Autism and psychosis: Clinical implications for depression and suicide

Abstract

There is increasing recognition of the co-occurrence of autism and schizophrenia spectrum disorders. However, the clinical significance of this on outcomes such as depression and suicidal thinking has not been explored. This study examines the association of autism spectrum traits, depressive symptoms and suicidal behaviour in individuals with psychotic experiences.

In two cross sectional studies, individuals from a non-help seeking university student sample and patients with first episode psychosis (FEP) service completed standardized measures of autism spectrum traits, psychotic experiences, depressive symptoms and suicidal thinking.

In healthy non-help seeking students, increased autism traits and increased subclinical psychotic experiences were significantly associated with depressive symptoms; a significant interaction effect suggests their combined presence has a greater impact on depression. In FEP, high autism traits and positive symptoms were associated with increased depression, hopelessness and suicidality, however there was no significant interaction effect. In FEP a multiple mediation model revealed that the relationship between autism traits and risk for suicidality was mediated through hopelessness.

Young people with subclinical psychotic experiences and all patients with FEP should be screened for autism spectrum traits, which may have significant impact on clinical outcomes. Tailored interventions for patients with high levels of autistic spectrum co-morbidities in FEP should be a priority for future research.

Introduction

Despite early recognition of similarities between autism and schizophrenia at the times of Bleuler and Kanner, historically these have been understood as two distinct disorders (Kolvin, 1971). Time course (onset in early childhood vs adolescent) and progression (stable vs progressive) are two possible distinguishing features (Chisholm et al., 2015, Wood, 2017). However, it is now recognised that autism spectrum disorder and schizophrenia spectrum disorders share not only some clinical similarities, but also a biological, particularly genetic, liability (Ruzzo and Geschwind, 2016). While autism has long been understood as occurring on a spectrum, with a continuum of severity and deficits identifiable from population to disorder level (Young et al., 2005), recently psychosis has also been suggested to exist on a continuum; positive symptoms are reported in the general population, can be used to identify those at heightened risk of developing psychosis with the ultra-high risk paradigm, and are seen in schizophrenia and other severe mental illnesses (Yung et al., 2006).

At the population level, autistic and psychosis traits are evidentially co-occurring; using the Avon Longitudinal Birth Cohort, Sullivan et al. demonstrated that poorer pragmatic language, a key deficit in autism spectrum disorders, was associated with later psychosis and depression (Sullivan et al., 2016). In the same cohort, Siebald and colleagues showed that children aged 8 diagnosed with pervasive developmental disorders, which included the DSM-IV classification of Autism (Association and DSM-IV., 1994), had heightened risk of later psychotic experiences with an odds ratio of 8 (Siebald et al., 2016).

Within clinical populations, the two disorders co-occur at higher rates than would be expected by chance. There is a reported mean incidence of schizophrenia spectrum disorders of 13.8% in autism spectrum disorder populations, and a mean incidence of autism spectrum disorder of 24.1% in schizophrenia spectrum disorder populations (Chisholm et al., 2015). Shared clinical features of autism spectrum disorders and schizophrenia spectrum disorders also include difficulty with social communication, emotional expression, salience and restrictive behaviours (Abu-Akel et al., 2017a, Abu-Akel et al., 2017b, Hommer and Swedo, 2015, Ruzzo and Geschwind, 2016). Depression is common in both autism spectrum disorder and schizophrenia spectrum disorders. It has been reported that 14–20% of young people with autism spectrum disorders have experienced a significant depressive episode by the age of 18 (Gotham et al., 2015). Likewise, up to 80% of patients with schizophrenia may experience significant depression (Upthegrove et al., 2010), and this has significant long-term consequences for autism spectrum disorders and schizophrenia, including suicide. Lifetime prevalence of completed suicide in schizophrenia is between 7 and 10% (Dutta et al., 2011, Häfner et al., 2005, Upthegrove et al., 2010), and the prevalence of suicidal behaviour in autism spectrum disorder is between 11 and 50% (Zahid and Upthegrove, 2017).

We have previously shown that depression, positive symptoms and hopelessness are known precursors of self-harm and suicide in schizophrenia spectrum disorders (Upthegrove et al., 2014). However, the role of autism spectrum disorder traits as a potential pathway to depression and suicidality in schizophrenia spectrum disorders has not been explored to date.

While there has been a considerable amount of recent recognition of the co-occurrence of autism spectrum disorder and schizophrenia spectrum disorders, and how this might inform aetiological understanding, the evidence of the clinical significance of this co-occurrence has yet to develop. This knowledge is needed for evidence-based treatment and management decisions. This study aims to determine the importance, in terms of depression and suicidal behaviour, of autism traits in individuals within the spectrum of psychotic experiences. Given that both disorders are associated with depression and suicidal behaviour, we predict that depression and suicidal behaviour would be increasingly prevalent when autistic spectrum traits and psychosis co-occur.

Hypotheses:

• Autism traits and subclinical psychotic experiences are associated with increased levels of depressive symptomatology in a healthy, non-help seeking population.

• Autism traits are associated with positive symptoms and their combined presence increases depression and hopelessness in people with first episode psychosis.

• Autism traits in people with first episode psychosis increase the risk of suicidality over and above other known risk factors.