A new paper from Eric Fombonne is in electronic print at the journal Pediatric Research. It will apparently be published in the paper version of the journal some time after April.
The title is ‘Epidemiology of pervasive developmental disorders’ and as the name suggests, Fombonne looks at all the available quality epidemiology he can find relating to PDD ’s.
This article reviews the results of 43 studies published since 1966 that provided estimates for the prevalence of Pervasive Developmental Disorders, including Autistic Disorder, Asperger Disorder, Pervasive Developmental Disorder Not Otherwise Specified, and Childhood Disintegrative Disorder.
Combining all these categories together Fombonne presents a prevalence of 60-70/10,000.
For autistic disorder, Fombonne says:
The correlation between prevalence and year of publication was statistically significant and studies with prevalence over 7/10,000 were all published since 1987. These findings point towards an increase in prevalence estimates in the last 15-20 years.
For PDD -NOS, Fombonne explains that it is next to impossible to get accurate prevalence rates as:
This group has been much less studied in previous epidemiological studies…
Again, for Aspergers, Fombonne says that AS specific epidemiological studies are sparse but, in something of a surprise:
By contrast, other recent autism surveys have consistently identified smaller numbers of children with AS than those with autism within the same survey. In 9 out of 10 such surveys, the ratio of autism to AS prevalence in each survey was above unity, suggesting that the prevalence of AS was consistently lower than that for autism. How much lower is difficult to establish from existing data, but a ratio of 3 or 4 to 1 would appear an acceptable, albeit conservative, conclusion based on this limited available evidence. This translates into a prevalence proportion for AS which would be ? to ¼ that of autism. We therefore used for subsequent calculations an estimate of 6/10,000 for AS, recognizing the strong limitations of available data on AS.
Lastly, for CDD:
Eight studies provided data on childhood disintegrative disorder (CDD). Prevalence estimates ranged from 0 to 9.2/100,000. The pooled estimate based on eight identified cases and a total surveyed population of 406,660 children, was 2.0/100,000. The upper-bound limit of the associated confidence interval (4.0/100,000) indicates that CDD is a very rare condition, with about 1 case to occur for every 103 cases of autistic disorder.
Fombonne then tackles the question everyone wants an answer to – is there an autism epidemic?
In order to answer this accurately, he explains that there has to be tight control over incidence estimates (the number of new cases occurring in a population over a period of time) and prevalence (the proportion of individuals in a population who suffer from a defined disorder). Failure to control these gives false results. Bearing this in mind, Fombonne goes through the five approaches taken so far to try and determine if theres an autism epidemic or not.
1) Referral Statistics. Trends in time for referral statistics are not reliable. They fail to control for things such as referral patterns, availability of services, heightened public awareness, decreasing age at diagnosis and changes over time in diagnostic concepts and practices. An example of the issues from referral statistics is:
Strong evidence of “diagnostic switching” was produced in California and in all US states indicating that a relatively high proportion of children previously diagnosed as having mental retardation were now identified as having a PDD diagnosis. Decreased age at diagnosis has also been shown to contribute to the rising numbers of children diagnosed with PDD. In the UK, Jick and Kaye (62) have shown that the incidence of specific developmental disorders (including language disorders) decreased by about the same amount that the incidence of diagnoses of autism increased in boys born from 1990-1997. A more recent UK study has shown that up to 66% of adults previously diagnosed with developmental language disorders would meet diagnostic criteria for a broad definition of PDD.
2) Comparison of cross-sectional epidemiological surveys If I’m understanding his point here (and please correct me if I’m not) Fombonne is saying that too many epidemiological studies are uniquely designed – not enough attempt to replicate a previous study – and hence:
The most convincing evidence that method factors could account for most of the variability in published prevalence estimates comes from a direct comparison of 8 recent surveys conducted in the UK and the USA. In each country, 4 surveys were conducted around the same year and with similar age groups. As there is no reason to expect huge between-area differences in prevalence, prevalence estimates should therefore be comparable within each country. However, there was a six-fold variation in prevalence for UK surveys, and a fourteen-fold variation in US figures. In each set of studies, high estimates derived from surveys where intensive population-based screening techniques were employed whereas lower prevalence proportions were obtained from studies relying on passive administrative methods for case finding. Since no passage of time was involved, the magnitude of these gradients in prevalence can only be attributed to differences in case identification methods across surveys.
3) Repeat surveys in defined geographical areas So this is the opposite of the above – these are studies where they are being replicated as closely as is possible. However, the issue here is that there are simply not enough of these studies to form a definite conclusion. However, it may be worth noting that in the two studies Fombonne highlighted as being carried out in exactly the same way in exactly the same place to exactly the same age cohort – but just at two different times one showed no increase in prevalence whilst the other showed no increase at 4 sites and an increase at 2 sites.
4) Successive birth cohorts This means in very large surveys with a wide age range, if the proportion of people who have autism rises this could be a rise in incidence and therefore a good hint that there is an epidemic. I say could as other possible causes need to be ruled out first.
...two large French surveys [used this method]. The surveys included birth cohorts from 1972 to 1985…, and, pooling the data of both surveys, age-specific prevalence showed no upward trend.
A US survey did show an upward trend but:
...the increase was not specific to autism. These analyses also showed a marked period effect that identified the early 1990s as the period where the prevalence estimates started to go up in all ages and birth cohorts, coinciding closely with the inclusion of PDDs in the federal Individual with Disabilities Educational Act (IDEA) funding and reporting mechanism in the US.
5) Incidence studies The few incidence studies did show incidence trends rising over short periods of time. As noted in point 4) above, this could be attributed to an autism epidemic. However –
...none of these studies investigations could determine the impact of changes over time in diagnostic criteria, improved awareness and service availability on the upward trend.
Contrary to what people who want there to be an autism epidemic, these are non trivial reasons. It stands to reason that if (for example) Birmingham, UK - the countrys second city, goes from having zero service availability and no means of diagnosis in 1960 to having numerous types of service availability both publicly and privately funded and a lot of means of diagnosis in 2000 there will be a lot more autistic people in Birmingham. A hell of a lot. When we then consider that the diagnosis criteria has widened massively than we go from a hell of a lot more autistic people to a whole hell of a lot. If we also consider that people who used to carry one kind of diagnosis are now being swapped to autism then we go from a whole hell of a lot to a descriptive term beyond my ability. This isn’t even science – its basic common sense. The only issue is – ‘a whole hell of a lot’ is not a very accurate measurement.
Fombonne closes by saying that – based on the available data – we still cannot really say one way or the other if there has been an autism epidemic. Remember when you read the quote below that its incidence that gives us an epidemic.
Current evidence does not strongly support the hypothesis of a secular increase in the incidence of autism but power to detect time trends is seriously limited in existing datasets. Whilst it is clear that prevalence estimates have gone up over time, this increase most likely represents changes in the concepts, definitions, service availability and awareness of autistic-spectrum disorders in both the lay and professional public. To assess whether or not the incidence has increased, method factors that account for an important proportion of the variability in prevalence must be tightly controlled. The possibility that a true change in the underlying incidence has contributed to higher prevalence figures remains, however, to be adequately tested.