By now most people interested in autism have read the CDC report, or at least read the news, regarding the new estimated prevalence rates of autism in the United States. Today I finally was able to read the full original report and have some brief general thoughts.

The report is based on the findings by the CDC Autism and Developmental Disabilities Monitoring Network. This network consists of a series of sites across the united states that calculate the rates of autism diagnoses for specific communities. The network first provided autism estimates based on data obtained in 2000 and then 2002. Last week’s report is based on data obtained in 2006. I have previously reviewed how the CDC prevalence rates for autism are obtained, so I will focus this post on highlighting some across state variability and differences between the 2002 and 2006 results.

In sum, the 2006 data came from 11 states (Alabama, Arizona, Colorado, Florida, Georgia, Maryland, Missouri, North Carolina, Pennsylvania, South Carolina, and Wisconsin). Teams at these sites reviewed the records of 8-year-old children living in specific communities. The teams reviewed medical/health and educational records for evidence of a probable autism diagnosis (education records were only monitored in 6 of the 11 states). When probable cases were identified, the records were then reviewed by clinicians to provide a final diagnosis based on DSM-IV criteria. The total number of ASD cases was then compared to the population of 8-year-olds for each target community.

The average ASD estimate across all sites was 9 per 1,000 children (1 in 111 children), but there was significant variability between the states:

Alabama: 1 in 166
Arizona: 1 in 82
Colorado: 1 in 133
Florida: 1 in 238
Georgia: 1 in 98
Maryland: 1 in 108
Missouri: 1 in 82
North Carolina: 1 in 96
Pennsylvania: 1 in 119
South Carolina: 1 in 116
Wisconsin: 1 in 131

Those sites that included a review of educational records had higher prevalence than those that relied only on health records:
Sites that included health and educational records: 1 in 98 children
Sites that included only health records: 1 in 133 children

Prevalence for boys alone:
Alabama: 1 in 110
Arizona: 1 in 53
Colorado: 1 in 87
Florida: 1 in 137
Georgia: 1 in 60
Maryland: 1 in 64
Missouri: 1 in 52
North Carolina: 1 in 59
Pennsylvania: 1 in 89
South Carolina: 1 in 70
Wisconsin: 1 in 79

The picture is much better for girls.

Prevalence for girls alone:
Alabama: 1 in 345
Arizona: 1 in 204
Colorado: 1 in 294
Florida: 1 in 1000
Georgia: 1 in 294
Maryland: 1 in 417
Missouri: 1 in 213
North Carolina: 1 in 294
Pennsylvania: 1 in 303
South Carolina: 1 in 385
Wisconsin: 1 in 435

Increases in ASD diagnoses from 2002 to 2006 among 8-year-old children:

Alabama: 82%
Arizona: 95%
Colorado: 27% (not statistically significant)
Florida: No 2002 data
Georgia: 34%
Maryland: 37%
Missouri: 66%
North Carolina: 60%
Pennsylvania: 58%
South Carolina: 43%
Wisconsin: 46%
AVERAGE: 57% increase.

A few last things to keep in mind:

- The report indicated that increases in prevalence was NOT due to increases in children diagnosed with PPD-NOS. That is, they found increases in the use of pure autism diagnoses too.
- The same diagnostic criteria was used in 2002 and 2006. The changes are NOT due to differences in diagnostic criteria.
- The report was not based on a nationally representative sample.
- Within State variability is so great that it is very likely that fluctuations in prevalence between states are due to methodological differences.
- HOWEVER, significant increases were also observed between sites that did not have changes in methodological procedures between 2002 and 2006.
- Thus, the increases from 2002 to 2006 are unlikely to be due to methodological differences
- There were no major changes from 2000 to 2002, which highlights the significance of the changes in diagnoses from 2002 to 2006.
- The study does not answer the question of “why”. We simply do not know why the prevalence rate of autism increased from 2002 to 2006.
- The new CDC estimates as more in line with a recent nation-wide autism prevalence study published in pediatrics.

The study concludes:

More children than ever before are receiving services for ASDs and are having symptoms of ASDs documented in developmental evaluation records. Even without fully understanding the complex causes of this increase in identified ASD prevalence, the impact on affected children, families, and communities is substantial. Prevalence estimates can be used to plan policy, educational, and intervention services needs for persons with ASDs. In addition to continued evaluation of ASD prevalence changes, major collaborative efforts are needed to improve research into what factors put certain people at risk and how to intervene to help reduce the debilitating symptoms of ASDs. Concerted efforts are essential to address the many needs of affected persons and to provide coordinated support services which improve daily functioning and long-term life outcomes

Kogan, M., Blumberg, S., Schieve, L., Boyle, C., Perrin, J., Ghandour, R., Singh, G., Strickland, B., Trevathan, E., & van Dyck, P. (2009). Prevalence of Parent-Reported Diagnosis of Autism Spectrum Disorder Among Children in the US, 2007 PEDIATRICS, 124 (5), 1395-1403 DOI: 10.1542/peds.2009-1522

Post to Twitter

Tagged with:

19 Responses to A closer look at the new CDC autism prevalence rates

  1. Socrates says:


    I’ve taken a rather more National Enquirer style approach to exploring this issue here.

    But I’ll wager you in all seriousness, $1000 dollars, that in a decade this statement will hold true for the US as well as the UK:

    In the UK, the rate of Autism is the same today as it has been for at least 3 generations.

  2. Professor Lopez-Duran thank you for your usual disciplined, objective look at the new CDC data.

  3. JulieL says:

    Nestor – I’m curious, couldn’t this still have more to do with physician diagnosis rates increasing vs actual autism rates increasing? There is considerable more general knowledge about autism, which would cause more parents to seek diagnosis by their physician. Also – hasn’t there been quite a lot of literature geared to pediatricians to help them diagnosis autism more effectively and earlier. If I remember correctly there was such a paper written in collaboration with Brazleton in the last year or so. (If my memory is correct) So to me, this still may not have anything to do with autism actually increasing in rates vs awareness. Also regarding Missouri – the worst rates. Are these rates from St. Louis City? I’m from St. Louis and the city is separate from St. Louis county, and the rates on all kinds of issues (violence, job losses, etc) often come from just the city, which is not the majority of the area. I’m wondering if in this study that happened as well; that the rates for Missouri are coming from the city, which is primarly poor African Americans. I just wonder how true these numbers are in representing the included states. Thanks.


    • Hi Julie, some of what you present is still a possibility. The data do not address those issues (e.g., greater health services seeking behaviors, greater attention by pediatricians, etc etc). What the data clearly addressed is that the changes were not due to changes in diagnostic criteria or increases in non-pure autism diagnoses (more PDD-NOS, etc).

      Regarding Missouri, while you are correct that the data came from St. Louis, the demographic hypothesis is not supported by the data. The Missouri sample was close to 70% white, and other States with significantly higher % of ethnic minorities in urban settings had significantly lower autism rates (e.g., Colorado – all from metro Denver with only 55% white; Florida – all from Miami with only 23% white; Georgia – all from metro Atlanta with only 38% white).

      Thanks. Nestor.

  4. K and J Investigations and Case Management » Cultural Symptoms: Diagnostic Spectrums and the Marketing of Autism says:

    [...] in children. Child Psychology Research Blog is a site we respect and has a post titled “A closer look at the new CDC prevalence rates” that gives even more detail and insight into these autism prevalence rates. Here is an [...]

  5. Rebecca Cochran Ph D says:

    Julie L’s post reflects the inaccurate view that Autism is a “poor” child’s dx, nothing could be further from the truth. The lion’s share of Autistic children come from middle to upper middle class socioeconomic families, with the majority of parents being college educated.

    It’s simple, Autism is increasing, the vaccinations are increasing. They have now gone after pregnant women and small children with flu vaccines. The Autism rate will continue to increase until they are willing to make a connection between the vaccines and Autism.

    • Dear Rebecca, thank you for your comment. However, please read the follow up post I wrote today. Vaccinations rates did not increase during the target period (1994 to 2000) and remained stable at around 90% coverage. There may be other arguments that can be made (e.g., that the types of vaccines used during that time changed – although I don’t know if there is data that would suggest that) but an “increase in vaccinations” can not explain the 50% increase in prevalence because there was no noticeable increase in vaccination rates during those years. In fact, in many of the states that presented an over 40% increase in prevalence, the vaccination rate during that period actually dropped.

  6. MG in NYC says:


    I would love to see your data on vaccination rates because I frankly don’t believe your assertion that they’ve been constant. If you’re making a bland statement about the rate of all children who have received SOME vaccines vs those that have recieved none, then maybe. However, it is pretty clear that children are now innoculated against more diseases, innoculated more often when adding booster shots to the count, and innoculated at earlier ages than was the case 20 years ago. It seems every few years, a new vaccine has been introduced into the schedule, e.g. Hib, rotovirus, Hep A, chicken pox, just to name a few in the last 10 years or so.

    There has not been a single meta analysis of the vaccine hypothesis using Bayesian techniques, whereas the predominant mainstream hypothesized cause, genetic factors, has been found inconclusive at best over the many meta analyses one can readily find in Google Scholar.

    This epidemic and crisis totally eats away at the credibility of medicine as a scientific field and as an institution. Here we have a devastating lifelong disease that seems to be growing at remarkable rates, and the best response the medical community can summon is a pithy “I bet it’s been like this for the last 3 generations”, a statement that is blatantly absurd to anyone in their 40s who sees autistic children today knows it was totally rare a generation ago.

    • MG, I provided the link to the CDC vaccination coverage yearly statistics. There you have the yearly coverage rates for all vaccines and all combinations of vaccines from 1990 to today. Feel free to review the data yourself.

      You stated:

      ‘However, it is pretty clear that children are now innoculated against more diseases, innoculated more often when adding booster shots to the count, and innoculated at earlier ages than was the case 20 years ago.’

      But that is not pertinent to the issue at hand. The issue is whether there were changes in vaccination practices between 1994 and 2000 (not 20 years ago to today) that could explain the 50% increases in the incidence of autism for children born in 1994 and 1998. I simply pointed out that the CDC vaccination data do not suggest that an increase in vaccination rates was observed for those two cohorts. I was very specific that I was referring to increases in vaccination rates, and not about changes in other vaccination practices ( I said “Thus, given the striking increases in prevalence rates among the 1998 children, you would expect that compared to those born in 1994, children born in 1998 received higher vaccination dosages, received more harmful dosages, or simply were vaccinated at a higher rate. I have some data on vaccination rates:” and then proceeded to provide the rates) Clearly, you could argue that the CDC data are wrong, and that is your pejorative.

      • One a more general note, one thing that is sad about the political debate on autism is how much the medical community is unfairly attributed with ideas and statements made by non-research professionals (activists, media, neurodiversity advocates, etc., etc.). For example, I am yet to hear a serious autism researcher say “I bet it has been like this for the last 3 generations” and dismiss the idea that autism may be truly increasing. That is a popular argument among activists that do not believe that autism has increased, but not among serious researchers; at least not among any of the key autism researchers with whom I have discussed this issue.

  7. MG in NYC says:


    A couple of points and comments. You write:

    I said “Thus, given the striking increases in prevalence rates among the 1998 children, you would expect that compared to those born in 1994, children born in 1998 received higher vaccination dosages, received more harmful dosages, or simply were vaccinated at a higher rate. I have some data on vaccination rates:”

    Actually, you can look at your post again and you DON’T write this passage, so I don’t know where your getting this. Changes in vaccination rates would not be a measure of toxicity as informative as the other measures I mention in my comment, so if you are sticking to just CDC data, it is statistically censored as a metric and not as informative.

    Regarding the political climate, you can the blame the mainstream medical community for this. You write:

    one thing that is sad about the political debate on autism is how much the medical community is unfairly attributed with ideas and statements made by non-research professionals (activists, media, neurodiversity advocates, etc., etc.). For example, I am yet to hear a serious autism researcher say “I bet it has been like this for the last 3 generations” and dismiss the idea that autism may be truly increasing.

    I guess it depends on who you consider a “serious” autism researcher. Two months ago, while having a discussion with the head of pediatric neurology at a top NYC hospital, that’s precisely the gist of what he said to me. When I brought up the issue of what could be behind the autism epidemic, he chided me for even calling it that, saying that diagnostic criteria had been much wider now and that autism was always with us in the form of Asperger’s. He then proceeded to tell me a fabricated statistic (that he probably believed was true) that Asperger’s patients represent about 75% of all spectrum cases, so as a result, it’s easy to see how that would have been missed several generations ago. Forget the fact that the math doesn’t work out even if that statistic were true, the statistic is outright false. And this guy was the head of pediatric neurology, where 1/3 of his patient are spectrum kids who have epilepsy.

    So, yes, the medical community is clued out and is responsible for promoting these ideas and statements. After all, that is where the activists and media get their talking points from.

    Finally, you should look at the #1 comment here, where Mr. Socrates plainly repeats the comment that you mocked. Is he not a “serious autism” researcher? If so, what is he doing trolling a research blog?

    • Sorry MG I was not very specific – I think we are talking about different posts. This specific post (“A closer look at the CDC…”) is not about vaccines and the discussion on vaccines was raised only in the comments. In the comments to this post I refer to a post I wrote today (“More thoughts on the CDC…”) where I did discuss the vaccine issue and presented the CDC data. That is the post I was quoting. In that post, I simply pointed out that if vaccines played a role in the increase in prevalence, then it is sensible to expect to see evidence of a change in vaccination patters (types, frequency, intensity, rates, etc) between 1994 and 2000 that would parallel the 50% increases reported by the CDC on Friday.

      Most of my readers are not serious autism researchers; most of my readers are parents. I do not know who Mr. Socrates is, but I don’t assume that a visitor is a serious autism researcher just because he/she leaves a comment on a research blog.

      We do agree on one issue…. the increase in prevalence in the 1994 to 1998 cohorts can not be due to changes in criteria, since both cohorts were evaluated using identical criteria. There may be other explanations for the 1994 to 1998 cohort change, but ‘widening of the ASD criteria’ does not seem to be one of them.

  8. Sarah says:

    You have to go back farther that 1994. You have to look at vaccine schedules from 1970′s and 1980′s compare them to 1994-present and then you’ll see a parallel in the 50% increase. The increase in autism has in fact been dramatic since the 1980′s.

    So, how does that relate to vaccines?

    There’s no question children are being vaccinated more aggressively and at a younger age than in the past. That has to have an impact.

    Compare the schedules: Past and Recent

    When you compare past vaccine schedule to more recent schedules and look at number and age at which children today get vaccinated it’s obvious that todays schedule is not only more aggressive targets young children but at least triple the number of shots. (See links to 1983 and 2009 schedules below)

    Also, I’ve yet to find a study done on the culmulative effect of these shots (in total) on a sensitive populations (such as immune compromised or mitochondrial disorder). It’s disconcerting that such as a study has not been done to ensure the safety of the actual schedule that so many pediatricians rely on.

    So what changed?

    In 1993, the Comprehensive Childhood Immunzation act was passed that called for a national vaccination program for children at the earliest possible age. This is about the time that the number of vaccines started to increase as did the autism rates. Here’s the Act.

    The Comprehensive Childhood Immunization Act of 1993

    Some related links:

    1983 Childhood Immunization Schedule:

    2009 Childhood Immunization Schedule

    Graph showing Autism Prevalance rates 1992-2008

    Autism Increase Not Due to Better Diagnosis
    Irva I. Hertz-Picciotto, Ph.D.
    University of California Davis

  9. Former CDC director, Dr. Julie Gerberding, to work for Merck « Crux of the Matter says:

    [...] apparently that is not the case. For example, read about the findings of this study, as well as this one. The crux of the matter is that we will never get to the truth as long as researchers and [...]

  10. Madeline says:

    “#14 Sarah Says:
    December 25th, 2009 at 9:01 am

    You have to go back farther that 1994. You have to look at vaccine schedules from 1970’s and 1980’s compare them to 1994-present and then you’ll see a parallel in the 50% increase”.

    I agree; you need to consider how *mothers* of these autistic children received more vaccines.

  11. megan says:

    I’m a mother of a 7 year old autistic boy and I’m also an epidemiologist. I don’t accept that the vaccine science is “settled” but I also think hat some of the arguments against vaccines are so misinformed that they make us (those questioning vaccine safety) look like a bunch of loonies.

    #14. What Dr. Lopez-Duran said is accurate but you don’t seem to understand it. If the issue is that vaccines given to KIDS increase the odds of those KIDS developing autism, then THERE MUST BE A DIFFERENCE between the vaccines given to kids born in 1994 and those born in 1998. IT IS NOT about changes in the 1980s or 70s. Never before was the answer so easily available. What the Dr. presented in the other post (that there were not changes in MMR vaccines) is not enough. Scientists must look at changes in the schedule and vaccine contents between 1994 and 1998. There is the answer.

    #16, Now this is a new conspiracy theory…. now it is not the vaccines given to kids that cause autism but the vaccines given to those mothers when they were babies???? Well, aside I guess in theory that is plausible. But you have a major problem. The 1994 and 1998 cohort of kids are strikingly different in their autism rates. Epidemiologically, these kids come from two very different population. However, at the parent level, 4 years is not enough to differentiate parental cohorts. Let me explain: the average and range of ages of parents who had babies in 1994 and those who had babies in 1998 are nearly identical. They are the SAME cohort of parents. This is a common population-based phenomena. Simply go to a McDonald’s playroom and look at the parents who have a baby and compare them to those who have a 4 year old. They are from overlapping cohorts if not the same. So saying that the increase in autism from the 1994 to 1998 kids is due to the parents vaccination does NOT HELP OUR CAUSE. it makes us sound like a bunch of tea party conspiracy theorists.

  12. henrysdad says:

    you tube has good reality videos on autism under autism and self injury and when you see these videos it is clear that autism is very serious and our legislators need to do more for this population

  13. BellaBlue352 says:

    Obviously, there is a genetic component involved, as most children with autism are male. Most likely, there is also an environmental trigger involved as well, as the rates of autism have gone up dramatically. If you look at the ingredients in vaccines, a good amount of them have extremely high levels of aluminum in them. It is widely known that children with autism very frequently have problems with their ability to process foods and other “toxins” in their bodies. We know that people that have impaired kidney function have problems with excreting aluminum out of their bodies and it accumulates in the brain and can cause brain damage. So if child is born with autism (but you do not know this–as they are only a day old–and since they have autism, they might have a tough time excreting the aluminum out of their bodies–and then you inject them with Hep B the day they are born (which has aluminum in it) what you might end up with is a child that now has brain damage, making them severely autistic. Why are researchers not looking at this as a possible cause. Most of the time, vaccines cause no damage in children that are born without the autism “gene”–but what about the kids that do! This makes complete sense to me as the possible link between the two.

Leave a Reply

Your email address will not be published. Required fields are marked *


+ 1 = six

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Looking for something?

Use the form below to search the site:

Still not finding what you're looking for? Drop a comment on a post or contact us so we can take care of it!

Set your Twitter account name in your settings to use the TwitterBar Section.