Generation Rescue has a long history of promoting bad science. They even have tried their hand at it themselves before, with a phone survey that was so bad it would have earned a college freshman in epidemiology a failing grade.
So when they came out with their own “study” of vaccination rates around the world, you can imagine I didn’t expect it to be good. In fact, I just avoided it altogether until they sent me an email telling me how good it was.
So I looked.
It was worse than I expected. Far worse.
The “study” is here. Generation Rescue (GR) looks at the vaccine schedules for multiple countries and compares this with the infant mortality rate and autism rates in those countries.
I read it and, Oh…my…god… I expected bad science and poorly/biased interpretations. Instead, what I found was pretty clear evidence that Generation Rescue is knowingly distributing misleading information.
Before you get worried that this post is way long and question whether you really want to read the details, here’s the short version:
1) They compare infant mortality rates between the US and other countries–even though it is clear (according to their own expert no less!) that the US uses different criteria for infant mortality and it isn’t accurate to compare the US infant mortality to that in other countries.
2) They compare autism rates amongst countries to show the US has the highest rate, suggesting that the higher the number of vaccines the higher the autism rate. They just “forget” to tell you that the prevalences for the other countries are from old studies. We can debate why the reported autism prevalence is going up with time, but no one debates that the older studies report lower prevalences than we see now. So, why does Generation Rescue compare prevalence in the US using 2002 data for kids born in 1994 with, say, a Finnish study using 1997 data on kids born as early as 1979? I consider them very biased, but not incompetent enough to miss those fatal mistakes in their study.
3) They claim that the US has the highest vaccination rates and the highest autism rates. They conveniently ignore prevalence from Canada and the UK, which have comparable prevalences to the US and much much lower numbers of vaccines. Yes, you read that right, they left out the well known studies that would show that their conclusions are nonsense.
The worst part is that it is almost certain that Generation Rescue didn’t make an honest mistake. These are so obvious that whoever wrote that “study” had to know he/she was producing what amounts to the lowest form of junk pseudoscience.
For those who want the gory details, here they are:
Infant Mortality Rates
Generation Rescue points out that the reported infant mortality rate is highest in the United States, which also has the most childhood vaccines. All well and good, but can we really compare the infant mortality rates from country to country?
When I type infant mortality rate into a google search, the first hit is a Wikipedia page which, as it turns out, addresses exactly this question.The answer is a resounding “NO”, we can’t compare the US infant mortality rate with that of other countries.
While the United States reports every case of infant mortality, it has been suggested that some other developed countries do not. A 2006 article in U.S. News & World Report claims that “First, it’s shaky ground to compare U.S. infant mortality with reports from other countries. The United States counts all births as live if they show any sign of life, regardless of prematurity or size. This includes what many other countries report as stillbirths. In Austria and Germany, fetal weight must be at least 500 grams (1 pound) to count as a live birth; in other parts of Europe, such as Switzerland, the fetus must be at least 30 centimeters (12 inches) long. In Belgium and France, births at less than 26 weeks of pregnancy are registered as lifeless.
So, who wrote that 2006 article in US News & World Report?
Yep, the same Bernadine Healy that is Generation Rescue’s favorite “mainstream” doctor.
One has to believe that GR saw that article in Wikipedia and the US News article. They are, after all, Google Ph.D.’s. Given the author was Bernadine Healy, they have to have considered it accurate, don’t you think? And, yet, GR conveniently forgets to mention the differences in how the US and other countries count infant mortality in their vaccines cause autism “study”.
Autism Rates 1: Autism Prevalence by country
Start with the conclusion of the Generation Rescue “study”:
This study appears to lend credibility to the theory that the U.S. vaccine schedule is linked to the U.S. epidemic of autism, particularly when compared to the published autism rates of other countries.
Given this bold claim, it is critical that they use good data for the autism rates. By “good” I mean that they need data that they can accurately compare to the CDC reported prevalence of 1 in 150. That data was taken in 2002 on 8 year old children. I.e. kids born in 1994. Since reported prevalence numbers are going up with time, it would be very misleading if they were to use, say, prevalence numbers from the early 1990′s, wouldn’t it?
Any prevalence that they use would have to use prevalence numbers from about the same time, on kids of about the same age.
Here’s their table comparing the autism rates.
Let’s take a look at the studies they cited for their numbers, shall we?
Iceland: Prevalence of Autism in Iceland. This 2001 study uses kids from birth years 1984-1993. I.e. most (if not all) of the kids are from the time before the big upsurge in autism diagnoses. Hardly a good comparison to the 2002 CDC study, eh?
For Sweden, they use a paper called, “Is autism more common now than 10 years ago?” from The British Journal of Psychiatry. Published in… 1991. That’s pre DSM-IV. Amongst other problems, they won’t be including the other PDD’s in the autism spectrum, like the CDC study does. Besises, the kids from the CDC study weren’t even born yet, it was so old! Is there any wonder that the Swedish study shows a lower prevalence?
For Japan, they use a paper titled Cumulative incidence and prevalence of childhood autism in children in Japan. The study uses data from 1994 on kids who were born in 1988.
Are you starting to see the pattern here? Time after time, GR is comparing US 2002 prevalence data to much older data from other countries. Let’s go on:
For Norway, they use the paper Autism and related disorders: epidemiological findings in a Norwegian study using ICD-10 diagnostic criteria. The paper was published in 1998 on children 3-14 years of age. Simple math suggests they had kids with birth years going back to at least 1984 in that study. Hardly a good comparison to kids born in 1994.
For Finland, they use Autism in Northern Finland. Here is an updated version from 2005. The study uses data from 1996-97, on kids up to 18 years old. I.e. they are using kids that were born as early as 1979. Also, they are using data on patients from hospital records who used “communal health services”. Sounds a lot like “inpatient”–one of the critiques that GR uses against studies from Denmark. Also, the Finland study didn’t include Aspeger syndrome, as that was a new diagnosis at the time. Hardly a good comparison to the CDC study.
For France, they use Autism and associated medical disorders in a French epidemiological survey. This uses “French children born between 1976 and 1985″.
For Israel, they use Autism in the Haifa area–an epidemiological perspective. This paper looks only at autistic disorder (no PDD-NOS, no Aspergers, no Rett’s no Childhood Degerative Disorder). Right off the bat that reduces the prevalence and makes it impossible to compare the the CDC 2002 study. The Israell study also is, you guessed it, based on kids older than the CDC study: children born between 1989 and 1993.
Last, Denmark. If you’ve been following the thimerosal debate, you know this is going to be ironic. They use Madsen’s paper, Thimerosal and the Occurrence of Autism: Negative Ecological Evidence From Danish Population-Based Data. Generation Rescue refers to this study (incorrectly, I might add) as “This one goes beyond useless”. I guess “useless” is only when it is used to refute the thimerosal hypothesis? Come on, GR, this level of hypocrisy is just painful.
There are some very well known studies that Generation Rescue somehow forgot to include in their “study”. Could this be due to the fact that they are very good counterexamples to the vaccine-hypothesis ? Let’s look at some and see, shall we?
United Kingdom: Pervasive Developmental Disorders in Preschool Children: Confirmation of High Prevalence ( study performed in 2002 with a prevalence of 1 in 170), and Pervasive developmental disorders in preschool children (study performed in 1998/9 with a prevalence of 1 in 160).
Canada: Pervasive Developmental Disorders in Montreal, Quebec, Canada: Prevalence and Links With Immunizations (birth years 1987 to 1998. Prevalence 1 in 154).
Wow, the United Kingdom and Canada have prevalence numbers comparable to those in the US!
So, let’s complete the comparison, shall we? What is the vaccine schedule like for the UK and Canada? Using the Generation Rescue “study” we get 20 vaccines for Canada and 21 for the UK.
Wow, that’s way less than the US (with 36), and they have the same autism prevalence as the US? How could that be? Is it, perhaps, that the autism is NOT related to the number of vaccines in a given country’s schedule?
Anyone doubt why GR left the UK and Canada off their table of Autism Prevalences Around the Globe? No, I am not giving them a pass that this could be an honest mistake.
To quote Generation Rescue’s top funny guy (Jim Carrey), “How stupid do you think we are?”