Prevalence of Co-Occurring Mental Health Diagnoses in the Autism Population: A Systematic Review and Meta-Analysis

Background: Autism spectrum disorder (hereafter ‘autism’) is a neurodevelopmental condition characterized by social-communication difficulties and repetitive/stereotyped behaviours. Autism has a prevalence of around 1% worldwide, and co-occurring mental health/psychiatric conditions (CMHCs) are common, where it is estimated that as many as 70% of individuals with autism have at least one CMHC, and almost 50% are diagnosed with multiple CMHCs. For autistic people, having CMHCs impedes quality of life and leads to poorer long-term outcomes. Accurate pooled prevalence estimates for CMHCs in autistic individuals are needed to enhance recognition and care. However, prevalence estimates in the current literature vary greatly across studies.

Objectives: (1) To determine the best estimates of CMHC in autism and; (2) to identify moderators that account for heterogeneity across studies.

Methods: We conducted a systematic review and meta-analysis according to MOOSE and PRISMA standards. Published studies from 1993 to September 2018 were screened, and articles were included if they were (1) published in English or French, (2) reported on the prevalence rates of CMHCs in individuals with autism and (3) reported confirmed clinical diagnoses of CMHC and autism using DSM, or ICD-based criteria. Abstracts and full-text articles were reviewed, and risk of biases was assessed. The overall pooled estimate of prevalence for different CMHCs in autism was determined using a random-effects model, and reported as a percentage with 95% confidence intervals. Estimates were then stratified by study design (population/registry vs. clinical sample based). Heterogeneity was investigated using random-effects meta-regression models.

Results: Out of 9,515 unique studies, 420 were selected for full text review. CMHCs with sufficient data for meta-analysis (83 studies) showed overall pooled estimates of: ADHD 33% (95%CI 29-37%), anxiety disorders consistent with DSM-IV, DSM-5 or ICD-10 classification 23% (19-27%), anxiety disorders consistent with DSM-5 classification only 22% (17-27%), sleep-wake disorders 13% (7-20%), depressive disorders 12% (9-14%), obsessive-compulsive disorder 10% (8-13%), disruptive, impulse control and conduct disorders 10% (8-13%), schizophrenia spectrum disorders 5% (4-7%), and bipolar disorders 5% (3-7%). Estimates in clinical studies were higher than population/registry-based studies for most CMHCs. Age, gender, intellectual disability and country of origin were associated with heterogeneity across studies.

Conclusions: Given that evidence-based interventions are available, careful assessment of mental health concerns is an essential component of care for all autistic people.