If you look at enough variables, you WILL find a correlation with whatever you are looking at.
Key paragraph:
Vinyl flooring is commonplace in Sweden, where only about 1 percent of homes have carpeted floors. But it is uncommon in U.S. bedrooms, so it may not be related to autism among American children. However, carpeting contains other contaminants, including pesticides and brominated flame retardants, which have been found to harm brain development in animal tests.
This makes me very suspicious, because it smacks of "our unusual theory does not bear out when we look at another country, so we will come up with another unusual theory to explain the effect there".
Also, while I am not familiar with Sweden, there are other places where vinyl/carpet flooring versus wood/tile correlates with socioeconomic factors.
In fairness, that handwaving is from Sciam's writers; the study itself is quoted on page 1, expressing caution against drawing conclusions: 'The data are far from conclusive. They are puzzling, even baffling, and not readily explicable at this time. However, because they are among the few clues that have emerged about possible environmental contributions to autistic disorders, we believe that they should be weighed carefully and warrant further study.'
If their insight is correct, it may be significant. Other data (eg the link about smoking in this article, and many other sources) suggest a degree of heritability, as well as noticing the correlation with asthma, which condition is also thought to be heritable. Although modern vinyl flooring may be low in phthalate emissions, they may well have been higher in the past. I grew up with linoleum in a lot of the house, and well remember its characteristic smell (along with those of vinyl car seats and a bunch of other plastics which were widespread in the 1970s). While I certainly wouldn't draw conclusions from that about having ADD - factors like parental smoking are likely to have played a more significant role - it's worthy of further study.
You're being uncharitable to the researchers. They have no theory for why they found this correlation - they weren't even looking for it. They noticed something unusual in their data, they reported it and invited further investigation. That's the responsible thing to do.
"The scientists were surprised by their finding, calling it 'far from conclusive.' Because their research was not designed to focus on autism, they recommend further study of larger numbers of children to see whether the link can be confirmed."
That's the money quote in the interesting submitted article.
The finding is quite intriguing, and I wonder how it will filter through the varied opinion-makers in the autism-issues blogosphere.
After edit: You're welcome for the link to Peter Norvig's article. As regular readers here know, it is one of my favorite links to post in comments here on HN. My further comment on the submitted article in this thread is that its second page refers to an increase in autism in California. Statements that autism has increased in recent decades are controversial,
because diagnostic criteria have changed over the same period, so how much real increase in the actual condition there has been may be much less than the increase in recorded diagnoses.
This is the most worst piece of a bad article:
"Rates of autism in California have increased seven-fold since 1990, a recent study found. Because genetics do not change that quickly, scientists suspect that chemical pollutants are probably playing a role. But there have been few studies attempting to pinpoint which chemicals, or combination of chemicals."
How could they not mention that the diagnostic criteria for autism changed over that time period? In addition to it getting slapped on shy kids, it's become a socially acceptable word for what used to be called mental retardation.
Agreed. I was surprised by this similar sentence in the article, stating that autism "has increased dramatically in children over the past 20 years."
Such an important component of the article shouldn't be misstated like that. That really should be written as "the number of children diagnosed with autism has increase dramatically over the past 20 years."
That's a crucial point. I wonder why that angle keeps getting overlooked when there are studies that support and explore it so well. Are the scientists dropping the ball? Are the writers?
One danger in any quotes about autism statistics is that sometimes folks use autism as shorthand for "autism spectrum disorders", which includes much more mild things like Asperger's syndrome. Plus, one could argue that some of the things they classify as "bad" within that "range" are actually good things. For example, I've noticed that the rise in so called autism or ASD roughly correlates with the rise of the Internet and web usage. Perhaps something about the web and our increased use of it has brought out a much more nerdy aspect of many people's brains, a wider swath of people than previously had engaged in nerd/geek/Asperger's-like behaviors and thought processes. That could explain the "cause" rather than some dramatic increase in some biological/chemical factor. Maybe it's just a social/intellectual shift.
It's sad, but my first thought at seeing yet another weak "link" between something and autism was, "well, better to have wacko parents shield their children from PVC floors than from vaccines."
>'It's sad, but my first thought at seeing yet another weak "link" between something and autism was, "well, better to have wacko parents shield their children from PVC floors than from vaccines."'
I guess you'd call me a wacko parent then.
When we were offered the swine flue vaccine I looked at the medical literature as, to my mind, using an untested (long-term tests) vaccine against an unknown risk seemed a bit chancy.
My research showed that there were associated with one of the factors (Thiomersal) apparent risks of elevating occurrence of autism. I ruled out the risk based on the evidence from the USA that removal of this factor over the last few years had not lead to a consequential reduction in the number of children suffering from autism. This seemed sufficient evidence for me to rule out the link and go on to look at the other constituents.
Once an intial, if now discredited, link between the chemical factor and autism was established, lacking refuting evidence, it doesn't seem "wacko" to me to exercise caution. A large part of this is that the vaccines have been successful in reducing incidence of the once common diseases making them known smaller risks to weigh against the then unknown risk of inducing autism.
The evidence in the OP shows a correlation. Phthalates certainly interact in negative ways with human biology already and so are a good thing to test - if I were fitting a floor I'd be more inclined to something phthalates-free given this result.
>Infants or toddlers who lived in bedrooms with vinyl, or PVC, floors were twice as likely to have autism five years later, in 2005, than those with wood or linoleum flooring.
Wood flooring is more expensive than vinyl, so.....really, maybe its just an indicator of economic status...
Vinyl flooring isn't cheap, floor boards are usually already there and sand paper isn't that expensive. A rug is a lot cheaper than vinyl were I am - perhaps very thin vinyl tiles would be cheap but vinyl requires a very flat starting surface so you need floorboards, hardboard, levelling compounds and then the vinyl on top. Most people can use sand paper but laying a levelled floor needs skills.
It's also been linked to asthma, allergies, endocrine disruption, obesity and diabetes. The federal government recently banned them in children's toys.
So even if this study turns out not to hold up, you should keep this stuff away from kids. The other worrisome chemical in plastic is BPA.
I've heard of a study that found a link between U.K. nuclear power plant sites and cancer ... including ones they were only thinking about placing a plant in.
[+] [-] frossie|16 years ago|reply
Key paragraph:
Vinyl flooring is commonplace in Sweden, where only about 1 percent of homes have carpeted floors. But it is uncommon in U.S. bedrooms, so it may not be related to autism among American children. However, carpeting contains other contaminants, including pesticides and brominated flame retardants, which have been found to harm brain development in animal tests.
This makes me very suspicious, because it smacks of "our unusual theory does not bear out when we look at another country, so we will come up with another unusual theory to explain the effect there".
Also, while I am not familiar with Sweden, there are other places where vinyl/carpet flooring versus wood/tile correlates with socioeconomic factors.
[+] [-] anigbrowl|16 years ago|reply
If their insight is correct, it may be significant. Other data (eg the link about smoking in this article, and many other sources) suggest a degree of heritability, as well as noticing the correlation with asthma, which condition is also thought to be heritable. Although modern vinyl flooring may be low in phthalate emissions, they may well have been higher in the past. I grew up with linoleum in a lot of the house, and well remember its characteristic smell (along with those of vinyl car seats and a bunch of other plastics which were widespread in the 1970s). While I certainly wouldn't draw conclusions from that about having ADD - factors like parental smoking are likely to have played a more significant role - it's worthy of further study.
[+] [-] scott_s|16 years ago|reply
[+] [-] tokenadult|16 years ago|reply
That's the money quote in the interesting submitted article.
http://norvig.com/experiment-design.html
The finding is quite intriguing, and I wonder how it will filter through the varied opinion-makers in the autism-issues blogosphere.
After edit: You're welcome for the link to Peter Norvig's article. As regular readers here know, it is one of my favorite links to post in comments here on HN. My further comment on the submitted article in this thread is that its second page refers to an increase in autism in California. Statements that autism has increased in recent decades are controversial,
http://www.sciencebasedmedicine.org/?p=4726
http://www.asatonline.org/about_autism/ontherise.htm
http://www.cdc.gov/ncbddd/features/counting-autism.html
because diagnostic criteria have changed over the same period, so how much real increase in the actual condition there has been may be much less than the increase in recorded diagnoses.
[+] [-] Oxryly|16 years ago|reply
[+] [-] chime|16 years ago|reply
[+] [-] mattmcknight|16 years ago|reply
How could they not mention that the diagnostic criteria for autism changed over that time period? In addition to it getting slapped on shy kids, it's become a socially acceptable word for what used to be called mental retardation.
[+] [-] aasarava|16 years ago|reply
Such an important component of the article shouldn't be misstated like that. That really should be written as "the number of children diagnosed with autism has increase dramatically over the past 20 years."
[+] [-] Oxryly|16 years ago|reply
[+] [-] mkramlich|16 years ago|reply
[+] [-] Legion|16 years ago|reply
[+] [-] pbhjpbhj|16 years ago|reply
I guess you'd call me a wacko parent then.
When we were offered the swine flue vaccine I looked at the medical literature as, to my mind, using an untested (long-term tests) vaccine against an unknown risk seemed a bit chancy.
My research showed that there were associated with one of the factors (Thiomersal) apparent risks of elevating occurrence of autism. I ruled out the risk based on the evidence from the USA that removal of this factor over the last few years had not lead to a consequential reduction in the number of children suffering from autism. This seemed sufficient evidence for me to rule out the link and go on to look at the other constituents.
Once an intial, if now discredited, link between the chemical factor and autism was established, lacking refuting evidence, it doesn't seem "wacko" to me to exercise caution. A large part of this is that the vaccines have been successful in reducing incidence of the once common diseases making them known smaller risks to weigh against the then unknown risk of inducing autism.
The evidence in the OP shows a correlation. Phthalates certainly interact in negative ways with human biology already and so are a good thing to test - if I were fitting a floor I'd be more inclined to something phthalates-free given this result.
But I guess I'm just a wacko ...?
[+] [-] pyre|16 years ago|reply
[+] [-] runT1ME|16 years ago|reply
Wood flooring is more expensive than vinyl, so.....really, maybe its just an indicator of economic status...
[+] [-] ars|16 years ago|reply
Lots of diseases are linked to economic status, but I've never heard of autism being one of them.
[+] [-] pbhjpbhj|16 years ago|reply
Vinyl flooring isn't cheap, floor boards are usually already there and sand paper isn't that expensive. A rug is a lot cheaper than vinyl were I am - perhaps very thin vinyl tiles would be cheap but vinyl requires a very flat starting surface so you need floorboards, hardboard, levelling compounds and then the vinyl on top. Most people can use sand paper but laying a levelled floor needs skills.
[+] [-] guelo|16 years ago|reply
It's also been linked to asthma, allergies, endocrine disruption, obesity and diabetes. The federal government recently banned them in children's toys.
So even if this study turns out not to hold up, you should keep this stuff away from kids. The other worrisome chemical in plastic is BPA.
[+] [-] mnemonicsloth|16 years ago|reply
"You see, every time the Boston Red Sox have won the World Series..."
[+] [-] hga|16 years ago|reply
[+] [-] JacobAldridge|16 years ago|reply
Alas, uncredited, but still a tick on my 'Big things to do' list.
[+] [-] rlpb|16 years ago|reply
[+] [-] jganetsk|16 years ago|reply
[+] [-] greenlblue|16 years ago|reply
[+] [-] joezydeco|16 years ago|reply
[+] [-] ohashi|16 years ago|reply
[+] [-] wdewind|16 years ago|reply