Only if it’s statistically significant. It could be a small enough effect that they don’t notice unless they’re looking for it (if you’re going to publish a finding from either extreme, you’re supposed to use a two-tailed test, so they’d presumably want something stronger than p = 0.05), but large enough to keep them from accidentally noticing the opposite effect.
Or alternately, it’s a large effect but the rarity of autism and of non-vaccinated kids makes it hard to reach statistical-significance given sampling error. So let’s see, the suggestion here is that the reason so few studies threw up a false positive was that the true effect was the opposite of the alternative, vaccines reduce autism.
Autism is… what, 0.5% of the general population of kids these days? And unvaccinated kids are, according to a random Mother Jones article, ~1.8%.
So let’s imagine that vaccines halve the risk of autism down from the true 1.0% to the observed 0.5% (halving certainly seems like a ‘large’ effect to me), autism has the true base rate of 1.0% in unvaccinated, and the unvaccinated make up 1.8% of the population. If we randomly sampled the population in general, how much would we have to sample in order to detect a difference in autism rates between the vaccinated & unvaccinated?
The regular R function I’d use for this, power.prop.test, doesn’t work since it assumes balanced sample sizes, not 1.8% in one group and 98.2% in the other. I could write a simulation to do the power calculation for a prop.test since the test itself handles imbalanced sample sizes, but then I googled and found someone had written something very similar for the Wilcoxon u-test, so hey, I’ll use the samplesize library instead; filling in the relevant values, we find for a decent chance of detecting such a correlation of vaccination with reduced autism, it takes:
a total n=90k. I’m guessing that most studies don’t get near that.
Of course, a lot of that penalty is going towards picking up enough kid who are both autistic and unvaccinated, so one could do better by trying to preferentially sample either of those groups, but then one gets into thorny questions about whether one’s convenience samples are representative and biased in some way...
As the original article says, if there was no effect, you’d expect a few studies to get p < 0.05 by chance. Similarly, if there was no effect, you’d expect a few studies to get p > 0.95 by chance, suggesting that vaccines prevent autism. If vaccines do prevent autism, then it would be even more likely to have p > 0.95.
Not all statistical analysis has to be preregistered. If a data has a trend that suggest vaccination might reduce autism I’m sure the researchers would run a test for it.
If the study is underpowered to find a effect in that direction it’s also like to be underpowered to find a effect in the other direction.
Can someone with more statistical expertise run a test to see whether the studies are underpowered to pick up effects in either direction?
There is fairly extensive data (not published in the peer reviewed literature) that groups which are unvaccinated have far lower autism rates than the general public.
UPI Reporter Dan Olmsted went looking for the autistic Amish. In a community where he should have found 50 profound autistics, he found 3. The first was an adopted Chinese girl who’d had vaccinations rushed before she was shipped from China and more here on the way to the adoptive parents. The second had been normal until developing classic autism symptoms within hours of being vaccinated. The third there was no information about.
http://www.putchildrenfirst.org/media/e.4.pdf
Olmsted continued his search for unvaccinated Amish with autism beyond that community, finding none for a long time, but eventually found a Doctor in Virginia who had treated 6 unvaccinated Amish people from various places with autism. 4 of them had very elevated levels of mercury.
A telephone survey commissioned by the nonprofit group Generation Rescue compared vaccinated with unvaccinated boys in nine counties of Oregon and California [15]. The survey included nearly 12,000 households with children ranging in ages from 4 to 17 years, including more than 17,000 boys among whom 991 were described as completely unvaccinated. In the 4 to 11 year bracket, the survey found that, compared with unvaccinated boys, vaccinated boys were 155% more likely to have a neurological disorder, 224% more likely to have ADHD, and 61% more likely to have autism. For the older boys in the 11-17 year bracket, the results were even more pronounced with 158 % more likely to have a neurological disorder, 317% more likely to have ADHD, and with 112% more likely to have autism. [15]
In addition to the Generation Rescue Survey, there are three autism-free oases in the United States. Most publicized are Amish communities, mainly studied in Ohio and Pennsylvania [16].The Amish are unique in their living styles in largely self-sustaining communities. They grow their own food. Although they have no specific prohibitions against medical care, very rarely do they vaccinate their children. In local medical centers available to the Amish, most centers reported that they had never seen an Amish autistic child. The only Amish children that were seen as a rule were those with congenital disorders such as fragile X. The one autistic Amish child that was discovered during the surveys was taken to a medical office for an ear infection where the child was incidentally vaccinated, probably without the mother’s consent.
The second is the Florida-based medical practice of Dr. Jeff Bradstreet. While treating several thousand autistic children in his practice, Bradstreet has observed that “there is virtually no autism in home-schooling families who decline to vaccinate for religious reasons” [17]
The third, the “Homefirst Health Services” located in Chicago, has a virtual absence of autism among the several thousand patients that were delivered at home by the medical practice, and remained non-vaccinated according to the wishes of the parents [18].
Clusters of autistic children have also been found among parents with occupational exposures to chemicals prior to conception [19], and in children exposed prenatally to organochlorine pesticides [20].
Reportedly the CDC has been surveying the vaccination status of the Amish for years, attempting to induce them to vaccinate (with some success I believe), and has consistently refused requests to include an autism question with their survey to gather data.
UPI Reporter Dan Olmsted went looking for the autistic Amish. In a community where he should have found 50 profound autistics, he found 3.
He went looking for autistics in a community mostly known for rejecting Science and Engineering? It ‘should’ be expected that the rate of autism is the same as in the general population? That’s… not what I would expect. Strong social penalties for technology use for many generations would be a rather effective way to cull autistic tendencies from a population.
I don’t reject the possibility there are other explanations for the observation that unvaccinated Amish have very low autism rates. I even offered one: that they also reject Glyphosate.
However, when it turns out that the rare cases of Amish with autism that are found mostly turn out to be vaccinated, or have some very specific other cause obvious that’s not present in the general population (high mercury), the case for vaccination being a cause becomes much much stronger.
And when you realize that other groups of unvaccinated also have low autism rates, the case becomes stronger.
And when you realize that injecting the aluminum into animal models causes behavioral deficits, and injecting vaccines into post-natal animals causes brain damage, in every study I’ve found, the case becomes stronger still.
And when you discover that the safety surveys don’t cite any empirical measurements whatsoever of the toxicity of injected aluminum in neo-nates, (or even injected aluminum in adults, for that matter), and don’t generally address the issue of aluminum at all, and don’t cite or rebut any of the many papers published in mainstream journals observing these things, or rebut or cite any of the half dozen or more epidemiological studies showing aluminum is highly correlated with autism, then I think you should conclude there is strong cognitive bias at work, if not worse.
The Amish are unique in their living styles in largely self-sustaining communities. They grow their own food.
The Amish vary greatly from one place to another. Here in Mercer County, they don’t grow much of their own food, and when they do, they can it. They do make their own milk, but they like fast food and packaged food. Storing ingredients without refrigeration, cooking fancy meals on a wood stove, and cleaning up after them with no hot running water, isn’t so simple.
Why are you responding to me? I just gave a possible explanation that I specifically said that I didn’t believe. You could post this in the main discussion to give credence to the hypothesis of the publishing bias explanation.
I could critique this if you want, although if you actually want to talk about whether or not vaccines cause autism I’d suggest posting in the open thread or starting your own post. This one is talking about publishing bias.
I don’t think this is likely, but one possible explanation is that vaccines prevent autism.
If that’s true, why didn’t one of the researchers publish a paper on that thesis? It should show up in the data they gathered.
Only if it’s statistically significant. It could be a small enough effect that they don’t notice unless they’re looking for it (if you’re going to publish a finding from either extreme, you’re supposed to use a two-tailed test, so they’d presumably want something stronger than p = 0.05), but large enough to keep them from accidentally noticing the opposite effect.
Or alternately, it’s a large effect but the rarity of autism and of non-vaccinated kids makes it hard to reach statistical-significance given sampling error. So let’s see, the suggestion here is that the reason so few studies threw up a false positive was that the true effect was the opposite of the alternative, vaccines reduce autism.
Autism is… what, 0.5% of the general population of kids these days? And unvaccinated kids are, according to a random Mother Jones article, ~1.8%.
So let’s imagine that vaccines halve the risk of autism down from the true 1.0% to the observed 0.5% (halving certainly seems like a ‘large’ effect to me), autism has the true base rate of 1.0% in unvaccinated, and the unvaccinated make up 1.8% of the population. If we randomly sampled the population in general, how much would we have to sample in order to detect a difference in autism rates between the vaccinated & unvaccinated?
The regular R function I’d use for this,
power.prop.test, doesn’t work since it assumes balanced sample sizes, not 1.8% in one group and 98.2% in the other. I could write a simulation to do the power calculation for aprop.testsince the test itself handles imbalanced sample sizes, but then I googled and found someone had written something very similar for the Wilcoxon u-test, so hey, I’ll use the samplesize library instead; filling in the relevant values, we find for a decent chance of detecting such a correlation of vaccination with reduced autism, it takes:a total n=90k. I’m guessing that most studies don’t get near that.
Of course, a lot of that penalty is going towards picking up enough kid who are both autistic and unvaccinated, so one could do better by trying to preferentially sample either of those groups, but then one gets into thorny questions about whether one’s convenience samples are representative and biased in some way...
As the original article says, if there was no effect, you’d expect a few studies to get p < 0.05 by chance. Similarly, if there was no effect, you’d expect a few studies to get p > 0.95 by chance, suggesting that vaccines prevent autism. If vaccines do prevent autism, then it would be even more likely to have p > 0.95.
Not all statistical analysis has to be preregistered. If a data has a trend that suggest vaccination might reduce autism I’m sure the researchers would run a test for it.
If the study is underpowered to find a effect in that direction it’s also like to be underpowered to find a effect in the other direction.
Can someone with more statistical expertise run a test to see whether the studies are underpowered to pick up effects in either direction?
There is fairly extensive data (not published in the peer reviewed literature) that groups which are unvaccinated have far lower autism rates than the general public.
UPI Reporter Dan Olmsted went looking for the autistic Amish. In a community where he should have found 50 profound autistics, he found 3. The first was an adopted Chinese girl who’d had vaccinations rushed before she was shipped from China and more here on the way to the adoptive parents. The second had been normal until developing classic autism symptoms within hours of being vaccinated. The third there was no information about. http://www.putchildrenfirst.org/media/e.4.pdf
Olmsted continued his search for unvaccinated Amish with autism beyond that community, finding none for a long time, but eventually found a Doctor in Virginia who had treated 6 unvaccinated Amish people from various places with autism. 4 of them had very elevated levels of mercury.
A telephone survey commissioned by the nonprofit group Generation Rescue compared vaccinated with unvaccinated boys in nine counties of Oregon and California [15]. The survey included nearly 12,000 households with children ranging in ages from 4 to 17 years, including more than 17,000 boys among whom 991 were described as completely unvaccinated. In the 4 to 11 year bracket, the survey found that, compared with unvaccinated boys, vaccinated boys were 155% more likely to have a neurological disorder, 224% more likely to have ADHD, and 61% more likely to have autism. For the older boys in the 11-17 year bracket, the results were even more pronounced with 158 % more likely to have a neurological disorder, 317% more likely to have ADHD, and with 112% more likely to have autism. [15]
In addition to the Generation Rescue Survey, there are three autism-free oases in the United States. Most publicized are Amish communities, mainly studied in Ohio and Pennsylvania [16].The Amish are unique in their living styles in largely self-sustaining communities. They grow their own food. Although they have no specific prohibitions against medical care, very rarely do they vaccinate their children. In local medical centers available to the Amish, most centers reported that they had never seen an Amish autistic child. The only Amish children that were seen as a rule were those with congenital disorders such as fragile X. The one autistic Amish child that was discovered during the surveys was taken to a medical office for an ear infection where the child was incidentally vaccinated, probably without the mother’s consent.
The second is the Florida-based medical practice of Dr. Jeff Bradstreet. While treating several thousand autistic children in his practice, Bradstreet has observed that “there is virtually no autism in home-schooling families who decline to vaccinate for religious reasons” [17]
The third, the “Homefirst Health Services” located in Chicago, has a virtual absence of autism among the several thousand patients that were delivered at home by the medical practice, and remained non-vaccinated according to the wishes of the parents [18].
Clusters of autistic children have also been found among parents with occupational exposures to chemicals prior to conception [19], and in children exposed prenatally to organochlorine pesticides [20].
excerpted from:
http://vactruth.com/2012/03/13/vaccines-human-animal-dna/
Reportedly the CDC has been surveying the vaccination status of the Amish for years, attempting to induce them to vaccinate (with some success I believe), and has consistently refused requests to include an autism question with their survey to gather data.
Its probably worth noting that Seneff et al, http://www.mdpi.com/1099-4300/14/11/2265 who have identified one pathway by which vaccines might be causing autism, have also in other work argued that glyphosate may invoke the same pathway, and the same groups may also be avoiding glyphosate. http://people.csail.mit.edu/seneff/WAPF_Slides_2012/Offsite_Seneff_Handout.pdf
He went looking for autistics in a community mostly known for rejecting Science and Engineering? It ‘should’ be expected that the rate of autism is the same as in the general population? That’s… not what I would expect. Strong social penalties for technology use for many generations would be a rather effective way to cull autistic tendencies from a population.
I don’t reject the possibility there are other explanations for the observation that unvaccinated Amish have very low autism rates. I even offered one: that they also reject Glyphosate.
However, when it turns out that the rare cases of Amish with autism that are found mostly turn out to be vaccinated, or have some very specific other cause obvious that’s not present in the general population (high mercury), the case for vaccination being a cause becomes much much stronger.
And when you realize that other groups of unvaccinated also have low autism rates, the case becomes stronger.
And when you realize that injecting the aluminum into animal models causes behavioral deficits, and injecting vaccines into post-natal animals causes brain damage, in every study I’ve found, the case becomes stronger still.
And when you discover that the safety surveys don’t cite any empirical measurements whatsoever of the toxicity of injected aluminum in neo-nates, (or even injected aluminum in adults, for that matter), and don’t generally address the issue of aluminum at all, and don’t cite or rebut any of the many papers published in mainstream journals observing these things, or rebut or cite any of the half dozen or more epidemiological studies showing aluminum is highly correlated with autism, then I think you should conclude there is strong cognitive bias at work, if not worse.
The Amish vary greatly from one place to another. Here in Mercer County, they don’t grow much of their own food, and when they do, they can it. They do make their own milk, but they like fast food and packaged food. Storing ingredients without refrigeration, cooking fancy meals on a wood stove, and cleaning up after them with no hot running water, isn’t so simple.
Why are you responding to me? I just gave a possible explanation that I specifically said that I didn’t believe. You could post this in the main discussion to give credence to the hypothesis of the publishing bias explanation.
I could critique this if you want, although if you actually want to talk about whether or not vaccines cause autism I’d suggest posting in the open thread or starting your own post. This one is talking about publishing bias.