Monday 4 June 2012

The ‘autism epidemic’ and diagnostic substitution

Based on: King & Bearman (2011) American Sociological Review, 76(2), 320-346; 
Data from birth and diagnostic records for all children born in California 1992-2000
Everyone agrees there has been a remarkable increase in autism diagnosis across the world. There is, however, considerable debate about the reasons for this. Three very different kinds of explanation exist.
  • Explanation #1 maintains that something in our modern environment has come along to increase the risk of autism. There are numerous candidates, as indicated in this blogpost by Emily Willingham
  • Explanation #2 sees the risks as largely biological or genetic, with changing patterns of reproduction altering prevalence rates, either because of assortative mating (not much evidence, in my view) or because of an increase in older parents (more plausible). 
  • Explanation #3 is very different: it says the increase is not a real increase - it’s just a change in what we count as autism. This has been termed ‘diagnostic substitution’ - the basic idea is that children who would previously have received another diagnosis or no diagnosis are now being identified with autism spectrum disorder (ASD). This could be in part because of new conceptualisations of autism, but may also be fuelled by strategic considerations: resources for children with ASD tend to be much better than those for children with other related conditions, such as language impairment or intellectual handicaps, so this diagnosis may be preferred.
In 2008, my research group published a study that documented one kind of diagnostic substitution. We contacted people who had taken part in our studies of children with specific language impairment years ago. We carried out a standard diagnostic observation procedure for autism with the young adults themselves and, where possible, interviewed their parents about their early history. We found a number of individuals who had been regarded as cases of specific language impairment ten or twenty years ago but who would nowadays be diagnosed with ASD. Although it’s possible that some people develop autistic symptomatology as they get older, in our cases the autistic symptoms appeared to have been present from early childhood - as indicated by the parental interviews. Around half of the sample had been identified as having ‘semantic-pragmatic disorder’ in childhood, but autism had been excluded because at that time, prior to publication of DSM-IV diagnostic guidelines, it was regarded as a very rare condition in which there were severe social and behavioural impairments. How many children would have qualified for ASD diagnoses had they been seen today? Well, it depends. I suspect few people appreciate just how flexible the diagnostic criteria are for autism, even when lengthy standardized diagnostic instruments are used. Although we used the gold standard diagnostic procedures (ADOS-G and ADI-R) we found they seldom gave the same answer. If we diagnosed ASD only when both diagnostic instruments agreed, 21% of cases met criteria. If we included anyone who met criteria for autism or PDDNOS on either ADI-R or ADOS, the rate shot up to 66%.
Last year, a fascinating study by Brugha and colleagues attacked the same question from a different angle. They did an epidemiological survey of a representative sample of adults from the English population, using the ADOS-G, and found that the rates of ASD were similar to those recently reported in children. Within the adult population, rates of ASD did not change with age. Thus, provided we stick to the same diagnostic criteria, then the prevalence of autism is the same for those born several decades ago, as it is for the current generation of children. Importantly, none of these adults with ASD had received a formal diagnosis.
Recently, we conducted a study with another group: children with an additional sex chromosome (i.e. trisomy). We had not intended to study diagnostic substitution: the goal was rather to understand more about the language difficulties that had previously been described in children with sex chromosome trisomies. The effect of an extra sex chromosome is relatively mild: most of these children attend mainstream schools and they do not have any obvious physical abnormalities. Indeed, they can be hard to study because many individuals with trisomies will be unaware of their condition. We gathered information by parental report, and did not do any direct evaluation of the child, but we did ask about whether the child had had any kind of diagnosis by a medical or psychological expert. We confirmed that there was a strong association with language problems in all three kinds of trisomy (girls with XXX, and boys with XYY or XXY), many of whom had had speech-language therapy. But we also found that 2/19 (11%) of boys with XXY and 11/58 (19%) of those with XYY had received an ASD diagnosis.
It is important to emphasise that most children with a sex chromosome trisomy did not have an ASD diagnosis, and many were not giving any cause for concern. Nevertheless, although they are only a minority of cases, the proportion with ASD is much higher than in the general population. We were really surprised at this because before publishing our study we had done a systematic review of the literature on children with sex chromosome trisomies, focusing on studies that avoided ascertainment bias. In these studies, not a single case of autism had been mentioned when discussing outcomes. So was our study a fluke? We are confident this is not the case, because this year two further studies from the USA have been reported (Ross et al and Lee et al, in press), both of which got results very similar to ours, though using different methods.
This research provides further evidence that diagnostic substitution has occurred, suggesting that children who in the past would have been diagnosed with language impairment are now being diagnosed with ASD. The only other way to explain the increased diagnosis rate in children with a known chromosomal abnormality would be if the trisomy acted as a risk factor, making children more sensitive to environmental factors that could cause autism. That’s a possibility, but it seems more likely that cases of ASD were missed in the past because more stringent diagnostic criteria were used, just as was found in our follow-up of children with SLI and in the epidemiological study of adults by Brugha and colleagues.
It is becoming clear that changing diagnostic criteria, increased awareness of ASD, and strategic use of diagnosis to gain access to services, have had a massive effect on the numbers of children with ASD. When I started studies in this area, I thought diagnostic substitution had happened but I did not think it would be sufficient to explain the increase in numbers of ASD diagnoses. But now, on the basis of studies reviewed here, I think it could be the full story.

PS: a slightly extended version of this blogpost was featured on PLOS Blogs on 8th June 2012.

Bishop, D., Jacobs, P., Lachlan, K., Wellesley, D., Barnicoat, A., Boyd, P., Fryer, A., Middlemiss, P., Smithson, S., Metcalfe, K., Shears, D., Leggett, V., Nation, K., & Scerif, G. (2010). Autism, language and communication in children with sex chromosome trisomies Archives of Disease in Childhood, 96 (10), 954-959 DOI: 10.1136/adc.2009.179747
Bishop, D., Whitehouse, A., Watt, H., & Line, E. (2008). Autism and diagnostic substitution: evidence from a study of adults with a history of developmental language disorder Developmental Medicine & Child Neurology, 50 (5), 341-345 DOI: 10.1111/j.1469-8749.2008.02057.x  

Brugha, T. (2011). Epidemiology of Autism Spectrum Disorders in Adults in the Community in England Archives of General Psychiatry, 68 (5) DOI: 10.1001/archgenpsychiatry.2011.38

Lee, N. R., Wallace, G. L., Adeyemi, E. I., Lopez, K. C., Blumenthal, J. D., Clasen, L. S., & Giedd, J. N. (2012, in press). Dosage effects of X and Y chromosomes on language and social functioning in children with supernumerary sex chromosome aneuploidies: Implications for idiopathic language impairment and autism spectrum disorders. Journal of Child Psychology and Psychiatry

Ross, J. L.,et al (2012). Behavioral and social phenotypes in boys with 47, XYY syndrome or 47, XXY Klinefelter syndrome.  Pediatrics, 129(4), 769-778. doi: 10.1542/peds.2011-0719



  1. Uta Frith has been trying unsuccesfully to add a comment to this post. I am not sure why this doesn't work, as I do not have restrictions on comments, but if anyone else has the same trouble, please contact me!
    Meanwhile, here is Uta's comment:

    Why is it so difficult for most people to believe that the massive increase in cases of ASD is not dues to an actual increase in children born autistic, but due to greater awareness, and wider diagnostic criteria and more sensitive assessment instruments? Could it be that we are extremely vigilant to avoid possible and previously unknown dangers - even when such dangers do not exist? Scare stories always find ready believers, and sensible and seemingly demystifying accounts are often less attractive. This is why the Bishopblog is a breath of fresh air.

    I too am persuaded by the data so far that there simply is no autism epidemic - but there is diagnostic substitution. Of course, as a cautious scientist I know that other explanations cannot be ruled out completely. Still, the fraction of cases these may conceivably explain would be vanishingly small compared to the cases explained by diagnostic substitution.

    Here is another conjecture: autism is a genetically based condition and the reason that many people still find this difficult to accept is that our understanding of how genetic factors work is very poor. This poor understanding adds to the suspicion that there is something sinister about an increase in cases. In fact, the increase means something good, since cases previously not diagnosed may well have been denied essential help.

  2. I appreciate this post because I have seen people who are respected in their fields (fields not related to autism) comment publicly that the increased prevalence, particularly in the US, *must* be largely attributable to "toxins" and "chemicals," rather than to the factors you describe here. Their rationale doesn't square with any data I've seen or even with common sense, as Dr. Frith notes above, but I've nevertheless been seeking expert analyses from people who *are* autism researchers to gauge my own inferences about the data. As I keep pointing out, the prevalence of autism as a clinical entity was zero before Kanner characterized it, and that wasn't because autistic people didn't exist; it was because they had been given other labels.

  3. Interesting post on a complex topic perhaps echoing other suggestions: (DSM-IV vs. DSM-III).

    I would however refer you to the series on autism published in Nature last year:
    and in particular this paper:

    What they seem to be saying is that the increase in cases of autism is likely complex and whilst diagnostic substitution is a contender, it is probably not the only reason for the increase.

    I offer no opinion as to what else might be driving the numbers of cases coming through aside from suggesting that we cannot rule out differing factors in different locations.

    I do think it is premature to discount 'environment' as potentially being involved. Particularly when the rise (and rise) of sciences such as epigenetics seems to be offering some interesting glimpses...

  4. This is almost completely off topic, but, this is the first time I've seen Autism Spectrum Disorder shortened to ASD. As a parent of a child with Atrial Septal Defect, which is very commonly abbreviated as ASD, I wonder about the confusion this might bring if it also becomes common usage. But, there it is on an NIH info page. Hmm... I will be extra cautious when filling out medical release forms for schools.

  5. Many thanks to all for your comments
    HardyoYo - I'm afraid ASD for autism spectrum disorder is extremely common usage - in the research literature at any rate. Let’s hope these acronyms are used in contexts where the meaning will be clear.
    Paul: thanks for those additional references. I agree it’s complicated and the Nature news feature does an excellent job in giving a balanced account in our current state of knowledge.
    To all my readers, let me clarify how I see the logic of this.. Basically, if rates of a disorder increase dramatically over a few decades, then we have to look for something in the environment that has changed that could explain it: changes in heritable factors would only occur over a much slower timescale (though an environmental factor could operate by causing an increase in mutations. Or you could get more mutations if the proportion of older parents increased). But the converse does not follow: if the rate of autism is actually stable when the diagnostic criteria are held constant, that does not prove that it’s all genetic. It’s entirely possible that there may be environmental factors involved and there may be more complex interactions such that the impact of the environment depends on the genotype.
    So I am definitely not ruling out a role for the environment, but I am questioning whether the prevalence rise is part of the evidence we should be using. Purely from the point of view of the science, it’s a shame if the increasing prevalence is just a red herring, because otherwise it could provide a valuable clue to potential environmental factors that might be involved: we could look for factors that correlate with rates of autism at different times and at different geographical locations. If, though, the true rate is stable, then we’d need to do just the opposite - look for environmental factors that have been consistent over the decades.
    The real bugbear in all of this is the failure of the child psychiatry community to come up with robust and objective diagnostic criteria. This is not for lack of trying, but I find it depressing that after years of developing ever-more complex diagnostic procedures, we still come down to “expert clinical opinion” as the basis for diagnosis. See:

  6. I am a firm believer that it is our 21st century societies with their chaos, unlimited choices, etc which are causing the increase. Etiquette, strict social rules, etc were de rigeur 2, 3 decades ago. An autistic person in the jungle will not have a lot of bother, one with 1 leg only will. A physically disabled person with 1 leg will not have a lot of bother in our society, an autistic person has more and more problems adapting to our chaotic society.

  7. Carine, why firmly believe anything without evidence? I have the feeling that there were always many autistics, mostly from historical tales of people who seem to match. The etiology, though, is up for grabs.

  8. Option 3 must have some applicability. This is not the first time new criteria and new attention have lead to an increase.
    ADHD and Bipolar Disease NOS experienced significant increased diagnoses in American children.
    Autism children experiencing severe meltdowns and anger have new hope though; search on "Hope for the Violently Aggressive Child" for more information.

  9. For me there is a 4th reason at that is just better recognition and greater knowledge so those that would have not been diagnosed 10 years ago are now being diagnosed. Also if you go back a little further the diagnosis of Aspergers was not around and when it was it was little know about and hard to get diagnosed. This meant that those, like me, we just left wondering with no real diagnosis confirmed, just be told it may be this or it may be that.

  10. in my opinion, a lot of what counts for 'autistic' in children - at least under 5 - is a misinterpretation of a child's reaction to anxiety. now, some of the services that an anxious child may need are the same as for children with ASD.

  11. To emphasise what blue4dolphin says above, greater knowledge is probably very largely responsible here. Asperger's Syndrome - which I was diagnosed with myself as an adult a few years ago - was hardly known in the English speaking world until the early 1980s, and even then it took years for it to filter down to public programmes of support. In my experience as well the ignorance level concerning Asperger's Syndrome remains especially high - even in the healthcare profession, and even amongst doctors. During the 1980s, someone who today would be diagnosed as Aspergic would be characterised as odd or eccentric if they were lucky, weird or other terms less friendly if they were not.

  12. I, like most, have heard the reports and seen the articles saying that the number of cases of autism are increasing dramatically. Most of what I have seen and heard have lead me to believe that there has to be an underlying cause for such an increase. Is there something in our foods, or perhaps there is an environmental factor that is responsible for this increase.
    I now have another theory to ponder. What you are saying makes sense. There has been a tremendous amount of media coverage on autism in the last several years and while these reports are inteded to "educate" the public, they have done just that. Anyone with any concerns or questions about autism can find an overwhelming amount of information available to them. Thus they know what questions to ask, they know the characteristics to look for and they won't allow their children or themselves to be misdiagnosed.

  13. I'm afraid ANOTHER person has had problems adding comments to my blog. If this happens to you, please contact me and I can put the comment up. This one from Bob Jensen today

    A new study just published reveals that diagnostic substitution has been occurring on a massive scale since the introduction of DSM-IV (1994) and ICD-10 (1994). Utah is one of the states included in the CDC autism monitoring network. The 2012 CDC Autism report found that Utah had the highest autism prevalence rates of all the states in the network, 1/44. The authors located 108 of 498 children who were examined in the 1980’s and were diagnosed as ‘not autistic’ using pre-1994 diagnostic criteria and found that 59% of the children not diagnosed with autism now met diagnostic criteria for autism using current diagnostic criteria for autism. The IQ scores in this group was significantly lower than the IQ scores in the Utah cohort included in the 2012 CDC autism monitoring network. Yet to be studied is the large group of children who pre-1994 were considered perfectly normal but who are considered to be shy, awkward or clumsy or may have been diagnosed with an oppositional defiant disorder.

    If you throw out too wide a net you catch all the wrong sorts of fish.

    Miller JS, Bilder D, Farley M, Coon H, Pinborough-Zimmerman J et al. (2012). Autism Spectrum Disorder Reclassified: A Second Look at the 1980s Utah/UCLA Autism Epidemiologic Study. J Autism Dev Disord. Jun 13. [Epub ahead of print]

  14. Very interesting article. My son was Dx'd with Semantic Pragmatic and never had any support as it was moved to ASD (10yrs ago) we have been repeated told throughout the last 10yrs he has SPD, Dyslexia and Dyspraxia and we are now awaiting testing for APD also. I am currently re-filling in the ADOS Test Forms as he is now 15yrs and needs support into his adulthood. He has great empathy, great humour, a lot of social understanding BUT terrible social language, struggles with same as peers and has a severe anxiety problem.
    We were told by Paeds he had too much empathy to be ASD and too much Comprehension. But due to lack of support he has struggled in school. The Ed Psych has told us he has Social Anxiety and CAMHS told us the same.
    However it would seem he would now qualify as ASD (if the boundaries have changed) and should have had sooooo much more support - which would have made a dramatic difference to his life. All the other ASD children have received OT, SALT, TA's and intense help. My son got nothing.
    But on another note I have run 2 ASD Support Groups over the last 10yrs and my son is very very different to every other child that has attended and been classed as a 'helper' with all the other parents asking him to help their child... It has been quite a lot of pressure on him but he enjoyed it to a point...
    It is sad that SPD is being re-classified SLI or ASD as there are in some cases children who are clearly not ASD losing support and vital services which in turn push them into a blackhole of helplessness, anxiety and recurrent fear of failure.
    My son understands ASD and isists he is not the same but is desperate to help himself and wants to lead a normal life which he will struggle doing as there is no recognition of SPD. Either he takes an ASD Dx which he hates to try and get help from Adult Services or he goes without again.
    Would you accept a Dx that is wrong just to get at least a tiny amount of support?
    This is the dilemma we now face along with thousands of others - but what happens when they tighten the ICD and DSM Criteria and refuse to reinstate SPD? Our children will again be totally left spiraling into the void.

  15. Thanks Tulahberry for your comment. It gets at the nub of the problems that arise when we try to assign people to diagnostic categories when the reality is that there are all kinds of variability, and many people don't fit neatly into the groups.
    For what it's worth, I found that when I followed up children into adulthood, those who had got a diagnosis of ASD got much better help and services than those who were diagnosed as having SLI or semantic-pragmatic disorder. The latter two diagnoses just weren't understood by anyone. An employer can, for instance, get recognition for employing someone with a disability if they have an ASD diagnosis, but not otherwise.
    I know that the people deliberating over DSM5 are very concerned about this issue. They were proposing a category of 'social communication disorder' which would have encompassed those who didn't meet ASD criteria but had pragmatic problems. However, this has been criticised because it's not clear who would provide services for such a group, and it's also difficult to determine the boundaries. I know the debate is still under way and hope that some kind of sensible solution will be found.

    1. Thank you for your reply. I apologise for my spelling mistakes (typing with the help of a 3yr old lends itself to a catalogue of errors!)
      Another issue of clustering SPD into ASD is that it detracts valuable support from people with ASD too. Specialised support is needed for both disorders and extending the boundaires of disorders actually weakens the vital support needed for both.
      I actually would go with a 'Social Communication Disorder' and hope to access support from Speech Therapists, Audiology, OT's, Literacy Support and Linguistic Support. From the adult SPD'ers on a forum I admin, it seems they all suffer from Social Anxiety too so CAMHS would also be a welcome help.
      People with SPD have so much to offer the workplace and feel stigmatised by being pushed into an ASD Dx just to access support and basic 'rights'. They invariably end up un-employed and socially more isolated by label-association rather than the disorder itself.
      My son copes prefectly well in a 'mainstream' rock-climbing club, goes to the Gym by himself, goes to sports clubs independently and is happy socially in a Trampolining Club, yet ask him to approach a member of staff and request a drink and he verbally falls apart... He rehearses the sentences, gets flustered, works out scenarios that may play out and eventually after much deliberating he has a go. If everything works out OK he is really happy but if it goes wrong he is devastated.
      On another note - my 3yr old daughter plays exactly like my son did, has the same sense of humour, enjoys the same games and is similar in nature to my son however she has perfect language ability. She would never be diagnosed Aspergers due to language ability even though she likes traffic lights and 'wet floor' signs - she is perfectly sociable and happily 'neuro-typical' Yet my son would be have been classed as having ASD for having a language disability and enjoying traffic lights and 'wet floor' signs. Sometimes we can "over analyse" normal behaviour and cluster 'signs' into something they are not just to fit square pegs into round holes.
      I really do hope that the DSM reviews its ASD policy and perhaps tightens the boundaries and accepts the SCD category.

  16. I agree with this post. With technology taking off in the fast pace it is going I believe we will see more and more cases being diagnosed with ASD. With the doctors now just really finding out more about this. I also feel that there are alot of conditions that are over diagnosed such as children with ADHD. Now days it seems every time a child is a little on the hyper side a school, parent, daycare ect wants to diagnose them with the condition. Likewise the ADHD statics would be on the rise. Same as ASD being on the rise. More physicans are quicker to give out the diagnoses. Parents want answers and the physicans have to give them some answers. There are alot of children who have ASD and ADHD and many other disorders. I do agree the policy does need to be tighten so they can make sure children have being diagnosed correctly.

  17. Jeffrey Williams1 October 2012 at 13:23

    Good post. I agree totally. Children must be diagonsed properly at a young age. As they grow older, they need all the support and aids they can get to help them live a life that is considered "normal"! I have found various items very useful, including educational pictograms and tasks boards. I´ve shared the link below for anyone else who may be interested.

  18. I am quite prepared to believe that diagnostic substitution is the cause of the increase in autism. However, I have some reservations about the 2008 study. The method of data selection is vulnerable to an analogous phenomenon to regression to the mean. If developmental language disorder (DLD) and autistic spectrum disorder (ASD) can be 'mistaken' for each other, then it is theoretically possible that those who were classified as ASD in the past would have been classified as DLD today. However, unless I have misunderstood, the sampling method employed in the 2008 study would never find such individuals. It could only find individuals who should be moved from DLD to ASD and never those who should be moved from ASD to DLD. We need to know that former exceed the latter in number to find the explanation convincing.

    This strikes me as being a classic example of a data-filter problem. For a recent case of such a filter, see my discussion of Ben Goldacre's Bad Pharma here
    Stephen Senn

    My declaration of interest is here:

  19. Thanks Stephen for your contribution.
    I think there is a difficulty with your argument, which is that the prevalence of autism is lower than that of developmental language disorder - this is the case now, and was even more so in the 1980s, when the prevalence of autism was estimated as around 1 per 1000. DLD prevalence depends on how it is defined, but a common estimate is around 3-5%.
    Thus even if many autistic children were rediagnosed as having language disorder, they would account for a tiny number. Also, in the 1980s it was unusual to get a diagnosis of autism unless you had pretty extreme symptoms - as explained in the paper, people used DSMIII which was far more stringent than contemporary diagnostic criteria. And finally, it is often explicitly stated that autism is a life-long diagnosis - so even if the individual's symptoms abate, they are still regarded as autistic - see for instance here:
    So while I can see that it is logically possible that there would be diagnostic movement in both directions - and you are right that our study would not have detected autism->DLD changes - in practice it does not seem plausible that an equal number of children would be rediagnosed from autism to DLD as in the opposite direction.

  20. Thanks, Dorothy. As I said in my original comment I find the diagnostic substitution theory plausible. However, being a statistician makes me a nit-picker (or to nit-pick that statement, there is a strong association between being a nit-picker and being a statistician!). So to me, there is a world of difference between saying 'the 2008 study suggests that diagnostic substitution could be an explanation' and ' if the following not unreasonable assumptions are made the 2008 study suggests etc.' The point is that it is clear that since the lowest possible estimate such a design can produce is 0%, the expected estimate it must produce must be greater than 0% even if nothing is going on. Thus the design fulfills the classic condition for being biased. In practice it must overestimate the contribution even if other arguments can be adduced to show that the degree of bias should be fairly small. In other words, the design used is not in general sound, although it might work in this case in view of other arguments that can be produced.

    Many interesting points are made in the 2008 paper but I don't think that that is one of them.

  21. Thanks for the post. You should take part in a contest for one of the best blogs on the web. I will recommend this site!
    moving services California

  22. Interesting and useful information that you have provided here on your post.

    autism ireland & gheel