A report in today's Scotsman gives extensive coverage to the claims of Bill Welsh, Chair of Action Against Autism. The article reports that
Official statistics show 825 pupils were diagnosed with autistic spectrum disorder in state secondaries in 2005, compared with 114 in 1999 – an increase of 623 per cent. Over the same period, the number of autistic youngsters in primary schools more than quadrupled, from 415 to 1,736.
According to Bill Welsh
This is not a Scottish problem; it is a developed world problem, because the same thing is happening in America, Canada, Australia and New Zealand. One common factor affecting children throughout the developed world is the vaccination issue.
"The number of young children who have this previously very rare diagnosis is reaching epidemic proportions and it is being ignored by the medical establishment.
Newspaper reports are hard to interpret. But taking this one at its word suggests that there has been an increase in the number of pupils who are diagnosed with autism while they are at secondary school. Did they regress aged 12 or were they autistic before then? In which case the "Hidden Horde" just got younger. Or are they part of the growing number of children who are being diagnosed at primary school and are now entering the secondary system.
The wording for primary children is more ambiguous. What proportion of these children are entering school with an autism diagnosis and how many are getting a diagnosis once they enter the school system? Could the figures be related to an 11% growth in the number of full-time equivalent (FTE) educational psychologists employed in Scotland since 1999. If so what are the autism figures for Glasgow where the number of FTE educational psychologists has fallen?
There are a number of problems using school data on autism to draw conclusions about prevalence. These have been widely discussed by bloggers. Notably, Joseph has discussed this in relation to the data from California and Orac has discussed the study by Paul Shattuck that questioned the usefulness of school administrative data in epidemiological studies of autism.
The data fluctuates wildly over school districts. In Scotland's case
Aberdeenshire, for instance, recorded a 533 per cent increase among primary pupils over the six-year period, while in the Borders the rise was a relatively small 137 per cent.
Variation and growth can be driven by a number of factors. School districts with poor provision and awareness do not go out of their way to identify autistic pupils. When I began teaching in 1983 my county had no officially diagnosed autistic pupils. But we could identify five in our special school of less than 40 pupil. Districts that do provide services attract parents. The recent drive for inclusion gives schools a financial incentive to identify pupils with special needs, including autistic ones. There is also the problem with diagnostic substitution discussed in Shattuck's paper. If autism is going up are other diagnoses going down? And while school data shows increases it still seriously under estimates numbers compared with the most recent epidemiological studies.
Given all these qualifications what are we to make of the figures? The report in the Scotsman links to an earlier story.
A study published today by Health Protection Scotland reveals that 25 postcode districts in Scotland have more than 20 per cent of nursery school children at risk of catching measles because they have not had the MMR jab.
So all we can say is that in Scotland MMR has gone down and autism has gone up amongst primary age children. It is also rising among teenagers who had their MMR a long time ago. The figures fail to demonstrate a link between MMR and autism and suggest to me that there is no link. But, as reported, they prove nothing. They certainly do not provide any basis for the claims of Bill Welsh and Action Against Autism. I will return to this when the actual wriiten answer to the parliamentary question is published.