Investigating mechanisms underpinning sleep problems and links to mental ill health in autism

Abstract

Autistic individuals experience high rates of sleep problems and mental ill health, which are associated with detrimental impacts including reduced quality of life and higher mortality risk. In the general population, sleep and mental health are bidirectionally linked; however, this relationship remains unclear in the autistic population. Treatments for sleep problems and mental ill health often fall short in the autistic population, partly due to insufficient understanding of the mechanisms behind sleep problems, their links to mental ill health, and the ways these issues evolve over time.

The over-arching aim of this thesis was to longitudinally characterise sleep and mental ill health in the autistic population, while exploring the underpinning mechanisms of sleep problems and their role in contributing to mental ill health. A multimethod approach was used to gain insights over two timescales. First, a national population-based cohort (Avon Longitudinal Study of Parents and Children; ALSPAC) was used to investigate long-term changes and associations from childhood to adolescence. Second, a pilot cohort study was conducted to gain preliminary insights into short-term associations between daily sleep and mood in autistic adolescents. This research was co-produced with a team of experts by lived experience.

First, in the ALSPAC cohort, I found that autistic and non-autistic young people displayed distinct developmental trajectories of sleep and mental health – captured as internalising difficulties – from childhood to adolescence, with the autistic group experiencing worse sleep and higher internalising difficulties from a young age. Between ages 5 and 7, autistic children appear especially vulnerable to sleep and mental health challenges, potentially marking a critical window for intervention.

Second, I identified heterogeneity in long-term night-time sleep duration trajectories in ALSPAC and found that having an autism diagnosis and certain autistic traits (i.e., high social communication difficulties, low speech coherence, high repetitive behaviour) were linked to poorer sleep trajectories.

This highlighted the importance of accounting for heterogeneity in both sleep patterns and manifestations of autism when researching and designing interventions for sleep problems in this population.

Lastly, by combining analytical approaches, I found that sleep and mental health may have a nuanced relationship in the autistic population.

There was no evidence for long-term bidirectional associations between sleep and internalising difficulties in autistic participants in ALSPAC. However, in the cohort study, I found preliminary evidence for a short-term, unidirectional association such poor sleep predicted worse next-day mood in autistic adolescents – but not the reverse. This points to temporal complexity in this relationship and suggests that there might be other factors that moderate this relationship in the long-term.

Overall, this thesis advances the understanding of developmental changes in sleep and mental ill health in the autistic population and provides mechanistic insight into the relationship between these issues in autistic individuals. These insights lay the groundwork for future research that could help improve treatment strategies for sleep and mental health in this population.