Ayoub/Yazbak/Fombonne

_Jonathon has posted an excellent summation which I entirely agree with on the Ayoub/Yazbak/Fombonne ‘affair’. I’ll include it here and shut off comments, including a link to Jonathon’s post for you to see references and leave comments._

_My one comment is to say that I think that Fombonne has a case to answer. I also hope that he will. I further think that the critical errors Jonathon highlights show clearly that the more outlandish claims that Ayoub and Yazbak have demonstrated correlation between MMR and autism are unfounded. They should’ve just stuck to the genuine and good quality criticism of Fombonne._

_The floor is now Jonathon’s: _

A Review of the Critiques of Fombonne et al. (2006)
Abstract

The critiques of Fombonne et al. (2006) do point some relevant problems with that study. However, they also use some considerably problematic arguments themselves.

Introduction

Fombonne et al (2006) attempts to correlate data between an increase in the number of cases of the Pervasive Developmental Disorders and vaccine use in school children residing in Montreal, Canada. That research failed to reveal any such correlation. In addition the researchers found a prevalence rate of the PDD’s which match what is seen in the US (Bertrand et al., 2001), the UK (Chakrabarti & Fombonne, 2001; and Chakrabarti & Fombonne, 2005), the Faroe Islands in Northern Europe (Ellefsen et al., 2006), and matches data on Autistic Disorder in Japan (Honda et al., 2005)

Recently, two web based articles have been released and promoted based out of the National Autistic Association website. These articles were written by Dr. David Ayoub & Monica Ruscitti for one letter and Dr. Edward Yazbak for the other. These authors raise important and damaging points which call into question the validity of Fombonne et al. (2006). I note however, that not all the points raised in these critiques have equal merit; in fact some of the points raised are most notable for their lack of merit. The points raised in the critiques are discussed below.

Review of Ayoub & Ruscitti

The authors point out that Fombonne et al. (2006) only uses data from only one of Montreal’s five school districts. This district is specifically an English speaking section. The authors accurately point out that a selection bias may have occurred here. In addition the authors note that Fombonne et al. (2006) claimed they could not gain access to the other school districts’ data. The authors then claimed they gained access to this data and make specific claims about a notably higher rate of autism in the school district Fombonne et al. used. I note that these data and analysis methods are absent.

The authors note that English is only the third most common language in Quebec, the largest group being French and the second largest being assorted foreign languages. The authors note that access is restricted to the English speaking school district by a law designed to help protect the French language. However, the authors simultaneously make the claim that this English speaking school is more inclusive than the others. This claim seems contradictory. It is possible that while the English speaking district is more exclusive in general it is in fact more inclusive towards students with disabilities. However this point is not explained.

The authors correctly criticize Fombone et al. (2006) for describing thimerosal laden shots as “nil”. The authors go on to describe vaccines in Canada that continue to use thimerosal. However, the authors fail to note that none of these vaccines are actually required any longer and for that reason exposure should not be comparable.

The authors mention that fourth graders largely participated in a voluntary vaccination program for Hepatitis B, which contains thimerosal. However, the DSM-IV mandates that Autistic Disorder first appear in a child by age 3, even if it is diagnosed later. Fourth graders do not develop autism. Also, and rather remarkably the authors cite Roy et al. (1999) which is concerned with the health and safety including vaccination based on high risk behaviors of street youth in Montreal. The median age of that study was 19.5.

The authors criticize Fombonne in the following terms:

“he ignored the fact that autism rates increased following a doubling of the MMR exposure after 1996 when a second MMR shot was added to the immunization schedule and chose to emphasize that a rise in PDD rates coincided with a decline in MMR coverage rates. Obviously the increased amount of administered viral load to the population was far greater influenced by a doubling of shots administered than by a marginal drop in immunization coverage rates. He likewise ignored the potential impact of mass measles immunization campaigns in Quebec that delivered a second dose of measles to a large number of infants and children throughout 1996. (11) The subsequent rise in PDD shortly after that campaign is clearly depicted in their figures and would lead us to believe this observation supports an association between PDD and MMR exposure.”

The above assumes that the drop in thimerosal was not only made up, but actually exceeded by the increase in MMR dosage. For this explanation to work, one must simultaneously assume that both thimerosal and the MMR can cause autism. A more parsimonious explanation would be that the rate of autism would have risen regardless, which is certainly what we see in other locations including the no longer mandatory thimerosal exposure in California.

The authors also state:

“Numerous reports of higher PDD rates among immigrants have been reported in Canada and other industrialized countries.”

This claim is demonstrated no where in any research. It even seems to contradict some actual research (Kamer et al., 2004).

The authors go on to assert:

“Finally the paper’s observation about rising PDD rates seems to contradict Dr. Fombonne’s well-known contention of the lack of evidence of an autism epidemic. In an Inserm interview, Dr. Fombonne said, “to declare an epidemic, or sensible increase of the prevalence, it would take incidence studies, always the same, year after year, but this data is not available in any country.” (12) The database we obtained from the MEQ represents the type of dataset Fombonne stated was required to detect true increases in PDD. According to one Montreal-based autism organization, data from the MEQ revealed an increase in annual PDD cases in Quebec from 410 (1990-1991) to 4,483 (2005-2006), a nearly 1,000% increase over 15 years. (13) This is staggering and is strong evidence of a real rise in neurodevelopmental illnesses that cannot possibly be solely genetic in nature but supports an environmental etiology.”

The above statement is presents a false problem. An increase in this data set may not be attributable to real change in the actual number of cases. The system could be open to a lack of control for the six threats to statistical validity. This is a well known problem in other administrative data sets in the world of autism.

Review of Yazbak

The author submitted an letter to the editor detailing concerns about uptake of the MMR rate in Montreal and the increase in the PDDs. The editor forwarded this comment from Dr. Fombonne:

“This person is known to pursue the MMR-autism agenda at all
costs in order to ‘demonstrate’ a link he strongly believes in. The only way ahead is to encourage him to do independent research. All controlled epidemiological research thus far has concluded to the absence of such a link.”

the editor of the journal had this to say:

“As a note, I believe the evidence of no link between MMR and Autism is sufficient. It’s not worth publishing more on this subject. We will not be publishing this exchange of correspondence.”

I disagree with Dr. Fombonne’s approach to this situation. It is not relevant that Dr. Yazbak believes in a vaccine etiology of autism. I also disagree with his failing to address Dr. Yazabak’s criticisms which are specific and serious.

I also heartily disagree with the editor’s refusal to publish Dr. Yazbak’ criticism. While it is certainly his job to ensure that all matters within the journal merit inclusion under a serious scientific aegis, Dr. Yazbak’s criticisms are specific and fall into known categories of scientific criticism. Whether or not he believes it has no beating on his responsibility to publish scientific criticism.

However, given the title Dr. Yazbak selected “Far-Fetched”, perhaps the editor had ground to refuse this letter or at least demand that it be re-titled. Pediatrics is a serious academic journal of high standard. In most such journals while a given level of sarcasm and dismissiveness is permitted, this seems to have crossed the line of acceptability.

Dr. Yazbak goes on to assert:

“When he was in France, Dr. Fombonne was a well known psychiatrist who published articles on psychiatric topics. He was still a psychiatrist when he moved to England …until Andrew Wakefield suggested that the link between MMR vaccination and autism should be further investigated and suddenly …Dr. Fombonne became a “psychiatrist / epidemiologist” and a consultant to the UK medical authorities on MMR vaccination and autism”

The Wakefield controversy began in 1998. So:

Fombonne, E. (1996). Is the prevalence of autism increasing? Journal
of Autism and Developmental Disorders, 6, 673–676.

Fombonne, E. (1997). The prevalence of autism and other pervasive
developmental disorders in the UK. Autism, 1, 227–229.

QED.

The author further asserts that:

“It is obviously customary to disclose sources of funding, Disclosing sources of “Non-Funding” on the other hand is unusual. In any case, it is nice to know that Dr. Fombonne’s research was never funded by the “Industry”.”

This is easily explainable in that Dr. Fombonne’s work has been informally criticized as being supported by the pharmaceutical industry, possibly with the intent that he would manipulate the data in favor of finding no association. This statement may have been given to help put such non-sense to rest.

Conclusion

It is important to remember that the weak criticisms in the critiques above do not remove the genuine and quality criticisms. The authors do point some genuine problems. The failure of both editor and Dr. Fombonne to make adequate response is also disagreeable.

By the same token some of the criticisms are remarkable for their lack of relevancy or factual basis. There are problems with these critiques that have a real potential to mislead others. It is to be hoped that the NAA and the authors will take steps to amend this, leaving their better criticisms intact.

See references and leave comments on Jonathon’s site.