Maternal Exposure to Childhood Abuse Is Associated with Elevated Risk of Autism

Background:  Adverse perinatal circumstances have been associated with increased risk of autism.  Women exposed to childhood abuse experience more adverse perinatal circumstances than women unexposed, but whether childhood abuse is associated with autism in offspring is unknown.

Objectives:  To determine whether maternal exposure to childhood abuse is associated with increased risk of autism, and whether possible increased risk is accounted for by higher prevalence of adverse perinatal circumstances among abused women.

Methods:  We identified children with autism spectrum disorders (ASD) of mothers in the Nurses’ Health Study II, which includes women from across the entire United States.  Children with ASD were identified by maternal report (n=451), validated with the Autism Diagnostic Interview-Revised in a subsample. Controls (n=52,498) were children of women who did not report autism in offspring; a single child was randomly selected if a control mother had more than one child.  Combined childhood physical and emotional abuse and sexual abuse was assessed via a questionnaire administered separately from the questionnaire about children with ASD. We used generalized estimating equations with a log link and Poisson distribution to estimate risk ratios and 95% confidence intervals (CI) for the association between exposure to childhood abuse and risk of autism in the children of mothers.

Results:  Exposure to abuse was associated with increased risk of autism in children in a monotonically increasing fashion. The highest level of abuse was associated with the greatest prevalence of autism (1.8% versus 0.7% in women not abused, P = 0.005) and the greatest risk for autism after adjustment for demographic factors (risk ratio=3.7, 95% confidence interval=2.3, 5.8).  All adverse perinatal circumstances were more prevalent in women abused except low birth weight.  Adjusted for perinatal factors, the association of maternal abuse with autism was slightly attenuated (highest level of abuse, risk ratio = 3.0, 95% confidence interval=1.9, 4.9).  In these models low birth weight (<5 pounds), gestational diabetes, smoking during pregnancy, abortion prior to birth, and intimate partner abuse in the calendar year before the birth year were all associated with higher risk of ASD, but toxemia, alcohol intake, and pregnancy duration were not.

Conclusions:  We identify an intergenerational association between childhood exposure to abuse and risk for autism in the subsequent generation. Adverse perinatal circumstances accounted for only a small portion of this increased risk.