One of the popular topics this week has been the discussion about children outgrowing their autism diagnosis. If you have missed these discussions, you may catch yourself up by visiting here, here, and here.
Before we jump in to the original article on which all of this discussion is based, we would like to point out that we have already discussed how children with autism can recover from autism. If you missed that post, you may read it here.
Ok, so what is all this talk about “outgrowing autism”?
We prefer to go to the original source to make sure people are accurately reporting what was published. Heather A. Close, Li-Ching Lee, Christopher N. Kaufmann, and Andrew W. Zimmerman authored the paper. The journal Pediatrics published the paper which is available online now and will be available in hard copy in February 2012.
The authors set out to describe characteristics and co-occurring conditions in young children, children, and adolescents.The authors also stated that they wanted to describe how characteristics and conditions may cause a change in the diagnosis of ASD.
The authors actually sought participants who either: a) currently had an autism spectrum disorder (ASD) diagnosis; or b) who had an ASD diagnosis in the past but no longer carry such a diagnosis. Interestingly, the authors did not speak to any of the participants directly. Rather, they pulled the data from a database that was collected as part of the 2007 National Survey of Children’s Health.
To complete the study, the authors opened the 2007 database and retrieved answers to two questions.
- Has a doctor or other health care provider ever told you that the child had ASD?
- Does that child currently have autism or ASD?
Once the authors retrieved the data for parents who responded to the two questions, the authors created two groups with the data. One group with a current diagnosis of ASD and a second group of children who do not have an ASD diagnosis but whose parents were once told that the child had ASD.
The authors then ran a number of statistical analyses with the data based on three age ranges of children.
- young children aged 3–5
- children aged 6–11
- adolescents aged 12–17
The authors also examined the data set to determine if any of the following conditions co-occurred in the children:
- attention-deficit/hyperactivity disorder
- learning disability
- developmental delay
- speech problems
- hearing problems
- behavioral or conduct problems
- seizures or epilepsy
Finding #1: Young children ages 3-5 with a current diagnosis of ASD were more likely to have a current learning disability. Might we add here that learning disabilities are often not diagnosed until age 8 or 9. Thus, we find it hard to believe that children ages 3-5 were diagnosed with learning disabilities. Perhaps they had developmental delays but the authors did not state that.
Finding 2: Young children ages 3-5 with a current ASD diagnosis were more likely to have current co-occurring conditions than children without a current ASD diagnosis.
Finding 3: Children ages 6-11 with a current diagnosis of ASD were more likely to have a past speech problem and current anxiety.
Finding 4: Adolescents age 12-17 with a current diagnosis of ASD were more likely to have current speech problems and 10 times more likely to have current seizures or epilepsy. Note: We would hope that the children have current speech problems since the definition of ASD diagnosis requires that speech problems be present!
Honestly, it is surprising that such a low-quality study would find itself in a prestigious peer-reviewed journal such as Pediatrics. It is even more surprising that it would receive subsequent attention from the press. This type of study hardly qualifies as a master’s thesis, let alone a study that is covered in national news.
The authors did not conduct an experiment. The authors did not verify if the children actually had autism. The authors did not review records to determine if a diagnosis existed in the child’s history. The authors based their entire paper on someone else’s data set.
The original data set the authors utilized is also full of issues. The authors noted that the majority of respondents were white, non-Hispanic, and that most of them had health insurance for at least one year. Not to mention that at 61% of the respondents had incomes over 200 times above the poverty level. Thus, the study results are skewed towards white, middle and upper-class families.
Finally, the authors based all of their conclusions and findings on parent report. Please do not think that we do not believe parents. We do. However, a parent may think their child has ADHD or learning problems but that does not mean that an actual diagnosis of ADHD or learning disability exists. Thus, when the authors discuss how other conditions may impact the autism diagnosis, they are merely speculating as their study did not prove, or control for the other conditions.
The bottom line here is that children in this study who had a history of ASD, may or may not have even had an ASD to begin with. The authors failed to verify this information. Moreover, the authors failed to ask if the children received intensive early intervention or biomedical intervention that may have resulted in recovery from autism.
Finally, the authors never stated that children outgrow autism. The authors themselves state that the children who no longer have a diagnosis could have been:
- misdiagnosed in the first place
- responsive to early intervention
- or they may have experienced developmental changes
So, we are asking you to please check the facts before spreading rumors. Children do not outgrow autism.