The Reasons Behind the Rise in Autism

Laurie Scudder, DNP, PNP; Susan L. Hyman, MD


May 06, 2014

Editor's Note: A just-released estimate of autism spectrum disorder (ASD) prevalence published by the Centers for Disease Control and Prevention (CDC) concluded that rates of ASDs are roughly 30% higher than previous estimates.[1] These new data put the figure at 1 in 68 children aged 8 years (or 14.7 per 1000) -- compared with a 2012 estimate that 1 in 88 children (11.3 per 1000) had an ASD. This rise was despite the fact that the criteria used to diagnose ASDs and the methods used to collect data have not changed, according to the report.

The new estimates are based on 2010 data from 11 sites participating in the Autism and Developmental Disabilities Monitoring Network, an active surveillance system that provides estimates of the prevalence of ASD and other characteristics among children aged 8 years. The age of 8 years was chosen to ensure that the majority of children with the condition would be included; the overwhelming majority of children with ASD will have been diagnosed by that age.

Medscape spoke with Susan L. Hyman MD, a professor in the Department of Pediatrics and Chief of Neurodevelopmental and Behavioral Pediatrics at the University of Rochester, and Chair of the American Academy of Pediatrics (AAP) Autism Subcommittee, about the survey and the implications for primary care pediatrics.

Medscape: Given that these were 2010 data, what would you expect today?

Dr. Hyman: The AAP first recommended developmental screening for autism in 2006.[2] As a result, I believe that children born in subsequent birth cohorts are likely to be identified earlier. The study just published was reporting data about children who were born in 2002, so these children were already 4 years of age when the initial recommendation for screening came out and were 8 years of age in 2010.

I think what we are likely to see with subsequent birth cohorts is an earlier age at diagnosis. With increased education for both pediatricians and educators, we may actually see the numbers go up, because these data are collected from surveillance of health and education records and depend on the written observations of the professionals in those settings.

Medscape: How do you expect the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) to affect this process of data collection?

Dr. Hyman: That's an excellent question. First let me describe how the CDC data are collected in order to answer.

CDC data are collected by surveillance. The researchers examine school and health records using a validated method of looking for key words to help identify children with autism. Sometimes the diagnosis is made by outside clinicians in the communities. By virtue of looking at high-risk populations, including children who have other developmental issues and who are receiving special education services in the schools, they come up with kids who aren't already diagnosed as well.

They've been using the DSM-IV, the prior diagnostic algorithm, to identify children with autism. What's happened over the decade of monitoring that has been completed is that educators and clinicians have become more aware of the symptoms of ASD as described in DSM-IV. In future monitoring, the CDC will compare the algorithm from DSM-5, the new diagnostic criteria, with the DSM-IV. In this way of population monitoring, they will be able to examine how the 2 algorithms compare, which will be quite informative.

How do I personally expect DSM-5 to affect this information? I think that we know a lot more about autism than we did in 1994, when DSM-IV was published. I think that DSM-5 needs a chance to be used in clinical practice so that we get a sense of its accuracy. I believe we understand now that sensory symptoms should be captured when making a diagnosis of autism, and that repetitive behaviors shouldn't be minimized as they had been previously. So the DSM-5 may actually capture more accurately what we know clinically to be autism.

Medscape: In your opinion, were these children missed by the medical community before, or were they diagnosed with other conditions?

Dr. Hyman: I think it's a combination of both. The diagnostic criteria in DSM-IV allowed for the capture of children on the diagnostic spectrum who had a high intelligence quotient (IQ) and were high-functioning -- what we used to call "Asperger syndrome" -- as well as children who were low-functioning and did not have language, those who in the past we said had profound intellectual disabilities. It became a way to capture social reciprocity -- social give-and-take independent of IQ -- so it widened the IQ range. People who previously would have been considered quirky or odd may have been better described as being on the autism spectrum, and that will persist through DSM-5.

The question that you're asking is: Were these people diagnosed with other things? The answer is yes -- people were diagnosed with profound intellectual disability, what we used to call "mental retardation." A large number of higher-functioning individuals were not diagnosed or may have been diagnosed with attention-deficit/hyperactivity disorder (ADHD) or mental health disorders, and the autism spectrum wasn't recognized.

Medscape: What every pediatric clinician really wants to know is, is this a true rise vs better identification vs a diagnostic shift? Or all of the above?

Dr. Hyman: It is a little bit of everything. I believe that we're better at identifying people with ASDs, and I also can't rule out that it's not occurring with increasing frequency. There has been a rise in other disorders -- for example, ADHD -- that isn't getting as much press as the rise in autism. We really need to examine the rise in all of these disorders in our culture, and in our country. It's extraordinarily complex.

There are a number of families who come in to see us, we diagnose the child, and the parent then questions whether either they or a sibling also has ASD. It is very common for the condition to be undiagnosed in adulthood. We don't have the capacity to look at the whole population to see whether or not the prevalence is greater in adults with these new criteria.

Medscape: What does current evidence say about reasons for this rise? Is there an environmental factor that is contributing?

Dr. Hyman: This may be true. It may also be true -- and unrelated -- that we are diagnosing better and more broadly. But there also may be other things that interact with genetic predisposition to allow the manifestations that we see as autism.

Lots of correlational studies have been published in recent years that have examined factors associated with autism. We don't yet understand the biology of causality, though the association of older maternal[3] and paternal[4] age with ASDs is an important, open question. Is maternal obesity, which is rising even faster than autism and ADHD, associated with autism?[5] Are there aspects of air pollution that are associated with ASD? There was a recent report about the use of pitocin augmentation in labor.[6] None of these are studies that would allow us to say anything about causation.

What these epidemiologic studies do is identify observational associations. It may be that whatever it is that causes autism results in these associations. Or it may be something that causes both. Or it may be just that obesity is rampant in the population.

A very recent study, published in the New England Journal of Medicine, documented abnormal areas of development in the brains of youngsters with autism and points to fairly early gestational differences in how the brain is being formed.[7] This supports our current understanding that most, if not all, of the influences related to the etiology of autism are prenatal. Whether it is genetic predisposition, things in the environment, or the internal environment of the mother -- we don't know yet. That's one of the very compelling reasons to support translational research.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.