What is Driving the Increased Prevalence of Autism Spectrum Disorder in the US?

During the NDC Symposium-Neurobiology of Disease in Children: Session I: Clinical Aspects at the 43rd Annual Child Neurology Society Meeting, held in Columbus, OH, October 22-25, Craig J. Newschaffer, PhD, delivered a presentation on the epidemiology of autism spectrum disorder (ASD), focusing on disease prevalence, environment, and risk factors.

Newschaffer is director of the A.J. Drexel Autism Institute, and professor of Epidemiology and Biostatistics at Drexel University in Philadelphia, PA,

According to information from the CDC, ASD prevalence in the US increased from 1 in 166 children in 2000 to 1 in 68 by 2010. Data from the CDC’s Autism and Developmental Disabilities Monitoring Network (ADDM) compiled in 2010 show that estimated ASD prevalence among children aged 8 years in some states was greater than 20 per 1,000 children (New Jersey). The national estimated ASD prevalence was nearly 15 per 1,000 children. Interestingly, according to the ADDM data, states that relied primarily on data from health care sources had lower prevalence rates than states that also used education records to compile their estimates.

ADDM data also shows that ASD prevalence rates correlate to socioeconomic status (SES), with higher SES associated with higher ASD prevalence. Among low SES cohorts, ASD prevalence was highest among white non-Hispanics; among mid SES cohorts, non-Hispanic African Americans had the highest prevalence of ASD. Asians had the highest prevalence among high SES cohorts.

Several explanations for the dramatic increase in ASD prevalence have been put forth, including changes in diagnostic approach and criteria and increased awareness of ASD among parents and health care professionals. However, Newschaffer suggested that “it is impossible to quantify the total effect of changing diagnostic tendency through the ‘retroscope.’”

He said the current best estimate of ASD prevalence among 8 year-olds in the US is 1 in 68, and that ASD is far more common in boys (the US prevalence ratio across gender is 4-5 to 1 male-to-female). He also said the US prevalence ratio across racial and ethnic groups is trending toward 1:1. Further, data suggest estimated US prevalence is still increasing, and that a major contributor to this is increased diagnosis among children with average or above intellectual capacity.

However, available data do not fully explain why prevalence is increasing. Newschaffer cautioned that ADDM data “are not invulnerable to changes in diagnostic practice, community awareness etc.”

There are many ongoing studies examining the genetic and environmental factors that contribute to ASD risk, as well as studies examining the interaction between and among these factors. Although the evidence supporting the possible link between them and ASD risk is not strong, several potential “exposure risks” have been identified in the literature, including older parental age, preterm or low birth weight birth, maternal medication use, pregnancy complications (including bleeding and gestational diabetes), and perinatal/neonatal health.

A recent paper by Lyall et al. in the International Journal of Epidemiology reviewed “the evidence on modifiable preconception and/or prenatal factors that have been associated, in some studies, with autism spectrum disorder (ASD), including nutrition, substance use and exposure to environmental agents.” Factors considered by the authors included maternal use of prenatal vitamins and supplements, maternal smoking and alcohol use, and prenatal exposure to air pollution. They concluded that “More work is needed to examine fats, vitamins and other maternal nutrients, as well as endocrine-disrupting chemicals and pesticides, in association with ASD, given sound biological plausibility and evidence regarding other neurodevelopmental deficits.”

Newschaffer identified several epidemiologic challenges going forward, including designing instruments for assessing exposure to risk-increasing factors, including biomarkers of exposure and exposure response. More work is needed to better understand gene-environment interaction and the effect it has on ASD risk.