Posted October 21, 2013
Alberto Ascherio, M.D., Dr.PH, Susan Santangelo, Sc.D., Marc Weisskopf, Sc.D., Ph.D., Harvard University, Boston, Massachusetts

Alberto Ascherio, M.D., Dr.PH, Susan Santangelo, Sc.D., and Marc Weisskopf, Sc.D., Ph.D. In the first year (Fiscal Year 2007) of the Autism Research Program (formerly the Autism Spectrum Disorder Research Program), Drs. Alberto Ascherio, Susan Santangelo, and Marc Weisskopf received an Idea Development Award to collaborate and research maternal risk factors for Autism Spectrum Disorder (ASD). Utilizing the unique resource of the Nurses' Health Study II (NHS II), the investigators, along with their research fellows, Drs. Kristen Lyall, and Andrea Roberts studied a broad range of potential maternal and environmental risk factors. Of interest, the researchers looked at dietary fats, particularly polyunsaturated fats, maternal stressors that have been associated with ASD, and prenatal exposure to air pollution with autism.

This multidisciplinary study resulted in several key findings. Looking into prenatal factors that might influence outcomes such as ASD, the research team found that mothers with gestational diabetes had an 80% increased risk of having a child with ASD. The use of in vitro fertilization and other assisted reproductive technologies was not significantly associated with having a child with ASD, nor was a history of infertility. It was noted, though, that for women aged 35 years and older, there were modest associations between use of ovulation-inducing drugs and having a child with ASD.

Maternal exposures, especially diet, have been an area of concern for long-term developmental outcomes for children. In this study, the researchers found that intake of high amounts of linoleic acid lowered the risk of having a child with ASD, while mothers with very low intake of omega-3 fatty acids had a significant increase in offspring risk of autism compared to women with higher intakes of omega-3s. Both linoleic and omega-3 belong to a group of fats called polyunsaturated fats, which are essential for brain development.

In further investigations, it was found that ASD risk increased with perinatal exposure to several hazardous air pollutants, including diesel particulate matter, lead, manganese, mercury, methylene chloride, and other metals. These analyses accounted and adjusted for demographic factors and other variables that may influence the association. Pollutants associated with ASD in prior research were more than four times as likely to be associated with ASD as other pollutants not previously related to ASD. Therefore, this study adds weight to the evidence that certain air pollutants, in particular heavy metals, may increase risk for autism.

The research team also looked at social factors that might influence long-term developmental outcomes, such as maternal exposure to abuse (physical, emotional, and sexual) in early childhood. Perinatal factors, including pregnancy complications, were more common in abused mothers, but accounted for only a small part of the association with ASD risk. The researchers also examined evidence for family patterns of social functioning and autistic traits in a large analysis of Social Responsiveness Scale scores. Parents of children with ASD were significantly more likely to have both parents with high scores (in top 20% of distribution) than were parents with typically developing children, demonstrating heritability of autistic traits. Analysis of key candidate genes, and how they may relate to the above findings, is under way.

Most previous studies have been limited by small numbers or use of data collected after the child's diagnosis, which can lead to biased results. A number of the analyses funded under this award, however, used data from NHS II questionnaires collected before the child's birth and included up to 507 cases and 22,098 controls. In addition to the size of the cohort and its wealth of data, validation studies have demonstrated that these educated nurse participants are highly accurate at providing health information, which suggests strength of maternally-reported information used.

This work provides suggestions of underlying pathways to an autism diagnosis. The next steps will be to determine the biological pathways that link these factors to autism, and also to replicate and expand novel findings. Knowledge of the underlying biology may aid in the development of treatments for those exposed. Future work that builds on these findings may also lead to dietary suggestions for expectant mothers or interventions for young children that may reduce symptoms; personal and public interventions to reduce air pollution exposure; and other public health strategies and policies aimed at reducing prevalence and burden on families.


Perinatal Air Pollutant Exposures and Autism Spectrum Disorder in the Children of Nurses' Health Study II Participants. Roberts AL, Lyall K, Hart JE, Laden F, Just AC, Bobb JF, Koenen KC, Ascherio A, Weisskopf MG Environ Health Perspect. 2013 Aug;121(8):978-84.

Maternal dietary fat and fatty acid intake and autism spectrum disorders. Lyall K., Munger K., O Reilly E., Santangelo S.L., Ascherio A. (2013). American Journal of Epidemiology, 178(2):209-20.

Pregnancy complications and obstetric suboptimality as risk factors for autism spectrum disorders in children of the Nurses' Health Study II. Lyall K., Pauls D.L., Speigelman D., Ascherio A., Santangelo S. (2011). Autism Research, 5(1):21-30.

Fertility therapies, infertility, and autism spectrum disorders in the Nurses' Health Study II. Lyall K., Pauls D.L., Santangelo S., Speigelman D., Ascherio A. (2012). Paediatric and Perinatal Epidemiology, Jul;26(4):361-72.

Association of maternal exposure to childhood abuse with elevated risk for autism in offspring. Roberts, A.L., Lyall, K., Rich-Edwards, J.W., Ascherio, A. & Weisskopf, M.G. (2013) Journal of the American Medical Association Psychiatry, 70(5):508-15.


Public and Technical Abstracts: Maternal Risk Factors for Autism Spectrum Disorders in Children of the Nurses' Health Study II

Top of Page