Autism Spectrum Disorders: Is Anything Left for the Environment?

Stoltenberg, Camilla

doi: 10.1097/EDE.0b013e31821d0b6d
Autism: Commentary

From the Norwegian Institute of Public Health, Oslo, Norway.

Correspondence: Camilla Stoltenberg, Norwegian Institute of Public Health, Marcus Thranesgate 6, PO Box 4404 Nydalen, N-0403 Oslo, Norway. E-mail:

Article Outline

Autism has a strong genetic basis.1 Abrahams and Geschwind have shown in a comprehensive review of autism genetics how several lines of evidence support genetic factors as the predominant causes of autism. A number of known de novo mutations and mutations inherited as part of rare syndromes increase the risk of autism, and account for at least 10%–20% of the autism spectrum disorders. Twin and family studies provide high estimates of the heritability of autism, some even higher than 90%.2 What then is the role of environmental exposures in the etiology of autism spectrum disorders? Can environmental factors cause or trigger autism in susceptible persons, or will genetic factors eventually explain it all? Although much remains to be discovered concerning the genetics of autism, we know more about the genes at play than we do about possible environmental factors.

A wide range of environmental factors have in fetal or early neonatal life been postulated as causal factors or triggers of autism.1,2 Most of these are speculative, in the neutral sense of the word, and have little or no empirical basis. The long list of environmental factors includes viral infections (particularly rubella); the vaccine for measles, mumps, and rubella (MMR); thimerosal (a mercury preservative used in some vaccines); parental age, and particularly paternal age; folate deficiency or other nongenetic one-carbon metabolism disturbances; hypocholesterolemia; exposure to prenatal testosterone; and hyperactivation of the immune system. Paternal-age effects have received considerable attention lately but are not necessarily strictly environmental. Such effects may reflect accumulating mutations in the spermatocytes, which again may or may not be environmentally induced.

In 2 articles published in the current issue of Epidemiology, Irva Hertz-Picciotto and her collaborators have investigated environmental factors possibly involved in the etiology of autism. One study is on month and season of conception,3 whereas the other is on prenatal vitamins.4 Both studies provide important new results and make use of valuable data sources. The significance of these 2 articles is that they bring legitimacy to the possibility of environmental factors in autism.

The study on month of conception includes more than 6.6 million children born between 1990 and 2002 in California, of whom 19,238 were diagnosed with autism before the age of 6. Children conceived in the winter season (December, January, or February) had a 6% increased risk of developing autism compared with those conceived in July. The study population is large by international and historical comparisons, and allows for analyses that are difficult or impossible to perform in most other countries. Although statistically significant, the effect is small and, as the authors well recognize, at best a rough indicator of true causal effects of environmental factors that are more prevalent during the winter. Little is known about what environmental exposures these might be, although infections are obvious candidates.

The analysis on prenatal vitamins and functional one-carbon metabolism gene variants is based on data from the Childhood Autism Risk from Genetics and Environment population-based case-control study of children in California. The study shows that periconceptional use of prenatal vitamins may reduce the risk of having children with autism, especially among genetically susceptible mothers and children. This study also illustrates the importance of including both genetic and environmental information in the analysis. This is the first account of a protective effect against autism for mothers who report prenatal vitamin supplementation. Altogether 429 cases and 278 controls were included, which is a fairly high number for this kind of analyses. Nonetheless, the analyses need to be replicated in larger studies, and in studies with collection of information on vitamin use prior to the outcome.

In addition to improved knowledge on the genetics of autism, we need large population-based registries and prospective pregnancy cohorts with biologic samples collected at baseline that can provide trustworthy and precise knowledge on environmental factors in autism.5 Slowly but steadily, such infrastructures for etiologic research are developing in several countries. These new resources will eventually allow us to appropriately address a range of old questions about the role of environmental factors in autism.

Back to Top | Article Outline


CAMILLA STOLTENBERG is a physician with a doctoral degree in epidemiology from the University of Oslo. She is responsible for the Autism Birth Cohort Study (ABC) in Norway (Ian Lipkin, PI) and related studies of autism and other neurodevelopmental disorders based on The Norwegian Mother and Child Cohort Study and national health registries. Since 2007, she has served as deputy director general at the Norwegian Institute of Public Health with special responsibility for strategic development of health registries.

Back to Top | Article Outline


1. Abrahams BS, Geschwind DH. Advances in autism genetics: On the threshold of a new neurobiology [review]. Nat Rev Genet. 2008;9:341–355.
2. Rutter ML. Progress in understanding autism: 2007–2010 [review]. J Autism Dev Disord. 2011;41:395–404.
3. Zerbo O, Iosif A-M, Delwiche L, Walker C, Hertz-Picciotto I. Month of conception and risk of autism. Epidemiology. 2011;22:469–475.
4. Schmidt RJ, Hansen RL, Hartiala J, et al. Prenatal vitamins, one-carbon metabolism gene variants, and risk for autism. Epidemiology. 2011;22:476–485.
5. Stoltenberg C, Schjølberg S, Bresnahan M, et al. the ABC Study Group. Perspective: The Autism Birth Cohort (ABC): a paradigm for gene–environment–timing research. Mol Psychiatry. 2010;15:676–680.
© 2011 Lippincott Williams & Wilkins, Inc.