Nov 19 2008

Somali Autism Cluster

Published by under Uncategorized
Comments: 25

Recently there has come to attention a potential cluster of autism cases among Somali immigrant in Minnesota and Sweden. If true, this could potentially be an important clue as to the pathophysiology of some types of autism.

Autism is unknown in Somalia, but the children of Somali immigrants in two communities in Minnesota and Sweden have reported higher numbers of cases than the general population. This suggests that there is an environment trigger – something they are getting or not getting in their new communities that is different than Somalia. This report indicates:

In Minneapolis, Somalis account for 6 percent of the city’s public school population, but make up 17 percent of early childhood special education students who have been labeled autistic, according to data aggregated by the Minneapolis Public Schools.

But we need to slow down – apparent clusters of diseases are reported all the time. Most of the time the clusters are not real, meaning they are just statistical flukes. So the first question to answer with any apparent cluster is – does this represent a real epidemiological phenomenon.

With regard to the Somali autism clusters this question has not yet been answered. There is enough anecdotal evidence to warrant more definitive investigation. Somali parents certainly believe they are experiencing something new, and some pediatric neurologists in these areas have had their suspicions also. But this is not enough to form a scientific conclusion – only to justify further research.

The true autism rate in Somalia needs to be investigated also. We should not assume that because the culture does not recognize autism it does not exist.

One problem with the cluster hypothesis is that other immigrant Somali communities have not experienced increased autism rates. If there is an environmental trigger causing the two identified clusters, why are there not clusters in these other communities?

If it turns out to be true that autism rates have significantly increased in some Somali immigrant communities, above what is seen in Somalia or in non-Somali in the same communities, then we can conclude that something is going on and a potential trigger should be sought.

It also has to be noted that autism is really a collection of diseases, not a specific disease. So we may be seeing a new entity that has clinical overlap in features and symptoms with recognized forms of autism.

We see examples of this in neurology all the time. For example, there is a cluster of ALS in Guam. This is likely due to a toxin from the local cycad seeds which are used by the locals (perhaps concentrated in fruit bats which they eat). But this form of ALS does not necessarily have anything to do with sporadic ALS outside of this cluster – different diseases with the same or similar clinical outcomes.

If the Somali autism clusters are real, one potential cause has already been proposed – vitamin D.  Dark-skinned Africans who migrate to northern climates often suffer vitamin D deficiency, which is made in the body through exposure to sunlight. It is possible that Somali immigrants are simply not getting enough vitamin D from the northern sun in Minnesota and Sweden, and low levels of vitamin D are increasing the risk of neuro-developmental disorders with features of autism (or autism itself).

Vitamin D deficiency has been recognized as a contributing factor in some neurological diseases, and has been raised as a potential autism trigger even before the apparent Somali clusters. This does not change the fact that there is copious evidence that autism is largely genetic. But environmental factors can influence the expression of genetic disorders, so even for a dominantly genetic disorder, an environment trigger can be playing a role.

Another intriguing possibility is that these isolated clusters are due to the “founder effect.” When a small isolated population happens to contain an individual or family with a mutation for a disease, subsequent generation may have a cluster of that disease because of increased marriage within a small genetic pool. This would explain why some immigrant communities have had apparent clusters and others have not – it all depends on the founding immigrants. This hypothesis can be tested by looking at family histories within these communities.

Incidentally, vaccines are certainly not playing a role in these alleged clusters. Thimerosal, the mercury-based preservative in some vaccines and a popular target for anti-vaccinationists, was removed from Swedish vaccines prior to the Somali immigration in the 1990’s.  It was also removed from childhood vaccines in the US by 2002.

It seems, however, that the whiff of a possible environmental trigger (like blood in the water) is enough to attract the anti-vaccinationists, even if that trigger is not vaccines.  Anti-vaccine crank David Kirby is already sizing up this issue for his usual hack treatment. He is trying to portray as yet another case of parents trying to look out for their children having to fight an uncaring system. Kirby has only a few themes, and this is one of them.

He writes:

When I first wrote about the large number of Somali children with autism in Minneapolis, back in August, I was contacted by a young mother named Idil, who told me she had been trying for more than a year to get Minnesota officials to pay attention to all the sick kids in the local Somali community.

He then follows with letters from a concerned Somali mom to these Minnesota officials. He is playing the – “If only scientists would listen to the moms”- card. Meanwhile, the issues seems to be getting appropriate attention. As I outlined above, there are many scientific questions that need to be answered.

I know that when you are a parent of a sick child the gears of science may grind maddeningly slowly, and no amount of funding or attention is enough. I also know that squeaky wheels get the grease – so I have no problem with parents squeaking away to advocate for their issues. I do have a problem with crank journalists exploiting the situation and trying to make it sound like a conspiracy.

Let the wheels of science grind away. These questions can all be answered. We need to ask all the questions, in proper order, step by step to see if these clusters are real, if so are they genetic or environmental, and if they are environmental what is the trigger.

What we don’t need are people with an agenda politicizing this issue.

No matter what the truth turns out to be, this is likely to be an interesting story and we are likely to learn something about autism.

25 responses so far

25 Responses to “Somali Autism Cluster”

  1. Darrenon 19 Nov 2008 at 2:13 pm

    One thing of note that you don’t mention:

    make up 17 percent of early childhood special education students who have been labeled autistic

    (emphasis mine, if it shows up on this blog…)

    It’s important to understand that is is not doctors or autism specialists that are applying these labels, but educators who might have a small amount of “special education” training.

    It’s entirely possible that these kids have other learning disabilities, or no disability at all (just a pre-school education that is insufficiently helpful in local public schools). They merely meet screening criteria to place them in special-education programs that cater to autistic children.

    This mis-categorization happens because schools simply can’t – and probably shouldn’t – diagnose disabilities. They simply use screening programs to select kids for special attention (which usually helps, even if there is no real disability… so I guess it’s a good thing) and report this information to the parents.

    It makes sense that a typically-poor immigrant community may not have the resources to pursue confirming diagnosis, so we may not know if this is a real cluster without some outside research and funding.

  2. Fizziziston 19 Nov 2008 at 3:25 pm

    Very interesting indeed. I look forward to seeing more research on autism so that we can shut up all these crank anti-vaccinationists.

  3. daedalus2uon 19 Nov 2008 at 4:44 pm

    There are a number of potential confounding issues that are all within the known environmental causes of autism. Exposure to stress in utero does result in an increased incidence of autism, so does exposure to certain pre-natal infections. Somalia is a very high stress environment, and if these children were conceived in Somalia and their mothers emigrated as refugees while the children were in utero, a higher incidence of autism would be expected. Living as an immigrant may be sufficient stress to increase it also.

    Somalia is an undeveloped region, the type of place where agents of the hygiene hypothesis are expected to be protective against some diseases of the developed world. Some of these may well be related to nitric oxide.

    A confounding factor that is not well appreciated relates to nitric oxide physiology and the presence or absence of a surface biofilm of ammonia oxidizing bacteria (the subject of my research and blog). These bacteria are removed by bathing, so moving from a region where people rarely bathe (such as Somalia) to a place where bathing is frequent (any developed region) will reduce the NO/NOx status of individuals and would (according to my low NO hypothesis of autism) shift individuals developing in utero under such conditions to be more on the autism spectrum.

    Vitamin D is not the only physiological species produced through photochemical effects in the skin. There are NO photochemical reactions too, S-nitrosothiols and other NO/NOx species can be produced (or destroyed) by photochemical reactions in the skin. There is an excess incidence of ASD individuals born in the July time frame, who had a first trimester during December, the darkest and coldest time of the year. Inhaling cold air does reduce NO production in the nasal passages.

    Many people living in the rural undeveloped world have a significant body burden of intestinal parasites. These parasites cause the body to generate NO from iNOS, and this NO does seem to help some inflammatory diseases such as inflammatory bowel disease (where oral doses of parasitic worms are substantially curative). People accustomed to a certain body burden of intestinal (or other) parasites might have adverse effects from their removal (or non-replacement).

    As an aside, calomel, mercurous chloride (HgCl) was once used as an anti-worming agent in Switzerland where hundreds of milligrams were given to children to rid them of worms (150 to 500 mg was the usual dose). This is not a misprint, the dose was milligrams and not micrograms. Some children developed what was called “pink disease” which was called a “hypersensitivity” to a few hundred mg of mercury as HgCl in single doses. This bears no relationship to the fictitious “hypersensitivity” claimed from exposure to 10’s of micrograms of mercury in thimerosal spread out over years.

  4. Perky Skepticon 19 Nov 2008 at 8:28 pm

    Very interesting post. I’m following the Vitamin D angle with great interest.

  5. blog-thing : Somali autism riddleon 19 Nov 2008 at 9:35 pm

    […] Whatever the actual rate of autism in Somali communities it requires careful investigation. I agree with Steve Novella that […]

  6. HODANon 19 Nov 2008 at 11:51 pm




    since you haven’t done your home work and the only thing you can com up with is you ignorance,let me tell you about our kids.

    Autism is created in the developed countries since they are afraid of death.,we on the other hand did not vaccinate our kids ,eat preserved food ,and our people don’t die of cancer every year ,just FYI we are natural people ,we eat natural food ,and we dont load medications with our selfs.

    Before you even comment on our kids ,you should know they were all born here.



  7. maliathon 20 Nov 2008 at 1:26 am

    There is a cluster of Somali immigrants in Amarillo, TX. I wonder if they have a higher incidence as well? The Texas panhandle is very sunny.

  8. […] Somali autism spark new leads? Could Somali autism spark new leads? – The rate of autism among Somali immigrants in Minnesota and Sweden at least appears to be […]

  9. HODANon 20 Nov 2008 at 10:22 am

    David kirby is an american HERO.He is amazind and does his job above and beyond.

  10. Steven Novellaon 20 Nov 2008 at 12:41 pm


    Small correction – I think you misspelled “Zero”.

  11. Potter1000on 20 Nov 2008 at 1:00 pm

    HODAN has spoken.

  12. […] and also to a theory that a Vitamin D deficiency can be linked to autism. Dr. Steve Novella at the Neurologica blog writes specifically about the notion of a “cluster” of autism cases being found: […]

  13. superdaveon 20 Nov 2008 at 4:07 pm

    6% is such a huge number. I had no idea there were such large pockets of Somalis living in the US. Hurray for diversity.

  14. Dread Polackon 20 Nov 2008 at 5:27 pm

    I live in Mpls and have a friend who taught at a school that was built specifically for Somali immigrants. I seem to remember him mentioning a high level of autism. I’ll have to ask him about it again.

    He didn’t, however, have anything to say about their cleanliness or lack thereof.

  15. daedalus2uon 20 Nov 2008 at 7:25 pm

    My point about bathing had nothing to do with cleanliness or with dirt or disrespect. There is extremely well replicated data that many of the diseases of the developed world, including allergies, asthma, heart disease, hypertension, diabetes, obesity and others are rare to non-existent in the rural undeveloped world. In trying to explain the gradient of disease along the gradient of development, people have coined the term “hygiene hypothesis”, to explain the differences observed as being in some unknown way related to “hygiene”.

    The “hygiene hypothesis” has been a topic of study for many years by many researchers. They have looked at disease organisms, parasites, dirt, and have not found a satisfactory answer. I think the agent of the hygiene hypothesis is the bacteria that I am studying, ammonia oxidizing bacteria living on the skin and metabolizing ammonia in sweat into NO/NOx. The reason they have not been found by previous researchers is that they are obligate autotrophs, they are incapable of growth on any media used to isolate pathogens. In the media used by the FDA to test for sterility they do not grow; they are “unculturable”. Unless they are specifically looked for with the proper techniques they cannot be found. They are slow growing with a doubling time of ~8 hours, about 30 times longer than heterotrophic bacteria such as E. coli. Unless you allow your culture to grow 30 times longer (months instead of days), you will not get the same degree of replication. If the kinetics of removal of autotrophic and heterotrophic bacteria are the same, bathing even once a week can wash them off faster than they can proliferate.

    It turns out that all the diseases that show a gradient along the development gradient are diseases associated with low NO. My explanation is because the loss of the bacteria I am working with causes a relative deficiency in NO/NOx and (in susceptible individuals) that shows up as low NO mediated disease. The details of which organ system is affected first depend on the idiosyncratic details of that individual’s physiology, genes, diet, epigenetic programming and activity level. Virtually all of the low NO disorders also mimic the symptoms of stress. Stress makes them all worse. That is because stress is a low NO state, and physiology invokes low NO as a compensatory reaction.

    Somalia is a lawless war zone. A lawless war zone is a very high stress environment. It would not be a surprise if individuals from a lawless war zone had symptoms of stress. Exposure to stress in utero is known to increase the incidence of ASDs. The UN reported that only 29% of the population had access to safe drinking water (in 2004). Presumably even fewer have access to sufficient water for bathing.

    My hypothesis is that autism is also a disorder of low NO. All the physical symptoms of ASDs are what would be expected if basal NO was lower. The neurological symptoms can also be explained by low NO. Many social pathways use NO as a neurotransmitter.

  16. mike stantonon 20 Nov 2008 at 8:24 pm

    As well as autism, Somali migrants have above average pulmonary disorders and diabetes. Is there a connection? If so it is likely to be a consequence of war, famine and unhealthy conditions in refugee camps and nothing to do with vaccines, vit D deficiency or anything else they have encountered in America.

  17. isleson 20 Nov 2008 at 8:29 pm

    If you are right, daedalus, what would the treatment or prevention implications be? Every time I see nitric oxide mentioned anywhere, I think of you. 🙂

    It’s not at all clear that there actually is an elevated rate of autism among the Somalis. The Minneapolis schools have good autism programs and parents in that area have a lot of choice about where to send their kids within the various public systems, so it could be that all the area families with autistic kids are attracted to Minneapolis schools. I wonder if the people who are researching this will compare the prevalences in various local school districts – if it’s higher in Minneapolis, it’s either an *extremely* localized environmental effect or an artifact of school choice.

  18. epguyon 20 Nov 2008 at 9:17 pm


    If you are truly serious about NO and autistic spectrum disorders, I recommend skipping the microbiology, which as you pointed out can tie you up for years, and head right to clinical trial. Find a high-risk population – i.e., identical twins of pts with autism – and randomize to isosorbide vs. placebo! If you’re not wild about oral nitrates in kids, then use an appropriate dose of topical nitropaste vs. placebo. If the argument is convincing enough, I am sure someone would fund you. I, of course, wouldn’t attempt this without appropriate IRB/Human Studies oversight.

    I’d have to say that I am skeptical about the hypothesis, but it is certainly testable. Even if your micro work showed an association, I’d still hold out for the clinical data before buying in.

    As for the Somali migration cluster – I can’t help thinking “pirate” exposure may have a protective effect with regard to Autism. (Sorry – there goes any credibility I may have had)

  19. daedalus2uon 20 Nov 2008 at 10:28 pm

    Isosorbide dinitrate doesn’t raise NO levels. I am quite sure that it would not work on autism. It doesn’t even work that well on heart disease. I think it is more of a trigger of ischemic preconditioning than an NO donor. All the organic nitrates seem to work that way, which is why they cause migraines and why they don’t prolong life.

    The only way to raise NO levels is with the bacteria I am working with. The problem (feature?) is that NO/NOx physiology is so complex, so coupled and so well regulated that it is virtually impossible to manipulate it pharmacologically without pretty major side effects. That is why there are only a handful of NO based therapies, inhaled NO (which only has local effects in the lung), and infused sodium nitroprusside (one of the few “real” NO donors which has a half life in the blood of only a few minutes).

    Organic nitrates are not NO donors. They do have NO/NOx effects, but those effects are complex and not well understood. NO/NOx physiology is so complex, there are at least thousands of pathways that involve NO, and they are all coupled and non-linear (and mostly (i.e. 99+%) unknown). NO isn’t blocked by anything, it diffuses everywhere, so all of these NO/NOx pathways are coupled; you can’t change one without affecting every single other one. Evolution has configured physiology to work that way.

    The only way that NO levels can be raised without side effects is to use a natural process that physiology is controlling. The natural process that humans evolved to use is these bacteria living on the external skin. It would be impossible for humans living in the wild to not have a biofilm of these bacteria on their skin. These bacteria are universally found in the environment, in every source of water, ground, surface, sea, mineral spring, etc.

    It is completely preposterous to suppose that we could artificially control something like NO physiology when we can’t even measure it in any tissue compartment at the time, concentration and length scales that we know are important. We know that all of those scale over at least 3 orders of magnitude.

    I would love to do a clinical trial, but I am only willing to do it in the ethically responsible way, unlike autism’s false prophets. Obtaining the resources to do it has been a challenge.

    I thought pirates prevented global warming.

  20. Dread Polackon 21 Nov 2008 at 10:43 am


    Just so we’re clear, my comment had nothing to do with yours. In fact, I never read your comment, I only skimmed through the comments and saw HODAN’s reaction. I was only making a (barely) humorous reference to him, not you.

  21. Calli Arcaleon 25 Nov 2008 at 5:13 pm

    HODAN, I don’t think any disrespect was intended. The hygeine hypothesis, while badly named, refers not to anyone being “dirty” but to first-world countries perhaps trying too hard to kill pathogens — or simply to the consequences of there not being very many pathogens around. This isn’t related to whether or not anyone bathes. It goes far beyond that, to clean drinking water (hard to come by in a war zone), adequate nutrition (also hard to come by in a war zone; my grandfather saw the same thing happening with European children in the 1940s), widespread use of disinfectants, excessive use of antibiotics, climate-controlled dwellings making us reluctant to go outside into nature, and so on.

    Regarding vaccines, the hygiene hypothesis should predict that people in overly clean environments should get *more* vaccines to make sure their immune systems get adequately challenged, at least in my opinion.

    Vaccines do not cause autism; do not be afraid of that. The numbers are in. Japan stopped some of their vaccines out of fear of autism; autism rates did not change, but measles rates went up and there were deaths.

    Preserved food has its benefits. One should get as many fresh foods as possible, but preserved food is vastly better than spoiled food, and not all preservation methods are as nasty as folks think. Just remember to do everything in moderation. Make sure you’re not getting too much sodium, for instance; that’s in most preservatives, including table salt (the most ancient and most natural of the preservatives). Processed meat is not, in and of itself, bad. It all depends on how its made, and learning that takes some determination and self-education.

    And don’t be afraid of medication either. Sure, it’s bad to have too much (that’s what’s creating “superbugs” like MRSA), but there are times when it can save your life. I just got over a case of strep throat. I’m on amoxicillin (an antibiotic) for it, and I’m very glad. It’s cured the disease. I just have to take the drug long enough to be sure I killed it all. I’ve also had lots of urinary tract infections, and while natural remedies like cranberry juice can help prevent them, there’s no substitute for antibiotics once a bacterial infection has set in.

    The trick is to remember “everything in moderation”. 😉

  22. HODANon 27 Nov 2008 at 9:20 pm

    Dr. Novella with all due respect dont be a hater,

  23. HODANon 27 Nov 2008 at 9:57 pm

    Dr. Novella

    you know and i know and everyone else for this matter that David kirby is far away from being ZERO.

    He is brilliant and very smart reporter.

    if you dont have anyhting nice to say about other professionals,

    then i recomend you not to say anything at ALL.

  24. HCNon 28 Nov 2008 at 6:45 pm


    Mr. Kirby went from writing for Gay Rights magazines like OutWeek and Advocate, and travel articles for the New York Times to become a contracted writer for a book (Evidence of Harm, see page 152 of “Autism’s False Prophets” by Paul Offit).

    How does this make him an expert on autism compared to Dr. Novella, who is a neurologist who actually went to medical school?

  25. son 06 Jan 2009 at 12:01 pm

    Just so you know. C Gillberg was the epicenter of an extremely vitriolic debate over scientific dishonesty and “medicalisation of an entire generation” a couple of years ago that ended in him destroying 15 years of research records that a court had ordered him to release for scrutiny. (He could just have chosen to turn them over to an impartial body for review!) For further reference see article in BMJ (

    (Also notice that Stockholm is at Anchorages latitude while Minneapolis is quite more sutherly…, before drawing too many conclusions on insolation and sunshine hours.)

    Talking about vitriol I noticed that many comments in this forum (blog) are pretty acerbic and authoritarian rather than reasoned skepticism. 🙁

    Kudos to David* for keeping to proposing his ideas instead of just attacking the messengers. As a matter of fact I just love his suggestion of disruptions/disturbances in NO-metabolism. A grand idea with both flair and viable biochemical/-medical and evolutionary explanations. (I would publish if I were he. At least in “medical hypotheses” just to get a broader audience.) Just blew me away 😉

    *For more see: Whitlock DR. Human microbiome: hype or false modesty? Nature 454(7205), 690 (2008)
    Whitlock DR, Chapter 8. Regulation of Oxidative Stress: Dysregulation or Bad Setpoint? Interactions with Basal Nitric Oxide, pp. 133-144 in Oxidative Stress: Clinical and Biomedical Implications (2007), Editors: Matata BM, Elahi MM.

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.