Publication: The global prevalence of food selectivity in children with autism spectrum disorder: a three-level meta-analysis
Loading...
Date
2025-07
Authors
Ilham, Nur Aazifah
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Food selectivity (FS) is a common feeding issue among children with autism spectrum disorder (ASD), typically characterised by food refusal, limited dietary variety, or preference for specific food characteristics. A pre-registered systematic review and multilevel meta-analysis (PROSPERO ID: CRD4202459844) using a mixed-effects model was conducted. The objective was to estimate the global pooled prevalence of FS among children with ASD and to explore potential sources of heterogeneity. Comprehensive searches were conducted across PubMed, Scopus, Web of Science, and ScienceDirect database up to 22 September 2024. Observational studies reporting prevalence estimates of FS in children with ASD aged 1–18 years were included. Risk of bias was assessed via Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Studies Reporting Prevalence Data. A total of 18,223 records were identified through database searches. After removing duplicates, 10,239 unique records remained, of which 315 full-text reports were assessed following title and abstract screening. Ultimately, 55 studies (including 3 from citation tracking), comprising 144 prevalence estimates, and 6,251 participants were included in the final meta-analysis. The overall pooled prevalence of FS among children with ASD was 45.2% (95% CI: 39.5% to 51.3%), with heterogeneity (I² = 94.8%, Q (143) = 4417.6, p < 0.001). Subgroup analysis by FS subtype revealed notable variability: general FS exhibited the highest pooled prevalence at 59.9% (95% CI: 50.5% to 68.5%), followed closely by food neophobia at 57.1% (95% CI: 42.8% to 70.2%). Limited food variety and food refusal were associated with intermediate prevalence estimates of 39.3% (95% CI: 29.6% to 49.9%) and 31.7% (95% CI: 15.4% to 54.1%), respectively, whereas sensory-based FS showed the lowest prevalence at 26.0% (95% CI: 18.2% to 35.7%). A test for subgroup differences confirmed that FS subtype significantly moderated prevalence estimates (p < 0.001). Despite accounting for these variables, residual heterogeneity within FS subtype subgroups remained high (I² = 93.2%), indicating the presence of further unexplained variation. Moderator analyses revealed no statistically significant effects for geographical region, income level, risk of bias, publication year, mean age, sample size, or proportion of male participants. Food selectivity affects approximately half of children with ASD globally, with substantial variation across subtypes. These findings highlight the importance of routine screening for food selectivity in children diagnosed with ASD or those undergoing assessment for suspected ASD.
Description
Keywords
Autism spectrum disorder , food selectivity