Autism Spectrum Disorders In Relation to Parental Occupational Exposures During Pregnancy

Background: To attempt to understand and explain the continuing rise in the number of children reported with autism spectrum disorders (ASD), it is important to examine the role of exposures to environmental chemicals.  However, reliably ascertaining exposure during the critical period of development, thought to be in utero, can be difficult because affected children are not identified until several years later. Chemical exposures that occur on the job have traditionally been at higher levels than in the general population; thus examining parental occupation as reported at birth of the child provides several advantages for studying exposure.

Objectives: To explore whether mothers of children with ASD are more likely to work in occupations with potential neuro- or repro-toxic exposures during pregnancy, in a population-based sample.

Methods: Subjects included 284 children with ASD who were identified through records-based systematic surveillance and 659 gender-matched controls, born in 1994 in the San Francisco Bay Area of California.  Parental occupation and industry were abstracted from birth certificates and potential exposure to toxic chemicals was identified by a physician certified in Occupational Medicine and checked by an industrial hygienist. Up to three exposures to any of seven chemical groups were also coded for each parent (exhaust/combustion, solvents, pesticides, heavy metals, cooling fluids, disinfectants, and auto paint), as well as higher likelihood of electro-magnetic field (EMF) exposure. Limiting analyses to parents with a usual occupation and industry indicating employment outside of the home or school, odds ratios (AORs) were calculated by logistic regression, adjusting for maternal age, education, and child race.

Results: Among the 60% of mothers who were employed, 11.3% of case mothers worked in chemically-exposed occupations compared to 4.3% of controls (AOR 2.8; 95% CI 1.4-5.5). The exposure categories with the highest and statistically significant AORs were exhaust and disinfectants, but metals and solvents had slightly elevated AORs as well. Including these four in the same model did not substantially alter the AORs for each specific exposure category, except for metals, which was no longer elevated.  EMF exposure was not related to ASD in these data and none of the mothers were exposed to cooling fluids or auto paint. Maternal occupation in the medical/dental field was highly related to ASD in offspring, but based on small numbers (AOR 11.3; 95% CI 1.3—99). Work in laboratories or as chemists was also three times more likely in mothers of cases than controls, but was not statistically significant. Fathers of cases were not more likely than those of controls to work in exposed occupations.

Conclusions: Although exposure assessment was rudimentary, use of birth certificates allows ascertainment of occupation before a child is diagnosed with ASD, which may affect subsequent maternal employment, and avoids issues of recall bias. These descriptive data suggest that maternal occupational exposure may be associated with increased risk of ASD and indicate areas for additional study. The findings will be discussed in relation to other literature on environmental exposures.