Correlation is not causation.
But sometimes it points out interesting things…
I’d seen an interesting RSA illustrated talk about education on Link TV. It had a chart in it of Autism rates by State. There was no particular date given for the data nor all the technical details needed to figure out from where it came. It claimed to show autism increasing as you moved eastward across the country, reaching a peak on the East Coast. I wanted to find such a map.
After some amount of looking, I didn’t find it. I did find a couple of charts of prevalence by State that were much more complex (and leads me to think the map in the talk was over simplified).
But as is often the case, I found something even MORE interesting along the way. That is the prevalence rate in Israel is incredibly low. Which begs the question: “Is there something about keeping Kosher that prevents Autism?” (More precisely ASD Autism Spectrum Disorder, that includes many variants including Asperger’s at the low end).
I’m going to quote some numbers from the wiki:
The rate of autism diagnoses in Canada was 1 in 450 in 2003. However, preliminary results of an epidemiological study conducted at Montreal Children’s Hospital in the 2003-2004 school year found a prevalence rate of 0.68% (or 1 per 147).
The most recent estimate states that up to 1 out of every 88 children, or 11.3 per 1,000, have some form of ASD.
A 2003 study reported that the cumulative incidence of autism in Denmark began a steep increase starting around 1990, and continued to grow until 2000, despite the withdrawal of thiomersal-containing vaccines in 1992. For example, for children aged 2–4 years, the cumulative incidence was about 0.5 new cases per 10,000 children in 1990 and about 4.5 new cases per 10,000 children in 2000. A subsequent critique of the Danish study by Mark Blaxill of SafeMinds pinpoints several discrepancies associated with the collection of autism data. Such discrepancies include changing the autism criteria in the middle of the study, 1994, from inpatients to outpatients, adding the numbers of a large Copenhagen clinic after 1992, etc.
Eric Fombonne made some studies in the years 1992 and 1997. He found a prevalence of 16 per 10.000 for the global PDD. The INSERM found a prevalence of 27 per 10,000 for the ASD and a prevalence of 9 per 10,000 for the early infantile autism in 2003. Those figures are considered as underrated as the WHO gives figures between 30 and 60 per 10,000. The French Minister of Health gives a prevalence of 4.9 per 10,000 on its website but it counts only early infantile autism.
A 2008 study found that inpatient admission rates for children with ASD increased 30% from 2000 to 2005, with the largest rise between 2000 and 2001 and a decline between 2001 and 2003. Inpatient rates for all mental disorders also rose for ages up to 15 years, so that the ratio of ASD to all admissions rose from 1.3% to 1.4%.
A 2009 study reported prevalence rates for ASD ranging from 0.21% to 0.87%, depending on assessment method and assumptions about non-response, suggesting that methodological factors explain large variances in prevalence rates in different studies.
The incidence and changes in incidence with time are unclear in the UK. The reported autism incidence in the UK rose starting before the first introduction of the MMR vaccine in 1989. A 2004 study found that the reported incidence of pervasive developmental disorders in a general practice research database in England and Wales grew steadily during 1988–2001 from 0.11 to 2.98 per 10,000 person-years, and concluded that much of this increase may be due to changes in diagnostic practice.
A 2008 Hong Kong study reported an ASD incidence rate similar to those reported in Australia and North America, and lower than Europeans. It also reported a prevalence of 1.68 per 1,000 for children under 15 years.
A 2009 study reported that the annual incidence rate of Israeli children with a diagnosis of ASD receiving disability benefits rose from zero in 1982–1984 to 190 per million in 2004. It was not known whether these figures reflected true increases or other factors such as changes in diagnostic measures.
Notice that is per million. or about 1.9 per 10,000. That’s about 1/10 th of most countries and 1/100 th of the US rate.
So what’s the deal with Israel?
A 2005 study of a part of Yokohama with a stable population of about 300,000 reported a cumulative incidence to age 7 years of 48 cases of ASD per 10,000 children in 1989, and 86 in 1990. After the vaccination rate of MMR vaccine dropped to near zero, the incidence rate grew to 97 and 161 cases per 10,000 children born in 1993 and 1994, respectively, indicating that MMR vaccine did not cause autism.
Studies of autism frequency have been particularly rare in the Middle East. One rough estimate is that the prevalence of autism in Saudi Arabia is 18 per 10,000, slightly higher than the 13 per 10,000 reported in developed countries. Estimates for ASD prevalence in Saudi Arabia are not available.
That Saudi number is particularly important as Halal dietary laws are very similar to Kosher (being based on the same parts of the Bible and with similar prohibitions). But it is an estimate based on unknown things. Not very usable, really.
In California, the UK, and Denmark, the rate knees up near the time when MMR is put into common use. In Japan they dropped MMR and rates didn’t drop. Germany has a drop for no identified reason.
All I can find is that vaccination rates in general in Israel may be lower, or later in life (not as infants with immature immune systems).
Finds a connection between a measles outbreak in the UK and low vaccination in Israel.
Since August 2007, there have been approximately 50 cases of measles in Israel, the majority of which have been serologically confirmed. Most cases have been concentrated around the Jerusalem area, with almost all patients from the ultra-orthodox community.
A three-year-old girl who was hospitalised in Jerusalem in mid-August with measles encephalitis was also ultra-orthodox and non-immunised, but she was neither a tourist nor from the Satmar community. Her contact was traced back to an ultra-orthodox un-immunised child who had travelled from London (with her parents) to a different wedding in Jerusalem that took place earlier in the summer.
Routine measles immunisation was introduced in Israel in 1967, and a two-dose schedule for the MMR vaccine at the ages of 12 months and 6 years was introduced in 1994 . Israel’s Ministry of Health has estimated that coverage for MMR in the ultra-orthodox community is between 60 and 70% [Y. Amitai, personal communication]. It is likely to be lower in the Satmar, a group who are known not to officially recognise state institutions. Nationally, an estimated 94-95% of the general non-ultra-orthodox population take up the first dose of the MMR vaccine according to Ministry of Health recommendations, and the coverage for the second dose is 95-97%.
So for the last 18 years they’ve had an MMR program, but some unknown percentage of folks are “reluctant” (and may only get vaccinations later in life).
In any case, it sure looks to me like there’s a very nice A/B test case waiting to be studied. Look at Christians in Israel to see if the food habits make a difference. Look at ultra Orthodox Jews in the USA and Europe to see if they have a different rate. Look at the various sub-populations inside Israel, with special attention to things they prohibit or with which they do not comply, to look for correlations. (BTW, this is also a very good example of just why uniformity is evil. Having a government mandated “everyone must” makes it much harder to figure out where you screwed up… having individual choice gives divergent populations and much better understanding.) Also compare diagnostic processes to see if it’s just a procedural thing.
Or maybe they just don’t put junk like Pink Slime in their food… or don’t have a lot of cats with toxoplasmosis… Or maybe they just don’t spend all day soaking in EMF fields from cell phones, TVs and everything else we stew in here in California. (I do find it funny when a friend is paranoid about his cell phone but lives a couple of miles from the airport radar under the aircraft radar approach ;-) Or maybe they just don’t dump a lot of pesticides on their ‘sand lawn’ ;-) and don’t like Crisco Hydrogenated Shortening and Fast Foods with trans-fats…
What is pretty clear to me, from a casual look at the numbers, is that there is something pretty well across the board in the “Industrialized West” that is harming kids, it ramps up in the ’70s to 90’s (with variation in when by country), and Israel has managed to dodge most of it.
It ought to be possible with that A/B set to figure out just what.