The prevalence of autism spectrum disorder traits and diagnosis in adults and young people with personality disorders: A systematic review

Objectives: Autism spectrum disorders and personality disorders are spectrum conditions with shared clinical features. Despite similarities, previous attempts to synthesise literature on co-existing prevalence and shared traits have employed a unidirectional focus, assessing personality characteristics of individuals with an autism spectrum disorder diagnosis. Here, we assess the prevalence of autism spectrum disorder diagnosis and/or traits among persons diagnosed with a personality disorder. Methods: We systematically reviewed the English-language literature following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, according to a pre-registered protocol (PROSPERO: CRD 42021264106). Peer-reviewed quantitative studies reporting the prevalence of autism spectrum disorder diagnosis or traits in persons with an established personality disorder diagnosis were included. Studies were critically appraised using the Appraisal tool for Cross-Sectional Studies. Results: Fifteen studies were identified, including 72,902 participants (median: 48, interquartile range: 30-77). Diagnoses included borderline, schizotypal and obsessive-compulsive personality disorders, and cohorts with unspecified personality disorder diagnoses. There was significant heterogeneity in diagnostic methodology and assessment tools used. We identified preliminary evidence of an increased prevalence of co-existing autism spectrum disorder diagnosis and traits among those diagnosed with a personality disorder, although significant limitations of the literature were identified. Conclusion: Our research suggests clinicians should consider conducting a careful developmental assessment when assessing service-users with possible or confirmed personality disorder. Future research directions may include larger studies featuring clinical control groups, an exploration of shared and differentiating behavioural-cognitive features of the two conditions, and investigation into potentially shared aetiological factors. Research investigating demographic factors that may contribute to potential diagnostic overshadowing would also be welcomed.