I've been a bit remiss when it comes to writing about the lunacy in which it is claimed that vaccines cause autism, allegedly due to the mercury in the thimerosal preservative that was in most childhood vaccines until the end of 2002, when it was removed from all but flu vaccines. It turns out that the class action suit by parents who think that vaccines caused their children's autism will be going to court in June. Hearings for this suit, known as the Autism Omnibus, will mark a new phase in the pseudoscientific pursuit of "compensation" for nonexistent "vaccine injuries." Even though science and epidemiology have clearly refuted the thimerosal/autism hypothesis (and, as I like to say, dignifying the idea as a "hypothesis" is something that now causes me to choke on the word, given how poor the evidence supporting the idea is and how copious the evidence refuting it is), there is actually a possibility that these parents might win, as Arthur Allen described in Slate earlier this week:
In June, the U.S. Federal Claims Court, across Lafayette Square from the White House, will begin hearings on 4,800 claims filed by parents of children on the autism spectrum who think that the government's vaccine program caused their children's disorders. The scientific consensus rejects the idea that thimerosal, a mercury-containing preservative used in vaccines, causes autism. Still, it's conceivable that some of the claimants could win, because the vaccine court requires a lower standard of scientific evidence than regular courts. And so the parents are trying to enhance the legitimacy of their arguments.
In April, the government-funded Institute of Medicine held a two-day workshop to discuss ways to research possible toxic causes of autism. Leading voices among the parents who believe in the thimerosal-autism link shared their views with Science publisher Alan Leshner, who ran the meeting, as well as senior government scientists. Two of the groups, Safe Minds and the National Autism Association, later issued a news release that appeared to distort the remarks of a CDC scientist to make it appear that he shared their views. The meeting probably wouldn't have taken place without the support of several members of Congress, including Sen. Joe Lieberman, D-Conn., and Rep. Dan Burton, R-Ind. Other activists have taken to harassing scientists whose results they don't like.
I've actually written about the latter incident to which Allen refers. He's talking about Paul Shattuck, whose study showing that the "autism epidemic" is very likely due to nothing more than broadening of the diagnostic classification criteria for autism and autism spectrum disorders (ASDs) resulted in the knives coming out for him by members of the mercury militia. What I particularly like about Allen's take on this is how he nails the intellectual dishonesty that is Mark and David Geier:
Then there is the activists' reliance on Dr. Mark Geier, a fixture as an expert witness in vaccine court, where he has testified about 100 times. Geier and his son, David, who holds an undergraduate biology degree, operate under various business names from a house in suburban Maryland. The special masters who run the vaccine court have tossed out their testimony on 10 occasions, and federal district courts have been similarly skeptical. One judge recently described Geier as "intellectually dishonest," and a special master called him "a professional witness in areas for which he has no training, expertise, and experience."
eier and his son have published several journal articles claiming to show a link between autism and vaccines containing the mercury-based preservative thimerosal. But the papers have been contradicted by study after study, and the mainstream medical community has proclaimed their work on the subject to be bunk. (The Geiers did not respond to e-mail and telephone requests to be interviewed for this article.)
"Bunk" is putting it kindly. As I've documented before, the Geiers perform highly dubious research that involves fitting curves to inappropriate data based on preconceived notions in a ludicrously simplistic manner (or, as I like to call it, "dumpster-diving"). More recently, they've been treating autistic children with Lupron, a very powerful sex hormone suppressing drug that is most commonly used to treat men with prostate cancer by blocking testosterone production or in women undergoing in vitro fertilization to suppress estrogen and other hormones and allow total control of hormone levels (and thus control of maturation of egg follicles and prevent premature rupture) by exogenously administered hormones. It's also occasionally used for chemical castration for various reasons. All the while, the Geiers have been publishing in crap journals like Medical Hypotheses, which intentionally publishes speculative "hypotheses," and the rabidly antivaccination "libertarian" wingnut publication Journal of American Physicians and Surgeons, which has in the past published anti-immigrant screeds and rants against peer review. When they venture into legitimate peer-reviewed journals, they somehow managed to find a journal where the co-editors are both listed as experts for the plaintiffs in the Autism Omnibus.
Allen's article is generally excellent and well worth reading, as he discusses the way the Geiers set up a rubber-stamp IRB to "approve" their "clinical trial" with the Lupron protocol, as I've discussed before. One thing that I've wondered ever since Kathleen Seidel informed me of the Geiers' mendacity is how the Geiers could get away with such an obviously dubious "study." Given the lack of action by the FDA over Jim Tassano's selling of home brew dichloroacetate, I'm beginning to understand. For one thing, the Geiers do not receive government funding, nor are they trying to get a drug FDA-approved, which means that they may not be subject to the Common Rule, although Maryland state law may say otherwise. For another thing, it requires that someone report them. Also, they are using an already approved drug for an "off-label" indication. They can't hide behind that excuse, however, because (1) there is no legitimate medical indication for the off-label use of Lupron in autism and (2) they are trying to publish their results in medical journals and going through the motions of having their research approved by an IRB. It is research. Incompetent, unethical, and vile research, but research. Worse, it's painful. As Allen points out, it's not unusual for autistic children under the Geiers care to get 60 injections a month.
One thing that Allen did report that I didn't already know is that the Geiers are now reportedly using Androcur in some children, a drug that is not FDA-approved and thus not available in the US, as Lupron is. It does, however, appear to be available from online pharmacies. If the justification for using a drug like Lupron in children is exceedingly tenuous, the use of a drug like Androcure is even more so. Besides not being approved for use in this country, its known side effects include depression, liver damage, and blood-clotting abnormalities.
Finally, lately, the Geiers have been trying to publish in more "respectable" journals, and, indeed, they've managed to get a couple of their usual low quality papers into actual peer reviewed journals. No doubt this is for the purpose of the Autism Omnibus hearings. Real experts recognize that Medical Hypotheses and JAPandS are justifiably not taken seriously. The first is simply a sounding board for "out there" ideas, and the seccond is utter crap. If they are to be taken the least bit seriously as "expert" witnesses, it is imperative that they publish in real journal, in particular in the J. Toxicol. Environ. Health and Neuro. Endocrinol. Lett, the latter of which is particularly disturbing because it references the Geiers' bogus IRB as having given approval to the study. I've been meaning to look at these papers in more detail. I think I've just been procrastinating because it causes me pain to actually read a Geier paper, with its invariable bad science and now, with their Lupron work, dubious ethics. I feel dirty after reading one. However, with the the Autism Omnibus getting into high gear, I may not have the luxury.
I may have to dive in and get dirty, just like Mike Rowe.
More on the topic:
I'm not familar with medical journals, but in the last year they have published the following which look like articles in decent journals:
Geier, D.A. and Geier, M.R. (2006) A clinical trial of combined anti-androgen and anti-heavy metal therapy in autistic disorders. Neuroendocrinology Letters, 27, 833-838
Geier, D.A. and Geier, M.R. (2006) A meta-analysis epidemiological assessment of neurodevelopmental disorders following vaccines administered from 1994 through 2000 in the United States. Neuroendocrinology Letters, 27, 401-413.
Geier, D.A. and Geier, M.R. (2006) A clinical and laboratory evaluation of methionine cycle-transsulfuration and androgen pathway markers in children with autistic disorders. Hormone Research, 66, 182-188.
Geier, D.A. and Geier, M.R. (2006) A prospective assessment of porphyrins in autistic disorders: A potential marker for heavy metal exposure. Neurotoxicity Research, 10, 57-63.
Geier, D.A. and Geier, M.R. (2006) An assessment of downward trends in neurodevelopmental disorders in the United States following removal of thimerosal from childhood vaccines. Medical Science Monitor, 12, CR231-CR239.
Geier, D.A. and Geier, M.R. (2006) An evaluation of the effects of thimerosal on neurodevelopmental disorders reported following DTP and Hib vaccines in comparison to DTPH vaccine in the United States. Journal of Toxicology and Environmental Health - Part A - Current Issues, 69, 1481-1495.
The above strikes me as a fairly decent publication record. In the couple of years prior to 2006, they were also publishing, with a similar frequency, in Prenatal Diagnosis, Autoimmunity, Medical Science Monitor, Medical Hypotheses, International Journal of Toxicology, Clinical and Experimental Rheumatology, Brain and Development, Hepato-Gastroenterology and Expert Opinion on Pharmacotherapy.
I'm not really making any comment on their science, rather that it does seem a little unfair to charge them with primarily publishing in shitty journals.
The leading court cases regarding standard of proof for Vaccine Compensation Act claims have held that claims should not automatically be thrown out on the basis that claimants can't point to supporting peer-reviewed medical literature. However, claimants must still support their claims by a "preponderance of the evidence." That is, they must show it is more likely than not that the vaccine in fact caused the injury, in the context of a proceeding where the opposition is presumably providing evidence to the contrary.
It seems to me the autism claimants may have considerable difficulty establishing causation when there is no chain of events leading from vaccination to autism that makes medical/scientific sense. But maybe it's just my low tolerance for bullcrap that makes me think that way.
SteveF: Take a look at some of the articles on the Neurovisersity weblog at www.neurodiersity.com. Kathleen Seidel goes to great lengths to document that much of what Geier publishes is regurgitation (a/k/a semi-plagiarized). The DOJ will have a field day with that on cross examination.
True that the claimant's must prove their claim by a preponderence of the evidence. One of the key things that they need to prove is biological plausibility. They cannot. Further, the Special Masters have ruled that they will apply the Daubert standard after hearing the evidence.
And, of course, the entire Omnibus may be derailed as described by Kevin Leitch.
Normally, where scientific research is published should make little difference. Unfortunately it does. The main difference in these journals is the caliber of papers they receive, who reads and cites them, and what peer reviewers are willing to peer review the papers.
If a paper has passed peer review, that does not mean that it is correct, it only means that it is "plausible", that is, there are no obvious glaring errors that are readily apparent to the reviewers who have reviewed it. The reviewers are people too, they have their own biases, they make errors, they can be fooled by authors, they can be fooled by themselves.
There are competing pressures on editors. They need to fill the journal they are going to publish with enough papers, they have to fill it with papers that readers will want to read, and who would be willing to spend money to read. Many journals charge for papers they publish, that is, to publish a paper, the author needs to pay what are called "page charges".
Scientific publishing doesn't make authors or editors or or peer reviewers rich. To review a paper is going to take some time. The paper has to be read carefully, and understood, the experiments have to be understood and checked, the reasoning checked and understood, the reviewer may need to review some of the cited papers, the reviewer needs to make sure that this is actually something new, and not something that has been done before.
In this case, the Geiers have made such a big deal out of their work being "peer reviewed", that the caliber of journal they are publishing in is relevant. Peer review has never been a "stamp of approval", or a "this is correct" mark. The real mark of a paper's value is how much it is cited in the future by others.
Scientific papers are meant to be the communication from one scientist to another scientist about scientific results. Non-scientists need to be extremely careful when using scientific papers.
I have read most of the Geier's papers, and they are crap. To me, they verge on fraud. The data is very selectivly used, the statistical methods are often used inappropriately. I recall one paper where two data sets were compared, one of 221, and one of 18. Each data set was characterized by 4 numbers (low, mean, standard deviation and high). For the two data sets they were (0, 4.06, 8.59, 58.65) and (0, 1.29, 1.54, 6.2). The differences in the mean (4.06 vs 1.29) were taken to be "significant", and representative and characteristic of differences between the entire groups. The distribution is obviously highly skewed. The standard deviation is higher than the mean.
To my mind, there is no way that a fair minded scientist could say that was a valid statistical analysis. That is not the way that science is supposed to be done. Science is supposed to be open and transparent, that the methods, the data, the analysis of the data, and the conclusions are supposed to be laid out in complete detail. Science is about the path, not about the conclusion. A "conclusion", or an "answer" reached via trickery or error is not science.
The only appropriate way to deal with such highly skewed data is to present the data in a graph. That would show what the actual distribution was and how appropriate is the comparison between the two goups. That the Geiers chose to not present the data shows (to me), that they were trying to justify a conclusion inapropriately, rather than to present hypothesis, data and conclusions.
TheProbe,
I'm not particularly trying to defend the Geiers. I don't know a great deal about this issue but instinctively side against conspiracy theorists, denialists and general woo-practitioners. However, I did think it reasonable to point out that the Geiers have published more in the mainstream literature than perhaps Orac implies.
On the subject of regurgitation, I did kind of think something along these lines, just from reading titles and quickly scanning abstracts. To play devils advocate to a certain extent, this is not uncommon practice. For example, it is reasonably common for a PhD thesis to be sliced up fairly thinly and then relatively similar papers sent out to different journals. This is often encouraged, particularly in the UK, by the need to have lots of papers out there in order to satisfy research assessment exercises (which I guess the Geiers aren't burdened by, not being in academia).
On the subject of citations of the Geiers papers, a key point as daedalus points out, my quick search suggests that they aren't having much of an impact. If something were highly significant, then a raft of citations would follow. In a number of instances, citations seem to come about as a result of the controversy that these guys build up (comment and reply, that kind of thing). There are a few instances of this in the Annals of Pharmacology, earlier in the decade for example.
Having said that, some of their papers are being cited in a more regular fashion. For example, their 2002 paper "Clinical implications of endotoxin concentrations in vaccines", was cited by 5 articles (plus self-cites). I couldn't be bothered checking them all out but at least one was a complimentary cite:
"Finally, the use of eukaryotic expression systems (i.e., Saccharomyces cerevisiae, [36] or Uncinocarpus reesii [41]) for production of the chimeric vaccine overcomes the potential adverse effects of endotoxin contamination associated with bacterially expressed proteins (18)."
(18 is the Geiers paper)
From:
Tarcha, E.J. et al. (2006) Multivalent recombinant protein vaccine against coccidioidomycosis. Infection and Immunity, 74, 5802-5813.
I guess it's possible that in certain circumstances, their science might be valid?
Even though science and epidemiology have clearly refuted the thimerosal/autism hypothesis (and, as I like to say, dignifying the idea as a "hypothesis" is something that now causes me to choke on the word, given how poor the evidence supporting the idea is and how copious the evidence refuting it is), there is actually a possibility that these parents might win,]...
I think this is largely due to the fact that many non-scientists, regardless of their intelligence or education, see this as an issue of competing scientific viewpoints, rather than bad science v. good science. SteveF's comments are rather indicative of this notion. Note that he is rather skeptical of woo, yet questions whether there is a grain or more, of truth in what the Geirs claim.
They have published a lot of really "sciency" sorts of papers, that many in the general public will accept on the basis of the abstract, if they even read that much. Even if they are predisposed to be skeptical of the Geirs claims, they will be hesitant to dismiss them outright. Others, who have a predisposition to assuming the worse about EBM and big pharma, will just be glad to find "evidence" that there is a huge conspiracy perpetuated by big pharma - really the same mentality behind the DCA hype.
To be honest, I was more than a little leary of vaccinating my son, before he was born. It wasn't until I had a long discussion with the doctor I grew up with, that I understood that the anti-vax hype was a crock. It wasn't that he wa the first doctor to tell me that, it was that he was the first one whom I trust and who (knowing me well) explained the flaws and cherry-picking in the anti-vax "studies." It was not actually reading the studies supporting vaccination that did it for me, for the most part they would be relatively meaningless to me. I work in construction with a fair grounding in structural engineering. I am planning on taking some classes in statistical analysis, but even that will only provide some ability to recognize bad data analysis. Without a background in the science involved, it will still be possible to be deceived by bad science.
In the end, like many others, I am dependent on the analysis of those who know a lot more about a subject and whom I trust. The problem when these issues go to court, is that judges and juries are required to listen objectively to the evidence provided by both sides. It still boils down to trust, but that trust is limited to what each side presents and how they present it in court. Neither judge, nor juries, have the right to depend on the word of people they know or have experience with, outside the courtroom. If the Geirs come off as more trustworthy than the experts slated against them, the actual quality of the evidence won't matter a bit.
As an aside, this is exactly why I have such a strong appreciation for science blogs and especially medical blogs.
Those papers, and I mean all of them, suck. The journals in which they are published, and I mean all of them, either suck or are crapping themselves because they now know just how bad they blew it by publishing "work" from individuals with pretend IRBs (their secretary, a reverend, and a gomer with a soft BA in History and biochem), who have been known to falsify credentials (affiliated with GWU - not!), who blatantly "borrow" ideas and phrasing (from themselves and others), who blatantly invent new science (mercury-mediated testosterone sheets), who have massive financial stakes in the pretend treatments (this one is key - the vultures have filed patent paperwork for their pretend cure using a chemical castration drug with TAP as a partner).
Those rubberstamping their crap, even in these crap journals, are complicit in the scam to endanger autistic children for fun and profit.
Although I find DuWayne's comment interesting (and agree with much of it), I am lightly offended by it. I am a scientist, or at least a PhD student, and can generally do a pretty good job of distinguishing good science from bad, particularly if I have taken the time out to investigate something.
In this particular instance, I have very little background whatsoever. My first post was simply a literature search, with the second being much the same. However, in the latter I did a mild bit of speculating right at the end. This was on the basis that some scientists do appear to have found at least one of the Geier's papers to be of some use. I have also just checked out one of the other references for their 2002 paper and it is similarly complimentary (and published by presumably reputable authors, in a decent journal).
Interestingly, this paper, whilst being anti vaccination is not about mercury and autism (the most controversial subject?), rather endotoxins. Perhaps when they are in other areas, they have a stronger case. Again, I am indulging in a little gentle speculation, but I don't think this is unreasonable, or indicates that I am incapable of distinguishing the existence of good vs. bad science.
I have read the Geier's papers carefully, and they are crap. There is lots of crap even in mainstream scientific journals. Usually that crap is due to simple error, and not due to fraud, or deliberate intent to deceive. In the case of the Geiers' work, it is a deliberate intent to deceive (IMO).
For the Geiers' work to be "valid", a lot of other stuff in the literature would have to be discarded.
I have posted an analysis of the false "mercury causes autism" idea (not a hypothesis because it is not consistent with much that is well known, an idea that has been falsified is not a hypothesis).
"SteveF just got fat Geier to draw wood"
I'm not entirely sure I understand this!
Daedalus,
Good take-down, thanks for the read. Have you read the paper that I was speculating on possible merit (my speculations of merit aren't especially worthwhile since I'm a geologist and haven't done any biology in years)? It's here in case you are interested:
Geier, D.A. and Geier, M.R. (2002) Clinical implications of endotoxin concentrations in vaccines. Annals of Pharmacotherapy, 36, 776-780.
BACKGROUND: A previous study suggested that high concentrations of endotoxin may be present in whole-cell diphtheria/tetanus/ pertussis (DTP) vaccine, and the scientific literature contains many studies examining the reactivity of whole-cell DTP vaccine. The medical and scientific communities have previously reported that the presence of endotoxin in commercial vaccines may have negative effects on vaccine recipients.
OBJECTIVE: To determine the endotoxin concentrations in whole-cell DTP, acellular DTP (DTaP), and DT vaccines and determine the clinical experience with each vaccine.
METHODS: To study the endotoxin concentrations in vaccines, the Limulus amebocyte lysate (LAL) assay was used. The vaccines analyzed with the LAL assay were whole-cell DTP vaccine lots manufactured by Connaught, Lederle, the Michigan and Massachusetts Departments of Health, and Wyeth; DTaP vaccine lots manufactured by Merieux and Takeda; and DT vaccine lots manufactured by Wyeth and Lederle. The incidence of adverse reactions following whole-cell DTP, DTaP, and DT vaccines were determined based on analysis of the Vaccine Adverse Events Reporting System (VAERS) database.
RESULTS: The results of the LAL assay showed that whole-cell DTP vaccines contained considerably more endotoxin than either DTaP or DT vaccines. The VAERS showed that statistically significantly more adverse reactions were associated with whole-cell DTP vaccine than DTaP or DT vaccines.
CONCLUSIONS: This analysis confirmed higher concentrations of endotoxin in whole-cell DTP vaccines compared with DTaP or DT vaccines. As high concentrations of endotoxin may be correlated with a higher incidence of adverse events, the switch from wholecell DTP to DTaP for routine vaccinations in the US seems well justified.
I seem to recall the Geiers took credit (on some level) for the switch to the acellular DTP, though a literature search may reveal a less integral role.
No surprise that there should be less (if any) endotoxin in DTaP but I'd be surprised to learn that the Geiers were the first to make this revalation. More likely this was meant to prop up an hypothesis du jour or entered as evidence in lawsuit.
Going way back, even Mark Geier's work on bacterial gene transfer is questionable.
The Geiers are trying to take credit for the "discovery" that vaccines without endotoxin have fewer side effect than vaccines with endotoxin? Huh? That has been known since forever. Why did the vaccine manufacturors take the endotoxin out?
The LAL test replaced the rabbit test, where rabbits were injected with stuff and their fever measured.
All injectables have been tested for endotoxin since forever. You can't have anything that is injectable unless you do.
Endotoxin is a (fairly) specific lipopolysaccharide (LPS) that forms the cell wall of Gram-negative bacteria. Essentially every Gram-negative bacteria has it (though there are some differences). Some, like from E coli are pretty bad, picograms will cause fevers. It is preventing endotoxin from being generated in multiple unit vaccines that requires preservatives to be used. A single bacteria can grow and produce a lethal dose of endotoxin in less than a day. You should read the CDC report on thimerosal
http://www.fda.gov/cber/vaccine/thimerosal.htm
These children died from endotoxin from bacteria growing in the vial of vaccine. I believe there was another death from that incident, I think the doctor who injected the children commited suicide.
SteveF,
When people write articles, sometimes they cite a paper after only having read the abstract. Luckily, my mentor did a good job of pounding it into my thick skull that this was not to be done.
This can happen when the writer is told that they need a citation for something, especially if it is supposed to be common knowledge. Do a quick lit search, put down a citation, move on. This really makes you look sloppy in the eyes of a knowledgeable reader. Admittedly, it can be hard to find articles that are some 70 years old. And without previous knowledge as to the bias of Medical Hypotheses or JAP&S, it can be easy to assume that they might be real journals.
I have seen the Geiers cited in a few otherwise excellent articles on possible adverse vaccine reactions. Typically, they get a sentence or two out of a massive review article. I was particularly interested in one of the articles they wrote, based on the review, and went looking for it.
My first Geier experience was... underwhelming. Poorly written, bad stats, etc. I googled the Geiers and found their press releases where in big font and scary bold, they claimed that vaccines were dangerous because, get this, given in a specific schedule, might help prevent childhood diabetes.
What? Prevention is a danger? They were trying to suggest in the press release that the vaccines were causing diabetes, and delaying vaccination would prevent it, but bad writing obscures this. Of course, there was something about how using the Geier's proprietary stats methods would be useful to the pharmaceutical industry.
A real scientist would have written this up with the help of a statistician (double and triple checking everything) and published it in Science! Instant celebrity, gauranteed career path for at least a decade as you seek out the cause, Mentions in the popular press as the person who cut childhood diabetes by X%. Science groupies!
But the article and stats were crap.
A quick look at the Maryland Code reveals this:
That sounds to me like Maryland law is simply requiring any human subject research to conform to the same standards as Federal regulations.
Not that I know what the Federal regulations are....
Cheers,
-Flex
I would like to use this for the next edition of Pediatric Grand Rounds. Please email me awesome_mom2061 AT yahoo DOT com if I have your permission to use it. Thanks!