A review of early interventions for children with autism

by Mitchel Fleming


For over 40 years there has been considerable interest in the provision of early intervention for children ‘at risk’ and for those with disabilities. The concept rose to prominence in the USA during the 1960s with the launch of the Head Start Programme that sought to enhance the development of young children from low.income families who were considered to be ‘at risk’. In the areas of learning disability and autism various approaches make claims of almost miraculous cures for children with disabilities if they enter certain treatment programmes early. As with all such claims it is wise to proceed with caution and to look for the scientific evidence that supports these claims. If an approach sounds too good to be true, then it probably is not all that it claims to be. However, before I briefly review the effectiveness of some of the more commonly used approaches I first want to summarise the general beliefs behind the concept of early intervention.

The aim of early intervention is to enhance the cognitive, emotional and social skills of young children ‘at risk’ or with a disability such as a developmental delay or autism. Most of these children are known to have a developmental trajectory that is less than that of a normally developing child. For example, a child who is two years behind at three years of age, may be four years behind at six, and six years behind when s/he reaches nine years. In such circumstances the child at nine is said to have a mental age of three. The concept of mental age is not very useful beyond the teenage years, as people’s mental age seems to reach a ceiling by the time they are 13–16 years of age.

As is apparent, a relatively small difference between a child’s actual age and his/her mental age early on is likely to become more pronounced with time unless intervention occurs. The hope of early intervention is to try and narrow this gap in development as soon as possible so that the child will develop along the same developmental line or trajectory as other normally developing children. If intervention is delayed for one reason or another there is a belief that the gap may have become too wide to bridge.

Also, there is a general assumption that the brain of a young child is far more amenable to change or has greater ‘plasticity’ than that of an older child. Most of the evidence for this assumption comes from research with animals where there are findings that suggest that early intervention can produce neurological changes in the developing brain. Based on these assumptions there has been an increased demand for the early recognition of developmental disorders, coupled with the delivery of effective interventions. The challenge, therefore, is to determine if these assumptions and beliefs are correct and to determine whether children who are diagnosed early and get interventions fare better than children who don’t.

Obtaining valid and reliable answers to the above questions is a complex matter. The challenge for all of us who are interest.ed in promoting the abilities of young children with disabilities is to distinguish between claims about interventions that work from the many false or essentially unverifiable claims. This has been the focus of scientific enquiry. There are various accepted scientific methods that enable us to determine whether a treatment is effective or not (see Bristol et al, 1996; Lonigan, Elbert and Johnson 1998; Green 1996).

Fortunately, a number of people have reviewed the existing literature on the effectiveness of different approaches commonly used with children with autism and have provided simple synopses of their findings. In 2002 Dr Linda Finnegan and Professor Alan Carr at University College Dublin (UCD) reviewed the literature concerning effective treatments for children with autism. Prof. Patricia Howlin, who is an advisor to the UK National Autistic Society, gave a summary of the current literature at the NAS International Conference in London last September in her presentation on ‘Assessing the effectiveness of early intervention programmes for young children with autism’. The last source is Prof. Richard Simpson and his colleagues at the University of Kansas who conducted a recent and detailed review of 40 commonly used approaches in the field of autism. Below I have attempted to summarise the evidence from the above sources concerning the most widely cited treatments. While each of these reviewers used different methods to deter.mine the effectiveness of various interventions, there are, however considerable similarities in their conclusions.

Interventions and treatments are divided into those that are supported by good scientific evidence, those that show prom.ise—but where the evidence is still limited, those where there is little or no convincing evidence to support them, and those that have tested negative or are clearly ineffective.

Treatments that are supported by good scientific evidence include various Applied Behavioural Analyses (ABA) approaches, such as Discrete Trial Teaching (DTT) and Pivotal Response Training, and Learning Experiences: an Alternative Programme for Preschoolers and Parents (LEAP). DTT came to prominence in the area of autism through the work of Prof. Ivar Lovaas at UCLA. DTT involves breaking tasks into small steps and then systematically teaching each of these steps using reinforcement until a certain criterion is reached before moving to the next step. Pivotal Response Training typically involves the careful selection of socially valid behaviours, which may be taught by modelling and their imitation reinforced, so that they are likely to help promote other skills not specifically taught. LEAP is an integrated 15 hours per week programme that uses behavioural techniques and which aims to develop play and interactional skills between preschool children with and without autism.

Interventions and treatments that show promise include Picture Exchange Communication System (PECS); incidental teaching; structured teaching; TEACCH (Treatment and Education of Autistic and related Communications handicapped Children); speech and language programmes, Circle of Friends; Hanen; augmentative alternative communication; assistive technology; joint action routines; play.oriented strategies; cognitive behavioural modification; cognitive learning strategies; social stories; social decision making strategies and sensory integration.

Limited or no supporting quality evidence exists for the fol.lowing practices: Gentle Teaching; Option method (also known as the Son.Rise programme); Floor Time approach by Greenspan; pet/animal therapy; relationship development intervention; Van Dijk curricular approach; Fast ForWord; cognitive scripts; Cartooning; Power Cards; Scotopic Sensitivity Syndrome (Irlen lenses); Auditory Integration Training; megavitamins; special diets; herb, mineral and other supplement; music therapy and art therapy.

Treatments and interventions that are not supported by evidence or that have been shown to have a negative effect and accordingly are not recommended include Holding Therapy; Facilitated Communication and the use of the drug secretin.

In summary, the evidence indicates that early interventions that are based on behavioural principles, applied preferably both in preschool/school and at home, and that focus on a child’s educational development can have a marked impact on the cognitive, behavioural and social adjustment of children with autism. There is no conclusive evidence to show that any one behavioural approach is better than another, or to indicate the optimal level of intensity, duration, structure or age of onset. Prof. Howlin advocates that for early intervention to be most effective it should last at least six-months; involve at least 15 hours per week (though 20–30 hours may be optimal); have a high adult.staff ratio; involve specially trained teachers; and begin early (perhaps at 2 to 3 years of age). In addition, Finnegan and Carr (2002) recommend that there should be a highly collaborative working relationship between parents, teachers and clinical staff; that the educational programmes should be tailored to the child’s individual needs; and that they should be well structured. Also, effective programmes should focus on enhancing skills in five key areas: 1) attending to aspects of the environment essential for learning, 2) imitation; 3) language usage; 4) imaginative play; and 5) social interaction.

Furthermore, the behavioural research literature indicates that slightly under half of the children who receive intensive early behavioural intervention will show very substantial gains in IQ and many will also show a substantial reduction in the expression of autistic behaviours. Most improvement in intellectual functioning, when it happens, seems to occur during the early stages of the programme (i.e., the first few months). Those who are most likely to make substantial gains appear to be children whose level of intellectual functioning was relatively good prior to the commencement of intervention (i.e. have an IQ of approximately 50 or higher). However, there is evidence to show that older children aged 4–7 years (Eikeseth et al. 2002) and a small number of children with very significant developmental delays do benefit considerably from early intensive behavioural intervention.