Critical Evaluations of "Impacts of Environmental Factors on the Prevalence of Autistic Disorder after 1979" , Deisher et al. 2014

The very shortest version: no, "fetal DNA in vaccines" isn't present; even it were, it would not be a likely cause of autism.

Note, September 19 2014: I have continued to add to critiques as they have been published

The Sound Choice Pharmaceutical Institute (SCPI) has some unusual notions about autism causation relative to vaccines. SCPA recently published a paper, Impacts of Environmental Factors on the Prevalence of Autistic Disorder after 1979, by Theresa A. Deisher and colleagues. In short, SCPI has the notion that there's "fetal DNA" left in vaccines which use viruses grown on human cell lines. Once the vaccines are injected into humans, the alleged DNA can somehow invade neurons, recomine with the subject's DNA, and damage the neurons.

This idea is biologically speaking, completely implausible.

This does not seem to deter Deisher and her company from advancing the idea, or pursuing the idea that vaccines can be made for sale without the use of human cell lines.

2014 Critiques of Deisher's ideas:

Abortion, Autism and Immunization: The Danger of the Plausible Sounding Lie by Genevieve H. at Rational Catholic (September 8, 2014)

There is no plausible way that [something injected after birth could cause autism] could happen. It makes absolutely no sense with what we know to be true about autism. Even though to the layperson it sounds convincing that foreign DNA injected into a developing child could somehow go to the brain and cause autism, there is no credible way that that could actually be the case. Not with the overwhelming evidence that autism begins before birth. This is yet another in a series of plausible sounding untruths about vaccines that are so convincing to the average, intelligent parent just trying to learn more....

I do not know why the authors of this article published the conclusions that they did, or why they chose to publish their work in such a questionable publication. In no way do I want to impugn their characters or call into question what are probably deeply held beliefs on their parts. However, deeply held beliefs do not make for rigorous scientific inquiry. And pro-life parents seeking to do the best by their children and by their culture deserve better than to have a plausible sounding lie masquerading as truth.

Religious fundamentalists try to prove fetal DNA in vaccines causes autism and fail by Orac at Respectful Insolence (September 9, 2014). This blog post goes into deep detail about the kinds of laboratory techniques used for finding RNA and DNA, and the shortcomings of Deisher's team, and then addresses the correlation issue

Autism is usually diagnosed between ages 2 and 4; so, unless the power of these evil tainted vaccines to contaminate the DNA of our precious children can also travel back in time, it’s hard to take correlations between these change points and vaccine introduction as anything more than spurious pseudo-correlations. It would be so hilarious if the consequences of such fear mongering weren’t so dire, although even then it’s still useful as a cautionary tale worthy of extreme mockery of how not to do linear regression and inflection point “hockey stick” analysis.

Dr. Deisher, vaccines and autism: three red flags, by Simcha Fisher at Patheos (September 11, 2014)

Deisher commits the cardinal sin of scientific research: she confuses correlation with causation. There is even some question whether even the correlation she posits actually exists.

This is bad science....

But if we are going to hesitate over any vaccine, let it be for medically and theologically sound reasons. Dr. Deisher has not provided either. Her study displays a conflict of interest; it banks on an emotional response to an ethical question; and most damningly, it relies on heavily flawed science. Catholics should keep their eyes open for other scientists whose research is sound.

Problems with Deisher’s study— Part I: The numbers by Laura C. at Rational Catholic (September 12, 2014)

If someone were to ask me to critically appraise the scientific merits of a study arguing that birth control usage caused earthquakes by changing the mating behaviors of fish, I might have had an easier time than I did with Dr. Theresa Deisher’s widely shared study. That may be over the top, but it is kind of true. In such a scenario, I would have been able to say to myself, like most other people in the scientific blogosphere are doing with the Deisher study, “This is just too absurd, too poorly done to waste any more time refuting it!” But, I can’t do that because you, my brothers and sisters in Christ, deserve to know how badly you are being deceived by a study that is so abominable, it would be an insult to bad science to call it bad science.

[in-depth analysis of Deisher's statistical failings]

Stay tuned for criticism of the biological plausibility and synthesis of Deisher’s conclusions…

Looking a little closer at the numbers— A supplement to Part I by Laura C. at Rational Catholic (September 17, 2014)

There are some things I want to make perfectly clear: I chose to operate under many of Deisher’s assumptions that I knew weren’t particularly or necessarily valid because I didn’t need to exit that paradigm to point out the flaws in the rationale. However, there are more reasons to think Deisher’s analysis is flawed from the outset and I’d like to address just a few.

Autism And Stem-Cell Derived Vaccines: Deisher’s New Paper, by Matt Briggs at William M. Briggs (September 18, 2014)

We’re finally back to Deisher, who from her paper does not appear to appreciate these and similar points. I find the paper poor in conception, argument, and quality, and regard her main contention as unproved (which is logically consistent with it is still possibly true).

Vaccines and Autism: This is Starting to Get Really Weird by Joseph Moore at Yard Sale of the Mind (September 19, 2013)--Evidently, feelings about Deisher and her paper, pro and con, are running quite high in a segment of the blogosphere I can best describe as Roman Catholics doing science. Mr. Moore describes.

Problems with Deisher’s Study— Part II: Biological Implausibility by Laura C. at Rational Catholic (September 24, 2014)

Deisher’s study is incredibly, incredibly thin in the realm of biological plausibility. This is surprising (or not) because she is making some novel, extraordinary claims. Yes, she has a nice bibliography salad, but the studies she cites do not directly, or even indirectly at times, support her central hypothesis that DNA from fetal cell lines is a direct environmental cause for increasing autistic disorder (AD) diagnoses. Further, even her unpublished research that has been cited in newsletters and pro-life media means very little in terms of the hypothesis.

Autism and The Sins of the Fathers by anonymous blogger at Pentimento (September 27, 2014)

The takeaway from Deisher's study -- at least as it's being expressed throughout the Catholic blogosphere -- is that autism must be cured (if not eliminated), and that, in fact, autism can be avoided (if not eliminated) if the rubella vaccine, which was derived from the stem cell line of an aborted baby more than fifty years ago, is no longer used. This assumes that autism is a Very Bad Thing, devoutly to be un-wished for, and that it's worth risking the deaths of countless babies (other people's babies; it always is) in utero to avoid it. This is not just theologically faulty; it's morally faulty.

Previous critiques of Deisher's work and assumptions

A Fishing Expedition at the Vaccine Court, by Sullivan (Matt Carey) at LeftBrainRightBrain Autism (July 14, 2013)

[Deisher] did however, apply for an NIH grant to perform this research. The petitioners claim that the controversial nature of the study resulted in it not being funded. The referee reports, however, were clear that the planned study was weak and Ms. Deisher’s skills were not strong in epidemiology and statistics (among other weak points).

Although petitioners make assertions to the contrary, the evidentiary record before the undersigned contains a withering assessment of Dr. Deisher’s ability to competently lead the proposed study. Petitioners here seek extraordinary relief, and the undersigned is reluctant to substitute her scientific judgment for that of the NIH reviewers—a panel of Dr. Deisher’s peers—who have found her proposed study to be critically deficient. In the undersigned’s view, the NIH reviewers’ comments merit weighted consideration.

Ms. Deisher’s previous research has focused on "changepoint” analysis of autism prevalence data. She follows the method set forth by two people at the FDA who presented such a changepoint analysis previously. I found that analysis lacking and submitted a comment to the journal on it. I find Ms. Deisher’s analysis lacking as well.

Iterative Hockey Stick Analysis? Gimme a break! By Mark Chu-Carroll at Good Math, Bad Math (April 29, 2010). In this post, Chu-Carroll analyzes the presentation of Sound Choice Pharmaceutical Institute's data, from a company publication.

Let’s summarize the problems here.

They’re using an iterative line-matching technique which is, at best, questionable.

They’re applying it to a dataset that is orders of magnitude too small to be able to generate a meaningful result for a single slope change, but they use it to identify three different slope changes.

They use mixed datasets that measure different things in different ways, without any sort of meta-analysis to reconcile them.

One of the supposed changes occurs at the point of changeover in the datasets.

When one of their datasets shows a decrease in the slope, but another shows an increase, they arbitrarily choose the one that shows an increase.

Thermonuclear stupid about vaccines from someone other than Jenny McCarthy, by Orac at Respectful Insolence (July 22, 2009)

....[Deisher's argument] doesn’t make a lot of sense, strictly from the standpoint of a temporal correlation. After all, these cell lines were derived over 40 years ago. If there was a correlation between DNA from these cells in vaccines and autism (or any other of the problems blamed on vaccines), wouldn’t it have started decades before the early 1990s?

.... why this doesn’t happen frequently from the many viruses humans are exposed to each and every day or after a blood transfusion. Or what about childbirth? There is almost always some mixing of fetal blood with the mother’s blood upon childbirth, meaning that the mother is exposed to fetal DNA from white blood cells and monocytes in the fetal blood. Why is it that mothers don’t all get anti-DNA autoimmune diseases after childbirth?

Note: It is true that Deisher spent from 1980 to 1990 as a student at Stanford University, first as an undergraduate and then as a PhD candidate in the Department of Molecular and Cellular Physiology. However, she has not been associated in any way with Stanford for 24 years, so any publication referring to her as a "Stanford scientist" should be viewed with deep suspicion.

Image source: http://www.collective-evolution.com/2014/09/05/human-fetal-dna-fragments-in-vaccines-are-a-possible-cause-for-autism-according-to-this-stanford-scientist/ Image description: Black on white headline reading "Human Fetal DNA Fragments in Vaccines are A Possible Cause For Autism --According to This Stanford Scientist."