Press Briefing Transcripts

CDC′S SURVEILLANCE REPORT ON AUTISM AND AUTISM SPECTRUM DISORDERS IN 14 COMMUNITIES IN THE UNITED STATES

Thursday, February 8, 2007 12:00 p.m.

OPERATOR: Good afternoon and thank you all for holding. At this time, your lines have been placed on listen

only until we open up for questions and answers. At that time, if you would like to ask a question, please press star followed by one on your touch tone phone. Today′s conference is being recorded. If you have any objections, you may disconnect at this time. I would now like to turn the conference over to Mr. Glen Nowak. Sir, you may begin.

GLEN NOWAK: Thank you. And thank you all for joining us today for this tele-briefing on CDC′s Surveillance Report on Autism and Autism Spectrum Disorders in 14 communities in the United States. And we will be joined today by Marshalyn Yeargin-Allsopp, who is an MD and Chief of our Developmental Disabilities Branch at CDC, and she works in our National Center on Birth Defects and Developmental Disabilities.

Joining Marshalyn this afternoon is Catherine Rice, who is a PhD Behavioral Scientist who also works at the National Center on Birth Defects and Developmental Disabilities at the CDC. After this, well sometime later this afternoon, about two or three hours after this press conference, we will have a transcript of this teleconference available on the CDC web site at www.cdc.gov/media, in other words, CDC′s media relations home page. With that I will now turn the conference over to Dr. Marshalyn Yeargin-Allsopp.

MARSHALYN YEARGIN-ALLSOPP, MD, CHIEF DEVELOPMENTAL DISABILITIES BRANCH, CENTERs FOR DISEASE CONTROL AND PREVENTION. Thank you Glen. And I would like to thank all of you again for joining us today. We would like to brief you on the report that CDC will be releasing in this week′s MMWR Surveillance Summaries on the "Prevalence of the Autism Spectrum Disorders" or ASD.

CDC is fully committed to the health and well being of children. Our highest priority is to protect the health of children and insure that all children can live, play and learn to their fullest potential.

ASDs are an urgent public health issue and affect the lives of many families and communities. CDC has been working on getting an accurate picture of the number of children affected by this disorder so that communities can plan accordingly.

There are many reasons why accurately estimating autism prevalence and trends is challenging. Because autism is a behavioral condition, children are often diagnosed at different ages and many are not diagnosed until they enter school. For decades, autism was believed to occur in four to five per 10,000 children.

Studies done in more recent times have been summarized as an ASD occurring in two to six per 1,000 children. As a general statement, we have used the figure of up to one in 166 children having ASD based on recent studies done in multiple countries. Today we will announce new statistics, and in the future we will continue to monitor these rates and gather more data to verify these statistics.

We know that approximately 17 percent of children have a developmental disability ranging from a mild disability such as speech and language impairment to serious developmental disabilities such as intellectual disabilities, cerebral palsy and autism. The more severe developmental disabilities affect approximately two percent of the individuals. The cost to American families runs in the billions and the personal costs are immeasurable.

CDC has been monitoring the prevalence of developmental disabilities since the 1980s and autism since 1996. Three previous CDC ASD studies looked at autism prevalence rates. First, in Metropolitan Atlanta. The Metropolitan Atlanta′s Developmental Disability Surveillance Program or MADDSP as we call it, in 1996 found that 3.4 per 1,000 three to ten year olds.

A community study in 1998 in Brick Township, New Jersey found that 6.7 per 1,000 three to ten year olds had an ASD. And the May 4, 2006 MMWR reported estimates using parent reported data from two national surveys.

Estimates of diagnosed autism was 5.7 per 1,000 4 to 17 year old children from the National Health Interview Survey and 5.5 per 1,000 4 to 17 year old children from the National Survey of Children′s Health in 2003 to 2004.

The study′s reported today in the MMWR will provide a more detailed picture of autism prevalence estimates from multiple areas of the United States. The first ever multi-site report from the Autism and Developmental Disabilities Monitoring Network or ADDM that collected data for the states for the surveillance year of 2000. And the second report provides autism prevalence estimate for the 2002 surveillance year for these six states and eight additional states for a total of 14.

Since the ADDM sites do not represent a nationally representative sample, the prevalence estimate should not be generalized to every community in the United States. Although accurate for the areas of the United States we studied, they may be lower or higher in other areas

.

However, these prevalence estimates can help these communities project how many children may have autism for planning and identification purposes. Taken together, results from the previous studies and the ones we are releasing today affirm that autism is a major public health concern that affects many families.

I will now turn this over to Dr. Cathy Rice, Behavioral Scientist here at CDC and the lead for the CDC ADDM project.

CATHERINE RICE, PHD, BEHAVIORAL SCIENTIST, NATIONAL CENTER ON BIRTH DEFECTS AND DEVELOPMENTAL DISABILITIES, CDC: Thank you very much. Autism is a developmental disability affecting many children and is a concern for those of us working in public health.

Service providers and families express concern about the growing number of children diagnosed with ASDs. There are many challenging questions that make it difficult to determine exactly what is going on.

Is this a change in the way ASDs are identified or is there an increase in the people at risk for ASDs such that there is a real increase in the condition. The fact that autism and related disorders are diagnosed by behavioral observation of development and not a medical test makes characterizing the population of people with ASDs challenging, especially when the criteria for defining ASDs have changed over time.

The reality is that we know more children are identified with an ASD than in the past and we need to do our best to get a better understanding of how many children are affected and to be able to track changes within the United States.

To better understand how many are affected by ASDs, the CDC has formed the Autism and Developmental Disabilities Monitoring (ADDM) Network. The ADDM Network provides population based estimates of the number of children who have an ASD and other developmental disorders, in order to characterize and compare the prevalence of ASDs in different sites over time.

As Dr. Yeargin-Allsopp stated, today′s report is the first report of the ADDM network and provides data from the first and largest multi-site study on ASD prevalence in the United States to date.

This report also provides important baseline information on the prevalence of ASDs in multiple areas of the United States and will enable us to better understand the changing landscape of individuals affected by ASDs over time.

I would like to tell you a little bit about the methodology, you should also be - there will be a fact sheet available that has some more information specific to the ADDM Network.

The ADDM Network provides prevalence rates for autism spectrum disorders as a group to include autistic disorder, Asperger′s disorder and "Atypical Autism" or "Pervasive Developmental Disorder, not Otherwise Specified," PDD-NOS.

For simplicities sake, I may use the term ASDs and autism interchangeably. The ADDM sites work collaboratively to maintain consistent surveillance methods based on CDC′s "Metropolitan Atlanta Developmental Disabilities Surveillance Program," or MADDSP. This methodology includes screening evaluation records at multiple sources that educate, diagnose and provide services for children with developmental disabilities and abstraction of detailed behavioral information on children who may have the features of an autism spectrum disorder.

Case status is confirmed by experts in the diagnosis of autism who systematically review the information contained in the evaluation records based on the current diagnostic criteria for the autism spectrum disorder. These current criteria are based on the diagnostic and statistical manual of the American Psychiatric Association. The current manual is the fourth edition, text revision, also knows as the DSM-IV-TR.

In addition, we have extensive quality assurance and training standards to insure that sites are applying the same methods across all of the projects. The ADDM Network does not rely solely on a child′s previous autism spectrum diagnosis or special education eligibility to determine if a child meets case definition for an ASD.

Instead, we look for a pattern of behavior associated with ASDs that are present up to the age of eight years in existing evaluation records. It is important to note that the ADDM Network is trying to describe the population of children affected by ASDs in areas and by subgroups, for example by boys and girls. And meeting a surveillance case definition of ASD is not the same as diagnosing a child followed by an in-depth clinical examination.

The project has monitored ASDs in children of age eight, because previous studies have shown that by this age most children with an ASD have been identified for services. Determining the prevalence of autism and other complex behavioral conditions is a challenging endeavor, especially if you are trying to go beyond counting children who have already been identified for service or diagnostic systems.

This project took a significant amount of initial setup time and coordination between the CDC, state agencies and multiple service providers to insure that the maximum number of children were counted. Now that these programs are in place, ongoing monitoring is continuing. And ADDM Network sites are currently finishing the 2004 surveillance year and have begun data collection for 2006.

A third paper publishing this issue of the MMWR Surveillance Summary, was an evaluation of the ADDM Network′s 2002 surveillance project. We evaluated the methods, because we wanted to apply the most scientific rigor possible to our methods and to continue to improve upon the effectiveness and efficiency of the system.

We hope this report will be used to interpret the ADDM Network data and also serve as a model for other public health systems that are interested in determining the prevalence of "Autism Spectrum Disorders.′ Estimates of the prevalence of ASDs were four to five per 10,000 children affected for many decades.

Previously reported summaries of ASD prevalences were up to one in 166 and represented a summary of multiple reports using various study methods to identify autism across several countries. However, the data reported today by the ADDM Network provides more specific information collected in the United States using the same methods across multiple sites.

Some detail on the results. For the first report, it covers the year 200 project where six sites participated. Those sites were in Arizona, Georgia, Maryland, New Jersey, South Carolina and West Virginia.

The second report includes an additional eight sites for a total of 14 for the 2002 project. The additional sites included were in Alabama, Arkansas, Colorado, Missouri, North Carolina, Pennsylvania, Utah and Wisconsin.

For the 2000 year from six sites representing approximately 4.5 percent of U.S. eight year old children born in 1992, 1,252 children were identified with an autism spectrum disorder. For the 2002 year, from the 14 sites representing approximately ten percent of U.S. eight-year old children born in 1994, 2,685 children were identified with an autism spectrum disorder.

The results of the study indicate that on average, 6.7 per 1,000 children who are eight year old in 2000 and 6.6 per 1,000 who were eight year olds in 2002 had an autism spectrum disorder.

This means that in these community studies, approximately one in 150 children born in 1992 and 1994 had an autism spectrum disorder or had an autism spectrum disorder. It′s important to note that the average of one in 150 children with ASD in these projects cannot be generalized to every community in the United States as we did find variability in ASD prevalence across the sites.

However, this estimate can be used cautiously to get a general estimate of ASDs in children for service planning purposes. I would like to provide you with a little more detail on the total ASD prevalence results across the site.

As I mentioned for 2000, the first report, the average ASD prevalence across the six sites was 6.7 per 1,000 eight year old children. However, the total prevalence of ASDs ranged from 4.5 per 1,000 in West Virginia to 9.9 per 1,000 in New Jersey.

The other four sites had more similar rates from 5.5 to 6.5 per thousand. For 2002 the average prevalence across all 14 sites was very similar to that of 2002 at 6.6 per 1,000. There was also a range from 3.3 per 1,000 in Alabama to 10.6 per 1,000 in New Jersey.

However, 12 of the 14 sites with ASD prevalence were in a tighter range from 5.2 to 7.6 per 1,000. When comparing the six sites that had data from the two different projects, from the two different years, prevalence was stable from 2000 to 2002 in four sites, but increased slightly in one site, Georgia and more significantly in a second site, West Virginia.

We also found across both study years that there were differences among boys and girls. Both study years found that autism prevalence was higher for boys age eight years than for girls at the same age indicating three to seven boys for every girl affected.

This is consistent with ranges found in other past studies. The data also showed differences among children by race or ethnicity. In most sites, ASD prevalence rates were similar for white and black children.

However, five of the 14 sites found a higher prevalence among white children compared to black children. For Hispanic children, prevalence tended to be lower in six of the 13 sites with data on Hispanic children. Sites with a sufficiently large population of Hispanic children are working together on a follow-up report of ASD prevalence among Hispanic children.

These data also showed an overlap between ASDs and cognitive impairment, also know as mental retardation or intellectual disability, defined as an IQ less than or equal to 70, between 33 percent and 62 percent of children with an ASD had cognitive impairment. Girls were more likely to have an ASD and cognitive impairment than boys in most sites reported.

Finally, the data showed a delay in ASD diagnosis. In both reports, the majority of children identified with an ASD had documented concerns by a parent or professional before three years of age, such as concerns about the child′s language, social or play development.

But the median age of earliest ASD diagnosis was approximately 4.5 to 5.5 years. Over the two year period, from 2000 to 2002, the delay in documented ASD diagnosis did not decline for the six sites included in both study years. As I noted before, additional information on these results can be found in the fact sheets provided and in the copy of the "MMWR Surveillance Summary," which should be available on line shortly.

In conclusion, these data provide important information on the prevalence of ASDs in areas of the United States and will be used to examine trends over time. Using the average prevalence found across all sites in both years, approximately one in 150 children who were born in 1992 and 1994 have an ASD in the communities included. We did find a range from a low of one in 303 children with autism to a high of one in 94 children with autism.

These findings help build awareness of the significant strain placed on families, health care and education systems in providing appropriate identification, intervention and support for people with an ASD. While the stability of ASDs in four of the six sites is encouraging, the increase in two sites is a concern. We cannot make conclusions about trends in ASD prevalence at this time.

However, continued monitoring of ASD prevalence in these sites will help us answer that question starting with children born in the 1990s. We are continuing to work with the ADDM Network to track the prevalence of ASDs in eight year old children in 2004 and 2006, and will report additional results as soon as possible.

While these studies did not investigate the causes of ASDs, CDC’s Centers for Autism and Developmental Disabilities Research and Epidemiology or CADDRE Network is conducting a multi state collaborative study to help identify factors that may put children at risk for autism spectrum disorders, and other developmental disabilities.

We recognize that ASDs are conditions of urgent public health concerns and we hope this information the ADDM Network has provided on the numbers of children affected will be part of the larger public and private effort to understand the impact of ASDs, the causes of the disorders and the most effective interventions to provide in order to help each individual reach their full potential. Thank you very much.

MR. NOWAK: We will now take questions.

OPERATOR: Thank you. We will now begin the question and answer session. If you would like to ask a question, please press star one on your touch tone phone. You will be prompted to record your name for proper registration. To withdraw your request, you may press star two. One moment please, for the first question. Our first question comes from Peggy O′Crowley with the Star Ledger. Please go ahead.

PEGGY O′CROWLEY, REPORTER, STAR LEDGER: Good morning. I wanted to ask you, I cover New Jersey for my paper and as you know, New Jersey′s rates were significantly higher than many of the other states.

And I′m wondering to what that might be attributed? Why are we seeing that jump in New Jersey compared to the other sites?

MR. NOWAK: I′ll have Dr. Rice answer that question.

DR. RICE: All right. OK, thank you. The causes of autism spectrum disorders are not known, but they do appear to be multiple complex genetic and environmental interactions.

But since we don′t know the causes of autism, we don′t know why the rates are higher in New Jersey. We think it′s very important that we were able to document that such a high rate is possible; however we need to continue working towards figuring out what′s causing the range of autism to be able to answer that question.

MR. NOWAK: Next question.

OPERATOR: Thank you. Our next question comes from Neil Osterweil with MedPage Today, please go ahead.

NEIL OSTERWEIL, REPORTER, MEDPAGE TODAY: Hi, Neil Osterweil with MedPage Today. Just a question of it′s - would it be possible at some point to break down the data by disorders within the autism spectrum, Asperger′s, PDD-NOS, et cetera?

MR. NOWAK: I′ll have Dr. Rice answer that question again.

DR. RICE: All right. We′ve looked at a variety of ways to break down the estimates by sub-type and determine that - the methodology that we use, because we′re looking over existing evaluation records over a child′s life time up to age eight. It′s a challenge to really apply those sub-types.

So what we′ve done instead is to look at other features of autism, such as the co-occurrence of cognitive impairment, level of language development and try to sub-type in those types of ways so there′s a little information about that in the report, but we don′t currently break it down by Asperger′s, PDD-NOS and autistic disorder.

MR. NOWAK: Thanks. Next question.

OPERATOR: Thank you. Our next question comes from Will Dunham with Reuters.

WILL DUNHAM, REPORTER, REUTERS: Yes, hello. I know that you have not given a nationwide rate, but is there any way you could give a ballpark estimate of how many millions of children nationwide you think fall under this category?

MR. NOWAK: Thanks. I′ll have Dr. Yeargin-Allsopp answer that question.

DR. YEARGIN-ALLSOPP: Yes. You′re absolutely correct. We have not given a national statistic and we have given a lot of caveats around the 1 in 150 children. However, if there are 4 million children born in the United States every year, which is the current estimate, if we apply these statistics to the population of children from birth to 21, then we get a number, of 560,000 children in the United States with an ASD.

MR. NOWAK: Next question.

OPERATOR: Thank you. Our next question comes from John Reichard with Congressional Quarterly.

JOHN REICHARD, REPORTER, CONGRESSIONAL QUARTERLY: Sorry, can you hear me? Yes. I just wondered, you mentioned, you know, the issue of service planning and prevalence statistics that could be used in planning services.

Could you explain what some of those services would be? And also what the impact would be on kind of overall funding for these services if communities, you know, throughout the country assumed 1 in 150 children?

MR. NOWAK: Dr. Rice will answer that.

DR. RICE: OK. Well, some of them, the services that apply prevalence disabilities, particularly the public education system - many children with autism are receiving special education.

And every year, the departments of special education in each state report how many children are classified under different types of disabilities to the Federal Department of Education.

And those estimates are used in terms of planning for resources needed, the types of education that those students may need. What we found though, is not all children with autism and for many reasons are classified as having autism, and so you might not be able to plan appropriately for the amount of services that you may need related to autism.

So for instance, that is one example in terms of providing education services. Others, each, most states have disability boards or disability services that provide potentially respite services or early intervention services, and that may also be helpful to have an estimate of how many children you should be expecting to be serving.

MR. NOWAK: We will take another question.

OPERATOR: Thank you. Our next question comes from Rick Weiss with the Washington Post.

RICK WEISS, WASHINGTON POST: Hi. Thank you. A narrow question and a broad question. The narrow one, you mentioned I think a higher prevalence in New Jersey in 2002 than 2000. But I think you did not include New Jersey as one of the couple of states that has seen an increase. So, should I presume that that increase number is not statistically significant?

And the broader question, you say in the abstract of this report that the numbers you are getting here are higher than previously believed. But at the same time, it sounds like a lot of what you said is within the range of what has previously been shown. What is the best way for us to characterize these new numbers compared to what we thought before?

MR. NOWAK: Dr. Rice, answer that? We could start with the first question I guess.

DR. RICE: OK, first in terms of New Jersey, you are exactly right. That wasn′t a statistically significant difference from 2000 to 2002. So that is encouraging that there was some stability, although the rates were higher than we would have expected there.

In terms of the second question, it is challenging to, we all want to have one number to point to in terms of autism. It is a challenge, because it is a behavioral condition, and depending on how you define it and the methods you use to count it, you may find slightly different numbers.

We see this average of one in 150 as really confirming the past, recent estimates of looking at up to one in 166 children having autism, recognizing that these more recent statistics are more accurate and complete than what we had in the past.

When we say is higher than was previously considered, what we are really talking about is the historical estimates that for decades, the common estimate of autism was considered to be four to five per 10,000 children.

So now that we recognize, it is much closer to the one in 150, we do realize it is more common than historically assumed and these data really reaffirm that particular to the United States autism is more common than we believed and is an urgent public health concern.

MR. NOWAK: The next question.

OPERATOR: Thank you, our next question comes from Mike Stobbe with the Associated Press.

MIKE STOBBE, ASSOCIATED PRESS: Thanks for taking it. This is just basically the question Rick asked. But I was just wondering if there is a best way to characterize your reaction when you got those results? It sounds like surprise is not the right word?

MR. NOWAK: I will let Marshalyn answer that question.

DR. YEARGIN-ALLSOPP: Your question was, whether we were surprised?

MR. STOBBE: Kind of shocked? What was your reaction when you got the 6.6 and 6.7?

DR. YEARGIN-ALLSOPP: Well I think that as Cathy said, you know, it was confirming some recent studies from outside the United States, using different methods in different areas around the world, where there are studies from the late 90s to the present that are reporting this higher rate, the six to seven per 1,000 children.

So it was confirming for us that that is the upper range of studies that have been conducted recently. The fact that the rate in New Jersey was higher though, makes us aware of the fact that in New Jersey and in several other communities there are in the literature a few other studies that are also reporting a rate as high as one percent of the population of children have autism.

So I think that what we′re saying is that where before we were saying there was a range, maybe thinking that it was more towards a lower end, we now feel more confident that the prevalence that in many communities is more towards the higher end of the two to six per 1,000.

MR. NOWAK: Next question please.

OPERATOR: Thank you. Again, as a reminder, if you would like to ask a question, please press star followed by one. Our next question comes from Tom Costello with NBC.

TOM COSTELLO, REPORTER, NBC: Hi. Good morning or good afternoon to you. Two questions, they′re quick ones. What other countries have reported the higher rate that you just mentioned and also, just for clarification, if you don′t mind, could you one time -- one more time go through how do you reach that 560,000 estimate? What′s the math on that so we′re completely accurate on this?

MR. NOWAK: I′ll have Catherine Rice answer that question, those questions.

DR. RICE: OK. In terms of the other two studies we′ve identified that have the rate of around one percent, one was published last year from England and the second was published actually back in 1999 from Sweden.

In terms of that estimate of approximately 560,000 0 to 21 year olds having autism, that comes from if you take the one in 150 as - this is what we′re assuming is the average prevalence of autism, so again, we didn′t study in every community so this is just an exercise, but certainly helps us get an idea of how many people may have autism.

If you assume that four million children are born every year, which is the current birth rate, and also assuming that even though we haven′t identified the rates of 100, one in 150 consistently because we haven′t had the data systems, but let′s just assume that′s what the rate was, you apply to four million births from 0 to 21 years you would get 560,000 children.

MR. NOWAK: And again Tom, that would represent the upper end of the estimate. Other questions?

OPERATOR: Thank you. Our final question comes from Elizabeth Lopatto. Please go ahead - with Bloomberg News.

ELIZABETH LOPATTO, REPORTER, BLOOMBERG NEWS: Yes, I was wondering what this means for future studies and for future research into autism spectrum disorders?

MR. NOWAK: I′ll have Marshalyn answer that question.

DR. YEARGIN-ALLSOPP: OK. Well I′m glad you ask that. We will continue with the autism surveillance with our ADDM Network and data are currently being collected for the 2006 study year. So we can expect reports from the 2006, 2008 and the 2010 study years, we′re funded through 2010 to continue with ADDM.

As Cathy mentioned, we are also - we were mandated by the children′s health act not just to look at the prevalence of autism in different communities across the United States, but to also look at potential risk factors and causes for autism.

So our CADDRE study that Dr. Rice mentioned will be launched this year. So that study is looking at six different areas throughout the United States. It′s a collaborative case control study looking at different risk factors such as reproductive and hormonal factors, immune factors, sociodemographic factors, genetics and others to see what′s different about children and families that have autism compared to typically developing children and children with other developmental disabilities.

In addition, we have the “Learn the Signs. Act Early.” campaign which is a campaign to increase awareness of developmental delays in children and to have parents, professional, health care professionals and child care professionals to act early in identifying these children and referring them for services. So we are, actually we are dong a number of different activities here focused on early identification, prevalence and risk factor for autism.

DR. RICE: Can I add something?

MR. NOWAK: Sure, Dr. Rice, add to that.

DR. RICE: I would just like to add, part of the help of having these types of estimates is to also recognize that these individuals are here living now. And in addition to recognizing the early signs, that these will also point to the importance of the research going on looking at how to best intervene and help support the individuals and families that are affected today. So, there is a lot of research going on in terms of education, support systems and service planning as part of the government and private response to autism.

MR. NOWAK: You actually could take two more questions if there are two questions.

OPERATOR: Thank you. Our next question comes of Liz Kowalczyk with the Boston Globe.

LIZ KOWALCZYK,BOSTON GLOBE: Hi, I just had one quick question on the issue of vaccines. Can we assume that this data would not provide any additional evidence one way or the other for vaccines being a cause of autism?

DR. YEARGIN-ALLSOPP: Since these children were born in, first of all the ADDM Network was not designed to answer that question specifically. So this was not a study to look at potential causes or risk factors for autism. It was a study just to determine the prevalence of autism and the characteristics of the children affected.

As we pointed out, these children were born in 1992 and 1994. So many of the issues related to changes in vaccine policy and number of vaccines, components of vaccines, obviously were not issues in the early 90s. So we don′t have the population that would be affected, for example, by changes in policy that occurred in the late 90s and in 2000.

MR. NOWAK: Next question.

OPERATOR: Thank you. Our next question comes of Adam Boylan, with U.S. News and World Report.

ADAM BOYLAN, US NEWS AND WORLD REPORT: Hi, I was just wondering how the process of identifying autism has changed over time and how that might relate to the increase?

MR. NOWAK: We will let Dr. Rice answer that question or start.

DR. RICE: OK, well in terms of over time, for many years autism was considered the more severe form of the disorder, autistic disorder. Over the time, actually it wasn′t until 1980 that the American Psychiatric Association included autism in the definition of childhood mental illnesses. Although, it is not considered necessarily a mental illness today, it is considered a neuro-developmental disorder. That is another major change in what has taken place, but there was no an official medical definition for autism until 1980.

And after that time, it was also considered a spectrum of disorders, not just classic autism but the inclusion of PPD-NOS or atypical autism, and Asperger′s disorder, which took place in 1994. So that made it very challenging to characterize the population, because what we called autism has changed somewhat.

Part of what we have tried to do with the ADDM Network method is not just look for how children are being classified, but look for the behaviors associated with what we define as the "Spectrum of Autism Disorders" according to the DSM-IV-TR today.

So, what is the wide range of behaviors that go along with autism and how can we find evidence of those behaviors in children with multiple types of developmental disabilities.

So part of what the ADDM Network hopes to do is to be able to have a consistent method to track more of the symptoms of autism from this point forward, get with the children in the first study born in 1992 and the second study born in 1994 and in continuing years as we continue to apply this method we will be able to look at how the symptoms are changing in the population over time.

MR. NOWAK: And if there′s one last question, we can take that.

OPERATOR: Thank you. We do have a follow up question from Rick Weiss with the Washington Post.

MR. WEISS: Hi, thanks very much. I want to make sure I understand correctly, you mentioned in passing that the time of diagnosis had not gotten shorter, so what I would infer from that is that to the extent you see increases, it′s not in this case because of quicker diagnosis and stacking things up that way.

And related to that, can you just speak to the methodology paper towards the end here, is there anything you′re learning about your own methodology that′s pushing you to change the way you′re doing this that might then make it more difficult for us to interpret your subsequent results when we want to compare them to these earliest results.

DR. RICE: Well...

MR. NOWAK: We′ll have Dr. Rice answer that.

DR. RICE: In terms of your first question about whether the age of diagnosis is related to the increases, we can′t really say that there are increases at this point. We need to continue to monitor over time to see if there are, but we know in terms of these two study years, we really weren′t finding that kids were being identified earlier.

In terms of your second of what we′re learning about our system, one is that certainly it is, although it′s a very comprehensive system and very complete that makes it a challenge in terms of being efficient and timely.

And now that the initial set up is there we′ll be able to continue on and move forward. One of the biggest things that we′ve learned is that because we are dependant on information that′s documented in how children are being evaluated for developmental disabilities, access to information on those developmental disabilities is very important.

And we look at information not just from a single source, such as an education system or a health clinic or a diagnostic clinic, but we combine information across sources. So what we′ve learned is that the variability in the quality of information in those records, the access to that information does make a difference in terms of us being able to confirm case status.

And some of our sites had some variation in terms of accessing records that did make a difference in there prevalence. But overall, we found a great deal of consistency across the sites when we were able to combine information.

MR. NOWAK: Thank you. And I guess that I would note that these materials will be available on the CDC web site ranging from the surveillance reports to the press release and in a few hours we will have the transcript.

I thank you for your time and attention this afternoon and look forward to working with you in the future.

OPERATOR: Thank you. This does conclude today′s conference call. We thank you for your participation. You may now disconnect.

END

####

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES