Apr 06 2011
Disparities in Autism Diagnosis
I have been following the literature on autism diagnosis both because of my general interest as a neurologist, but also because it is at the center of the vaccine-autism myth. There is no question that autism diagnoses have been on the increase over the last 20 years. There are some who assume that this increase in the number of diagnoses being made represents a true increase in the incidence of the disorder – hence an autism “epidemic”.
However, careful analysis reveals that the increase in diagnosis is largely explained (perhaps completely explained – but a small real increase cannot be ruled out) by increased awareness of autism and an expanded diagnosis. (See my earlier posts for a full explanation.)
A recent study sheds further light on this issue. The study was not addressing the broader question of the causes of the increase in autism diagnosis, but rather was focusing on socioeconomic disparities in diagnosis. The results, however, do lend support to the conclusion that the rising autism diagnoses was largely due to increased awareness, rather than a true increase.
Peter Bearman and others were looking for trends in the California system for autism diagnosis between 1992 and 2000. What they found is that autism diagnoses increased disproportionately in wealthier families, but more significantly in wealthier neighborhoods.
“At the height of rising prevalence, which involved children born between 1992 and 1995, kids whose parents had fewer economic resources simply weren’t diagnosed as often as wealthier children— wealthier kids were 20 to 40% more likely than poorer children to be diagnosed,” said study coauthor Marissa D. King, an assistant professor of Organizational Behavior at Yale University’s School of Management “Among children born in 2000, however, parental wealth alone had no effect on the likelihood that a child would be diagnosed.”
Even though the effect of wealth alone disappeared over time, the neighborhood effect remained strong:
According to the study, on average among children born between 1992 and 2000, a child from a poor family that lived in a more affluent neighborhood was close to 250% more likely than a child from an equally disadvantaged family living in a poorer neighborhood to be diagnosed with autism.
This supports other research that shows that the risk of getting an autism diagnosis increases with access to other families with a child with the diagnosis, but not necessarily physical closeness. The researchers conclude from all this that parents talking to parents is the significant variable increasing the probability of getting a diagnosis – parents telling other parents what to look for, and how to navigate the service system. Wealthier neighborhoods have more of an infrastructure and more opportunity for this interaction.
In addition, as further support, Bearman found that wealthier families had a higher proportion of autism diagnoses considered to be at the subtler end of the spectrum – indicating a greater sensitivity in screening for and making the diagnosis.
The totality of this research supports the conclusion that the increase in autism diagnoses during this time was largely driven by diagnostic practices – knowledge of the diagnosis, the availability of services, and parents discussing the diagnosis and how it relates to obtaining services.
In short – we see a sociological pattern of increase in the autism diagnosis over this time, not a biological pattern.
Reference: Socioeconomic Status and the Increased Prevalence of Autism in California.
Marissa D. Kinga and Peter S. Bearman
14 Responses to “Disparities in Autism Diagnosis”
Leave a Reply
You must be logged in to post a comment.
Steve
The link you provided did not have the ability to see the actual study. Was racial disparities controlled for, given the high probability for a racial split when looking across socioeconomic lines. Arguably this could relate more to a biological basis than diagnostic practices. Was this addressed in the body of the research.
Here is the reference: http://www.asanet.org/images/journals/docs/pdf/asr/Apr11ASRFeature.pdf
They did control for race:
“Our second set of analyses incorporates all proxies for socioeconomic status available from the birth certificates that we did not previously use, such as race and whether a child’s mother was born outside of the United States. Addition of these covariates
does not alter our results and the variables are generally insignificant”
Although it is far outside my specialty I find the autism situation fascinating. It seems to encompass so many pitfalls in thinking and in research that it will someday be a great example and teaching tool (more so once specific causes are narrowed down). I love reading about all the recent studies and appreciate the analysis here and over at SBM.
I highly recommend a recent talk I caught on BookTV (sorry for no link but last I checked it wasn’t working on BookTV’s website) by Paul Offit. In that talk he covers a past example of DPT vaccine and a show in the 80′s linking the vaccine to many conditions. It very neatly parallels what we see with the vaccine – autism debate. The show included many heart wrenching tales of parents stating that their child was fine and soon after the shot exhibited symptoms and seizures that hadn’t occurred before. Sound familiar. It turns out that in that instance the cause was eventually (years later) 100% linked to a specific gene variant. Dr Offit comes across as so reasonable that I can’t figure out how the pro disease camp paints him as some dark lord of vaccination.
thank you for the link.
It also went over the gender gap similar to ADHD which has always puzzled me, and apparently not yet identified. What was very compelling was how the lesser severity cases were much higher in the wealthier communities. It is the Police/crime syndrome, hire police offices and the crime rate goes up because more criminals are caught.
There are also disparities in access to diagnostic services. Many developmental pediatricians and neuropsychologists do not participate with insurance. A friend of mine in Southern California was recently quoted $3K for a diagnostic evaluation, which she does not have. Her daughter’s school will not evaluate her.
If you don’t have that kind of money, chances are your child will need to have significant issues with functioning before you can get a diagnosis. Well off families have the money to access services, as well as the knowledge to push for it.
I think the evidence is becoming more and more clear- the rise in diagnosis of this problem is not the same as a rise in actual incidence. That is to say- the same number of people have the problem, it’s just that more of them are being named.
This seems a prototype for understanding the kinds of things that happen when a new condition is recognized and named– including some hysteria around the ‘epidemic of new cases’ and the search for causes. (It is truly unfortunate that vaccines got smeared in this case.)
Thank-you for continuing to cover this.
The following is a fairly recent review of autism prevalence looking at the question from a multinational perspective:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2997266/?tool=pubmed
They point out the same issues Dr. Novella outlined above and in prior posts which account for the majority if not all of the increase in prevalence, though also interpret the data with caution, since an actual increase in incidence still cannot be excluded with available data.
I think it is important to mention that whether or not there are actually more cases, there are certainly more cases being identified. Recent CDC estimates 1 in 110 children in America are affected (1 in 70 boys). Services for these children and their families are not keeping up with the pace of increased diagnosis, unfortunately.
One other issue which I wonder about is the contribution of overdiagnosis. Since the ASDs don’t have a biomarker to make a definitive diagnosis, and diagnoses are based on clinical observations and parental reporting on standardized questionnaires, I wonder how often concerned parents coax themselves and the developmental specialist into a diagnosis. From the perspective of the clinician, what is the incentive to not diagnose a “questionable” case, when the evidence from the literature suggests that earlier diagnosis and intervention improves outcomes? Better to overdiagnose than to delay a diagnosis and miss a window of opportunity to treat…
A couple of years ago Prometheus did a comparison of (rising) autism diagnoses and (falling) mental retardation diagnoses, finding that diagnostic substitution accounted for most of the increase in autism:
http://photoninthedarkness.com/?p=158
Any (or a combination) of several root causes might be responsible for such a phenomenon. One is suggested by the study you’ve discussed — wealthy, networked families are more likely than the poor and isolated to hear about autism, and to find it a better explanation than old-fashioned mental retardation for what’s wrong with their kids.
Could comparisons be made with diagnoses for Dyslexia? This condition seems to have mushroomed in the last 30 years or so; the definition becoming ever broader.
Getting diagnosed can be very difficult in some areas, my youngest son has autism it took a long time to get him diagnosed. For a couple years people (teachers, relatives, and friends) would tell us he was really “odd”. We knew he was a little odd but we dismissed it because he was acting just like his dad did growing up.
When he was younger many of the traits could be dismissed as just he age, however as he grew it became much more obvious that there might be problems. So we took him to our pediatrician, when we told him we suspected autism his response was “So, what do you want me to do?” We had no idea and neither did he. The second doctor we asked for help was not a pediatrician but just a general practitioner, her response was “he talks, how can he have autism?”
It only got worse from there, we searched for months trying to find someone who could diagnose him, one psychologist basically gave him him the diagnosis without even looking at him, and didn’t understand that we really wanted to know for sure what was different with him. It took about 9 months before we found a doctor who specialized in this area who could do the testing and give a real answer.
It turns out he was acting “just like Dad” of course Dad has autism too. My sons autism is a little more severe than my own, but it does explain quite a bit in my life.
If we were rich it would have take a lot less time, most of our problems stem from the fact that any provider we found had to be approved by our insurance. We had to drive almost 100 miles one way for multiple visits. If we were poor or destitute it would have been totally impossible. As it was, it was very difficult.
I’m not 100% sure why this study shows that increased autism diagnosis is a result of increased awareness of autism and an expanded diagnosis. Wouldn’t richer communities also have a higher rate of diagnosis if the rate of autism was actually rising? If autism really was increasing, why wouldn’t the wealthier parents telling other parents what to look for etc. also result in higher rates in wealthy areas?
Anecdote Alert:
I work in a private pediatric clinic where one of the services offered is autism diagnosis (I’m actually the one who usually administers the ADOS, under supervision of a psychologist). We’ve had a lot of meetings with local doctors, social workers, and pediatric rehab centres. At one of the centres, we were told by the social worker that they will accept a child into their autism program if the diagnosis was based on *either* the ADOS *or* the ADI-R.
That’s an unreliable way to make the diagnosis. You really need both tests, plus a developmental history and in some cases other developmental testing, before you can make a solid diagnosis. Meanwhile, our psychologist has occasionally gotten referrals for therapy for kids diagnosed with autism based on one test or the other, only to actually meet the child and figure out pretty quickly that the child is *not* autistic, but has some other sort of issue — cognitive delay, severe auditory processing disorder, anxiety, etc.
Meanwhile, the social worker at the rehab centre has told us she’s been accused of being too strict, and pressured to accept a “diagnosis” of autism based on an assessment involving no formalized testing.
Does this sort of thing happen elsewhere? If so, that would contribute to the apparent rise…
Whilst obviously as a method for measuring “whatever” becomes available/more refined logically the numbers of discovered “whatevers” increases.
To conclude from this there is no biological component isn’t very logical however.
IF indeed the increased awareness causes an increase in diagnosed cases, it surely means there are more cases then previously assumed.
Due to the slow reproductive cycle of homo sapiens it’s way to early to make any kind of educated guess, let alone conclusion that a biological component isn’t behind an increase since evidently there never was a correct baseline to begin with.
In a few centuries it’ll become more clear, my bet is on an environmentally driven adaptive process due to the fact that modern society is no match for the stratagems contained in the limbic system.
Thus the limbic system gets a smaller and smaller part to play, hence the changes in the white matter/corpus collosum as observed in autistic brains.
my 2 cents
There was a study out of Yale recently that took into account children not yet diagnosed
http://www.dailyrx.com/news-article/autism-may-be-far-more-prevalent-once-thought-13522.html
The samples were from Korea and Canada & so a good sociological cross section.