Table 1.
Summary of study characteristics.
| Study ID, ref. | Study design | Sample size | Age (years) | Main findings | Limitations |
|---|---|---|---|---|---|
| Bakkaloglu et al. [73] | Case–control study | N = 71 | 2–4 | 7/30 autistic children and 7/39 age and sex-matched controls reported allergic symptoms (p = ns). | A small sample size of 32 children with autism and 39 controls. |
| Chen et al. [3] | Longitudinal follow-up study | N = 21 756 | 0–3 |
Early childhood atopic disease increased the risk of ASD (HR 3.40, 95% CI: 1.95–5.93). HR of ASD increased with the number of atopic comorbidities. 1: HR 2.14 (95% CI: 0.90–5.11) 2: HR: 2.70 (95% CI: 1.44–5.05) 3: HR: 4.08 (95% CI: 2.24–7.43) 4: HR: 4.29 (95% CI: 2.25–8.19) |
ASD incidence was likely underestimated as the sample only included those who sought medical attention. |
| Gurney et al. [12] | Population-based cross-sectional study | N = 85 272 | 3–17 |
Atopic prevalence in ASD cohort vs. control Hay fever: 27% vs. 16.2% OR: 1.6 (95% CI: 1.1–2.4) Food allergy: 14.1% vs. 3.2% OR: 4.5 (95% CI: 3.0–7.0) Eczema or skin allergy: 14.9% vs. 9.2% OR: 1.4 (95% CI 1.0–2.1) |
Unvalidated parental report of atopic manifestations and no subtype analysis of ASD. |
| Liao et al. [7] | Population-based case–control study | N = 32 956 | 0.08–3 |
HR of ASD increased with the number of eczema-related doctor visits. 1: aHR 1.15 (95% CI 1.06–1.24) 2–3: aHR 1.18 (95% CI 1.07–1.29) 4+: aHR 1.40 (95% CI 1.25–1.56) |
Data may be confounded by patterns of help-seeking behaviour. |
| Magalhães et al. [74] | Case–control study | N = 45 | 7–18 | 86.6% of Asperger patients were atopic (allergic symptoms, elevated serum IgE, high eosinophil count or positive skin prick test), compared to <7% of controls | A small sample size of 15 Asperger patients, 15 atopic neurotypicals and 15 non-atopic neurotypicals |
| Miyazaki et al. [11] | Systematic review and meta-analysis | N = 10 380 | 2.5–18 | Prevalence of asthma and allergic rhinitis were significantly increased in the ASD cohort (OR 1.69; 95% CI: 1.11–2.59 and OR 1.66; 95% CI: 1.49–1.85 respectively) | ASD diagnostic criteria varied across studies (e.g. DSM-V vs. ADOS vs. ICD-10) |
| Tsai et al. [75] | Systematic review and meta-analysis | N = 1 055 837 | 0–18 |
Prevalence of AD was significantly higher in individuals with ASD than the control group OR: 1.485 (95% CI: 1.20–1.83) |
Heterogeneity in diagnostic criteria, age, race and disease severity across studies. |
| Xu et al. [5] | Population-based, cross-sectional study | N = 199 520 | 3–17 | Children with ASD were more likely to report food allergy (11.25% vs. 4.25%), respiratory allergy (18.73% vs. 12.15%) & skin allergy (16.81% vs. 9.84%) than a non-autistic comparison group. | Retrospectively collected self-reported data may have enabled misreporting due to recall bias |
| Yaghmai et al. [10] | Cross-sectional study | N = 92 642 | 0–17 |
Prevalence of ASD was significantly increased in children with eczema (2.19%) compared to non-autistic controls (0.89%). OR ASD: 2.73 (95% CI 1.94–3.84) |
Parental reporting subject to misclassification bias |
| Zerbo et al. [76] | Case–control study | N = 33 390 | 3–26 |
Allergies were more frequently diagnosed in ASD cohort (20.6%) vs. controls (17.7%) OR: 1.22 (95% CI 1.13–1.31) |
Did not account for heterogeneous ASD subtypes. |