Amidst the furor over autism in America, some very simple facts are getting lost in the rhetorical fog of medical denial, corporate self-dealing and civic irresponsibility. They're worth repeating. Autism was once very rare in the United States and required "discovering" by Leo Kanner among a small group of children born in the 1930s. When researchers first measured American autism rates, they were lower than surveys coming from other parts of the developed world, sometimes less than 1 in 10,000. Today, it's nearly impossible to find an American who doesn't know a family touched by autism and rates are over 1 in 100 in some areas of the country. It doesn't take a genius to derive a short list of broad-based environmental exposures that have changed rapidly enough to give us some pretty good ideas about causation.

Meanwhile, the moral cowardice of the American scientific community has been on full display in the midst of the obviousness of this scourge and its likely causes. Scientists, academics and public health officials get this pained expression on their faces when the subject of the autism epidemic comes up, as if someone has passed gas audibly in the middle of a tea party. "Can't we talk about something more civilized?" their faces seem to say. "There must be an explanation for all these pesky numbers that won't force me to change my beliefs or behaviors. Why can't all you autism advocates just go away?" After years of seeing these faces, I've learned to recognize the look and have gotten better at managing my own frustration over the incoherence of the denialist impulse. Still, the stark reality of these simple facts needs reinforcing and like any stubborn problem new facts constantly emerge to provide new reinforcement.

As if on cue, the latest autism tragedy has emerged: the Somali community of Minnesota. This cluster of families is just the latest chapter in the long history of political refugees seeking shelter in America. In this case the Somalis are escaping an African nightmare of a civil war that has lasted nearly two decades and emigrating to American in large numbers looking to replace their nightmare with the American dream. For many of these families, however, the move to America has led them straight into a new nightmare, one where overt physical violence is replaced with a new danger: one full of smiling faces and the outward appearance of order, but where the threat of injury to their families is no less real. This new danger reaches into their homes not through the barrel of a gun but through the tip of a needle and it leaves no visible wounds, just the inflamed brain tissues of their infant children.

Out of Africa and into autism. Welcome to America, our Somali friends.

For the American autism community, the rapidly evolving Somali experience in America is unfolding in familiar form: first with their own rising awareness of the autism anomaly as inexplicably high numbers of autism diagnoses show up in their children, followed closely by organized denial by public health authorities of both the rising numbers and the obvious potential causes. American parents are accustomed to the evidentiary arguments and the debating points. But in the case of the Somali anomaly, the evidence is even starker and bears repeating.

1) Autism has always been rare in Africa, with low rates that have surprised researchers.

2) Most autism in Africa occurred in elite families with access to Western health services.

3) Among Africans who migrate to Western countries, autism rates are remarkably high. These immigrants face unusual risks of over vaccination.

What does it take to connect the dots here? But let me take you all through a quick tour of the evidence base on autism in Africa (see a short list of references at the end of this essay). It's deeper and more conclusive than most of you might know.

Very low autism rates in Africa

Over thirty years ago, a man named Victor Lotter took a tour of Africa looking for autism cases. He didn't attempt a full prevalence study; instead he simply went to visit "collections" of mentally handicapped children in institutions in hopes that he would be able to find evidence of autism in these high concentrations of mentally impaired children. "It was clear from...preliminary inquiries", wrote Lotter, "that autistic behavior does occur in...Africa." So he visited nine cities in six African countries (Ghana, Nigeria, Kenya, Zimbabwe, Zambia and South Africa) in search for as many cases of autism as he could find. Although not a full scale population survey, Lotter personally screened over 1,300 mentally handicapped children during a two year period.

Lotter was well suited for the investigation. His 1966 survey in Middlesex County England was the first of its kind: the oldest population prevalence survey of autism ever published. His prevalence estimate of 4.1 per 10,000 (1 in 2400) defined the generally accepted disease frequency of autism that was repeated for decades to follow. (For many of us, when our children were first diagnosed in the 1990s, Lotter's rate of 4-5 per 10,000 was still the rate of autism most of us were quoted when we wanted to know how unlucky we really were).

What Lotter found in Africa surprised him, namely that "the number of autistic children found was much smaller than expected." Only 9 of the 1312 mentally handicapped children he saw in nine cities were autistic (a rate of 1 in 145, nearly the same as the autism rate today in the entire US childhood population!). He had expected to see over 1 in 20. Around that time, clinicians in Nigeria and later in Kenya confirmed that autism indeed was present among African children but found it rare enough that it was worthwhile for them to give detailed profiles of just four and three cases, respectively.

From our current vantage point, the obvious inference to draw from this low rate of autism in Africa is that the rate of autism was low because the exposure to early childhood vaccination in the general African population was low as well. But at the time these early surveys were published, the autism controversy that was raging focused more on Bruno Bettelheim's "refrigerator mother" theory than on vaccines, and so all these early authors tread carefully on the question of parental backgrounds (see below) and never even considered the vaccination question.

And for the most part, that is where the African prevalence information has been left. But the search for African cases has continued sporadically, and has led to an interesting recent twist. A couple of years ago, one of the most doctrinaire "autism is genetic" American researchers, in a major break with orthodox doctrine, conceded the idea that there might be such a thing as regressive autism. Not in America, of course, but in Tanzania. A group led by Susan Folstein recently examined reports that previously typical children could acquire autism after a malaria infection. In a group of 14 autistic children, Folstein's group conceded that in at least three cases "the relationship between onset of autism and severe malaria seems clear" and that in four additional cases there was a possible relationship. That implies that as many as half of the autism cases seen in African populations could be acquired and not genetic. This was surprising enough, but it was not the only unusual implication. If malaria can cause autism, then of course Africa should be full of autism cases!

The researchers reluctantly conceded this point, although the Folstein paper is a veritable symphony of contradictions between orthodox autism doctrine and their evidence. After first reporting, true to form, that "autism in the Western world has a primarily genetic etiology, although the genes responsible for the majority of cases are still not known", Folstein's team then went on to crow that theirs was the "first report to identify Plasmodium falciparum as an etiologic agent of autism." But this discovery led the authors straight to an obvious anomaly: "one logical consequence of the frequency of cases with infectious etiologies is that the prevalence of autism should be higher in Africa than it is in the West." And if not, as should be obvious to anyone, then maybe there was something else going on in these cases in addition to their malaria infections.

Clearly, the link between autism and malaria makes no sense at all if autism rates are low in Africa. How do the author's deal with this contradiction? Their response strikes a familiar chord: they ignored it.

"The prevalence of autism in Africa...is unknown."

Most African autism cases occur in elite families

Going one level deeper in the surveys of autism in Africa reveals another critical bit of evidence. Unfailingly, when autism was found by early researchers in African families, it occurred with overwhelming frequency in elite families. Lotter reported that "there was amongst all the children originally selected as possible cases an excess of the elite...this excess was even greater amongst the autistic children than amongst non-autistic children." Lotter defined as elite as "any child who had been born abroad [Britain, Europe or North America], or had lived for any period abroad, or whose parent(s) had lived for any period abroad, or whose father had a non-manual job." Lotter's finding found support from other researchers as well: all four cases in the Nigerian report came from elite families as did the three Kenyan children, the parents of whom included a medical doctor, an engineer and the "chairman of a parastatal organization."

Although none of the studies provide evidence on the point, the obvious inference to draw from this social class finding is that the children of elite families were at higher risk of vaccination. But again, this treatment risk was never considered. Instead, in the era of Bruno Bettelheim, the main question that emerged from this finding was whether or not autism was "class related."

Not surprisingly, this interpretation of autism delighted some academic observers, who were eager to blame, if not simply the parents, then society and social class for autism. "Infantile autism", wrote Victor Sanua in a 1984 paper that reviewed the evidence of autism in Africa, "appears to be an illness of Western Civilization, and appears in countries of high technology, where the nuclear family dominates." The idea that the risk of Western Civilization might be real but boil down to something quite specific seems to never have occurred to Dr. Sanua.

High autism rates in African immigrants

Well before the Somali anomaly in Minnesota, autism surveys noted an increased risk of autism among African immigrants. The first hint that immigrants to Western countries might have higher risk of autism came not from Minnesota but from Göteburg Sweden. In 1991, Christopher Gillberg, one of the more prolific autism survey authors, published a study entitled, "Is autism more common now than ten years ago?" in which he made the following observation. "Almost 60% of all new children with autism in the urban region detected between 1984 and 1988 were born to immigrant parents. Almost all of these parents had been born in non-neighboring countries and more than half came from southern Europe, Asia, Africa and South America. So far, this represents an unusual distribution of immigrants as compared with the Göteburg population in general."

In subsequent studies, Gillberg went on to investigate the idea that children born to parents of immigrants to Sweden (including but not limited to African immigrants) had higher autism risk. Gillberg didn't consider the vaccination risk in any of these papers. Instead, he raised for the first time the rather ridiculous notion that "men with Asperger syndrome (whose children would be more likely than others to develop autism) might marry women from other cultures, who, in turn, might not initially be as aware of the social and communication deficits shown by these men as native women would be." (I've called this hypothesis the "geeks get lucky" theory elsewhere but that's a subject for another day). As far as the African immigrants, Gillberg looked directly at three Ugandan immigrant families. In a detailed study of these specific cases, Gillberg rejected the parental mating theory while also not ruling out the idea that the Ugandan mothers were exposed to a novel virus while pregnant and that this exposure provoked autism in their children. The suggestion that the virus might be a vaccine rather than a wild type virus never came up.

Only recently has Gillberg even raised the specter of the vaccination hypothesis: this time indirectly in a 2008 analysis of Somali immigrants in Göteburg. Most importantly, this new study found a rate of autism among Somali immigrant families (1 in 142) that was three to four times higher than the rate in the non-Somali group from Göteburg (1 in 526). In their paper, the authors commented that the Göteburg Somali community was suspicious of the MMR vaccine and had a reduced vaccination compliance rate (70% vs. 95%). They also noted that they had collected medical records ("antenatal, perinatal, and postnatal data, medical information") for all the Somali immigrants. That would seem to set the stage for a rather obvious analysis: what was the nature of the vaccine exposure in the children and mothers of these Somali immigrants?

But despite collecting all the information in Somali vaccine records, there was no vaccination risk analysis of any kind disclosed in the paper.

The obvious risk that immigrants to any Western country, both Sweden and America, face is over vaccination. As vaccination programs have spread around the world in recent years, future immigrants are increasingly likely to be vaccinated in their home countries. When they travel, they are forced to receive another round of vaccinations in their home countries before they leave. When they reach their new countries, their previous vaccination records are generally not recognized as valid and they often must be vaccinated again. This unique migration risk is especially relevant for population groups that can influence autism risk: women of child-bearing age, pregnant women and infants. It's hard to know what kind of havoc these redundant treatments wreck on the immune system of such targets when they receive excessive vaccine doses. All we know is that children of modern immigrants are at high risk of both over vaccination and of autism. And that no one has ever bothered to investigate over vaccination as a specific risk factor of obtaining the medical entry visa to Western Civilization.

* * *

By now, the familiar rituals of denial in Minnesota are as predictable as they are pathetic. Autism rates can't be low in Africa, "we simply don't know what the prevalence is." Autistic regression can't be a reality in the US but when it's reported after a small number of malaria infections in Tanzania, then even the most hard core genetic determinists are willing to consider the possibility - in Tanzania. When higher rates of autism show up in African women who immigrate, certainly it can't be the experience these women had while immigrating that might shed light on their infant's risks, no it must be the innate autistic features of their husbands (Swedish geeks who got lucky and fooled their Ugandan brides). And, of course, when the Somali community in Minnesota begins to mobilize around the possibility that the excessive rate of vaccination to which they've been subjected during their immigration process might have increased their children's risk of autism, they are dismissed with the same contempt with which the American parents have been treated.

If it weren't so tragic, it would make for high comedy. What are these people thinking?

At the end of the day, there must be a moral accounting here. Lotter found 1 in 138 in a population of mentally handicapped children. By contrast, in Somalian immigrants in Minnesota, we are finding 1 in 28 in the entire population of children. The Somali anomaly is no anomaly at all. Instead it is a bright beacon that shines an uncomfortable light on the root causes of autism. I can just see those puckered faces of public health officials.

When will they open their eyes to see the simple facts?


References

Studies of autistic children in Africa
1. Lotter V. Childhood autism in Africa. J Child Psychol Psychiatry. 1978;19(3):231-44.
2. Dhadphale M, Lukwago MG, Gajjar M. Infantile autism in Kenya. Indian J Pediatr. 1982;49(396):145-8.
3. Longe, C.I. Four cases of infantile autism in Nigerian children. African Journal of Psychiatry 1976;2:161-75.
4. Khan N, Hombarume J. Levels of autistic behaviour among the mentally handicapped children in Zimbabwe. Cent Afr J Med. 1996;42(2):36-9
5. Mankoski RE, Collins M, Ndosi NK, Mgalla EH, Sarwatt VV, Folstein SE. Etiologies of autism in a case-series from Tanzania. J Autism Dev Disord. 2006;36(8):1039-51.
6. Sanua VD. Is infantile autism a universal phenomenon? An open question.
Int J Soc Psychiatry. 1984;30(3):163-77.

Studies of autism among families of African immigrants
1. Gillberg C, Steffenburg S, Schaumann H. Is autism more common now than ten years ago? Br J Psychiatry. 1991;158:403-9.
2. Gillberg C, Schaumann H, Gillberg IC. Autism in immigrants: children born in Sweden to mothers born in Uganda. J Intellect Disabil Res. 1995;39 ( Pt 2):141-4.
3. Gillberg IC, Gillberg C. Autism in immigrants: a population-based study from Swedish rural and urban areas. J Intellect Disabil Res. 1996;40 ( Pt 1):24-31
4. Barnevik-Olsson M, Gillberg C, Fernell E. Prevalence of autism in children born to Somali parents living in Sweden: a brief report. Dev Med Child Neurol. 2008;50(8):598-601.


Mark Blaxill is Editor at Large for Age of Autism.