Skip directly to search Skip directly to A to Z list Skip directly to page options Skip directly to site content

Key Findings: Have Secular Changes in Perinatal Risk Factors Contributed to the Recent Autism Prevalence Increase? Development and Application of a Mathematical Assessment Model

Photo: Child playing with blocksThe journal Annals of Epidemiology has published a new Centers for Disease Control and Prevention (CDC) article: "Have Secular Changes in Perinatal Risk Factors Contributed to the Recent Autism Prevalence Increase? Development and Application of a Mathematical Assessment Model." CDC’s Autism and Developmental Disabilities Monitoring (ADDM) Network estimates that an average of 1 in 110 children in multiple communities in the United States who were 8 years old in 2006 had autism spectrum disorder (ASD). That estimate represents a 57% increase in the prevalence of ASD since 2002 (1). The reasons for the increase are not completely understood. For this study, researchers developed a mathematical model to estimate how changes in pregnancy and birth factors during the period from 1994 through 1998 might have contributed to the subsequent 57% increase in ASD prevalence. The model developed for this study will be useful for future research and can be used to assess other risk factor scenarios. You can read the abstract of the article here. Following is a summary of the findings from this article.

Main Findings From This Study

  • This mathematical model showed that, for any given risk factor to be responsible for the 57% increase in ASD prevalence, it would have to have:
    • Occurred among a large percentage of the population of babies born in 1994,
    • Increased substantially during the period from 1994 through 1998, and
    • Had a strong association with ASD.
  • For each of the six pregnancy factors assessed (preterm and very preterm delivery, low and very low birthweight, multiple birth, cesarean delivery, breech presentation, and in vitro fertilization or other assisted reproductive technologies (ARTs), the likely effects on the recent ASD increase observed for the period from 2002 through 2006 were minimal. These specific factors were chosen for study, because they all have increased in recent decades.
    • The data showed that each pregnancy factor likely accounted for less than 1% of the total 57% increase in ASD—that is, these risk factors would have caused the ASD prevalence to increase only from 6.0 to 6.1 children per 1,000.

These findings suggested that it was very unlikely that any single environmental factor could have been responsible for such a large increase in ASD prevalence during such a short time period.

The model developed for this study will be useful for future research and can be extended to assess various risk factor scenarios. Though this study didn’t rule out that the perinatal factors assessed might put an individual at a higher risk for ASD, it did confirm that these factors alone were not likely responsible for the 57% increase in overall ASD prevalence observed among the population from 2002 through 2006.

About Autism Spectrum Disorders and This Study

ASDs are a group of developmental disabilities that often are diagnosed during early childhood and that can cause significant social, communication, and behavioral challenges over a lifetime. They include autistic disorder, Asperger disorder, and pervasive developmental disorder not otherwise specified. CDC’s Autism and Developmental Disabilities Monitoring (ADDM) Network estimates that an average of 1 in 110 children in the United States had an autism spectrum disorder (ASD) in 2006. That estimate represents a 57% increase in the prevalence of ASD since 2002 1.

The reasons for the increase are not understood completely. We at CDC know that, in the past decades, part of the increase was because of the changes in diagnostic criteria and the addition of autism as a special education category in public schools. Also, it is likely that increases in the prevalence of ASD are explained partially by greater awareness of autism by doctors, teachers, and parents. But, with the limited data currently available for many risk factors, we also cannot rule out that part of the increase might be related to an increase in one or more of the risk factors for developing ASD.

Studies have found that several pregnancy-related factors could be risk factors for ASD 2. For this study, researchers developed a mathematical model to estimate how changes in select pregnancy factors (preterm and very preterm delivery, low and very low birthweight, multiple birth, cesarean delivery, breech presentation, and ARTs (fertility treatments such as in vitro fertilization)) might have contributed to the 57% increase in ASD prevalence. These specific factors were chosen for this study because they have all increased in recent decades.

The mathematical model developed to estimate a risk factor’s likely effect on prevalence is based on how common the risk factor is among the population, how much the risk factor increases during the time period of interest, and how large the association is between the risk factor and having an ASD.

For this study, researchers used data from the National ART Surveillance System (NASS) and the National Center on Health Statistics’ Vital Statistics databases to calculate how much these risk factors had increased among the population from 1994 through 1998. Researchers reviewed the scientific literature to obtain the effect sizes, that is the strength of the risk factors associations with ASD. Combining information on the prevalence of these risk factors among the population, increases in the prevalence of the risk factors, and the strength of the association with having an ASD, the model produced estimated percentages of each risk factor’s contribution to the 57% increase in overall ASD prevalence.

Further Understanding This Mathematical Model

Following is an example of how this model could be applied to two different hypothetical risk factors. In each case, we assumed the risk factor increased substantially during a short time period. Both Risk Factor 1 and Risk Factor 2 increased in prevalence by 50% from 1994 through 1998.

Risk Factor 1

Given:

  • Risk factor 1 occurred among 5% of the babies born in 1994,
  • Risk factor 1 occurrence increased to 7.5% of the babies born in 1998, and
  • Children born with this risk factor were 2.5 times more likely to be diagnosed with an ASD than children who did not have the risk factor.

Then

  • Risk factor 1 could bring about a 4% increase in ASD―that is, the increase in risk factor 1 could have caused the overall ASD prevalence to increase from 6.0 to 6.2 children per 1,000.
Risk Factor 2

Given:

  • Risk factor 2 occurred among 20% of the babies born in 1994,
  • Risk factor 2 occurrence increased to 30% of the babies born in 1998, and
  • Children born with this risk factor were 5 times more likely to be diagnosed with an ASD than children who did not have the risk factor

Then

  • Risk factor 2 could bring about a 22% increase in ASD―that is, the increase in risk factor 2 could have caused the overall ASD prevalence to increase from 6.0 to 7.3 children per 1,000.

Thus, even if a risk factor for an ASD were to increase substantially, it would not necessarily have a large influence on ASD prevalence.

Autism Spectrum Disorders: CDC Activities

At CDC, we are working together with our partners to address the growing needs of thousands of families affected by ASDs and to understand the rising prevalence of these disorders.

  • Monitoring ASD. The ADDM Network produces ASD prevalence reports that help us understand the number and characteristics of children with ASDs living in several U.S. communities. While many studies on ASD focus on small groups of individuals, the ADDM Network monitors these conditions among thousands of children from diverse communities across the country. This ongoing, population-based approach allows the ADDM Network to monitor changes in the prevalence of ASD and characteristics of children with these conditions, such as the average age of diagnosis and disparities in identification. These data can help communities to direct their outreach efforts to those who need it most.

    Learn more about the ADDM Network »
  • Researching Risk Factors or Causes of ASD. CDC also is searching actively for the causes of ASDs by exploring key areas such as the roles of genetic and environmental factors. Like the ADDM Network, CDC’s Study to Explore Early Development (SEED) is population based and is active in six states across the country, enrolling over 3,700 children and their families in the largest study of its kind in the United States. SEED is an important tool that we at CDC are using to understand what puts children at risk for ASDs and likely will give us many clues to help unravel the mysteries that still surround our understanding of ASD.

    Learn more about SEED »

  • Educating Parents and Health Care Providers. The earlier a child is identified with an ASD, the more he or she can benefit from early intervention services that research has shown can improve a child’s development. CDC’s Learn the Signs. Act Early. program is having an effect on the lives of families now by helping parents and providers recognize the early warning signs of ASD and other developmental delays and by promoting early screening, evaluation, and treatment, when needed.

    Learn more about the Learn the Signs. Act Early. program »
Reference for Key Findings Summary
  • Schieve LA, Rice C, Devine O, Maenner MJ, Lee L, Fitzgerald R, Slay Wingate M, Schendel D, Pettygrove S, Van Naarden Braun K, and Durkin M. Have Secular Changes in Perinatal Risk Factors Contributed to the Recent Autism Prevalence Increase? Development and Application of a Mathematical Assessment Model. Annals of Epidemiology. 2011 Dec; 21(12):930-45. Epub 2011 Oct 13.
Other References
  1. Centers for Disease Control and Prevention. Prevalence of autism spectrum disorders—Autism and Developmental Disabilities Monitoring Network, United States, 2006. MMWR Surveill Summ 2009; 58(No.SS-10).
  2. Kolevzon A, Gross R, Reichenberg A. Prenatal and perinatal risk factors for autism: a review and integration of findings. Archives of Pediatric Adolescent Medicine. 2007; 161: 326-333.

E-mail Your Friends

"Children with autism spectrum disorder are not being diagnosed as early as they could be. Learn the signs of autism and get help if you’re concerned."

Send an E-mail

Share on Facebook

A young girl playing with blocks.

"Children with autism spectrum disorder are not being diagnosed as early as they could be. Learn the signs of autism and get help if you’re concerned."

Share on Facebook

Share on Twitter

"Children with autism spectrum disorder are not being diagnosed as early as they could be. Learn more."

Share on Twitter

TOP