A recent discussion online was held between Mrs Isabella Thomas and Jan Goodenough, both parents of youngsters with an Autism Spectrum Disorder. Mrs. Thomas is parent of two boys who took part in the frequently cited study of 12 children with ASD and gastrointestinal disease which was originally reported in the Lancet medical journal and then subsequently withdrawn (Wakefield et al. 1998). The discussion between the two parents referred to the work of Professor Simon Baron-Cohen, and reference was made to an interview that Jan Goodenough carried out with Professor Baron-Cohen for Mensa magazine in August 2010. At the time of publication the article came to the attention of an ex-colleague of mine, Dr Tilly Storr, a member of Mensa. Our response, which is clearly relevant to the recent discussions, was not submitted for publication at that time but has since been sent to Mensa. The response highlights several important issues: the likely contribution of environmental factors to the observed rise in autism diagnosis, the well reported association between autism and gastrointestinal (GI) disease, the possible benefit of dietary interventions in autism and related conditions, and the public right of access to factual information.
In the August 2010 issue of Mensa magazine, Professor Simon Baron-Cohen is asked by Jan Goodenough about the genetics of autism: specifically, he is asked, if autism is a genetic disorder then, ‘....how can there be identical twins where one has autism and the other does not?’ In his response Professor Baron-Cohen refers to there being 60% concordance (agreement regarding presence/absence of ASD) in identical twins vs. 5% in non-identical twins – an estimate which is based on three major twin studies: Ritvo et al., 1985; Steffenburg et al., 1989 and Bailey et al., 1995). The suggestion is that autism is not explained entirely by genetics; indeed Professor Baron-Cohen acknowledges this by saying in the Mensa interview that the findings mean‘...there are environmental factors, and probably interactions between genes and those environmental factors...’ involved in autism. The studies Baron-Cohen refers to are on twins born before the 1990s, and pre-date the recent rise in autism diagnosis and actually report a wider range of identical-twin concordance than that cited by Baron-Cohen: from 60-90% compared to as low as 0% in non-identical twins.
However, whilst Professor Baron-Cohen cited these older studies, much lower concordance in identical twin pairs, and higher concordance in non-identical pairs had been reported at the time of writing (Croen et al., 2002; Constantino and Todd, 2003). Croen and colleagues presented a population-based California study at the International Meeting for Autism Research (IMFAR, 2002) which concluded that ‘..heritability estimates from previous studies may have overestimated the role of genetics and underestimated the role of environmental factors in the etiology of autism.’
It surely cannot be the case that Professor Baron-Cohen is unaware of these later studies; at the time of the Mensa interview he was the chair of INSAR – the organisation responsible for the annual IMFAR conferences in which the Croen et al. study was reported. As such, one would expect him to be aware of IMFAR archives and Croen et al’s work. It is surprising and curious, therefore, that he fails to refer to any of these studies in the Mensa interview.
Why might the studies provide crucial information? Perhaps Richard Lathe has the reason. In his book ‘Autism, Brain and Environment’, Lathe, (2006), points out that, in comparing autism-twin concordance estimates from the late 1980s to those of 2003/4, a critical consideration is that a real change over time would provide strong evidence for an increasing environmental contribution to onset at a time when autism diagnoses have appeared to increase, and when debate has raged as to the possible factors involved. Crucially, the data suggest that the emerging ‘new paradigm’ autism phenotype is one which is increasingly determined by environmental factors that were not present, or were different, at the time the original twin studies were carried out.
In ‘Autism, Brain and Environment’ Lathe covers this issue, and its implications, in some detail. Professor Baron-Cohen provided a very positive review of Lathe’s book on its back cover, yet again fails to refer to this relevant work in his Mensa interview.