Children exposed in utero to pre-existing type 1, type 2 diabetes or gestational diabetes diagnosed by second trimesters are at increased risk of developing autism spectrum disorders (ASD). Here we assessed the role of maternal glycemic levels using hemoglobin A1c. Data were from 35,819 singleton children born in 2012 and 2013 with maternal HbA1c measured during the first two trimesters of pregnancy (median=9.0 weeks, interquartile range (IQR) 6.7-11.4 weeks). All children were from an integrated healthcare system and enrolled in the health plan by age one. Children were tracked from age one until December 31, 2017. Risk of ASD associated with maternal HbA1c levels were estimated by hazard ratio (HR) using Cox regression. Potential confounding by maternal age, maternal obesity, parity, education, race/ethnicity, history of comorbidity, and sex of the child were assessed. During a median of 4.5 (IQR of 4.0-5.0) years after birth, 707 (2.0%) children were diagnosed with ASD. Univariately, the HR of ASD associated with each 1% increase of HbA1c was 1.23 (95% CI 1.08-1.41; p=0.002). When analyzed categorically, compared to exposure to HbA1c <5.7% (n=30,419), the HR (95% CI) associated with exposure to HbA1c 5.7% to 5.9% (n=4,172), 6.0% to 6.5% (n=871) and >6.5% (n=357) were 1.24 (1.01-1.54; p=0.04), 0.98 (0.61-1.59; p=0.94) and 2.17 (1.30-3.62; p=0.003) respectively. None of the potential confounders changed the model estimates by more than 10% except for maternal pre-pregnancy BMI and race/ethnicity. Adjusting for both reduced the corresponding HRs to 1.08 (0.87-1.35; p=0.46), 0.79 (0.48-1.28; p=0.33) and 1.79 (1.06-3.00; p=0.03) respectively. Thus, exposure to hyperglycemia defined by HbA1c >6.5% in the first two trimesters was associated with an increased ASD risk in offspring. Improving glycemic control during pregnancy may reduce the risk of ASD in children, and monitoring HbA1c may assist in identifying at-risk children for early intervention to reduce ASD disabilities.


A. Xiang: None. T. Chow: None. M.P. Martinez: None. D. Getahun: Research Support; Self; Bayer AG. K.A. Page: None. T.A. Buchanan: Research Support; Self; Allergan, Apollo EndoSurgery. R. Feldman: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at