Preexisting diabetes and gestational diabetes as risk factors for autism spectrum disorder: an umbrella review.

BACKGROUND

Maternal diabetes during pregnancy, including preexisting type 1 diabetes (T1DM), type 2 diabetes (T2DM), and gestational diabetes mellitus (GDM), has been investigated as a potential risk factor for autism spectrum disorder (ASD) in offspring. However, the consistency and strength of evidence across published meta-analyses remain uncertain. This umbrella review aimed to synthesize and critically appraise the existing meta-analytic evidence on this association.

METHODS

Following the PRISMA 2020 guidelines, we conducted an umbrella review of meta-analyses identified through PubMed, Web of Science, and Google Scholar from inception to March 31, 2026. Eligible studies quantitatively assessed the association between maternal diabetes and the risk of ASD. Of 636 records identified, 11 meta-analyses met the inclusion criteria. Data on effect estimates, heterogeneity, and potential bias were extracted. Study quality was assessed using AMSTAR 2, and overlap among primary studies was evaluated using the corrected covered area (CCA). Given the substantial overlap and methodological variability, the most robust meta-analysis for each exposure was prioritized, and findings were synthesized descriptively without conducting a de novo meta-analysis.

RESULTS

Across included meta-analyses, maternal diabetes was consistently associated with increased ASD risk in offspring. Preexisting diabetes showed a more consistent and stronger association compared with GDM, which also demonstrated a positive but more variable association. Heterogeneity was commonly moderate to high, and publication bias was variably reported. Overall evidence ranged from suggestive to weak, with higher credibility observed for preexisting diabetes than for GDM.

CONCLUSION

A consistent and statistically significant association was observed between maternal diabetes-including both preexisting diabetes and GDM-and ASD, with evidence suggesting a risk gradient according to diabetes type. These findings highlight the importance of maternal metabolic health during pregnancy and support further research into the underlying mechanisms and potential preventive strategies.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261334307, 2026CRD420261334307.