Government not intending its autism over-diagnosis claim

A report from  the Australian National Audit Office (ANAO) said that in February 2014 the Department of Social Security (formerly FaHCSIA) funded $50,000 worth of research to establish "the extent of [ASD] over-diagnosis nationally". 

A Deputy Secretary of DSS wrote back that

this wording is not intended to presuppose the findings of the Autism CRC report.

The Department apparently accepts (does not deny) their wording clearly pre-supposes the existence of autism over-diagnosis. However, the actual intent is now unknown. It is hard to imagine how a contract could be signed without anyone noticing this (unintended?) presupposition. A4 will be disappointed in the Autism CRC if it undertook research based on the questionable premise that parents/families and/or professionals are fabricating diagnoses of a medical condition. Credible research would not presume over-diagnosis exists; it would first need to establish that over-diagnosis really exists.

A4 asked a number of questions (see below) on 6th April 2016. The Government sent a letter (15 April, see below) but didn't directly address the questions asked. The Government's response pre-released some information; it says:

This report does provide a partial answer to your questions. According to the report, 17 percent of diagnosticians surveyed have, on at least one occasion, diagnosed Autism when the person did not meet the full criteria. However, 88 percent of this group indicated that they had given the diagnosis because they thought the individual did have Autism and the assessment did not represent the individual's usual presentation. 

It is not clear from this whether they are talking about DSM-IV or DSM-5 criteria for "Autism". In any case, there are relatively few questionable diagnoses. And the rate of over-diagnosis, the "false-positive rate", identified is a fraction of cases from 12% of 17% ... that is a fraction of cases diagnosed by 2.04% of diagnosing clinicians. These data indicate that over-diagnosis is not a significant contributor to the growing rates of autism diagnoses. 

The Department's letter says:

the Department does not have evidence to suggest [increasing autism diagnosis rates] is due to over-diagnosis. 

The letter goes on to say

It is plausible that increased diagnosis rates stem from a greater understanding of Autism, as well as the development of improved diagnostic tools leading to earlier diagnosis.

This latter statement is false. The increased numbers are far too large for these reasons to account for all the increased ASD diagnoses that have been observed. In particular, the increase in numbers due to earlier diagnosis of autistic children is minor.

The Government should be much more concerned that fewer than 30% of children diagnosed as autistic by age 15 years were diagnosed in time (under 7 year of age) to access government early intervention programs.


Note: the report has been published and can be downloaded from http://a4.org.au/sites/default/files/Aut...

see also:


From: Bob Buckley (A4 Convenor) [mailto:convenor@a4.org.au]
Sent: Wednesday, 6 April 2016 6:39 AM
To: PRATT, Finn
Cc: GRAHAM, Carolyn
Subject: over-diagnosis of autism in Australia

Mr Finn Pratt
Secretary, Department of Social Security

Dear Mr Pratt

subject: over-diagnosis of autism in Australia

Autism Aspergers Advocacy Australia, known as A4, is a national grass-roots organisation for systemic advocacy for autism spectrum disorder (ASD). A4 is a member of Disability Australia and is associated with the Australian Federation of Disability Organisations (AFDO).

A recent report about early intervention services funded from DSS for children with disability, Australian National Audit Office (Mar 2016) Early Intervention Services for Children with Disability, ANAO Report No. 24 2015-16 (link here), says in Paragraph 3.23:

... In February 2014, DSS allocated $50 000 to investigate autism diagnostic practices in Australia, with the aim of establishing the extent of over-diagnosis nationally. The Department advised Government that this would assist with developing a more consistent standard of autism diagnosis which may lead to a reduction of expenditure for HCWA with fewer new children becoming eligible. The research was incomplete in February 2016.

We understand from this that DSS says that autism and related disorders are now being over-diagnosed in Australia. Apparently, DSS has no question about whether over-diagnosis is happening; the question is “to what extent?”.

In 2015, A4 obtained Centrelink data from DSS showing 2.06% of Australia children aged 5-15 years (or 2.25% of children aged 10-14 years) were formally diagnosed autistic and received Carer Allowance (child).

There is strong evidence that historically, autism was chronically under-diagnosed.

Data from a range of sources, including Centrelink (via FaHCSIA and DSS) and the Australian Bureau of Statistic (seehttp://a4.org.au/prevalence2015), show the number of children diagnosed with autism increased rapidly since the 1990s. Similar increases are observed overseas.

In the past there was chronic under-diagnosis of autism and a substantial part of the increase was due to increased awareness and improved diagnostic practice. The most reputable research journals question whether these factor account for all of the growth in autism numbers: seehttp://www.nature.com/news/the-prevalence-puzzle-autism-counts-1.9280

Note: a small blip in diagnosis rates in 2013 can be seen when the diagnostic criteria for autism changed from those in the DSM-IV to the recent DSM-5.

We are very interested to know how over-diagnosis of autism occurs.

Of the children born in 2000, 1,570 were diagnosed with Autistic or Asperger's Disorder in 2006 – when they were 6 years old – and 5,569 were diagnosed by age 15 years (in 2015). Fewer than 30% of children diagnosed as autistic would have been eligible to access early intervention. The situation is similar today: there were 4,657 children born in 2007 (aged 6 years in2013) who were diagnosed with Autistic Disorder, Asperger's Disorder or ASD and this rose to 6,413 children who were diagnosed in 2015 (aged 8 years), 38% growth in the number diagnosed in just 2 years. Unless over-diagnosis is close to 70% of those diagnosed at age 15 years and/or is sometimes before 6 years of age, most autistic children miss their opportunity for early intervention. Under-diagnosis of ASD is more likely in children under 7 years of age.

The ASD community has no interest in, gets no benefit from, over-diagnosis. The ASD community wants accurate/reliable and timely diagnoses for autistic children so they can access the early intervention services they need. A4 does not want over-diagnosis of ASD; A4 supports/encourages best practice in diagnosis generally and autism diagnosis in particular.

Please answer the following questions.

  1. Is there any evidence that the introduction of the Helping Children with Autism (HCWA) package caused an increase in ASD-related diagnoses, or even over-diagnosis? If so, what is that evidence?
  2. What causes “over-diagnosis nationally” of autism? What is the evidence of over-diagnosis of autism?
  3. When (what date) did ASD diagnoses in Australia tip over from under-diagnosis to "over-diagnosis nationally"?
  4. Are adults being over-diagnosed with autism too?
  5. Is over-diagnosis of autism in children equally likely at all ages?
  6. Who is doing the research described in the quote from the ANAO Report? When will the research be completed?
  7. The revised diagnostic criteria for ASD in the DSM-5 was intended to improve the reliability of ASD diagnoses. Did the revised criteria help create "a more consistent standard of autism diagnosis" in Australia?

We look forward to your response to our questions.

Please feel free to contact me via email or telephone 04xx xxx xxx you would like any further information.

--
Bob Buckley
Convenor, Autism Aspergers Advocacy Australia (A4)
website: http://a4.org.au/