Association between greenspace exposure before, during, and after pregnancy and autism spectrum disorder in offspring

Greenspace exposure in the period surrounding pregnancy may influence autism spectrum disorder (ASD) risk in offspring by reducing risk factors or mitigating effects through various pathways. Current research is limited but suggestive.

We explored the association between greenspace exposure during pregnancy and ASD risk, assessing potential periods of susceptibility: 3 months preconception; first, second, and third trimester; and 3 months post-birth.

We conducted a nested case-control study within the Nurses' Health Study II (NHSII), a US prospective cohort followed up biennially. Cases were children of NHSII participants who were maternally reported to have ASD (n = 245). Controls were randomly selected from the full cohort and frequency matched to cases by birth year (n = 1526). Greenspace exposure was assessed using the satellite-based normalized difference vegetation index (NDVI) measured within 270 m and 1230 m radial buffers of the residential address of participants. Temporally matched, time-linked NDVI was used to calculate greenspace exposure for each potential period of susceptibility. Multivariable adjusted logistic regression models were applied to obtain effect estimates.

Greenspace exposure during pregnancy was inversely associated with ASD risk when NDVI was measured within a 270 m radial buffer of the residential address. Specifically, an interquartile range (0.144) increase in NDVI during the first trimester decreased the odds of ASD by 25% (odds ratio=0.75, 95% confidence interval: 0.56, 0.99) in a model adjusting for other time periods of exposure. There was no other 3-month exposure period significantly associated with ASD. Adjusting for PM_2.5 did not change results. The analysis of NDVI measured within a 1230 m radial buffer showed weaker and inconsistent associations.

This study found that greenspace exposure during pregnancy was inversely associated with ASD, with the first trimester being a critical exposure period. Implications for urban and city design provide compelling reasons to increase research in this field.

Ours is the first study to report an inverse association between greenspace exposure during pregnancy and autism spectrum disorder risk in offspring that was specific to the first trimester. An interquartile range increase in satellite-based normalized difference vegetation index exposure (0.114) within a 270 m radial buffer of the residential address decreased the odds of ASD by 25% (odds ratio=0.75, 95% confidence interval: 0.56, 0.99). Future research is warranted to confirm these findings in other populations and explore the pathways by which greenspace may mitigate risk. Implications for urban and city design provide compelling reasons to increase research in this field.