52 / 67
50
Post-Doctoral Fellowships
Association between Air Pollution and Incidence of Autism Spectrum Disorder
in the Nurses’ Health Study II
Autism spectrum disorder (ASD) is a developmental disorder with increasing incidence worldwide,
yet with unclear etiology. Recent studies suggest that environmental factors in fetal development
are part of the etiology. The Nurses’ Health Study II (NHS II) is a large cohort of female nurses in the
United States that have been followed with biennial questionnaires since 1989. The autism project
in NHS II includes participants from across the US, and utilizes a nested case control approach,
eliminating many of the potential biases associated with comparisons in previous studies.
The aim of the current project is to explore the association between maternal exposure to particulate
matter (PM) air pollution and risk of ASD. Mother’s address at birth was geocoded and linked to
ambient PM
10-2.5
and PM
2.5
estimates, using the generalized additive models of these pollutants for
the continental US. These models use ambient PM data from air quality and research monitoring
networks, meteorological variables from weather databases and site characteristics from GIS.
Average exposure measures were calculated for the period the mother was pregnant with a case or
control child in the sample, and for adjacent 9-months periods. Subjects were NHS II participants’
children born between 1990-2002 with ASD (n=245) and without ASD (n=1,522). Participants
were randomly selected using frequency matching for birth years and ASD diagnosis was based on
validated maternal report.
We found that PM
2.5
exposure during pregnancy was associated with increased odds of ASD, with
the 4th exposure quartile having an adjusted odds ratio (OR) of 1.71 (95% CI: 1.10, 2.66) compared
with the 1st quartile. The association was stronger when limiting the sample to those who did not
change residences around the time of pregnancy (OR=2.02, 95% CI: 1.15, 3.54). ASD associations
with PM
2.5
exposure nine months before or after the pregnancy were substantially lower. During
pregnancy, PM
2.5
exposure during the third trimester had the strongest association (OR=1.49 per
inter-quartile range increase in PM
2.5
[4.40 µg/m
3
], 95% CI: 1.20, 1.85) with ASD. There was no
association between PM
10-2.5
and ASD. In conclusion, higher exposure to PM
2.5
during pregnancy,
and especially during the third trimester, is associated with risk of ASD.
Research publications
(1)
Raz, R., Weisskopf, M.G., Davidovitch, M., Pinto, O., & Levine, H. (2014). Differences in Autism Spectrum Disorders
incidence by sub-populations in Israel 1992–2009: A total population study.
Journal of Autism and Developmental
Disorders
. Advance online publication. doi:10.1007/s10803-014-2262-z
Fellow
Raanan Raz
Harvard University, USA
Supervisor
Marc Weisskopf
2012-2014