Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 May 19; 334(7602): 1027.
PMCID: PMC1871759

Rise in prevalence of autism in children continues to baffle researchers

Despite three recent studies showing that the number of diagnoses of autism spectrum disorders has almost doubled in the last seven years to around 1% of children, researchers say that it is still impossible to say whether the rise is due to a genuine increase in the number of new cases or to other factors.

At an Open University conference on autism research in Milton Keynes, Tony Charman, professor of neurodevelopmental disorders at the Institute of Child Health in London, cited his research describing the rise (Lancet 2006;368:210-5).

In the study Professor Charman and colleagues from Japan, Australia, and the United Kingdom estimated that the prevalence of autism in 57 000 children in the South Thames region of London was 116 cases per 10 000 children. Two other studies, conducted around the same time, put prevalence at 117 cases and 90 cases per 10 000 children (Journal of Child Psychology and Psychiatry 2005;46:572-9; Office for National Statistics, Mental Health of Children and Young People in Great Britain, London: ONS, 2005). All three studies showed a rise from the prevalence found in studies from 2000, which put the figure at around 60 cases per 10 000.

“Is it becoming more common? There is no definitive evidence that it is,” said Professor Charman.

He continued: “What might explain the apparent rise? It could be a widening recognition, a broadening definition of the diagnosis, more thorough studies, or the inclusion of children with average as well as below average intelligence, but other explanations cannot be ruled out, including a true rise in incidence.”

Comparing studies was fraught with difficulty, he added, because of methological differences between them.

“In our own study, our narrowest definition and our broadest definition give you four and a half times different rates in terms of prevalence (25 per 10 000 for autism and 116 per 10 000 for total autistic spectrum disorders), which is remarkably high within one study,” he said.

“So it is not that we are deliberately sidestepping the issue of whether there has been a true increase—it is a very difficult question to answer. If there were studies that demonstrated that there had been a true increase we would be happy to say it, but there aren't.”

Professor Charman added that prevalence might continue to change as children with the behavioural phenotype of autism but no biological or genetic markers for the disorder grow older and no longer have behavioural traits associated with the disorder.

It might be that such people will no longer be classed as having the disorder, which could lead to a fall in prevalence, or that cases defined through biological causes—such as genetic variations linked to autism—will keep the prevalence high.

Issues of prevalence and definitions of autism aside, parents' and individuals' own reports of the difficulties affecting them are real. He said, “Whatever the science ends up showing us, the clinical issues are there for all of us to see.”

Articles from The BMJ are provided here courtesy of BMJ Publishing Group