Promoting identification of people with autism: Developing a behavioural screen for people with learning disability

Metcalfe, Dale Robert (2020) Promoting identification of people with autism: Developing a behavioural screen for people with learning disability. Doctoral thesis, Northumbria University.

Preview

Text (Doctoral Thesis) metcalfe.dale_phd.pdf - Submitted Version Download (5MB) | Preview

Abstract

Autism is a lifelong developmental condition, diagnosed on the basis of persistent social communication deficits and repetitive, restrictive patterns of behaviour. Autistic people form a heterogenous group, but many experience challenges compared to non-autistic people. Identifying and diagnosing autistic people can be beneficial in helping them to better understand themselves, and to access support and interventions. Not all autistic people are diagnosed however, and one group which is disproportionately affected is people with learning disability. Screening tools have the potential to facilitate and speed up diagnosis by highlighting those who should be assessed in more detail. Many existing screening tools are largely inaccessible to people with learning disability due to the level of language required to complete them. The overall aim of the thesis is to facilitate the diagnosis and support of people with autism, in particular those with learning disability, by developing more accessible screening tools.

Chapter One provides an overview of the main issues to consider in the screening and diagnosis of autism, in those with and without learning disability. Chapter Two outlines the adaption of the Autism- Spectrum Quotient (AQ), with the aim of developing a more accessible version (AccAQ). The language of the AQ is simplified, and line drawings are added to enhance understanding. The results indicate that the original AQ and the adapted versions are equivalent in many ways, but the AccAQ still requires the person completing it to have some verbal communication, comprehension, and literacy skills. Chapter Three reports on a systematic review into the psychometric properties of autism screening tools used with people with learning disability. The review found that the majority of tools were lacking comprehensive validity and reliability data, were not specifically designed for use with people with learning disability, and required the input of an informant.

Chapters Four to Six discuss the development of a new autism screening tool, which was designed specifically for people with learning disability and does not require the input of a third-party informant. Chapter Four reports on the results of interviews with stakeholders about the current autism diagnostic pathway, the role of screening as a part of the diagnostic process and the properties that would be desirable in a new screening tool to ensure it is both accessible and useful. Overall, this study highlighted the need for such a tool to have good psychometric properties, be clinically useful, user friendly, cost-effective, and minimize demands on users and those being screened.

Chapter Five explored the literature that provided a theoretical basis for the inclusion of particular items within a behavioural screening tool. These are items that will prompt reactions which are indicative of the presence or absence of autistic traits: the concepts of empathy, mimicry, and contagion. These ideas are operationalised in a series of pilot studies which use existing video clips as stimuli and ask participants to self-report their responses to the videos. The results of a machine learning analysis, using cforest, showed that the self-reported responses were broadly predictive of the person’s AQ score and provided proof of concept of the idea of a behavioural screening tool. Chapter Six extended this work. Here a set of custom stimuli were created and were viewed by autistic and non-autistic participants, with and without learning disability. Their reactions to these were self-reported and video recorded, and subsequently coded. The results indicated that, while scores on the behavioural screen could not discriminate between autistic and non-autistic participants, or predict self-reported AQ score, they did predict informant AQ scores.

Overall, the thesis makes a novel and significant contribution to the literature by identifying tools which are currently available to screen for autism in people with learning disability, alongside adapting the AQ to be more accessible to a wider range of people. It provides proof of concept for a behavioural screening tool for autism, which is specifically designed for people with learning disability, that integrates stakeholder views and theoretical literature during its initial development and testing.

Actions (login required)

View Item

Downloads