IAN Research Report #2 - July 2007
The goal of the Interactive Autism Network (IAN) project, which launched on April 2, 2007, is to accelerate the pace of autism research by linking the families of individuals with autism and autism researchers. We now present our second report on project accomplishments and preliminary findings.
Over the course of IAN's first four months we have reached out to two very important groups: families of children with Autism Spectrum Disorders (ASDs) and autism researchers. (We are preparing to include adults with ASDs, as well, as we discuss in detail later in this report.)
Families of Children with ASDs
Mothers and fathers of children on the autism spectrum from across the United States have registered with IAN Research, filling in questionnaires over the internet to create a massive data-set addressing a variety of autism topics. They have shared information about their children with an Autism Spectrum Disorder (ASD), non-affected siblings, and themselves. So far, over 17,000 individuals (moms, dads, children with ASDs, and non-affected children) are taking part in IAN Research. Many of them are not only sharing data, but are willing to be contacted about additional research studies that cannot go forward without research participants.
Thank You, Families! It is due to your enthusiasm, and the generous gift of your time and your stories, that researchers will be able to make new discoveries.
We held our second IAN Research Conference on June 19, 2007, inviting researchers from all over the country to learn to use IAN to support, expand, or enhance their projects. They asked questions, provided feedback, and helped begin to design the next generation of IAN questionnaires. (Read our IAN Research Conference Report.)
Within days following this dynamic introduction to the project, five researchers had applied to use the IAN data, and nineteen had applied to recruit subjects through IAN.
Thank You, Autism Researchers! It is due to your willingness to try this new tool, and to help us improve and hone it, that accelerated progress will be possible on a wide variety of fronts.
"A Unified Genetic Theory for Sporadic and Inherited Autism" — First Research Paper to Use IAN Data
On April 18, 2007, only two weeks after the Interactive Autism Network's official launch, Dr. Michael Wigler of the Cold Spring Harbor Laboratory (CSHL) was the first researcher to use IAN data. He and his team have now published their groundbreaking article, "A Unified Genetic Theory for Sporadic and Inherited Autism," in the Proceedings of the National Academy of Science. This is the very first publication based, in part, on IAN data.
The article describes a new genetic model for understanding how autism is acquired. Unlike previous genetics efforts, which were largely focused on looking for an autism susceptibility gene – gene or genes that cause autism in many families around the world – the new model sees the underlying cause of autism as the spontaneous, new mutation of a gene. This intriguing theory therefore has the potential to link genetic and environmental factors in autism.
In the majority of cases, according to this theory, individuals acquire autism through a spontaneous mutation affecting genetic material within their parent's sperm or egg cell. In other words, a gene is altered within a single sperm or egg cell at some point during the parent's life. If a child is conceived with that particular sperm or egg, the child may be autistic. (The mutation was present in only one sperm or egg cell of the parent, but it will be present in virtually every cell of the resulting child.)
Although they account for the majority of autism cases, such mutations are fairly rare, and are unlikely to happen to the same family twice. This makes sense when you consider that the mutation was in only one sperm or egg cell. You'd have to have another spontaneous autism-causing mutation in another sperm or egg cell to have a second child with autism.
However, the authors of this paper theorize, autism is inherited in a minority of cases. In such instances, an autism-causing genetic mutation is already present in the family. A parent who has inherited it carries a mutated gene in virtually every cell in their body, including half of their sperm or egg cells. They will pass the altered gene on to their offspring 50% of the time in what is known as an autosomal dominant pattern. (For an explanation and chart explaining autosomal dominant inheritance, please see the Genetics Home Reference.)
In a family like this, we would expect to see multiple cases because male children would have a 50/50 chance of developing an ASD. (As we know, boys develop autism much more frequently than girls. Therefore, women carrying such a gene are much more likely than men who do so to function normally and to have children. They can then pass the gene to their sons and daughters. Similarly, male children inheriting the gene are more likely to develop ASDs than daughters who might inherit it.)
These findings may account for gaps in previous genetic models of autism. They also provide a fascinating possible link between genetic and environmental explanations for autism, for one must ask: What in the environment may be causing genes to mutate in the first place?
The researchers working on this theory hypothesized that, if they were right about spontaneous mutation and autism, a certain mathematical pattern would be found in large sets of autism data. They found it first in the Autism Genetic Research Exchange (AGRE), which contains data only from families with more than one child with an ASD. They then used data from the Interactive Autism Network (IAN) to check their theory against a data set including both families with more than one person with an ASD and families with only one person with an ASD. They found exactly the pattern they had predicted – a major scientific finding.
We had many questions about this new theory, and knew you probably would, too. Read our expanded write-up, as well as a list of questions and answers regarding the theory’s implications.
You can also access the original article in the Proceedings of the National Academy of Science: “A Unified Genetic Theory for Sporadic and Inherited Autism.”
IAN Data: Preliminary Findings
With time, more and more qualified researchers will use the IAN data, and we will share their findings with you. Investigation takes time, however, and publication takes even longer. In the spirit of sharing some results with you sooner rather than later, we now take a preliminary look at some of the IAN data.
In our first IAN Research Report (May 2007), we reported on the Autism Spectrum diagnoses of children in the IAN database, as well as their ages, whether their parents felt they had lost skills at any point during their development, and whether they had been diagnosed with mental retardation or motor delays. We reported on what types of schools the children were attending, and the top twenty treatments their parents were using to try to help them. We also illustrated to what extent children with ASDs, as well as their family members, had been diagnosed with or treated for disorders of attention or mood (such as ADHD, Depression, Anxiety, or Bipolar Disorder).
In the current report, we will take a look at:
Children with ASDs in IAN Research
- Gender Ratios
- Issues of Attention and Mood by Gender
- Areas of First Parental Concern
- Intellectual Disability and IQ
If you would like some help interpreting the graphs that follow, you can take a look at "Making Sense of IAN Charts."
According to the DSM-IV, autism is four to five times more common in males than in females. What gender ratio are we finding in the IAN Research data?
Figure 1. Children with ASDs in IAN Research -- Male to Female Ratios
Except for Asperger's Syndrome -- a category in which we have 9 boys for every girl -- the IAN Data does match the DSM-IV specified ratio of 4 or 5 boys to 1 girl. (See Figure 1.)
What might be going on with Asperger's Syndrome? Is it simply that girls with ASDs tend to be more severe cases, and so less likely to receive an Asperger's diagnosis? Or do girls with Asperger's present differently, and so are being missed?
(For some thoughtful insights on this topic, see Dr. Tony Attwood's paper, "The Pattern of Abilities and Development of Girls with Asperger's Syndrome," on his website at http://www.tonyattwood.com.au/.)
In our first research report, we looked at issues of mood and attention, but across ASD diagnoses and within families. We now report on such issues as these impact children with ASDs according to gender.
We are considering only children age ten and over because including younger children would be likely to dilute the percentages by mixing in individuals so young they would not be very likely to have been diagnosed with, for example, Depression.
Do older boys and girls with an ASD experience about the same level of mood or attention issues?
Figure 2. Children with ASDs in IAN Research -- Issues of Attention and Mood by Gender
From the IAN data, as shown above in Figure 2, it appears that the girls and boys with ASDs in IAN Research are fairly similar when it comes to being diagnosed with or treated for ADHD or ADD, Anxiety, or Depression. The only significantly different category was Bipolar Disorder. Thirteen percent of girls, but only 8% of boys, had been diagnosed with or treated for Bipolar Disorder. Are girls with ASDs really more likely than boys to suffer from Bipolar Disorder? Or is there some diagnostic bias by gender, in which girls with meltdowns are more likely than boys to receive a diagnosis of Bipolar Disorder? More research on this question and others related to "comorbid" conditions is very much needed.
One question we asked in IAN Research was: "What was your first concern about your child's development?" As parents look back, what do they say was their first worry about their child's development or behavior?
To begin, we looked at the overall answer to that question, irrespective of ASD diagnosis.
Figure 3. Areas of Parental First Concern
Twenty-five percent of families reported that delayed or abnormal babbling or speaking was the first aspect of their child's behavior that led them to worry. There was a great deal of variation in parents' area of first concern, however. (See Figure 3.)
Below, you can view the same data by diagnosis.
Figure 4. Area of First Concern (Autism and PDD-NOS)
Figure 4, above, shows that parents of children with Autism or PDD-NOS tended to share similar, though varied, areas of first concern. Parents of children with Asperger's Syndrome, on the other hand, tended to have areas of first concern that were somewhat different. (See Figure 5, below.)
Figure 5. Area of First Concern (Asperger's Syndrome)
It is interesting to note that the largest category chosen by parents of children with Asperger's was "Other," and that many parents of children with Autism or PDD-NOS chose this, as well. When researchers see that, they know that at least one crucial category was missing from the possible responses they provided with their questionnaire. What might this have been?
To find out, we looked at the comments entered by parents who had chosen "Other." (The IAN questionnaires provide parents a space in which to type additional thoughts or explanations when they feel a simple "yes" or "no" answer isn't adequate to convey what they mean.)
Many who had chosen "Other" described meltdowns, rages, tantrums, anxiety, or constant crying as an area of first concern. Additional areas of first concern mentioned under "Other" were seizures, or children who couldn't sit still or focus. Any future questionnaires on this topic will need to include categories that capture these.
In our last research report, we displayed pie charts showing the percentage of children with ASDs in IAN Research who had also been diagnosed with intellectual disability. Our data showed, and continues to show, that 10% of all children with ASDs in IAN have been diagnosed with intellectual disability, with that figure going up to 15% if only children with autism are considered.
There has been considerable debate in the field about how many children with ASDs are also intellectually disabled. Recent estimates have ranged from around 50% to as low as 15%, down from estimates during an earlier era that went as high as 80%. The decrease in the number of children diagnosed as intellectually disabled is thought to be due both to the broadening of the category of ASDs to include individuals with higher IQs, and to the successes of early intervention. 1
Nevertheless, our figure was definitely on the low side, so we decided to dig a little deeper into the IAN data. First, we took a look at IQ scores. About half the children with ASDs in IAN have had a formal IQ test, with a spread of IQs as follows:
|IQ Score||Percent of Children with any ASD Diagnosis||Percent of Children with an Autism Diagnosis|
40 or below
Total with IQ of 70 or Less
N for Table = 1,520 (In other words, 1,520 children with ASD participating in IAN had had an IQ test according to their parent)
Our figures for intellectual disability and IQ presented us with a bit of a conundrum. Parents had told us that only 10% of the children had been diagnosed with intellectual disability, but then told us that 28% of those tested had an IQ of 70 or less, which is the definition of intellectual disability. (See Table 1, above.) What could explain this?
Upon further investigation, we discovered that 18% of parents who had said their child had not been diagnosed as intellectually disabled also said that their child's IQ was 70 or below. To understand why, we looked again to parents' comments.
Some parents told us that their child's IQ had been in the intellectually disabled range, but nobody had ever officially given them the "intellectually disabled" diagnosis or label. Others said their child couldn't cooperate with the testing process, whether because of an inability to focus or meltdowns, and was given an intellectual disability diagnosis when they hadn't really taken the test. Still others reported that IQ tests had yielded different results at different times, describing 20 point spreads in test scores. Many objected to the IQ test, declaring that they didn't believe such tests can accurately measure the IQ of children who are non-verbal or possess little language, or who have social problems that make it hard for them to successfully interact with "some stranger" giving the test.
Clearly, many parents feel that their child's intelligence and ability cannot be easily measured by standard means. Researchers trying to discern what percentage of individuals with ASDs have what level of cognitive ability will need to be especially careful to address all of these issues.
Sampling: Are We Reaching A Representative Population?
One important issue for any researcher gathering data from and about people is whether they are reaching a representative population. In other words, if you are polling county voters, and 5% of the people taking part in your poll are African-American, when 35% of the voters in the county are African-American, you are not going to have a representative sample. Your results may be skewed and inaccurate to the extent that you are over-representing some voters and under-representing others.
IAN is trying to gather data from all over the United States. How can we begin to gauge if we are reaching people from all across the country, from all socio-economic strata, from all racial and ethnic groups, both urban and rural?
The fact that we are an internet project will make us accessible to many who could never have taken part in a traditional study at a clinic or university. Residents of rural areas or of any place far removed from a research center, as well as parents unable to take time away from home or work, can more easily take part in IAN. Still, there are some who do not have access to high speed internet services, or who lack basic computer skills, and so might find the project inaccessible. One question we wanted to explore was: how are we doing reaching a representative population?
Having collected data on families' race and ethnicity, we wanted to evaluate how we were doing as far as reaching minority populations.
Figure 6. Race and Ethnicity – U.S. vs. IAN Participants
From the data displayed in Figure 6, above, it is clear that we are under-representing several groups and will need to work harder in this area. We will have to find ways to get the word out about IAN to African-American, Hispanic, and Asian families, as well as develop ways for any family that does not have access to a computer to participate, perhaps via volunteer groups who will help inform under-represented groups about IAN, and who can assist willing participants to complete the IAN questionnaires. It is also our hope that we will be able to provide IAN in Spanish at some point in the future. We have received many requests for this in our initial months.
One factor that might help us determine if our study is reaching the broadest population possible is parental education. How does the educational level of the mothers and fathers taking part in IAN compare with that of men and women in the United States in general?
Figure 7. Education -- Mothers in IAN vs. Women in the U.S.
Figure 8. Education – Fathers in IAN vs. Men in the U.S.
Looking at Figures 7 and 8, it is apparent that mothers and fathers taking part in IAN are more highly educated than men and women in the general U.S. population. For example, consider Ph.D.s. Only 3% of men and 2% of women of parenting age in the general population have one, while 11% of fathers and 5% of mothers taking part in IAN do. This may be because recruiting for IAN began with advocacy organizations. It may be that those taking part in advocacy tend to be more educated than the general population. It is also likely that those with more education are more likely to have internet and e-mail access, and a certain comfort-level working with a computer-based project like this one.
This is far less the case now than in the past. At one time, the "digital divide" between the "wealthier, more educated, and plugged in" and the "poorer, lesser educated, and cut off" was much larger. According to an April 2006 PEW Internet and American Life Project Report, 73% of all American adults are now using the internet, and that figure goes up to 84% for 30-49 year olds and 88% for 18-29 year olds --those of parenting age.
That, however, still leaves 16% of 30-49 year olds and 12% of 18-29 year olds without access to the internet.
The IAN team will need to identify any barriers that are in the way of parents with less education (including the likelihood they hear about the project in the first place) and find ways to overcome these.
We were also interested in participation in IAN throughout the United States. Are all states participating to the extent you would expect based on their population?
The answer is no! Figure 9, below, shows a great deal of variation in participation in IAN by state. Please note: this map does not show total participation, but how close a state has come to reaching its "fair share" of participation in IAN based on its population.
What do we mean by this? If 1 of every 150 children in the United States has an ASD, and they are more or less evenly distributed, then California is going to have a lot more of them than Maryland. Our goal for California, therefore, will be higher than our goal for Maryland. Even though California may actually have more IAN participants than Maryland, it may still be behind on reaching our population-based goal for it.
The shading on the map in Figure 9, therefore, represents the "fair share" percent for each state (based on population) of our initial goal of 30,000 consented participants by the end of our first year. It does not represent the flat number of participants in any one state. (See the map key, at left.)
Please note: we are talking about initial goals only. We still need families from each and every state!
Maryland, Missouri, and New Jersey have exceeded initial goals (based on their population), with Washington, New Hampshire, Massachusetts, Connecticut, Pennsylvania, and Virginia approaching the initial goals set for them. On the other hand, participation is far lower than we had hoped for in much of the South and West.
Figure 9. Participation in IAN by U.S. State – Fair Share by Population (Hawaii and Alaska not shown)
If you see that your state is under-represented, please consider taking action to get the word about IAN to as many families and professionals in your area as possible. It is so important that voices from every part of the country are heard.
And remember... This is an initial goal only, which we hope to exceed by far. Even if you are in a state that has reached its initial goal, we still need you!
Our Next Goal: Including Adults with ASDs
From the very beginning of the IAN project, we heard one message loud and clear: Adults with ASDs and their families wanted very much to participate in IAN! We wanted to let you know that we are currently developing "Adults with ASDs" questionnaires, and hope to launch an Adult edition of IAN in early 2008. There will probably be at least two versions: one for those who are able to consent to become research subjects and fill out their questionnaires themselves, and one for those under guardianship to be filled out by a guardian. (It may be more complicated still, with individuals under guardianship who, nevertheless, are capable of filling out their own forms, or individuals who want to fill them in with the assistance of another person.)
We still have a way to go to work out the legal and technical details, but are making progress, and will keep you posted regarding the premiere of IAN for Adults.
More Questionnaires Coming
We will let you know, both by e-mail and on the IAN Community, as new questionnaires come on line.
Having Trouble Finishing? Can We Help?
If you have registered for IAN Research, but have not yet finished your questionnaires, we understand. It takes time and peace to fill out the IAN questionnaires, and both can be in short supply for a parent with a child on the autism spectrum.
We appreciate that you are trying, and ask that you please do whatever you can to finish. It is truly important! Until your forms are "finalized" – that is, you click the button indicating you've checked them over and you're done – your data is not included in analysis.
Please finish your questionnaires, keeping in mind that you can do a little at a time, if necessary, coming back to the forms as you are able. Don't hesitate to contact us if you have any questions or become stuck in any way. We are here to help!
Write us at ResearchTeam@ianproject.org.
Our Sincere Thanks!
We are so grateful to each and every one of you who has taken the time to register for IAN, filled out questionnaires, or spread the word to others about the project. The more people participate, the more valuable IAN will become as a resource for autism researchers, and the more progress we will make.
Families and Researchers alike, you are our IAN pioneers. Thank you so much for helping to make these first months a success.
The analyses presented here by the Interactive Autism Network are preliminary. They are based on information submitted via the Internet by parents of children with autism spectrum disorders (ASDs) from the United States who choose to participate. They may not generalize to the larger population of parents of children with ASDs. The data have not been peer-reviewed -- that is, undergone evaluation by researchers expert in a particular field -- or been submitted for publication. IAN views participating families as research partners, and shares such preliminary information to thank them and demonstrate the importance of their ongoing involvement.
We encourage autism researchers investigating these topics to apply for access to the IAN database. Contact firstname.lastname@example.org.