inhalants individually. After a month patients can reintroduce exclusions one per week to test tolerance and determine individual tolerance both to excluded factors and the dose managed.
Development of thinking with regard to symptoms diet changes
In the 1970’s hyperactivity and activity were emphasised, with some discussion of aggression. Remember that even the idea that diet could have any affect on behaviour was controversial then.
In my early work parents reported decreases in activity and also that the child was “nicer to live with”. As well they reported decreases in anxiety, tearfulness and bedwetting. Even more interesting were reports of decreases in physical symptoms such as eczema, rashes, diarrhoea, headaches, tummy aches, and hypo-activity [lethargy], in the presenting child or immediate family members. This made sense when it was realised that the diet Feingold used had been used by dermatologists, especially in the treatment of urticaria.
Many other researchers were also broadening their assessment with attention and learning aspects receiving more emphasis. In the 1980’s a Sydney Royal Prince Alfred Hospital team researched changes in headaches, migraine, mouth ulcers, IBS [Irritable Bowel Syndrome], as well as eczema, rashes, and ADHD [attention deficit hyperactivity disorder]. An additional area of change was reported by a Canadian team. It was halitosis, that is, bad breath, which was also being reported, along with other strong body odours in my patients. Most of the physical symptoms reported to decrease were what would be described as allergic, but some, such as migraine, were non allergic. The London team also reported diet helping the above symptoms as well as epilepsy, and I noted that where reflux was present in a food sensitive family it improved with the other symptoms.
By the 1990’s there were more surprising results. The area many reseachers found as most changing with diet treatment of ADHD was mood. Research using before and after questionnaires showed that symptoms such as irritable, touchy, cranky, and wingy or whiney were the symptoms that changed most. The main ADHD symptoms of restlessness, poor concentration and impulsivitity decreased too. Other problems that decreased were poor concentration, impulsivity, unreasonableness, restlessness, argumentativeness, tantrums, uncontrollability, aggression, sleep problems and excitability. The detail of all the research can be found in my Review article published in the Journal of Paediatric and Child Health, Vol 33 in 1997.
Parents reported that children whose mood changed quickly improved, and even when they got very upset when on the diet they were more easily calmed. Those who had been described as “happy-high” or “off with the fairies” were easier to get through to. Both those with outward directed talk such as “It’s his fault, he started it!”, to those with inward directed talk such as “I’m no good, I’ll never be able to do it”, changed more towards the normal range. Anxiety and depression symptoms in this group were reduced with some parents even reporting reduction in statements such as “I wish I were dead!” If you transpose these ideas to adults you can see why parents or adults who began diet investigation for allergic symptoms or migraine could report less anxious and depressed feelings.
In summary it can be said that diet factors “aggravate the underlying condition” whether this is ADHD, ASD, physical symptoms, attention, sleep or mood, and that this happens only in those who are susceptible.
Diet treatment is not dependant on what diagnosis patients have, as most diet treatments are, but on susceptibility to the cluster of symptoms mentioned here. ADHD symptoms do change, but irritable, touchy, cranky change most. Excessive projective thinking, and excessive introspective thinking were reduced.
Development of thinking with regard to the susceptible group.
Ideas developed over time in this area as well. In the 1970’s research about suspect food chemicals and additives was limited to the hyperkinetic syndrome. By the 1980’s it had expanded to those with atopic [allergic] symptoms. However diet responders also included those with non-atopic migraine and IBS. In my research ADHD children with an atopic family history, and those who could report seeing a definite reaction to a food in the past, especially if it was chocolate, were more often in the responder group. Another interesting phenomenon was the increased incidence of “supertasters” and “supersmellers”: that is those with greatly increased sense of taste and smell, usually known as “fussy”, in the diet responders. Families also report that diet responders often had higher breath or body odour which often improved on diet.
Diet responders are not an easily defined group. Food chemical intolerance, like allergy, is a multi-system disorder which runs in susceptible families. Families which contain atopic symptoms as well as migraine, headaches, IBS, mouth ulcers and carsickness are more likely to be food sensitive. Note that recent research has shown that there is an increased likelihood of depression occurring in atopic families, so this should be included in symptoms investigated. However there are some chemically sensitive individuals who have no family history of the usual cluster of symptoms. It is useful for families to fill out a “Family sensitivity history” of physical, attention and mood symptoms with suspect foods or smells. Each family has its own picture of symptoms and suspect items, in both sexes. In fact different family members may be attending different doctors for a variety of symptoms, such as ADHD, migraine and allergies, so that none realises that the overall picture shows that the family could be food sensitive.
Overall a patient is more likely to be food sensitive if she or he has seen a reaction to something eaten, comes from a family with the cluster of symptoms mentioned above, or a sensitivity to salicylates [usually aspirin], and notes increased sensitivity to smells both in food and in the environment. It must be said that at this time the only way to know if someone is food sensitive is to run an elimination diet for four weeks and challenge for up to seven days with excluded suspect chemicals.
Development of thinking with regard to the biochemical mechanism.
By now it will be no surprise that, since the chemicals needing exclusion were not clear and the susceptible group is ill defined, the thinking with regard to the biochemical mechanism was also not neat and tidy. Feingold had hypothesised a toxic mechanism but that was disproved. The public health problem of whether the suspect additives should be banned had been solved. Additive colour, flavour and preservatives did not need to be banned. But even the research that disproved the hypothesis also showed that some children did react. Researchers could see that there was an allergy component so the immune system was involved. But there was also an idiosyncratic component as different individuals reacted to different chemicals. It seemed to be like an adverse reaction to a medication whereby most people benefit from the desired effect, but some have adverse reactions. The other issue that was becoming clear was that the susceptibility did run in families.
Reactions showed features that suggest aspects of the biochemistry. There were cumulative effects whereby the reaction occurred after intake of several different suspect substances. Cravings were often reported. In fact I used the presence of cravings as an indicator that I had not worked out all that needed to be eliminated in the early days of diet investigation. There are withdrawal symptoms when the elimination diet is begun. These happened earlier and were clearer in very young children. Reactions are dose dependant. The threshold changes depending on the additive effect of different contributions from the total body load of problem items for each individual. When a reaction to colours, flavours, benzoates or salicylates occurs the effect is reduced with the use of urinary alkalisers which increase their excretion.
Most suspect chemicals have a chemical structure that is described as aromatic. As such most have a smell. Various mechanisms have been proposed. (See my thesis for a review.) One idea was that a leaky gut allowed chemicals to be absorbed. However the gut is not always dysfunctional. Also the use of gut enzymes does not change tolerance. The problems are not related to any nutritional deficiency. Both well nourished and poorly nourished people react, and supplements do not improve tolerance.
What is the most likely mechanism at this time? Most suspect chemicals are aromatic. Many food sensitive people have higher than usual body odour whether as bad breath, high urine or faeces odour or a greater need for deodorants. It is known that salicylates, additive colours, flavours, benzoate preservatives, and also amines are broken down in the body by enzymes called sulphotransferases, with different ones primarily breaking down salicylates and amines. Research on food sensitive autistic children found evidence of problems in the metabolism of phenolic [aromatic] compounds. They found that the amount of sulphur in the plasma and being excreted differed greatly from the normal population. So the clinical research and the toxicological research both point to some inborn error of metabolism, whereby sulphur containing enzymes are not functioning as well as in the normal population. Just as our biochemistry was not geared for the high energy low activity of our modern lifestyle, it probably was also not geared for the high dose of flavours currently eaten, particularly in some individuals.
In conclusion it can be said that sensitivity to suspect chemicals in food “aggravates the underlying condition” in susceptible people. Food sensitivity should be considered in medicine as part of a holistic approach, considering ADHD, ASD, allergic and other symptoms as well as mood changes. The thinking with regard to the diet itself, the symptoms it changes, the susceptible group, and the biochemical mechanism, have all evolved over the last thirty years, and will continue to do so in this challenging area of scientific investigation. Dietary investigation should be used, preferably using the book Are you food sensitive? and the additional care of an experienced dietitian, and monitored as it applies to each sensitive individual.