This short paper provides some more thoughts about the efficacy of intervention with young children with autism, reinforcing the desirability and efficacy of early intervention, the need for individual planning of component elements, the significance of parental partnership, and the dangers in assuming that a programme reported to bring about significant improvements in some children will be equally effective among others.
The work of Christie and Sandler (2002) of the Early Years Centre in Nottingham may have been published over two years ago, but is still salient. It describes a project carried out over a three-year period to develop and evaluate a diagnostic and intervention package for young children with autism.
It was not the intention to test some specific methodology by means of a controlled experimental design, but to determine what could be described as good practice in working with young children and their families, keeping in mind the reality of parental responsibilities towards both the child(ren) and to other aspects of their day to day lives, and while seeking to dovetail existing services with any further innovation and to make the most efficient use of available time/resources.
Existing principles seen as critical included the achieving of early diagnosis and equally early intervention (with the diagnostic process linked to the planning of intervention), and working closely with parents.
This latter point was seen as all the more critical in the light of the converging evidence for the accuracy of parental recognition of early signs of disorder in their children (commonly well before their third birthday) along with the reliability of screening devices (such as the C.H.A.T.); but also in the light of the statistics which suggest that a formal diagnosis is frequently not made until the child is in his/her fifth year.
The authors were also aware of the common concern among parents of young autistic children that they were left in a kind of limbo following the diagnosis and before support or intervention strategies were offered. A further principle, therefore, was the desirability of accompanying diagnosis with practical help and of enabling parents to develop skills in managing their children to complement the support available from professionals. It is suggested that the absence of this kind of support has been one significant reason for the parental quest for access to existing home-based programmes such as the Option Approach or Applied Behavioural Analysis.
The existing experience of the authors also led to their emphasising key attitudes for the establishment of intervention, such as the need to recognise and respect the individuality of each child and family, responsiveness to their perceptions of needs, and ensuring that work with the parents was a true partnership.
A sample of children of mean age 2years-4months was the target group and they were assessed via play-based methods and observations by which to gain measures of communication, reciprocal and pretend play, and problem-solving. These data were reinforced by the use of cognitive measures and language checklists.
Subsequently, the families were visited by the research psychologist on a weekly basis for around a two-hour session in which she would work with the child alongside the parent in structured play activities to emphasise sharing and turn-taking, receptive language, and expressive language. The activities would be followed up and continued by the parents.
Supportive information was provided through a series of booklets offering advice about communication, play techniques, structure, etc..
About once per six weeks, consultations would be held at the centre for further discussion and planning, plus access for the children to musical interaction by which to extend the scope of the play sessions at home. Weekly workshops were provided by which to introduce and practise more structured approaches and behavioural strategies, along with access to theoretical issues and to the opportunity to express feelings and concerns.
The evaluation included reassessment of the children along with a scanning of the pattern of information set out in parental diaries concerned with the interventions, and parental interviews were held.
What emerged from the various evaluative exercises as “ critical components ” in the parental view were as follows …. (from interviews after 6 months) :
…. (from interviews after 12 months) :
A comparison group of 7 families whose children with autism were in their first term of school and who had received no previous autism-specific support were consistent in saying that what they would have ideally liked was more information, practical help at home, and support in coming to terms with the diagnosis.
There was also a wish that diagnosis and intervention could be more closely linked, and they suggested an enhanced role for health visitors in developing the capacity to recognise warning signs and symptoms.
The authors summarised the main elements of the feedback into 3 basic sections …
1. Support needs to follow on rapidly from the diagnosis, with information available in such a form as to allow parents to appreciate the nature and implications of autism, and to ensure that diagnosis is rapidly followed by direct contact, ideally at home, to establish interventions.
2. Allowing for flexibility while maintaining the integrity of approaches, and having the opportunity to practise techniques during most of the day, albeit at a lower level, as well as within the specific play sessions.
3. Providing regular support at home so that parents are ensured feedback, and that the children can practise their communication and other skills in a secure and familiar setting.
Home-Based Early Intervention (Portage Model)
The study by Kaderoglou (2000) is also not immediately recent, but is of interest given the focus not only upon early intervention with children with autism but also the use of a modified programme based upon the Portage early curriculum.
The introduction describes autism as manifesting itself like an iceberg in that the immediately observable disability or disadvantage is probably related to communication and interaction, but beneath this may be a complex and severe range of difficulties which will vary from one child to another even if they all share the basic characteristic of insufficient or distorted communication and a non-reciprocal style of linguistic behaviour.
There is the further complication that some children have certain skills that can mask the autistic problems, or parents may be unable or unwilling to accept the autistic needs, or professionals may be cautious about diagnosing significant developmental difficulties in very young children.
All of this may inhibit the establishment of autism-specific intervention at an early age when this may be critical not only in maximising the child’s progress but also in helping the parents to plan and set up the routines and structures which will be beneficial to the child and to other family members.
Seven children with ASD were selected to receive a home-based programme with the parents to act as co-therapists. The criteria included the age of the child (below 42 months), and the willingness and availability of the mothers or fathers to commit themselves to the project.
All children had been rated as having severe autistic symptoms according to the C.A.R.S.,with progress measured in terms of the extent of acquired skills from baseline according to the Portage Developmental Checklist which covers language, cognition, socialisation, self help, and motor skills.
Reynell and Griffiths’ Checklists were administered by physicians blind to the ongoing intervention as a means of cross-checking estimates of progress.
The actual intervention involved a combination of strategies from 5 early intervention programmes …. Portage, TEACCH, Lovaas Behavioural Approach, PECS, and the Carolina Curriculum for Toddlers with Special Needs.
The critical principle was the involvement of the parents as partners in the planning and implementation of the intervention which was monitored via weekly meetings, with parents given a weekly activity chart containing clearly defined targets. The Portage curriculum was the major source of strategies, with enrichment from the other sources such as “readiness” activities from the work of Lovaas (1981).
21 hours per week were set aside for instructional activities alongside access to a mainstream school place (albeit with no additional support available during the school hours).
Observations indicated that all cases followed largely the same pattern of progress over the course of the year of intervention. The domains upon which there was the most impact were those concerned with cognitive and language skills, while the social domain showed the least change.
The significance of the observations was said to lie in the concept of critical periods for development with the risk that neural pathways are either not fully developed or are lost through non-use.
The earlier the intervention, the less the risk of failure to establish neural pathways and the greater the probability of reversing neuronal loss.
With stimulation, other neurons appear to take over the function of deficient areas.
In support of such thinking, the authors noted that it was the youngest child (34 months) who made the most marked progress in language and cognition.
The oldest child (42 months) followed the same pattern of progress but language onset was delayed compared to the other cases, and PECS was used for a prolonged period.
As well as relatively young age, success predictors also included good socio-economic status of the family and a high level of parental education; and greater capacity or willingness for communication in the child at baseline assessment, including a degree of eye contact, the use of some whole words, and responsiveness to one-step directions.
The opinions of the parents were gathered through structured interviews and there was consensus that the children’s development was boosted largely as a result of the carefully designed programme as well as the training and support made available to them as parent-therapists.
Particular reference was made to the need to establish daily living rules and clear structures for positive social adjustment in the children.
Also seen as significant was the personal relationship developed between the parents and the advisors giving the capacity and confidence to take some control over, and an active part in, their children’s development and education.
Access to information about the roles and availability of other services was reported.
The overall conclusion stressed that parents can be very successful in the role of therapist/teacher given access to a partnership with a professional with whom to plan an intervention which can be flexible in taking account of the particular needs of the child.
The further implication is for intervention to be implemented as soon as possible after diagnosis given the significance of a structured approach in these critical early years.
The initial study completed by Beadle-Brown et al (2004) was a follow-up to the reported success of the Applied Behavioural Analysis, set out by Lovaas in his 1987 work, including substantial gains in measured cognitive ability.
The study examined the factors affecting parental choice of an early intervention
(with a focus upon parenting style); and the outcomes for children participating in programmes following ABA or SonRise (Option) principles.
21 families took part in the study, 10 in the SonRise group, 6 in the ABA group, and 5 controls. All the children (18 boys and 3 girls) had a diagnosis of ASD, and average age at the start of the intervention was 47 months (with a range of 25 to 68 months).
The hours devoted to the ABA intervention ranged from 18-39, and the range for the SonRise intervention was 4-60 hours !
The majority of families were also undertaking additional interventions such as dietary regimes, PECS, speech and language therapy, and cranial osteopathy.
The period between baseline and follow-up assessment ranged from 6 to 9 months, with measures including the Vineland Adaptive Behaviour Scales, Leiter Performance Scale, Reynell Language Scale, parental questionnaires, and observations of social interactions
In respect of the findings from this initial study concerning factors in choice of intervention, it had been hypothesised that ABA choice would reflect a more directive style of parenting while SonRise choice would reflect a more permissive style; but this pattern was not observed. There was a mixture of parenting styles in each group with the tentative conclusion, therefore, that parenting style does not influence choice of intervention (with the acknowledgement that a pattern may exist but that the questionnaire methodology was not sufficiently sensitive to detect it).
It appeared that word of mouth was the most common route by which parents became aware of interventions, and it was recognised that most parents would not have access to formal evaluations of interventions published in research journals.
In respect of outcomes, none of the 3 groups showed any significant changes (either for the better or worse) in adaptive behaviour or personal/social sufficiency.
With regard to changes in symptoms (behavioural difficulty), there was observed to be a positive change among the SonRise children in respect of the “quality of social interaction” with a shift from “aloof” to “passive”. This reflected an increased willingness in the children to accept interaction with others albeit still showing reluctance to initiate social contact.
In respect of changes in cognitive skills, the only observable changes were in terms of the number of items attempted which provided a measure of improvement in attention span or in willingness to participate. All groups showed some improvement in the number of items attempted but the difference did not reach a statistically-significant level.
There were also found to be no significant improvements in language usage or social behaviour.
The authors concluded that, contrary to anticipation, there were no real changes between or within groups, with the children on ABA or SonRise not outperforming the control group.
The authors referred to the small sample sizes as partial explanation for the lack of significant outcomes (and one might also note both the short period of intervention, and the low intensity in some cases), with the plan to extend this current pilot study.
However, the conclusion drawn by the authors that some children may respond better than others to early intervention can be seen as helpful in reinforcing the principle of there being no single type of treatment that is appropriate for all children.
Instead, it is necessary to recognise individual and familial strengths, weaknesses, and circumstances in planning an individual programme whose various elements are chosen to reflect these permutations of variables - especially the idiosyncratic profiles of disabilities and comorbidities displayed by the children.
* * * * * * *
Beadle-Brown J., Dorey H., and Murphy G. 2004 Early intervention for autism study. Canterbury : the Tizard Centre (University of Kent)
Christie P. and Chandler S. 2002 A diagnostic and intervention package for young children with autism. Good Autism Practice 3(1) 2-13
Kaderoglou L. 2000 Home based intervention in high functioning autistic children. Presentation to the International Special Education Congress, 24-28 July 2000 University of Manchester
Lovaas O. 1981 Teaching Developmentally Disabled Children : The Me Book.
Austin, Texas : Pro-Ed
© Mike Connor 2005.
Back to NAS Surrey Branch Welcome Page