The aetiology of Autism Spectrum Disorders and the relationship with associated psychiatric difficulties.

Student thesis: Doctoral ThesisDoctor of Philosophy


Studies of the aetiology of Autism Spectrum Disorders (ASD) have been limited so far due to the rare nature of samples allowing detection of the total genetic and environmental influences. Most studies to date point towards strong genetic influences. However, two recent reports found that shared environmental influences mattered more than genetics.
The first two empirical chapters of this thesis describe a set of innovative analyses designed to provide new evidence on the aetiology of ASD, and to make sense of the existing contradictory evidence. The new evidence comes from the Social Relationships Study (SRS). For the first time, a single model was fitted to data recognising the Broad Spectrum as an ASD subgroup. The results revealed strong genetic influences and little support for shared environmental factors. The second empirical chapter reports on the first meta-analysis of ASD twin studies to redress which familial effects are more important: genetic or shared environmental factors? Again, the results showed the importance of strong genetic influences. Moreover, the strength of shared environmental factors was dependent on the prevalence values used to correct for selecting individuals on ASD affection status. These results re-affirm that ASD is under strong genetic influences and that detecting shared environmental factors may be a statistical artefact.
The second half of this thesis deals with the aetiology of associated psychiatric difficulties in autism, the study of which has gained significant momentum in the last decade. To extend this work, the third and fourth empirical chapters utilise data from the SRS as well as data on autism traits from the Twins Early Development Study (TEDS). Similar aetiology profiles were found for comorbid Emotional symptoms and Hyperactivity across the two samples. However, the aetiology of comorbid Conduct problems differed and appeared to be dependent on the severity of autism manifestations. This could possibly be explained by the fact that these problems are conceptualised differently in children with ASD and those with milder manifestations.
There are two main implications of the findings from this thesis for future research. First, the next-generation sequencing studies ought to increase their efforts to include samples across the whole autism spectrum as the genetic risk is very likely to be shared across clinical cases, those showing Broad Phenotype and individuals with mild autistic behaviours. Second, validation is needed of the observed relationships of comorbid psychiatric difficulties in ASD, using independent cohorts; such data are important for the future design of ASD-appropriate comorbidity measures and interventions.
Date of Award1 Feb 2016
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorFrancesca Happe (Supervisor) & Fruhling Rijsdijk (Supervisor)

Cite this