Pre-school autism intervention and autism development: a longitudinal follow up of the Preschool Autism Communication Trial (PACT)

Autism is a severe and persistent developmental disorder, starting in childhood and often extending throughout life and affecting about 1% of children in the community. Autism is diagnosed from about 2 years of age onwards. Parents find adjustment to diagnosis and the management of their child with autism very challenging; use of health, social care, and specialist education resources increases progressively through the pre-school and early school years with costs of supporting children with autistic spectrum disorders in the UK estimated at £2.7 billion per year. However, the additional impact of autism on families is probably far higher than this in terms of intense parental involvement and stretching of resources to care for other children and to sustain employment. Rates of family breakdown are higher than the average.
Early intervention is thought to be most beneficial and there is increasing evidence that programmes of appropriate advice and training improve child communication and behaviour. However, there has been a lack of research into the longer term effects of early intervention, if any, on the development and wellbeing of children with autism and also on family stress and family function.

We have conducted the largest early intervention trial to date of a treatment with parents to help them communicate better with their child with autism. This study (Green et al 2010) showed that, parents who took part in the intervention, were successful in adapting their style of interacting with their child, and in turn the child communicated more with the parent. In spite of these improvements in parent-child communication, children who received the intervention did not improve more than those who had the usual services in how they interacted with adults outside the family, that is the communication improvements did not seem to generalize. However our initial follow up was only at the end of the 13 month intervention - a very short time in the context of development for a child with autism. Therefore in this current study we wish to follow-up the same children about 3 years further on to see how our early intervention has affected longer term developmental outcomes. We will be doing the follow-up in the early school years which is a key time in the development of a child with autism, when many associated difficulties of social integration, and mental health problems such as anxiety, come to the fore. We will be looking at the child's language and communication, other symptoms of autism and social integration. We will also look at parent perceptions of their child's functioning and family life. We are interested in whether parents who took part in the intervention feel more confident and optimistic than those who did not.
This study will be to help us understand, for the first time, the impact of early intervention on later child communication, the presentation of autism in the early school years and family life, school adjustment and family need for services. This will help clarify how development proceeds into middle childhood, and give an indication of any necessary additional intervention required either in the pre-school years or later, to maximise the best developmental outcomes in autism. The findings will potentially impact the way early intervention and intervention in the pre-school and early school years is organised, ideally with the integration of medical, social care and education, and methods of supporting families as their children grow older.

Green J. Charman, T,. McConachie, H. et al (2010). Parent-Mediated Communication-Focused Treatment for preschool children with Autism (PACT); a randomised controlled trial. The Lancet, 375(9732), 2152-2160.

We plan 3 year follow-up of at least 80% of the existing PACT cohort to test mediation/process hypotheses arising from observational studies. The unique strength of this proposal is that the initial randomisation in the PACT trial allows for a causal interpretation of subsequent effects in a way that is not possible in observational studies. We use a "hybrid trial" design(1,2); consisting of 1) initial randomised allocation to two parallel groups, 2) an intervention in one arm targeted at key theory-based variables and including analysis of mediation, and 3) the current follow-on observational study of subsequent development. Our findings on treatment effect and causal mediation in the original PACT trial, together with convergent findings from independent longitudinal observation studies in relevant areas of autism development, generate causal hypotheses on the effect of parent child dyadic communication (altered by the intervention) on downstream child and family outcomes, including language and communication, autism symptoms, and family functioning. Follow-up also includes our own innovative service-user generated family functioning questionnaire and measurement of longer-term intervention effects on parental care burden and service usage. This innovative hybrid trial design has not been undertaken before in autism and will enable the causal analysis of some key developmental hypotheses related to early autism development, as well as measurement of generalized effects of intervention that may have taken time to develop. This approach will be innovative in the autism literature and have significant application in service development and intervention planning.
1) Howe et al. (2002) Dev Psychopath, 14: 673-94; 2) Green et al. (2008) Brit J Psychiat, 192: 323-325

Autism is a lifelong neurodevelopmental disorder that in many cases brings considerable burden and cost to the individual, their family and society. The proposed study has two main non-academic beneficiaries (i) individuals with autism, their families and those who care for them; and (ii) professionals, managers and commissioners in NHS, education and social care services who responsible for the care of individuals with autism.

Individuals with autism will benefit from the research as the study will test the medium-term effects (and both the limits of these effects and service and family influences on them over time) of the PACT preschool parent-training intervention in terms of improved language and communication abilities, improved parent-child interaction, reduction in autism symptoms, reductions in mental health difficulties and its impact on family quality of life. Developmental studies in autism show that many of these characteristics measured in mid-childhood are predictive of outcomes through adolescence into adulthood. Thus, knowing the medium-term effects (and limits) of a treatment such as the PACT intervention holds some promise to reduce burden to the individual, their family and society across the lifespan in the future.

Parents will benefit not only potentially from improvements in their child's communication, mental health and social functioning, and their interaction with their child, but also from the contribution the study can make to the literature on early intervention in autism. Despite the considerable need for evidence-based early interventions in the autism field the evidence base is sparse and very variable in quality. As a result, many parents (and some practitioners) seek out and access interventions, often paid for from their own resources, for which no positive evidence of benefit exits and in some cases evidence of potential harm. As the largest and more rigorous psychosocial intervention RCT conducted internationally to date, the proposed work to understand the medium-term effect of the time-limited intervention and the developmental sequelae and influences on outcome over time has the potential to significantly inform the evidence-based literature in the autism field to the future benefits of families and practitioners.

Practitioner and commissioners with the NHS who are responsible for individuals with autism and their families are faced with a weak and variable evidence-base on which to make decisions about service provision. The NICE guidelines on management are due to be published in November 2013 and several of the applicants are on the Guideline Development Group. The publication of these guidelines will highlight to practitioners and policymakers the need for publically funded services for families with autism to be based on sound clinical evidence. Whilst this follow-up will not be completed in time for inclusion in these guidelines, it will be able to inform future decision-making about autism services to the benefit of those responsible for commissioning and delivering these services and ultimately to the families under their care.