The co-occurrence of autism and avoidant/restrictive food intake disorder (ARFID): a prevalence-based meta-analysis

Objective

Avoidant/restrictive food intake disorder (ARFID) is a feeding and eating disorder characterized by extensive avoidance and/or restriction of food. Existing research demonstrates that ARFID is over-represented in Autistic populations and vice-versa, with both groups exhibiting shared characteristics. This meta-analysis investigated the co-occurrence between ARFID and autism via determination of autism prevalence in ARFID populations, and ARFID prevalence in Autistic groups. 

Method

This review systematically identified literature evaluating those with ARFID and Autistic individuals. Literature was searched for using SCOPUS, MEDLINE, and Web of Science. Selected publications included Autistic individuals and those with ARFID who either received a formal diagnosis of autism and/or ARFID or met clinical threshold cut-off scores on validated autism and/or ARFID questionnaires. Prevalence was reported in proportion-based values alongside 95% confidence intervals (CIs).

Results

This meta-analysis identified 21 studies (kARFID = 18 papers; kAutism = 3 papers) comprising of n = 7442 participants (nARFID = 1708; nAutism = 5734). Prevalence of autism diagnoses was 16.27% in those with ARFID (95% CI = 8.64%–28.53%), and ARFID prevalence in Autistic groups was 11.41% (95% CI = 2.89%–35.76%). Gender and ethnicity served as significant sources of heterogeneity in ARFID papers. There was insufficient data to provide comparator values or prevalence across study population and distinct underpinning drivers of ARFID. 

Conclusions

Meta-analytic findings highlight significant rates of co-occurrence between autism and ARFID, suggesting that in clinical settings, it may be beneficial to consider screening Autistic individuals for ARFID and vice-versa. Future research should further investigate co-occurrence across ARFID profiles, gender, and ethnicity.