Professor Tony Charman's quarterly research brief: prevalence of autism

12 replies (Jump to last post)


Is autism on the rise? Professor Tony Charman has written a report summarising the research exploring the prevalence of autism over the years, which you can read in the evidence base section here.

What do you think? How can we get accurate figures for the prevalence of autism? Are there the resources to meet the needs of increasing numbers of people diagnosed as being on the autism spectrum? We'd love to hear your thoughts...

Edited on February 23, 2017 - 9:20am

February 10, 2012 - 5:30pm

A new study of prevalence of autism in adults published a few weeks ago by the NHS suggested that it is 1.1%.  So maybe autism isn't on the increase. 

But on the other hand, we're being told to expect 20% more children in primary schools in the next decade, and that will presumably mean a corresponding rise of young people with autism.

So we still need to be preparing for more people with autism needing services and support.

February 15, 2012 - 1:47pm

I have worked with ASC pupils in minstream ed. for 10 years, from those who get along fine with only low levels of support to those who need high levels - 1-1, group work, specialist intervention etc. At present in a school population of approx. 1000 we have 25 pupils with an official diagnosis, at least 6 on the Autism Pathway and the same again of those who we think may be Autistic if parents would consent to referrals being made to the relevant agencies to do initial assessments. I don't think there is any greater prevalence in society of Autism/Asperger's but more a greater awareness and diagnosis of children who previously may have been treated as 'naughty', 'awkward, or 'a bit odd'. We do need to be more aware and diagnose as early as possible so interventions and support can be put in place for child and family, as well education. As more older people are being diagnosed this needs to be a priority. My son was diagnosed as Asperger's at 31 after years of my asking paediatricians, educationalists, doctors what was wrong and being told he was very intelligent but 'just a bit odd'.

March 04, 2012 - 12:11pm

Hi from the Czech Republic to everybody. Just to present you with the data from my country, resent research studies of my team also suggest a rapid increase in the prevalence of ASD. I totally aggree with the article of Professor Tony Charman concerning the numbers and the reasons for the prevalence. The sad thing is that we really do not know whether the raise is real or the diagnostic process has improved. Me personally, I believe both options are true. What we know for sure is, that the number of children with ASD in our schools is high and we really have to start worrying and providing them with a proper care and support.  

August 24, 2012 - 9:09am

The change in numbers is social construction - there is no 'real' rise in autism - just a change in definition, culture and language about the issue.

In the past one was often poorly diagnosed with/as something else (speaking from personal experience).

Keith Ingram

August 24, 2012 - 12:23pm

Is there a possible link to the recent Icelandic  research published in Nature.
This revealed a direct correlation between the number of mutations or slight alterations to the DNA, of the child and the age of their father.
The results indicate that a father aged 20 passes, on average, approximately 25 mutations, while a 40-year-old father passes on about 65. The study suggests that for every year a man delays fatherhood, they risk passing two more mutations on to their child.
What this means in terms of the impact on the health of the child is unclear. But it does back studies that also show fathers are responsible for mutations and that these mutations increase with age.
And, for the first time, these results have been quantified and they show that 97% of all mutations passed on to children are from older fathers.
"No other factor is involved which for those of us working in the field is very surprising," said Dr Stefansson.

Since the average parent age has been gently increasing over past decades this MIGHT have an effect.  Lots more work to do and I guess if this was a significant factor countries with lower average parent ages would have lower prevalence rates. I have no idea if that is true.

August 24, 2012 - 12:32pm

Not that genetic mutation should always be seen as pathological - but although the age of parenthood has increased, then if this were the case, we would expect similar increases in other 'genetic mutations' (and it is debatable if autism should be described as such rather than natural variation of geno/phenotype) - which I don't think is the case.  Also - it is not worth comparing prevalence rates in different countries as to the vast differences in how these rates are collected in terms of definition and reliability of statistics.  Emile Durkheim had similar issues with his study of Suicide back in 1897...

Keith Ingram

August 24, 2012 - 12:56pm

Oh I agree completely, it is interesting but not surprising that some news groups tagged autism and ignored other genetic mutations.  And does autism fall into that group. It seems likely we will always have, in the words of the late Douglas Adams "Clearly defined areas of doubt and uncertainty" and ultimately does it matter that we don't accurately know.  As a professional who manages services I don't need to know exactly how many adults or children are out there needing support.  They are out there, they do need support so as long as the numbers seeking service broadly match expectation then I should simply get on with it. 

Even local initiatives can seem to have an effect.  We have anecdotally had an increase in younger children in our area being picked up and referred to specialist education support.  This seemed to be a result of awareness training delivered to professionals from early years settings such as nurseries.  But whether this is real or coincidental is impossible to say without a much more rigorous approach. 

August 24, 2012 - 3:06pm

Hi Keith,

The article unfortunately isnt open access but you can view the abstract here

You can view the whole article if you are a subscriber or if you make a one off payment.


November 01, 2012 - 11:26am

At the risk of being completely and utterly wrong I have been thinking long and hard about prevalence.

Diagnosis is on the rise for whatever reason;we do have an aging population in the wealthy world in particular which would obviously link closely with the general rise in genetic mutation and so a rise in all genetically linked conditions but awareness is also growing and contributing to the rise.

Across the world we are constantly reassessing the prevalence levels - 1:100 was a shock when it was first announced, now we're looking at 1:88 in the States and some studies have suggested as high as 1:36 (Korea) which, as a primary teacher originally, gives me the frame of reference that says - almost every class of children in the country will include a child on the spectrum.

As we understand more about autism the number of people with more subtle presentations will clearly rise. We are now seeing girls diagnosed more frequently as this happens too so we are seeing a whole section of the population previously largely excluded enter the statistics. The changes in the DSM V may exclude some people who would have recieved diagnosis under DSM IV but  it will also see many of those with diagnoses such as PDD-NOS brought into the statistics, we will have to see how that pans out.

How far will the rise take us?

Will our understanding and diagnostic competence eventually see us looking at an even greater percentage of the population being labelled as ASC? Could we end up seeing the recognition of 2, 5 or even 10% of the population being on the spectrum?

Will the figures reach such significance that we have to rethink our approach entirely and look at autism, not as an unusual and challenging disorder, but as the norm for a percentage of the population with only some of those people presenting as having an actual disorder?

In other words, are we looking at a a condition that many people live with without significant problems but which creates difficulties for some which means they need intervention?

I am not a proponent of neurodiversity and autism being the next step in evolution - nothing which creates such difficulties for those living with it could be seen in that way in my view - but I do wonder whether, as we explore neurology, we will see different brain structures as part of the typical pattern of human development and begin to acknowledge that all people on the planet can be seen simply as belonging to a neurological type with each neurological presentation throwing up its own needs and difficulties as well as its own strengths, each essential for human development in its own way but creating problems for individuals within each group.

Will we then perhaps start to look at supporting neurological types in education, etc, so that the focus for learning and teaching for each group can be adapted to the specific learning style and needs of that group, instead of looking at all groups occuring outside the majority presentation as needing special measures to be taken on an individual basis?

These are just ponderings, I have no evidence and no reason to suggest, at the moment, that this may be the case, but I do find the idea intriguing.


November 01, 2012 - 12:00pm

Hello Zemanski,

There is already a term for the 10% idea - the 'broader autistic phenotype' - if one measures anything across a population (whether a bell curve distribution or not) one will find the line drawn is a somewhat arbitrary one - yet it does suggest a line for a recognised 'condition'.  It is not a 'disorder' if it also confers advantages, or if one (like I do) do not compare people to the norm as some kind of ideal.  I don't like the term 'condition' either - but that is another matter.  We do have significant challenges to live with in society - often due to the lack of understanding and discrimination we face interacting with our dispositions.  This does not mean we need 'interventions' to make us more 'normal' as these tend to make everything worse and devalues the dispositions and subjectivities we already have.  What people may need is social support in some way (yet many of us get nothing approaching something that works for us), and what we need is a far better understanding of what it is to be autistic.  Although this is impossible for the non-autistic, ideologies need to move far beyond the view of abnormality=pathology.  Neurodiversity is not the next step in evolution in the slightest (most likely), but nor is it a dysfunction to be eradicated - but a natural part of diversity.  Typing people is social construction - and not a reality.