Friday, 14 March 2014

Environmental exposure and autism incidence?

So: "After controlling for ethnicity, gender, and socioeconomic factors, the strongest predictor of ASD [autism spectrum disorders] was the rate of male congenital malformations of the reproductive system, used as an approximate measurement for exposure to teratogens, based on extensive epidemiological evidence".

That was one of the conclusions reached by the study by Andrey Rzhetsky and colleagues [1] (open-access) which looked at spatial incidence patterns of autism and intellectual disability (ID) based on an insurance claims database in the United States.

The media have been pretty much all over this study as per headlines such as: "Growing Evidence That Autism Is Linked to Pollution" and "Autism disorders greatly linked with environmental factors, study claims". That last headline comes complete with a picture of someone spraying crops, I assume intended to link something like pesticide use and the increase in cases of autism which we've witnessed over the past decades. I'm slightly unsure however that this is the best picture to use in this particular case though (see here).

The details of the Rzhetsky study are pretty widely available (including in the paper which is open-access) but a few points might be of interest:

  • As the publishing journal title suggests, this was a study based on statistics and data. Researchers based their analysis on the Truven Health Analytics MarketScan Commercial Claims and Encounters Database which includes data on some 138 million people dating back from 1995. They scanned approximately 105 million patient records collected between 2003 and 2010 for the purposes of their study.
  • As well as collecting data on variables like gender and ethnicity, they looked at various variables according to "eligibility in special education programs under the Individuals with Disabilities Education Act" to ascertain cases of autism and intellectual disability, and the spatial incidence patterns according to county-level geography within the United States.
  • Using data on the rate of congenital malformation of the reproductive system (genitals) for males and females and other variables related to viral exposures as "county-level environmental indicators", they examined whether such surrogate markers of environmental exposures tied into the rates of autism.
  • Results: yes, quite a few of them. "Adjusted for gender, ethnic, socioeconomic, and geopolitical factors, the ASD incidence rates were strongly linked to population-normalized rates of congenital malformations of the reproductive system in males (an increase in ASD incidence by 283% for every percent increase in incidence of malformations, 95% CI: [91%, 576%], p<6×10−5)". What this translates as is that for every 1% increase in male genital malformation, there was a 283% increase in autism in the same area. This compared with only a 94% increase for ID.
  • Big quote coming up... "The estimated county-level random effects exhibited marked spatial clustering, strongly indicating existence of as yet unidentified localized factors driving apparent disease incidence". Ergo something(s) environmental potentially driving the increase in cases of autism.

I note some quite sensible discussion about this study on the Autism Speaks website (see here) which rightly applauds this work but with caveats, including how one arrives at a diagnosis of autism differing from place to place. Remember too that this was not a study looking directly for metabolites of candidate exposure chemicals (and the mis-use of that word!) in mums, dads or children. 

Still, I find myself intrigued by the results from Rzhetsky et al and where they perhaps point to with the autism research agenda in mind. I think back to the Landrigan paper / editorial [2] (see here) a while back which listed the top 10 chemicals, and important mixtures of chemicals, thought to be causing "developmental neurotoxocity". Certainly things like endocrine disruptors have a presence on that list, pregnancy exposure to which could lead to offspring genital malformations [3] but again, one needs to be slightly cautious about interpreting the strength of such a connection and the various factors influencing any greater risk. That being said, the evidence is mounting that exposure might increase the risk of offspring autism as per the data from de Cock and colleagues [4] and Windham and colleagues [5] and the requirement for a lot more study in this area.

I'm going to finish with a quote from one of the media stories on this paper: "The takeaway is that the environment may play a very significant role in autism, and we should be paying more attention to it,” Rzhetsky said. “It really shifts the emphasis from genetics to more of an environmental side. We should definitely take into account environmental factors".

I struggle to disagree with that sentiment, although, as we've seen with the air pollution - autism correlation work, would be minded to suggest we don't chuck baby and bathwater out at the same time [6] and keep an open mind about that new sheriff in town that is [transgenerational] epigenetics [7] ...

[Update: I forgot to also mention the study by Braun and colleagues too...]


[1] Rzhetsky A. et al. Environmental and State-Level Regulatory Factors Affect the Incidence of Autism and Intellectual Disability. PLoS Comput Biol. 2014;  10(3): e1003518. 

[2] Landrigan PJ. et al. A research strategy to discover the environmental causes of autism and neurodevelopmental disabilities. Environ Health Perspect. 2012 Jul;120(7):a258-60.

[3] Morales-Suárez-Varela MM. et al. Parental occupational exposure to endocrine disrupting chemicals and male genital malformations: a study in the Danish National Birth Cohort study. Environ Health. 2011 Jan 14;10(1):3. 

[4] de Cock M. et al. Does perinatal exposure to endocrine disruptors induce autism spectrum and attention deficit hyperactivity disorders? Review. Acta Paediatr. 2012 Aug;101(8):811-8.

[5] Windham GC. et al. Autism spectrum disorders in relation to distribution of hazardous air pollutants in the san francisco bay area. Environ Health Perspect. 2006 Sep;114(9):1438-44.

[6] Frazier TW. et al. A Twin Study of Heritable and Shared Environmental Contributions to Autism. J Autism Dev Disord. 2014 Mar 7.

[7] Wolstenholme JT. et al. Transgenerational effects of prenatal bisphenol A on social recognition. Horm Behav. 2013 Nov;64(5):833-9


ResearchBlogging.orgRzhetsky, A., Bagley, S., Wang, K., Lyttle, C., Cook, E., Altman, R., & Gibbons, R. (2014). Environmental and State-Level Regulatory Factors Affect the Incidence of Autism and Intellectual Disability PLoS Computational Biology, 10 (3) DOI: 10.1371/journal.pcbi.1003518