by Mitchel Fleming


You could be forgiven for thinking that the country is experiencing an autism epidemic. Just a few years ago autism was considered to be a rare condition. Now it is mentioned almost daily in the media. We hear about new educational approaches that promise positive outcomes, but also about dire situations in which families of children with autism find themselves, and their inability to get the services they want from the state.

We are not the only country experiencing this apparent epidemic. The situation is the same in the UK, the USA and Japan. For example, a State Report released a few years ago in California announced that there had been a 273% increase in the number of children with autism requesting services over an 11-year period. The reason for this apparent explosion in the reported incidence of autism, and not of other conditions such as cerebral palsy and epilepsy, is not clear. Various explanations have been suggested. Some researchers blame it on pollution and vaccines. For example, the Centers for Disease Control and the Agency for Toxic Substances and Disease Registry in the US responded to local calls in the Brick Township, New Jersey, to investigate an ‘autism cluster’, when it was discovered that the town had 40 cases among 6000 children. Parents in the township had attributed the high incidence of autism to the local water supply which they believed was polluted by a nearby landfill, but federal investigators were unable to identify any pollutants. Similar concerns existed in Granite Bay (east of Sacramento, California), where another autism cluster of 22 cases in a group of 2930 school children was identified. However, no environmental causes were detected. For some years now many have had concerns about a link between the MMR vaccine and the onset of autism, but quality research has consistently failed to detect any causal link, and epidemiological studies have not shown that the higher apparent incidence of autism now can be explained by the introduction of the MMR vaccine. Indeed, the latest study conducted in Japan shows that children who did not receive the MMR vaccine were slightly more likely to develop autism, compared with those who were vaccinated (Honda et al. 2005).

If the above reasons don’t account for the increase in autism, what does? Many researchers think that the apparent increase in autism may be due to an improved awareness among healthcare professionals and the public about autism, a change in the criteria for diagnosing autism and, in particular, the introduction of a new and more inclusive diagnostic category of conditions called Autistic Spectrum Disorders (ASDs), different ways of estimating the prevalence of autism, and a change in diagnostic practices whereby people who were once diagnosed with an intellectual disability are now more likely to be diagnosed with autism.

Firstly, there is no doubt that there is a greater awareness now than before about autism. This has occurred because of increased media coverage, easier access to information such as the Internet, and greater emphasis on the recognition of autism during the training of healthcare professionals.

Secondly, Kanner’s (1943) original criteria for diagnosing autism have been replaced by Wing and Gould’s (1979) somewhat more inclusive ‘Triad of Impairments’, and it is difficult to ascertain what impact this change in diagnostic criteria has had on the estimated prevalence of autism over the past quarter of a century. The triad of impairments includes impairments in socialisation, in verbal and non-verbal communication and restrictive and repetitive patterns of behaviour. For a diagnosis of autism to be made these impairments must be present by the age of three years and must be evident when the person’s overall mental or developmental level has been taken into account. Furthermore, the introduction by Lorna Wing of the concept of an ASD, which includes atypical autism (sometimes referred to as Pervasive Developmental Disorder, Not Otherwise Specified or PDD-NOS), Asperger Syndrome and ‘classical’ autism, has greatly increased the number of people regarded to be affected by the condition. Atypical autism is a very broad diagnosis that can be given if a person displays a qualitative impairment in at least one of the triad of impairments used to diagnose autism at or after the age of three years. Of all the ASDs, it is, however, Asperger Syndrome that has shown the greatest rate of increase in recent years. Most clinicians now consider Asperger Syndrome to be synonymous with High Functioning Autism (HFA) however there is a lack of consensus as to how best to diagnose this condition. Overall, this latitude and confusion with the diagnosis of ASDs contributes, I believe, to considerable variation in estimates of ASDs.

Thirdly, there are different ways that researchers determine the prevalence of a condition. The simplest way is to count the number of people who are diagnosed with autism or an ASD in a region. This is known as the case-census method. The accuracy of this approach depends on how effective clinicians in a region are at diagnosing autism and ASDs. Not surprisingly, studies that have used this approach have found a significant increase in the rates of autism over the years, which may simply indicate that clinicians have got better at recognising the condition. A more reliable way is the ‘tiered’ approach, where all children in a defined population are screened for autism and ASDs using standard methods. There are fewer studies that have used this method, but typically they have obtained much higher prevalence rates than those that used the case census method.

Finally, evidence is emerging in some countries which show that children are now more likely to receive a diagnosis of autism instead of intellectual disability. For example, Croen et al. (2002) noted in California that the increase in the number of children with autism mirrored almost identically the decrease in the number of children diagnosed with an intellectual disability over an eight-year period. This finding may suggest an improvement in the detection of autism, or it may indicate that other factors exist which account for this diagnostic substitution. Undoubtedly, enhanced and more intense service provision for children with autism, coupled with some reports of better outcomes for autistic specific interventions, has had an influence. Autism and intellectual disability are, of course, not mutually exclusive diagnostic conditions. In fact, most people with a diagnosis of autism will also have an intellectual disability. Fitzgerald et al. (2003) found that 81% of people with autism in Ireland had an intellectual disability, and 56% of these had a moderate, severe or profound intellectual disability.

In conclusion, it is very difficult to know whether there genuinely is an increase in the incidence of autism. What is certain is that the estimates for autism of 4.5 per 10,000 of the population, as noted in the first study conducted in the UK in 1966, have risen considerably. Today, based on the tiered approach, the most accurate overall population estimate for the prevalence of autism in children is approximately 31/10,000, and for ASDs it is 60/10,000- The most recent population study of children and adults with autism in Ireland (Fitzgerald et al. (2003) used a case-census method and found an estimate for core autism of 5.2/10,000, and for ASDs of 9.1/10,000- The prevalence rates, however, for ASDs increase to over 15/10,000 for young children aged between 3 and 6 years, which may suggest that there has been an improvement by clinicians in their detection of autism over the past few years in Ireland. Nevertheless, there is still a considerable gap between the number of people expected to have an ASD and the number of cases reported. I suspect with time an increasingly large number of people will be diagnosed with an ASD and eventually there will be a smaller discrepancy between the expected incidence of autism and those reported. It is only, I believe, when the number of reported cases of autism exceeds the expected numbers, that we should start to get alarmed about an autism epidemic. We are not even close to this situation today.