These notes summarise continuing research evidence concerning the postulated link between the onset of autism and the MMR vaccination. Converging views point to the lack of evidence for this link, although the issue of particular vulnerability to negative environmental factors such as food additives or pollutants is noted.

The significance of diet is further explored with claims made for the potential benefits of the exclusion of certain elements.

Dysfunction in the prefrontal cortex is highlighted as a specific source of certain autistic impairments within the category of "mentalizing" ; and the final section summarises a survey of practices concerning diagnosis and provision for young children with autism, with implications for "adhocracy" in provision, and for the lack of any one intervention that will be appropriate for all the children affected.

M.J.Connor April 2001


Incidence of Autism ( and the [non]significance of the MMR vaccine )

There is converging evidence that there has been a marked increase in the incidence of autism (or in the frequency with which autism is diagnosed) over the last 10 + years. Much debate continues about the possible role of the MMR vaccine in the aetiology of autism, but the trend analysis carried out by Kaye et al (2001) strongly supports the view that this vaccine is not linked with increased incidence.

These authors examined cohorts of children born in the UK during the years between 1988 and 1993, as well as identifying a sample of children who were diagnosed with autism between 1988 and 1999 when they were 12 years of age or younger. They were able to confirm that there was a clearly observable increase in incidence ..... from 0.3 per 10,000 person-years in 1988, to 2.1 in 1999.

However, it was also demonstrated that the prevalence of the MMR vaccination had remained virtually identical for each of the successive birth cohorts, with the implication that exposure to the vaccine cannot explain the rapid increase in the incidence of autism.

The authors suggested that this increase may be largely explicable in terms of the greater recognition of various levels of autism such that children with mild forms of disability are now described as falling within the autistic spectrum and are, therefore, included within the autistic incidence rates.

Nevertheless, it was further acknowledged that there may be environmental factors which are producing a real increase in the incidence of autism, particularly in respect of the more severe forms.

A similar conclusion was drawn by Dales et al (2001) who analysed data relating to children in California, and compared the number of cases of autism reported during the years 1980 to 1994 with the rates of MMR vaccination over that same period. This analysis indicated that there was a sharp increase in the incidence of autism over these 14 years, but the percentage of children who received the MMR vaccine showed only a small increase during this time. Further, it was found that the rapid increase in cases of autism began before the small increase in the rates of MMR vaccination, and that the increase in cases of autism was maintained even after the MMR vaccination rates had stabilised and remained largely constant year by year.

These findings further undermine any suggestion that the increase in reported autism can be linked to the use of the MMR vaccine. The authors accept that it remains unclear whether MMR vaccination might be associated with a tiny proportion of autism cases, but they emphasise that there is no correlational evidence from their study to show that the vaccine is a factor in autism.

Yet another recent study has argued that there does not appear to be any link between exposure to measles-containing vaccines and the development of inflammatory bowel disease. Davis et al (2001) analysed the vaccination histories of children born from 1958 to 1989 in 4 health areas, comparing those with diagnoses of colitis, Crohn's Disease, or proctocolitis against a large control sample. No association was found between the MMR vaccination and increased risk for inflammatory bowel disease ; children vaccinated before 12 months of age or between 12 and 18 months were at no greater risk than unvaccinated controls. Children vaccinated after 18 months actually showed a slightly reduced risk of developing this form of bowel disease.

Despite evidence of the above kind, there is still a core group of parents and researchers who maintain the view that the vaccine, when administered at a very young age, may trigger an autoimmune reaction which leads to the onset of autistic symptoms. Legal action is being taken against the manufacturers of the vaccine, with a large number of cases already in the legal system in the UK.

Kaye et al and Dales et al in an interview with Medscape (Hung 2001) continue to make the point that, if the MMR vaccine were a major cause of the increasing incidence of autism, the risk of autism in successive birth cohorts would cease rising within a few years of the vaccine's being in full use.

The dilemma is that the success of the MMR vaccination depends upon very wide coverage .... about 95% of the population needs to receive it if the spead of the diseases is to be prevented. Currently in the UK, there is a MMR usage which falls short of this figure (it is down to an average of around 88%, and as low as 75 % in some areas ) with the accompanying risk to children's health.

Meanwhile, there continue to be occasional research papers which reinforce the anxiety over the MMR vaccine. For example, the Medscape article cited above refers to a paper in the March issue of Archives of Disease in Childhood which suggests a link between the MMR vaccine and a skin reaction caused by leakage of blood into the subcutaneous tissues.

As a a further example, one might refer to the work of Jyonouchi et al (2001) who cited evidence for atypical immune responses among children and young people with autistic spectrum disorder compared to a group of normally developing controls, and who found that the ASD sample also showed a high incidence of adverse reactivity to dietary proteins including soy, milk, and wheat.

The authors suggested that children with ASD may be predisposed to other adverse reactions to certain foods, bacteria, and vaccines, and may have an increased risk for autoimmune reactions. (However, while there may be evidence to support parental claims that children with autism are more prone to illnesses such as colds and influenza and are more sensitive to dietary problems, one might comment that this is not to suggest that there is any causal relationship between such vulnerability and the development of autism.)


Dietary Issues in Autism

It would appear that there is a general acceptance that there has been an real increase in the incidence of autism over recent years, and it is commonly held (including in the above quoted papers) that environmental factors will prove to be relevant. Significant among environmental factors is diet and what goes into food production in the first place.

A presentation at an international conference ( Kniker 2001 ) described how the elimination of dairy products or gluten, and some other food components, can markedly impact upon the symptoms of children and adults with autism.

Kniker reviewed a series of earlier studies which showed that a large proportion of children with autism may demonstrate significant improvement when dairy products and gluten are removed from the diet, with the explanation that some poorly-degraded food proteins leak from the gut into the blood, and certain peptides can adversely affect brain functioning.

Findings from ongoing studies were reported involving a sample of 28 diagnosed autistic individuals placed in residential settings. Baseline data were obtained and behaviour monitored over a period in which dairy products, food colouring, grains, and caffeine were removed from the diet.

More than 30% of the sample were described as showing marked changes in their behavioural symptoms. 5 individuals showed major improvements in respect of their (behavioural) symptoms such as less disruptiveness, greater alertness and retentiveness, and improved sociability and cognition ; while the other 5 deteriorated. It was held that the latter finding may be explicable in terms of the negative impact of other foods which those individuals would not normally eat during the period when elements of their "standard" diet were not available.

The conclusion supported the view that the increase in the incidence of autism does implicate environmental factors and, while genetic background will have some influence, diet, vaccination, and pathogens may all have a role.

(Again, one might comment that there appears to be a core group of children who, for reasons which are not yet clear, are more vulnerable than normally developing children to infections, and negative environmental features including diet and pollutants. Such children may also show a heightened risk for autistic spectrum disorder. In other words, in some children with ASD, the environmental factors may somehow interact with this heightened suceptibility. It is not a matter of diet, or vaccine, or pollutant alone which underlie the autistic symptoms .... and the implication is further to explore how to identify elements of this hypothesised susceptibility at the earliestr possible stage).

The significance of environmental factors, including diet, is emphasied by Shattock (2001) who also expresses concern that the available evidence so far does not prove one way or the other whether the MMR vaccination is relevant to the marked increase in cases of autism over the past decade. It is his view that such a sharp increase, especially as it involves greater incidence of the more severe forms of ASD, is unlikely to be a matter of changes in diagnostic practice or greater sensitivity towards autistic symptoms.

Shattock explores what significant changes have been observable over the recent past, and he draws attention to the increased use of pesticides, changes in food intake, plastics in food and drink containers, vaccinations, pollutants including heavy metals, etc..

He also refers to the possible significance of illness during pregnancy, childhood illness, and food intolerances.

However, he acknowledges that one does not know to any degree of confidence what are the factors underlying a given child's vulnerability nor what are the triggering agents which may impact upon that child ....( and one might draw the implication that any pattern of endogenous and exogenous factors which prove significant may be unique for any given individual, hence the difficulty in making general assertions about aetiology ).


ASD and the Prefrontal Cortex

Shallice (2001) describes how knowledge about the organisation of the brain has been greatly enhanced over the last few years as a result of advances in techniques such as functional imaging ; and he refers to the flood of information on various types of cognitive task where there can be observed significant differences in activation in complex patterns across prefrontal regions.

The example is given of episodic memory which used not to be considered to be strongly related to frontal functions, but which is now seen to involve several anterior regions which are differentially activated according to subtle changes in the actual task.

However, reference is made to the continuing difficulty in specifying precise processes which link, for example, perceptual input and motor output, partly because the tasks which lead to prefrontal activation may involve a whole series of sub-processes. Shallice suggests that progress in understanding brain functions will be all the more enhanced if the above kind of information is supplemented by data emerging from studies of the effects of brain lesions.

The research completed by Stuss et al (2001) concerned Theory of Mind, or mentalizing ... the cluster of abilities which enable a person to recognise and understand the thoughts or feelings of others ... and it followed up the demonstration on the part of Baron-Cohen et al (1985) that such abilities are selectively impaired in autism, with implications for some specific brain basis.

Earlier studies investigating frontal lesions, such as that of Stone et al (1998), suggested that this impairment in Theory of Mind was linked to damage in the inferior medial area ; and a more recent study by Channon and Crown (2000) found a left frontal effect in understanding Theory of Mind stories.

The study by Stuss et al uses 2 tasks, both requiring the participants to make an inference about the location of a toy which they cannot see from the direction in which the experimenters point.

In the first task, 2 experimenters are involved, and they point to different places. The participants need to recognise that, as they are sitting directly alongside one experimenter, both they and this person cannot see the toy, so that it is appropriate to rely upon where the other experimenter is pointing. Participants with frontal lesions produced a much higher error rate on this task, with implications for the significance of the frontal lobes in mentalizing.

In the second task, there is an element of deception in that the one experimenter points to the wrong position, and impaired and unimpaired individuals could be clearly differentiated in terms of their capacity to deal with the situation presented, and the role of the right medial prefrontal area was highlighted.

Findings of this kind have been supported by evidence of differential activation patterns from imaging studies, and Shallice concludes that there is significance in the convergence of evidence from two very different methodologies in respect of determining the physical basis of certain complex and higher-level cognitive processes.


Early Intervention for Children with ASD

Research completed by the NFER (Evans et al 2001), funded by the Local Government Association, set out to highlight the level of identification of autistic spectrum disorders among children up to the age of 7 years, and the pattern of provision made for them. A further target was to examine how the effectiveness of interventions was assessed. Information was gathered by a series of visits to a sample of Local Authorities.

The following was the range of types of provision for children between the ages of 2 and 7 years :

The research could not clearly identify the number of children with ASD given the lack of clear records complicated by variations within diagnostic practices and terminology ... with some common reluctance to use the term "autism".

In most educational settings, a range of approaches was found to be in use, with each having some benefits, but no one approach was found effective for all the children with ASD. The most critical feature was that of the expertise and understanding of ASD on the part of teaching and support staff ; the actual setting seemed to be of less importance.

A further critical element was the level of communication and coordination among all the professionals concerned ; and it was noted that there could often be a gap between diagnosis and the offer of support, especially if certain types of provision could only be accessed via a statement which introduced a time delay at a crucial stage. Parents commonly expressed concern at what they described as a sense of isolation, and difficulty in having the child's needs recognised and adequate provision established.

Variation was observed in the availability of staff training but, as indicated above, the expertise of the staff dealing with the children was held to be crucial to the ability of an LEA to make satisfactory provision.

Monitoring of the interventions was found not to be consistent such that there was little basis for determining effectiveness or value for money ... there was an absence of baseline and repeat assessments.

The research team recommended a number of features which would improve the coordination of action, and the effectiveness of support.

Such recommendations included :

In sum, the authors found great variation in the quality and quantity of provision, and they referred to a trend towards "adhocracy" in making arrangements to meet pressures and demands as they arise in local situations. Provision was a matter of what was already available in the locality rather than seeking closely to match provision to the child's needs.

The conclusions referred to the great variation among children all diagnosed with ASD, and among parents in respect of their aspirations and wishes, and they restated the point that staff expertise and knowledge are the critical issues, not where the child is educated.

Discussion with one of the co-authors highlighted the point about the difficulty or inconsistency in determining the effectiveness of interventions, notably in respect of the absence of some meaningful initial and baseline set of measures, and a dependence upon qualitative or even anecdotal data rather than some consistently re-applied series of measures and observations. This applied particularly to home-based interventions and it was acknowledged that this section of the report is the least complete.

The final point was an acknowledgement of the potential value of individualised "packages" which might include, for example, home-based programmes alongside supported nursery or playgroup placements ; but such arrangements reinforce the need for the presence of a key worker to ensure coherence among various strands of provision and communication among all the individuals and agencies concerned.


M.J.Connor          April 2001



Baron-Cohen S., Leslie A, and Frith U. 1985 Does the autistic child have a theory of mind ? Cognition 21 37-46

Channon S. and Crawford S. 2000 The effects of anterior lesions on performance of a story comprehension test. Neuropsychologia 38 1006-1017

Dales K. et al 2001 Subject : The link between the MMR vaccine and autism. Journal of the American Medical Association 7 March

Davis R. et al 2001 Subject : Measles-containing vaccine and inflammatory bowel disease.

Archives of Pediatrics and Adolescent Medicine.

Evans J., Castle F., Barraclough S., and Jones G. 2001 Making a Difference : Early Interventions for Children with Autistic Spectrum Disorders. Slough : NFER

Hung M. 2001 Subject : MMR vaccine and legal action.

Jyonouchi H. et al 2001 Subject : Aberrant immune response in autism. Presentation to the Experimental Biology 2001 Congress. 2 April. Orlando : Florida

Kaye J. et al 2001 Subject : MMR vaccine and autism rates. British Medical Journal ( 10 February )

Kniker E. 2001 Dietary changes and autistic symptoms. Presentation to the 12th. International Conference on Autism. 29 March. Durham : UK

Shallice T. 2001 Theory of mind and the prefrontal cortex. Editorial : Brain (March) 247-248

Shattock P. 2001 Environmental factors in autism. Quoted variously in the media, including daily press, eg Independent March 2001, and radio, during the Durham Conference

Stone V., Baron-Cohen S., and Knight R. 1998 Frontal lobe contributions to theory of mind. Journal of Cognitive Neuroscience 10 640-656

Stuss D., Gallup G., and Alexandrer M. 2001 The frontal lobes are necessary for theory of mind. Brain 124

This article is reproduced by kind permission of the author.

© Mike Connor 2001.

Back to NAS Surrey Branch Welcome Page