Somali autism riddle
Sometimes the world of autism produces an unusual event that challenges existing notions about the condition. These events need investigating in case they turn out to be significant. But too often their novelty value attracts premature speculation in the media. In July 2008 there were press reports that drew attention to a cluster of autism cases in the childen of Somali refugees in Minneapolis. In March 2008 a study was published that drew attention to high rates of autism among Somali children in Sweden.
There was a brief flurry of interest in the online autism world. Kristina Chew’s article provoked much discussion on Autism Vox. David Kirby weighed in of course with Is Autism an “American Disease?” I found his article interesting because he drew attention to the little known phenomenon of higher than average autism rates amongst the children of immigrant families, especially among non-white migrants to Europe and North America. Joseph has written a good introduction to the research literature on this. Despite going back over thirty years the literature is not extensive. But it does suggest that cultural and linguistic differences might be at least as significant as the ethnicity of the migrant populations.
Unfortunately Kirby confuses the question by trying to connect vaccines to immigration. But it is a very muddled connection. According to Kirby foreign migrants come mainly from countries which still use thimerosal in childhood vaccines and have good vaccination records (92% coverage in Mexico for example). He claims that
“many if not most of these children are routinely revaccinated on entry into the United States.”
Even assuming that Kirby is correct, what has this to do with the Somali children? They were all born and vaccinated within the USA. And the most dramatic increase in autism is in Somali children enrolled in early years programmes whose vaccination schedule would have been largely thimerosal free. If the Somali phenomenon is part of a broader trend affecting the children of immigrants then it is clear that vaccination history is not a commonality that they all share.
Kirby mentions vitamin D as another potential commonality. Lots of migrants are dark skinned and come from tropical or sub tropical climate zones. They come to countries in temperate climate zones with much less sunshine. Vitamin D deficiency is a recognized problem. But it does not cause autism.
A search of PubMed for autism and vitamin D deficiency found 3 articles: two were case studies of rickets in autistic children with restricted diets. One was about possible vitamin D deficiency in autistic Somali children in Sweden. But it is only a brief note (less than two pages) in Swedish and there is no abstract available in English. I did turn up another article, this time in Medical Hypotheses, that suggested a link between vitamin D and autism. It argues that lack of sunlight, particularly UVB penetration in northern climates makes dark skinned people more prone to vitamin D deficiency and hence more prone to autism. This link to autism has been described as no more than “an interesting speculation,” by Dr. Wendy Roberts, a professor of pediatrics at the University of Toronto. It is also interesting that the author of the speculation is founder of the Vitamin D Council whose mission is
to put an end to the worldwide epidemic of Vitamin D Deficiency.
Perhaps it is not about commonalities. What if the Somalis are a special case? Over on Age of Autism Kirby speculates that it could be more “evidence” of mitochondrial influence on autism.
I wonder if the fact that Somalians reflect a different mitochondrial haplogroup than most of the rest of Africa (M1 rather than L) might provide a clue here?
I had no idea what haplogroups were before now, but I do know that the distribution of mitochondrial haplogroups as depicted in this mitochondrial map of the world is far too complicated to support Kirby’s “wonderings” without some very hard data. What if we compare the data from Sweden and Minneapolis?
The first thing to note is that there is no relation between the two datasets. In Sweden an epidemiological study found a rate of 0.19% among children with a degree of learning difficulty(Europe)/mental retardation(USA). This is consistent with the consensus figure of 20 in 10000 for ASD amongst children with additional learning difficulties. The comparable figure for Somali children in Sweden is 70 in 10000. In Minnesota there is no reliable epidemiological data for autism. There is the administrative data for children in receipt of autism services. But this is based on teacher assessment and does not require diagnosis by a trained clinician. A clinical diagnosis requires
A total of six (or more) items from (1), (2), and (3), with at least two from (1), and one each from (2) and (3):
The three categories comprise the famous Triad of Impairments. In Minnesota, providing you satisfy their criteria for the first part of the Triad, meeting their criteria for either of the other parts is sufficient to access autism services. This is much less demanding than the DSM-IV criteria and probably explains why Minnesota has more than twice the national average of autism cases on its books. Even though the figures for Somali children in Minneapolis, at 1 in 28 or 3.6 per cent. are at least 4 times the national average they are only twice the average for Minnesota. Compared with the Swedish figures of 20 in 10000 for natives and 70 in 10000 for Somali immigrants, Minneapolis has a baseline figure of 180 in 10000 for natives and 360 in 10000 fo Somali immigrants.
All we can say with certainty is that
- no matter what criteria you use, the children of foreign migrants are more likely to be identified as autistic than native children.
- the higher the “going rate” for autism in a specific community, the higher the corresponding rate will be in the migrant community.
- clinical criteria are more parsimonious than administrative criteria.
Finally, the Age of Ignorance makes the important points that
- Somalia is a war zone and the refugees who have made it to America have suffered unspeakable hardship and trauma along the way.
- The children of psychologically distressed parents frequently display evidence of psychological distress themselves.
Whatever the actual rate of autism in Somali communities it requires careful investigation. I agree with Steve Novella that
What we don’t need are people with an agenda politicizing this issue.
I concur completely with the Age of Ignorance conclusion
One further comment. What Kirby, et al, are doing with the poor people of Somalia, i.e. using them to further their agenda, disgusts me. The people of Somalia are living in some of the poorest conditions in the world. They are being subjected to war, war where people die, where people are psychologically damaged with PTSD, depression, etc. They are uprooted and forced to leave their homes to protect their families. It is well known that children of the victims of war, e.g. survivors of the Holocaust, develop psychological problems.
Kirby and his acolytes on the blogs Age of Autism, Injecting Sense and Adventures in Autism are trying to exploit the situation in Somalia. The people of Somalia have suffered mass murder, rape, torture, famine, disease - all the usual features of a bloody civil war. Half of all children under five in 1993 died in Somalia. A third of all refugees have been tortured. Maybe having traumatized parents who are strangers in a strange land and growing up between two cultures is a more significant environmental difference for these children than vaccines or sunlight.
