Abstract
IMPORTANCE: Gestational diabetes has been associated with risk of neurodevelopmental disorders (NDD). An improved understanding of this association can inform prevention strategies and elucidate underlying mechanisms. OBJECTIVE: To determine associations between prenatal glucose intolerance and NDD and examine differences by gestational timing and child sex. DESIGN, SETTING, AND PARTICIPANTS: This population-based case-control study examined data from electronic health records from mother-child pairs in an integrated health system in northern California. Children born January 1, 2011, to December 31, 2018, and their mothers were eligible; children were followed up for outcomes through 2023. Data were analyzed from February 2024 to March 2025. EXPOSURES: Gestational diabetes was determined from routine prenatal test results and categorized as diagnosed early (less than 24 weeks), standard (24 to 28 weeks), or late (more than 28 weeks) in gestation. Prenatal subclinical impaired glucose tolerance (IGT) was defined by elevated glucose screening tests and neither GDM diagnosis nor treatment. MAIN OUTCOMES AND MEASURES: Autism spectrum disorder (ASD) and developmental delay were determined from medical records. Adjusted odds ratios (aOR) for associations between prenatal exposures and NDD were estimated using multivariable logistic regression models, adjusted for child sex, birth year, maternal age, race and ethnicity, education, parity, gestational age at prenatal care entry, and prepregnancy body mass index. Effect modification was evaluated by GDM diagnosis timing and sex. RESULTS: A total of 4546 mother-child pairs (median [IQR]) age of diagnosis: ASD, 3.0 [2.0-5.0] years; developmental delay, 2.0 [1.0-3.0] years; 2697 male children [59.3%]) were included in the study, of which 403 mothers (8.9%) had GDM and 64 (1.4%) had IGT; 683 children [15.0%] had ASD, 2054 [45.2%] had developmental delay, and 1809 [39.8%] were controls. GDM was not associated with increased odds of ASD (aOR, 1.15 [95% CI, 0.83-1.60]) or developmental delay (aOR, 1.24 [95% CI, 0.98-1.57]) overall. In sex-stratified analyses, GDM was associated with increased odds of ASD only among females (females: aOR, 2.05 [95% CI, 1.15-3.56]; males: aOR, 0.93 [95% CI, 0.62-1.37]; P for interaction = .04). When assessed by timing, early GDM was associated with increased odds of ASD among females (aOR, 3.23 [95% CI, 1.11-8.91]) but not among males (aOR, 0.78 [95% CI, 0.38-1.56]; P for interaction = .02). There were no associations between standard or late GDM and ASD in either sex. Prenatal IGT was associated with increased odds of developmental delay among females only (females: aOR, 3.25 [95% CI, 1.34-8.68]; males: aOR, 1.07 [95% CI, 0.50-2.39]; P for interaction = .08). CONCLUSIONS AND RELEVANCE: In this case-control study, GDM was associated with NDD in a gestational timing- and sex-specific manner. IGT associations with NDD were also sex-specific, adding to a body of research demonstrating influences of prenatal IGT on child outcomes.