Social influence and the autism epidemic

28 May

Prof. Peter Bearman at the University of Columbia has spent considerable effort studying the autism data from the California Department of Developmental Services.

I wrote about Prof. Bearman’s paper Social influence and the autism epidemic before. It just popped up in pubmed.

When I checked the website for the center where Prof. Bearman’s team works, I found the press release from the paper:

Study: “Social influence” playing role in increased autism diagnoses

Social influence plays a substantial role in recent increases in autism diagnoses, according to a study in the March, 2010 issue of the American Journal of Sociology.

The study, by researchers from the Institute for Social and Economic Research at Columbia University, found that children living near a child who has been previously diagnosed with the disorder are far more likely to be diagnosed themselves in the following year. The proximity effect is not due to environmental factors or contagious agents, the study found. Rather, it is due mainly to parents learning about autism from other parents who have a child diagnosed with autism.

“We show that the likelihood of getting an autism diagnosis is clearly associated with person-to-person transmission of information,” said Peter Bearman, a sociologist who authored the study along with Ka-Yuet Liu and Marissa King. “Parents learn about autism and its symptoms; learn about doctors who are able to diagnose it; and learn how to navigate the process of obtaining a diagnosis and services from parents who have already been through the process with their own child.”

The researchers stress that the results do not mean that autism is not real or that it is overdiagnosed. “Our study doesn’t address the underlying cause of autism,” Dr. Bearman said. “We are describing the mechanism by which the number of diagnoses is increasing. It could be that the real incidence of the disorder is only now being uncovered. I think that is a reasonable message from this paper.”

In California, where this study was conducted, the number of autism cases handled by the California Department of Developmental Services increased 636 percent between 1987 and 2003.

The Columbia University team looked at data on over 300,000 children born between 1997 and 2003 throughout California. The team found that children who live within 250 meters of a child with autism have a 42 percent higher chance of being diagnosed with the disorder in the following year compared with children who do not live near a child with autism. Children who lived between 250 meters and 500 meters from a child with autism were 22 percent more likely to be diagnosed. The chance of being diagnosed decreases significantly the farther children live from another child with autism.

The study used several tests to determine whether these results could indeed be explained by a social influence effect, or if environmental toxicants or a virus are to blame. For example, the researchers looked at children who live close to each other, but on opposite sides of school district boundaries. These children are likely exposed to the same environmental conditions, but their parents likely belong to different social networks. The research shows that the proximity effect only exists when parents reside in the same school district. Children who live equally close to a child with autism but in another school district were no more likely to be diagnosed with the disorder than children who do not have a neighbor with autism.

The results also showed the proximity effect to be strongest among children on the milder side of the autism spectrum. That is also consistent with a social influence explanation, Dr. Bearman says. “Parents of severely disabled kids are more likely to recognize the disorder without needing input from social contacts, so we would expect to see a weaker social influence effect there, and that’s exactly what we found.”

The Strength of Social Influence

The data set used in the study allowed the researchers to judge just how strong the influence effect is compared to other factors that may be driving the epidemic. For example, previous studies have found a link between autism and parents’ ages. Parents today are having children later in life, and that could be causing autism cases to increase. Other studies have found that parents’ education plays a role as well. Better educated parents may be more likely to obtain a diagnosis for their children.

The Columbia team found that each of these factors plays a role in the epidemic, but the social influence phenomenon was the strongest. The researchers estimate that the proximity effect explains about 16 percent of the recent increases in autism diagnoses. Put another way, if no child lived within 500 meters of a child with autism, there would be a 16 percent reduction in autism diagnoses. That effect was stronger than the other factors tested. For example, the mother’s age explained about 11 percent of the increase. The mother’s education accounted for 9 percent.

The study was funded by the NIH Pioneer award for innovative health research.

What caught my eye was that last paragraph, where they note that 16% of the increase can be accounted for from this social influence phenomenon, but also that 11% was due to increased maternal age and maternal education accounts for another 9%.

This is in addition to diagnostic substitution, which accounts for about 25% of the increase in the CDDS client caseload.

One of the footnotes in the Social Influence paper also caught my eye:

The strong positive association of autism (during the period of increasing prevalence, 1992–2000) with socioeconomic status (King and Bearman 2009b), coupled with the increasingly negative socioeconomic status gradient for MR (arising from differential abortion rates for fetuses identified as chromosomally damaged through amniocentesis), has led to a relative stigma reversal and a consequent decline in the rate of MR diagnoses.

King and BEarman, 2009b is a “working paper” at Columbia University (“The Increased Prevalence of Autism.” Working paper. Columbia University, Paul F. Lazarsfeld Center for the Social Sciences.)

One feature that is very interesting in the paper is the how the level of “functioning” of the child relates to the social influence effect. The basic result of the paper is that when a family lives near another family with an autistic child, they are more likely to obtain CDDS services under the autism label. That effect is stronger for “high functioning” autism. To me, at least, this seemed counter intuitive. Here is a graph from the paper showing that the social influence effect means that a child with “high-functioning” autism is about 40% more likely to be registered with the CDDS if he/she lives close to a family with an autistic child.

The trend with proximity is very clear, as shown in the figure below. The probability of a child be registered with the CDDS with an autism label drops off very smoothly and clearly with distance to a family whose child was recently diagnosed. This is not seen for mental retardation.

As the authors point out:

Compared with children who are 501 meters–1 kilometer away from their nearest neighbor with autism, those in close proximity (1–250 meters) to a child with autism have a 42% higher chance of being diagnosed with autism in the subsequent year. Proximity of 201–500 meters increases the chance by 22%. In contrast, being farther away from a child with autism reduces the chance of a diagnosis.

I know this is a repeat in many respects. But I also have found this work to be very interesting and thought it worthwhile to bring up these points I didn’t discuss before.

7 Responses to “Social influence and the autism epidemic”

  1. Joseph May 29, 2010 at 01:20 #

    One thing that should be emphasized is that the authors only looked at one mode of information diffusion in this study: neighbors. They didn’t look at relatives. They didn’t look at coworkers. They didn’t look at the internet. They didn’t look at active sources of information, like when there are awareness campaigns in the community. They didn’t look at pediatrician awareness.

    Presumably, these all add up to a much stronger effect than the authors encountered.

  2. Katie May 29, 2010 at 15:51 #

    Why did they have to keep using the word ‘epidemic’? Especially since this study refutes much of what the ‘epidemic’ hysteria claims? Ugh.

  3. Alex plank May 29, 2010 at 15:54 #

    Were you also at imfar?


  1. The Hard-Wired For Influence E-Book. | - May 29, 2010

    […] Autism Blog – Social influence and the autism epidemic « Left … […]

  2. Tweets that mention Autism Blog - Social influence and the autism epidemic « Left Brain/Right Brain -- - May 29, 2010

    […] This post was mentioned on Twitter by Brandon Blietz, jamie davis. jamie davis said: Autism Blog – Social influence and the autism epidemic « Left …: “We show that the likelihood of getting an auti… […]

  3. The Myth of an Autism Epidemic – Keep Kids Healthy - July 31, 2017

    […] Social influence and the autism epidemic […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: