ADHD is estimated to be found in 5–10% of the school age population [345,000–690,000 in England (Taylor et al., 1991)]. Poor concentration affects learning, and the symptoms of ADHD increase the risk of disruptive behaviour, interfering with learning in children with ADHD and in their classmates.
Studies selecting children with formal ADHD diagnoses are outnumbered by studies examining more general behavioural problems, which may be termed hyperactivity. Extrapolating the findings of the latter studies to children with ADHD needs to be done with caution.
Food additives. In the 1970s, allergist Benjamin Feingold in San Francisco claimed a ‘30–50% benefit’ in young patients with hyperactivity eating a diet free from certain artificial flavourings and colourings (Feingold, 1975). Controlled trials are now supporting these anecdotal findings. A meta-analysis (Schab & Trinh, 2004) examined 15 placebo controlled trials addressing the effects of artificial food colourings on the behaviour of hyperactive children (average age across the studies 7.9 years, some with formal ADHD diagnoses). This found that artificial food colourings promoted hyperactivity in hyperactive children.
Two more recent studies examined community based samples of children, rather than clinical samples. The first study (Bateman et al., 2004) concluded an adverse effect of food colourings and preservatives on the behaviour of 3 year olds, an effect not influenced by the presence or absence of prior hyperactivity. The second (McCann et al., 2007) found that that the addition of food additives to the diet of both 3 year olds and 8/9 year olds selected from the general population led to increased hyperactivity.
Studies of exclusion diets, where children only eat a limited range of ‘natural’ foods have suggested that such diets may reduce symptoms of hyperactivity. A recent randomised controlled trial of children with ADHD (mean age 6.2 years) given a supervised elimination diet demonstrated improvement in behavioural symptoms in 70% of children. (Pelsser et al., in press). The study was small, however, involving only 27 children and it was clearly not possible for families to be unaware that their children were in the ‘exclusion diet’ group.
Other researchers have examined the effects of restrictive ‘oligoantigenic’ diets (Egger et al., 1985; Schmidt et al., 1997) or ‘few foods’ diets (Carter et al., 1993). The studies have systematically reintroduced excluded foods after a period of limited diet. The aim is to identify items in the diet which provoke behavioural problems, and the studies suggested that some children will benefit from individually tailored diets eliminating some foods or additives. This study noted that it is not just artificial additives that can lead to behavioural problems.
The Food Standards Agency has produced guidelines recommending that eliminating certain artificial food colours and the preservative sodium benzoate may reduce hyperactivity (Food Standards Agency, 2007b). By contrast, the most recent draft of the UK National Institute for Health and Clinical Excellence guidelines on ADHD management (NICE, 2008) specifically state that ‘eliminating artificial colouring and additives from the diet is not recommended as a generally applicable treatment for ADHD’. The guidelines recommend referral to a dietitian where food or drink appears to affect behaviour, and ‘further management (such as elimination of specific foods) should be jointly undertaken by the dietitian, mental health specialist or paediatrician and the family’.
Sugar in the diet has been linked anecdotally to childhood hyperactivity, particularly by parents. Induced hypoglycaemia in non diabetic adults (McAuley et al., 2001) was found to cause attentional dysfunction without altering non verbal intelligence. An abrupt fall in blood sugar levels following an insulin surge responding to sugar consumption may in theory replicate this effect. However a meta-analysis of 16 studies (Wolraich, Wilson & White, 1995) examining the effects of sugar on behaviour or cognition in children concluded that there was no clear evidence of either being significantly affected by sugar intake.
Essential fatty acids. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are fatty acids found in small quantities in the normal human diet and are known as omega 3 fatty acids. They are precursors to brain hormones which function in the release of some neurotransmitters and are important for neuronal membrane flexibility. The human body can convert another fatty acid found in a wider range of foods, alpha-linolenic acid (ALA), to EPA and DHA. This conversion process is inhibited by the presence of saturated fats (present in high quantities in processed food). It is also inhibited by reduced dietary co-factors (including zinc, magnesium, manganese, vitamins A, B3 and B6), alcohol, smoking and diabetes. Diets favouring processed foods, which are high in saturated fats and low in micronutrients, may lead to low EPA and DHA levels. Low dietary levels of fatty acids can be increased by dietary supplements containing omega 3 fatty acids (EPA, DHA) or by eating oily fish, the main dietary source of EPA (Richardson, 2006).
Colquhoun and Bunday (1981) observed that children who were hyperactive (although not necessarily diagnosed with ADHD) had similar clinical symptoms to children with essential fatty acid (EFA) deficiency (excessive thirst and urination, dry skin or eczema and asthma). Mitchell et al. (1987) found that some children with ADHD did have significantly lower serum fatty acid concentrations than control subjects and hypothesised that a subgroup of children with ADHD may have altered fatty acid metabolism, as their dietary intake was not necessarily lower than that found with control subjects.
Recent theory (Hallahan & Garland, 2004) suggests an inverse relationship between plasma omega 3 levels and behavioural disturbance in impulsivity disorders, which includes ADHD. Colter, Cutler and Meckling (2008) studied 11 adolescents with ADHD with an age matched control group, and found the ADHD group to have lower blood levels of DHA than controls, despite equivalent dietary intake. The authors suggest that there may be differences in fatty acid metabolism between the two groups irrespective of diet, a possibility which warrants further research.
Supplementation studies show mixed results (Hallahan & Garland, 2004). A study of 8–12 year olds with ‘ADHD related symptoms’ (Richardson & Puri, 2002) claimed better results using EPA with DHA in supplements given for 12 weeks. The participants, selected from a school for children with literacy problems, did not have formal ADHD diagnoses.
The UK Oxford–Durham Study (Richardson & Montgomery, 2005) gave essential fatty acid supplements [EPA, DHA and GLA (Gamma-linolenic acid, an omega 6 fatty acid)] to children with developmental co-ordination disorder (DCD) as reported by their teachers. There were no significant treatment effects on motor skills. As a secondary finding the group receiving dietary supplements showed statistically significant improvements in hyperactivity symptoms and in reading and spelling abilities compared to the placebo group. Formal ADHD diagnostic assessments were not carried out, although hyperactivity levels were high at baseline; nor were children assessed for baseline deficiency of fatty acids.
Stevens et al. (2003) gave fatty acid supplements (predominantly DHA) to children with behaviour associated with ADHD. The results did not show clear improvement in their behaviour. In another study (Hirayama, Hamazaki & Terasawa, 2004) some of the children did have a diagnosis of ADHD, but again DHA did not improve ADHD related symptoms, a conclusion shared by earlier researchers (Voight et al., 2001). Sinn and Bryan (2007) saw some improvement in parent rating scales of behaviour when 7–12 year olds with behavioural problems received fatty acid supplements, but teacher rating scales were unaffected. A more recent study (Sinn, Bryan & Wilson, 2008) of 8–12 year old children with ADHD described some improvement in attention with polyunsaturated fatty acid supplementation but micronutrients (multivitamins/minerals) were taken in addition to the fatty acids. The results obtained from these studies cannot provide unequivocal evidence of the effectiveness of fatty acid supplementation in ADHD.
Itomura et al. (2005) gave fish oil (predominantly DHA) to 9–12 year old children with physical aggression and impulsivity, without clear evidence of benefit.
The omega 3 supplements used in the above trials led to few reported adverse effects. Some have been reported elsewhere, such as decreased glucose tolerance, diarrhoea, abdominal pain, bloating, nausea, flatulence, fatigue, somnolence, thrombocytopenia, skin irritation, vitamin toxicity and oxidative damage in large doses (Radia, 2005). There have been concerns about the heavy metal content of fish oils, particularly of mercury which is known to have potentially neurotoxic effects. This is a particular problem with fish caught close to shore, so fish oil obtained from ocean fish is recommended where used.
A topic barely covered by research to date is whether fatty acid supplementation leads to effects upon the behaviour and educational performance of school aged children who do not have preexisting behavioural or attentional problems.
There is some evidence for iron deficiency being associated with ADHD (Konofal et al., 2004) and beneficial results have been claimed for iron supplements but with inconsistent results in trials (Sever et al., 1997).