Panel Paper: Are Systemizing Parents More Likely to Have Children with Autism Spectrum Disorder? Evidence from Denmark

Saturday, November 4, 2017
Toronto (Hyatt Regency Chicago)

*Names in bold indicate Presenter

N. Meltem Daysal1, Todd Elder2, Judith Hellerstein3, Scott Imberman2 and Chiara Orsini4, (1)University of Southern Denmark, (2)Michigan State University, (3)University of Maryland, (4)London School of Economics


Medical research has theorized that Autism Spectrum Disorder (ASD) is an extreme realization of systemizing personality traits - a genetically passed aversion to uncertainty and preference for perceiving the world in terms of ordered systems. It has further been suggested that increased assortative mating (AM) on systemizing could explain some of the recent increase in ASD incidence - from 0.7% to 1.5% of children in the US from 2000 through 2012. However, empirical investigations of this risk factor for ASD have generally been restricted to data with small samples from select populations and limited information on systemizing traits. To address this gap in our understanding, we use detailed registries data from Denmark and proxy for whether a parent or grandparent is a systemizer through their occupational and educational field choices. Thus we ask whether, conditional on education level and a rich set of data on economic statuses, a child with parents in systemizing fields are more likely to be diagnosed with ASD. In addition to the large population and information set, Denmark provides an ideal setting for this analysis as it has ASD rates and growth similar to that in the US and universal insurance coverage combined with substantial financial benefits for disabled children limits differences in tendencies to seek diagnoses.

Preliminary results using field of education for people with some post-secondary education indicates that having a father in a systemizing field increases ASD likelihood by 21%. Mothers have a positive but statistically insignificant impact, but the likelihood increases by 36% if both are systemizers. However, while AM based on systemizing has not changed over the analysis period while ASD incidence doubles, casting doubt on AM being an explanation for increased incidence. Additional work will also look at occupations and use O-Net occupational skills databases to more finely define systemizing.