These notes begin with a reference to the need for early identification (and for implementing early intervention), and attempts to produce a simple and valid screening instrument for use with very young children.

There follows a description of attempts to identify the genetic loci of autism, and of research differentiating children who are later diagnosed with autistic spectrum disorder from normally-developing children in respect of neonatal blood analysis. Meanwhile, further evidence for a lack of any link between the MMR vaccination and autism is set out.

The final section deals with the self-report of (adult) individuals with high functioning autism indicating their at-risk status for negative emotional experiences; and advice is summarised concerning the needs of students about to move into the senior stage of schooling and the means of easing this transition.

M.J.Connor September 2001

Early Identification

Over the last decade or so, there has been an apparent increase in the incidence of autistic spectrum disorder, stimulating debate whether there are actually more cases of children with this condition or whether the incidence has remained much the same but with greater accuracy in diagnostic procedures along with enhanced sensitivity towards signs and symptoms.

The work of Fombonne and Chakrabati (2001) could be said to support both perspectives, but what matters is that they highlight how autistic spectrum disorders are more common in the UK than has been appreciated previously.

They collected data on a sample of more than 15,000 children between the ages of 2.5 and 6.5 in one large county, and organised standardised interviews and psychometric testing by a multi-disciplinary team in the case of those children who showed initial signs of some pervasive developmental disorder.

The prevalence of pervasive developmental disorders in this population was 62.5 per 10,000 children of whom the majority (79%) were boys. For autistic disorders, the prevalence was nearly 17 per 10,000. These figures are markedly higher than typical and previous estimates.

The implication drawn emphasised continuing to develop systems by which to identify as efficiently as possible the children with atypical social and language development in order to gain a fuller assessment and access to effective treatments in the critical early years.

(Meanwhile, there is recognition of a further condition - Childhood Integrative Disorder or Heller's Syndrome - which occurs in 3 and 4 year olds, characterised by a deterioration over a period of several months in intellectual, language, and social functioning after previously normal functioning. The cause is unknown but it has been linked with neurological anomalies with possible relevance of seizure disorders.
This condition is recognised by loss of social skills, bowel and bladder control, expressive or receptive language, motor skills ; and by a lack of play skills, failure to develop peer relationships, and impairment in nonverbal behaviours.
It is significant because the presenting symptoms are similar to those of autism, and treatment/intervention will be largely the same as those for autistic spectrum disorder.)

In a similar vein to the conclusions of Fombonne and Chakrabati, a report issued in America in June 2001 by the National Academy of Sciences' Committee (on educational interventions for children with autism) has held that children should be screened for autistic spectrum disorders as early as 2 years of age, and that those showing initial signs of risk should be followed up with appropriate diagnostic and educational assessment.

The report goes on to highlight the spectrum concept of autistic disabilities and the range of levels and types of observable signs and symptoms. It is noted that there is no single behaviour which is invariably typical of autism or any disorder which falls within the autistic spectrum, nor any behaviour which would automatically exclude an individual child from an ASD diagnosis.

The conclusion of this committee is that interventions of a relatively intensive and long term nature should be systematically planned as soon as a child is suspected of having an autistic spectrum disorder, and that the parents/carers should be provided with appropriate information and services by which to help them support the child and assist with or complement the planned intervention. Regional resource centres are recommended.

Their final point refers to the need for clearly defined objectives and a system of ongoing assessment by which to measure the effectiveness of the intervention in terms of the child's access to educational facilities and integration within the local community, and of the impacts upon family functioning.

The shared implication behind the above summaries is the need to develop some efficient but relatively simple screening system by which to identify as soon as possible those children at risk and to gain access to further assessment and interventions as appropriate.

This goal underlies the work of Gillham et al (2000) who begin by describing how traditional diagnostic systems have depended upon a categorical approach. They cite the example of "social dysfunction " which may be diagnosed by means of various criteria, such as the presence or absence of a qualitative impairment in social interaction.

The problem is that these categorical approaches may differ widely in the range or nature of signs or symptoms which are seen as significant. Further, the way autistic spectrum disorder is demonstrated through observable behaviour will differ according to age and developmental level.

The example is given of some failure to show symbolic play which might reflect a significant problem at a mental age of 7 but not at a mental age of 2. Similarly, problems with communication may present as echolalia in a five-year whereas adolescents may speak fluently enough but have minimal understanding of the subtle aspects of social conversation.

The danger, as set out by Bishop(1989), is that confusion may arise from this lack of developmental perspective, and that diagnosis may remain limited in accuracy.

The authors go on to describe how standardised measures have been developed to provide clear diagnostic criteria for various developmental disorders by evaluating an individual's performance in relation to those expected from chronological or developmental age.

In respect of some normative assessment of skills, particularly social functioning, there is reference to a number of studies which have demonstrated that the Vineland Adaptive Behaviour Scales (Sparrow et al 1984) can measure delays in social and communicative development in children with autism.

In a previous study of their own (Volkmar et al 1993), these authors evaluated the ability of the Vineland Scales to diagnose autism.

Following the use of the scales with a large and normative sample, it was possible to predict expected social and communication skills according to chronological or mental age and child gender. Applying these data to children who met DSM criteria for autism and to children with other developmental disorders ( such as language disability or mental retardation ), it was found that there was a greater discrepancy between observed and predicted scores in the autistic group than in the developmentally disordered group.

A discrepancy in scores relating to socialization was found to be a particularly powerful predictor of an autistic diagnosis.

The aim of the present study was to replicate findings that delays in socialization relative to actual age and mental age could distinguish autistic from non-autistic children.

The study also extended earlier investigations in that there was included, along with a group of children with autism and a group of children with non-pervasive developmental disorders, a further group of children diagnosed with pervasive developmental disorders other than autism, with the prediction that such children would also show delays in socialization but that the children with autism would be the most impaired.

The sample included children between the ages of 4 and 13 assigned to one of three groups, and involved the use of the Vineland Adaptive Behaviour Scales from which raw scores were drawn and analysed rather than standardised scores given the level of developmental impairment among the children concerned.

The findings demonstrated a consistency with earlier studies, confirming that individuals with autism displayed deficits in communication and socialization relative to their peers.

In this present sample, children with autism also displayed significantly poorer daily living skills and more serious maladaptive behaviours than children with the other developmental disorders ; and the differences were still observed even after controlling for differences in mental age between the groups.

However, even if the groups did differ in terms both of adaptive and maladaptive behaviour, it was a failure to gain adaptive skills, particularly in the area of socialization, which was most strongly related to clinical diagnosis. These results were held to be consistent with the view that an impairment in social skill development is more central to autism than the presence of unusual or deviant behaviours.

The authors concluded that standardised measures of adaptive behaviour, such as the Vineland Scales, may be useful in screening for autism even if further studies with larger samples are required in order to identify more precisely the socialization cut-off score for discriminating autism from other disorders.

It is also argued that, while one might directly compare in future research the use of the Vineland with other diagnostic instruments such as the Autism Diagnostic Interview, administration of the Vineland does not require extensive training or direct face-to-face assessment of the child as do these other instruments. Rather, it is simply a matter of a semi-structured interview administered to a parent or carer, and it is viable as an initial instrument for routine use.

Using these regression methods, Vineland and IQ scores may, thus, provide a useful screening methodology for autism, following which the children scoring below an appropriate cut-off point may be referred for more extensive diagnostic evaluations.

The authors conclude that evaluating a child's socialization in the light of his/her developmental level may be a more efficient initial approach than traditional categorical diagnostic systems.

This approach may also allow for recognition of the way symptoms change with age ; and the use of standardised and validated assessment material, such as the Vineland Scales, may be a significant first step to a more accurate system for identifying children with autism.

The work of Robins et al (2001) also begins with the recognition that currently available diagnostic material is not very satisfactory for the detection of autism or other pervasive disorders in very young children. They suggest that this may be because of the way symptoms vary from case to case, social and language deficits or delays may not be recognised until the child begins to interact with peers in a nursery or school setting, and motor milestones are usually within normal limits.

There is no standard and easily administered screening instrument for young children.

This being so, children with autism are rarely identified before they are 3 years old although there is evidence (Siegel et al 1988 ) that there is a significant delay between the age of the child when the parents first become concerned and the age when evaluation begins and the diagnosis is made.

It is recognised that this delay causes additional stress to parents as well as as waste of valuable time during which an intervention could have been in place. This is all the more significant because of the converging findings that early intervention is of the essence, and that more positive prognoses apply to those children who can be helped to develop language and symbolic play skills before the age of 5 years. Interventions may show diminishing returns as the children get older.

A review of studies by these present authors indicates how the primary deficit in autism has been variously viewed as a language impairment, a deficit in arousal modulation and sensory responsiveness, undeveloped theory of mind, delayed motor functions, and limited social/emotional development. They go on to suggest that the Modified Checklist for Autism in Toddlers ( M-CHAT ), an extension of the original checklist devised by Baron-Cohen et al (1992), includes items for all these areas or their precursors ; and highlighting which items are most correlated with a diagnosis of autism would provide a means for early and simply-administered screening for autistic disorders among very young children.

It is their view that all existing diagnostic instruments are inappropriate for detecting signs of autism or pervasive developmental disorders in children of around 18 months to 2 years. The reasons include the length or complexity of their administration, the lack of validity outside a school-age sample of children, a dependence upon a narrow sample of behaviour in a clinical setting, a focus upon particular characteristics that commonly fail to appear before 3 years of age, and a lack of standardisation. Importantly, these measures ( such as the Childhood Autism Rating Scale, or the Autism Diagnostic Observation Schedule ) are used only after particular children have been identified as having developmental disorders rather than administered to all children.

The CHAT was developed to help identify early signs of autism at 18 months by assessing developmental milestones. The original items included nine questions asked of parents and five involving observations of the child at home by a health visitor. Three items were identified as significant ..... pretend play, gaze monitoring, and pointing at an object of interest.

In the current study of Robins et al, the participants consisted of children assessed during routine paediatric or GP check-ups at 18 or 24 months along with a sample of children screened through early intervention service providers between the ages of 18 and 30 months. None of th children had been diagnosed previously with a developmental disorder.

The original M-CHAT had 30 items ( although after analysis and modification the current checklist has 23 items where the respondent simply has to answer "yes" or "no" to questions like ..... Does your child ever bring objects over to you to show you something ?" or " Does your child enjoye being swung or bounced on your knee, etc. ?" ).

If child failed 5 of the original 30 items, the family received a follow-up call. If the child failed 3 or 4 items, a clinician examined the checklist in order to identify and follow up any child where the failed items were of concern. The parents of children seen to be at risk were invited to take part in a further evaluation which included the use of "major" measures such as the Childhood Autism Rating Scale or the Bayley Scales of Infant Development, while a full history of the child was also taken.

The outcomes indicated that the M-CHAT was reliable in that the children who were later diagnosed with autism or pervasive developmental disorders failed more items than any other children and were significantly different on all items except for two, namely the enjoyment of being swung or bounced, and whether the child was walking.

Six items that were found particularly well to discriminate related to joint attention (pointing to something, following a point, and bringing things to show parents) ; social relatedness (an interest in other children, and imitation) : and communication (responding to name).

It was pointed out that the most powerful predictors of autism or pervasive developmental disorder in these very young children are largely negative symptoms, i.e. the failure of normal behaviours to mature, rather than positive and specific signs, such as finger twiddling, which may appear later.

The authors conclude that, while further and larger-scale validation studies are appropriate, the M-CHAT appears a valuable means of early screening to enable clinicians to detect autistic features in very young children.

There is a note of caution in that there are risks in basing decisions solely upon parental reports given that some parents are poor observers or judges of behaviour especially if they have limited opportunities to measure their own child against other children. In other words, where there are some concerns about a child's development there should be referral on for more formal or diagnostic evaluation even without a checklist score above some cut-off point.


On a different tack, it is already well established that genetic features appear significant in autistic spectrum disorder, and research continues to identify specific chromosomal influences.

The work of Alexander et al (2001) has shown that regions of 4 chromosomes appear to be linked with the development of autism. Such findings confirm the role of genetics in autism and are decribed as a major step in narrowing the search for the particular genes involved.

The research involved screening the DNA of more than 150 pairs of siblings with autism, with the outcome of strong evidence that 2 regions on chromosomes 2 and 7 contain genes involved with autism. Some evidence was found for autism-related genes on chromosomes 16 and 17, albeit with a weaker correlation.

Chromosome 7 has been shown to be associated with many language disorders and its link with autism has been demonstrated in some previous studies.

It was concluded that, because of a size of the study and the strength of the correlation found, there is strong evidence that this language-disorder chromosome is significantly involved with the development of autism.

However, in the light of the wide range of symptoms observed in individuals with autism, there is still a widespread view that this condition is the result of a complex interaction between several different genes involved with brain signalling and development ; and it is also likely that a range of environmental factors play a role.

Neatly linking the issues of aetiology and early identification, studies completed by Nelson et al (2001) have shown how unusually high levels of certain neuropeptides, ie particular compounds of amino acids which operate to stimulate neural growth, appeared to differentiate, from normally developing children, those individuals who go on to develop autism or pervasive developmental disorders.

The authors found that a gastrointestinal peptide was most significant in separating children diagnosed with autism from control children in their study of neonatal blood samples in children who went on to develop autism or other severe conditions or who developed normally. They argue that these neural growth factors are crucial in the production of new neurons and the organisation of those cells into particular networks.

It is possible that the unusually high level of these particular substances in autistic or mentally retarded children may disrupt the normal process of cell migration and differentiation during the early stages of the development of the central nervous system.

Meanwhile, there is continuing support for the already overwhelming argument that there is no link between the MMR vaccine and autism.

In America, the Federal Committee on Immunisation Safety reported in the Spring of 2001 that parents should not prevent their children from being vaccinated, and that there should be no change in official guidance. While no vaccine is claimed to be totally safe, the MMR vaccine is said to be as safe as a vaccine can be.

This advice is based upon a review, conducted by committee members, of all epidemiological research studies with any bearing upon this issue. It was acknowledged that there just might be rare cases where a link exists, and that most of the published studies did not set out to identify rare side effects, but current advice continues to be that the benefits of the vaccination outweigh considerably any possible risks.

A similar conclusion was drawn by Davis et al (2001) who examined the incidence of seizures in a massive sample of children, comparing experience immediately after the MMR vaccination against the experience of children with no recent vaccination.

It was found that this vaccination has the effect of increasing the risk of febrile seizures between one and two weeks after the event, but there was no increased risk of non-febrile seizures. Further, there was no difference found between children who did experience some of these vaccine-associated seizuress and children demonstrating similar seizures who had not been vaccinated.

The conclusion here was that this transient-only increase in the risk of seizures should not obscure the benefits of vaccination which is linked to no long-term adverse consequences.

The Experience of Autism

In their description of the nature and symptoms of autism, Jones et al (2001) refer to the significance of absent or limited emotional responsiveness, but they note that the majority of research studies involve third party observation where parents or siblings describe their perceptions of the individual with autism. There is not much evidence from studies where the emphasis is upon the perspectives of the people with autism themselves.

Their own study involved an analysis of internet-based first-hand descriptions of five people diagnosed with high functioning autism. An analysis of the contents revealed four major and shared themes.

The most common emotional theme observed, shared by four of the five individuals, was a sense of alienation with a pervasive view of themselves as different. This was linked to strong feelings of loneliness and exclusion.

(It was interesting to note one comment to the effect that an eventual diagnosis of autism helped to reduce in part some of these feelings given the realisation that there are others with similar difficulties, facing similar circumstances, and likely to be having similar experiences.)

Among three of the individuals, there was perceived to be a strong sense of frustration as a result of being autistic and of the impact upon day-to-day living.

There was reference to the feeling that other people cannot recognise or appreciate the differences or disabilities that characterise autism so that the individual concerned cannot break out of a limited world.

A central element of this sense of frustration was linked to social and language difficulties which appeared to be long term and pervasive.

In addition, there was mention of the particular difficulty with social interaction resulting from the volume of sensory-perceptual stimuli that come into play during what might be thought by normally-functioning individuals to be simple or routine social interaction, but where the person with autism cannot filter out the all the information that is irrelevant.

The third theme was that of depression, with a significant finding that the experience of depression was primarily before a diagnosis such that there is a failure to understand the reasons why one has feelings of being so different to other people.

One individual commented that a person who has moderate to severe autism may display sufficient symptoms that other people recognise the disabilities and are sensitive to them. Individuals with high functioning autism, on the other hand, may not have their difficulties appreciated so that they may suffer ridicule or exclusion.

Finally, there was confirmation of earlier reports (such as that of Grandin 1992) that the emotion of fear or apprehension is dominant in people with autism. In particular, this related to difficulties experienced with social interaction or an all-too-keen concern for what others may be thinking along with a belief that the individual with autism is constantly being watched.

The authors conclude that emotional issues can be important and highly relevant to people with autism, in contrast to much of the literature that highlights absent or undeveloped emotions in autistic individuals compared to normally developing controls.

However, there is also the implication that the emotions which are experienced most keenly are unpleasant, and it is those individuals towards the higher-functioning end of the spectrum who may be most at risk for emotional difficulties.

With respect to young people still at school, Adreon and Stella (2001) describe how the transition to senior schooling can be challenging for any student because of the changes encountered.

These include increases in school size and student numbers, access to many different subject teachers, increased expectations upon achievement and behaviour, and more demands in terms of class work and homework. At the same time, social expectations increase and peer relationships become more complicated.

It is argued that the students who lack the necessary skills to cope with these scholastic and social demands may suffer significant problems in adjustment and in self-esteem.

The authors then go on to argue that individuals with high functioning autism or Asperger syndrome with their characteristic weaknesses in social skills and communication will be particularly at risk at this time of transition to senior school. While they cope well with certain academic challenges, they may still have great difficulties in coping with the social or behavioural demands upon them in this setting. The anxiety experienced as the time of transfer approaches is likely to be the greater among this sample of students.

The recommendation is that the young people with ASD/Asperger syndrome will benefit from careful and proactive planning related to the transition process. While school visits to provide some measure of induction have become routine and will benefit the young people in question, they may not be sufficient in providing the necessary support. Rather, one needs to identify the whole range of issues which might give rise to anxiety to them even if they might be taken for granted by non-autistic students.

Induction should, therefore, include visiting all the classrooms where they will spend time and meeting all the different teachers, while rehearsing some of the day-to-day routines that they will need to develop ( such as the use of lockers ), and getting to learn the way around the school.

It is likely that a high level of support will be necessary for the intial period to enable the young people to become confident in the new setting.

The suggested first step is that of a meeting shared between staff from the junior school and staff from the senior school in order to highlight the kind of support deemed necessary to enable the students to manage successfully and happily in the first few weeks of the new school.

For example, it may be appropriate to identify another student who can act as a "buddy", and there should be opportunities for learning and discussing the school rules and routines. It is stressed that this kind of preparation should be in hand before the actual move to the new school. Meanwhile, it is important that all the teachers are given some awareness of the nature of autistic spectrum disorder and how it impacts upon behavioural and academic performance.

In respect of classroom arrangements, one might consider issues such as where the student has to sit, the use of visual clues to support learning, reduction in the size or number of demands, and help with the day-to-day issues such as taking notes.

Something which all staff would usefully keep in mind is the difficulty that students with Asperger syndrome have in facing unexpected changes in the daily routine, shifting attention from one task or topic to another, and following oral instructions of any length.

These authors note the particular difficulties that have already been described (see, for example, Connor 2000 ) in respect of unstructured times such as moving between classes, break time and lunch time, and the periods immediately before and after school.

These times are difficult because the social demands continue but the rules for how to behave are less clear. It is possible that many of the other children will have already formed friendship groups and the ASD student may be somewhat isolated and unsure what to do.

It is at times like these that the concept of a "buddy" may be most appropriate ; alternatively one might seek to provide a range of legitimate opportunities for the students with ASD to be involved with some activity in the school. Clubs, such as those dealing with ICT, may be particularly useful.

Finally, there is a reference to difficulties linked to PE or games, and it is suggested that the students in question might be allowed a particular and specific role such as the score-keeper or equipment organiser while ensuring that there is no risk of the student being ignored when teams are organised via peer selection.

The conclusion from these authors is that one should never underestimate the difficulties that students with autistic spectrum disorder/ Asperger syndrome have with socialising, communicating, and adapting to change, all of which put them at an enhanced risk for adjustment problems thus necessitating great care in planning and implementing transition arrangements.


M.J.Connor September 2001


Adreon D. and Stella J. 2001 Transition to middle and high school : increasing the success of students with Asperger syndrome. Interventions in School and Clinic 36(5) 266-271

Alexander D. et al 2001 Autism may be associated with specific regions of 4 chromosomes. (To appear in the American Journal of Human Genetics 69(3) 570-581)

Baron-Cohen S., Allen J., and Gillberg C. 1992 Can autism be detected at 18 months ? British Journal of Psychiatry 161 839-843

Bishop D. 1989 Autism, Asperger syndrome, and semantic-pragmatic disorder : where are the boundaries ? British Journal of Disorders of Communication 24 107-121

Connor M. 2000 Asperger syndrome and the self reports of comprehensive school students. Educational Psychology in Practice 16(3) 285-296

Davis R. et al 2001 MMR vaccination not associated with long term adverse outcomes.

Fombonne E. and Chakrabati S. 2001 Incidence of autistic spectrum disorders. The Journal of the American Medical Association 285 3093-3099

Gillham J., Carter A., Volkmar F., and Sparrow S. 2000 Towards a developmental operational definition of autism. Journal of Autism and Developmental Disorders 30(4) 269-278

Grandin T. 1992 An inside view of autism. In Schopler E. and Mesibov G. (Eds) High Functioning Individuals with Autism. New York : Plenum Press

Jones R., Zahl A., and Huws J. 2001 First hand accounts of emotional experiences in autism. Disability and Society 16(3) 393-401

Nelson K. 2001 Very early markers of autism and mental retardation. Annals of Neurology (May)

Robins D., Fein D., Barton M., and Green J. 2001 The modified checklist for autism in toddlers. Journal of Autism and Developmental Disorders 31(2) 131-144

Sparrow S., Balla D., and Cicchetti D. 1984 The Vineland Adaptive Behaviour Scales. Circle Pines, MN : American Guidance Service

Volkmar F., Carter A., Sparrow S., and Cicchetti D. 1993 Quantifying social development in autism. Journal of the American Academy of Child and Adolescent Psychiatry 32 627-632

This article is reproduced by kind permission of the author.

© Mike Connor 2001.

Back to NAS Surrey Branch Welcome Page