Prof. DeSoto discusses mercury and autism

3 Aug

A recent issue of the journal Acta Neurobiologiae Experimentalis (ANE) focused upon autism. Not just autism, but autism causation with papers on vaccines, acetaminophen and, of course, mercury. The idea for this focus edition came from Professor Dorota Majewska who holds the EU Marie Curie Chair at the Institute of Psychiatry and Neurology in Warsaw, Poland. The authors for this focus issue are largely the same as those from a conference Prof. Majewska organized in 2008, Autism and Vaccinations.

One of the papers in the focus edition of ANE was the paper by Hewitson et al., that we have discussed at length here at LeftBrainRightBrain.

Another paper in this focus edition is Sorting out the spinning of autism: heavy metals and the question of incidence by M.C. DeSoto and R.T. Hitlan. DeSoto and Hitlan gathered some attention for a paper a few years back where they analyzed an existing data-set, that by Ip et al.. D’oC and Interverbal discussed this paper at the blog Autism Street, starting with A Tale Of Two Tails. In that piece, D’oC and Interverbal discuss the statistical analysis used by DeSoto and Hitlan. Prometheus at the Photon in the Darkness blog also discussed the DeSoto and Hitlan paper in Winter Potpourri. Pure Pedantry blog at ScienceBlogs also discussed this study in Mercury, Autism, and a Note on Scientific Honesty. Perhaps the best analysis of the original DeSoto and Hitlan paper was performed by EpiWonk, an epidemiologist.

The recent paper by DeSoto and Hitlan, Sorting out the spinning of autism: heavy metals and the question of incidence, is basically a review article. It has been touted as support for the mercury hypothesis with a commonly quoted phrase,

Fifteen were offered as evidence against a link between exposure to these metals and autism. In contrast, a sum of 43 papers were supporting a link between autism and exposure to those metals

I somehow doubt the authors intended the debate to boil down to counting papers. It would be a weak support, and rather ironic at that as this paper is placed in exactly the sort of journal that leads to large numbers of papers supporting the heavy-metal/autism link. The current DeSoto and Hitlan paper is in a focus issue on autism in ANE which selected papers which support autism as vaccine injury. Many papers on the mercury appear in lower impact journals and by authors such as the father-son team of Geier and Geier (which if I counted correctly account for 19 of the 43 articles on DeSoto and Hitlan’s list). If you are unfamiliar with that team, the neurodiveristy.com blog has many articles on the team such as Significant Misrepresentations: Mark Geier, David Geier & the Evolution of the Lupron Protocol (Contents).

That said, I was planning to avoid the recent DeSoto and Hitlan paper. It isn’t really new (adding to the number of articles on toxins and autism without adding to the knowledge base). I was going to avoid the paper, that is, until Prof. DeSoto gave an interview for the Age of Auitsm blog. I don’t understand why the Age of Autism considers Prof. DeSoto to be an expert on so many areas of autism and the environment. The breadth of her work is not great. Below is an exchange which shows what I mean. Prof. DeSoto was asked to comment on the recent study by Hewitson et al., comparing vaccinated and unvaccinated monkeys.

Q: There is a study published in Acta Neurobiologiae Experimentalis alongside yours that deals with vaccinated and unvaccinated primates. Do you have a reaction to the study or its conclusions?

Dr. DeSoto: All the primates were vaccinated, the difference was whether there was a heavy metal additive. This is a potentially important study. There are a few weaknesses that prevent strong conclusions. The size of the control group is small (apparently n=2). Given that rhesus neural development within the brain region of interest is not all that well documented, a larger control group would have been desirable. This weakness is acknowledged by the authors.

Isolating the infant monkeys shortly after birth is a significant change from normal environment. The severing of the maternal bond and being raised essentially alone (only visual contact was maintained with the peer infants) affects every aspect of development – including neural development. There is evidence that brain volume is specifically affected by isolation. The rearing situation in the study, in my mind, is not very comparable to normal development, especially if the outcome of interest includes brain volume.

That said, this is the only study that has compared the net effect of multiple vaccination additives on brain development. Above all, I have to editorialize and say this seems difficult to understand (that is – why is this the only study?). If some scientists and some parents question the safety of the vaccine schedule, such studies as this one are the way to investigate the concerns.

Now, the one study that exists (even if there are caveats that go with pilot research) suggests there are differences. Whether one is of the opinion that individually testing vaccines is as good as testing the combined effect or not – at this point it is imperative that additional studies be conducted on the additive effect of the full vaccine schedules.

To be clear and to repeat, if one thinks that the vaccines with additives given in close succession have no effect on neural development– this ought to be established empirically. One thing that I noticed in the study is the main effect for difference in brain volume (no time effect). It should be noted that this suggests the early administration of additive-containing vaccine (first four rounds) was a culprit of interest.

Prof. DeSoto did not take a careful look at the Hewitson et al. study. How do I know this? In the above interview, DeSoto states:

“All the primates were vaccinated, the difference was whether there was a heavy metal additive”

The paper states, “”Four infants were assigned to the unexposed study group and received saline injections according to the schedule in Table I””. The differences included the heavy metal additive, as well as all the ingredients that make a vaccine differ from saline.

What amazes me is that the interviewer at the Age of Autism missed that as well. Even though AoA has touted the Hewitson study greatly, they don’t appear to have read it closely.

This is not a minor detail. It is key to the study design and conclusions.

Another comment:

Given that rhesus neural development within the brain region of interest is not all that well documented

I think that Prof. DeSoto can be excused for not realizing that there is a study tracking the development of precisely the amygdala in macaques. This is because Hewitson et al. did not include that reference (which was easily found in a pubmed search).

“The size of the control group is small (apparently n=2)”

The control group was 4. One was excluded for “scheduling reasons” and the other for unknown reasons. This was a major problem with the study. Fatal, one might say, as the brain sizes of the control group didn’t grow between the two time periods tested (about 4 months and about 6 months of age) for the monkeys. At the same time, their amygdalas shrank. This was a big warning sign that something was amiss with the control subjects, but this was ingored by Hewitson, et al.. Based on this faulty premise, Hewitson et al. claimed that the brains and amygdalas of the vaccinated monkeys were on an abnormal growth path. It is amazing that Prof. DeSoto missed that.

A fact that I am not surprised that Prof. DeSoto missed is that in a previous IMFAR abstract on this group, Hewitson et al. came to the exact opposite conclusion: that the brains of the vaccinated monkeys did not grow as fast as the unvaccinated monkeys.

Back to the recent DeSoto and Hitlan paper. They make the following statement:

It is worth noting that there have been only three empirical articles directly comparing those with and without an ASD on mercury levels in the body to a control group of normally developing matched controls that report that report no link (Ip et al. 2004, Soden et al. 2007, Hertz-Piciotto et al. 2010). While, the most recent article appears to be the strongest, lacking any obvious errors or flaws (we think that this recent article does provide at least some legitimate evidence contradicting the hypothesis that autism and heavy metals are linked), the other two are seriously flawed.

In the end, this mention of the Hertz-Picciotto study is why I decided to write about the DeSoto piece, and in the process bring in the interview.

Part of what made the Hertz-Picciotto study strong was the fact that they controlled for fish consumption. Correct me if I am wrong, but I believe this is something that Ip did not do, nor did DeSoto and Hitlan in their re-analysis. I don’t see mention of fish consumption in the recent DeSoto and Hitlan paper.

Again, I’ll point out that the analysis by EpiWonk was thorough and clear. I wish he had published it. I don’t think consider fish consumption to state that the DeSoto/Hitlan re-analysis of the Ip data is likely not thorough enough to make the conclusions they draw.

The fact of the matter is, the Ip data just aren’t that profound. It was worthwhile to do a re-analysis given the errors in the Ip dataset and paper. But it was three years ago that DeSoto and Hitlan did their re-analysis. In the meantime, Hertz-Picciotto et al. have a better dataset and a more thorough analysis.

DeSoto and Hitlan editorialize a bit in their paper:

If a person has publicly staked his/her career on a certain position being right, it may become harder to keep a truly open mind, even when new data become available and even when the original intent was to be objective. A way this bias might manifest itself is an overstatement or slight misstatement of results. We feel that both sides have been guilty of this, and this happens when a person becomes so confident in the correctness of his/her own view that he/she no longer reviews evidence to the contrary. Unconscious bias may exist even in the best scientists.

This begs the question of whether DeSoto and Hitlan are as guilty of those they chide. Re-analyzing the Ip data is not staking their career on a certain position. Repeatedly publishing on such a limited dataset does make this reader start to question whether some piece of their reputation is now tied to this position. With apologies to Prof. DeSoto, but the fact that her misimpressions of the Hewitson et al. paper are skewed towards the mercury hypothesis makes me wonder even more.

The autism research community needs to have fresh eyes looking at questions and data. DeSoto and Hitlan did well to reanalyze the Ip data once the mistakes were shown. They just appear to this observer to have (a) overstated the interpretation of their analysis and (b) gotten very quickly in to exactly the sort of rut they accuse others of being in.

About these ads

33 Responses to “Prof. DeSoto discusses mercury and autism”

  1. KWombles August 3, 2010 at 23:57 #

    Excellent post, Sullivan.

  2. Emily August 4, 2010 at 00:20 #

    This is a great post, and I’m glad to see Epi Wonk’s breakdown, too. I consulted a completely autism-disinterested statistician regarding the Hewitson paper (after I blogged it myself http://daisymayfattypants.blogspot.com/2010/07/in-name-of-science.html), and here’s the response I got, confirming what I (not a statistician) had thought, too:

    “I don’t like that large of unbalanced cell sizes because it loads the findings in your favor and muddies the findings you do have. They even state this in their paper (well, the former), which is surprising to me. They did handle data analysis correctly for analyzing an unbalanced experimental design by using GLM; however, most unbalanced designs are only off by a few data points at most (e.g., 9 vs. 7 or 10 vs. 6). I think 12 vs. 4 is a little TOO uneven. You’re looking at a 75%/25% split there. Who’s to say the differences they found weren’t due to differences in the actual monkeys or at least partially due to that? Obviously, the repeated measures helps address this issue, but with more subjects comes more variance so of course you are more likely to see movement in the experimental group because there are 3X the number of subjects = 3X the amount of variance. I would have liked to see effect sizes to help clarify this issue.”

    Of course, it’s really a 12:3 split and really 4X when you exclude the control monkey that was dropped.

    Problematic. All around.

    • Sullivan August 4, 2010 at 00:39 #

      Emily,

      thanks for the link. I really enjoy your writing.

      Keep in mind, they ended up 9:2. For some reason 3 of the vaccinated animals and another of the controls were not reported in the end. One, two…where did three go again?

  3. Emily August 4, 2010 at 04:38 #

    Really? I read that paper about three times–evidently more X than their control–and didn’t catch that. Sigh.

  4. RPClarke August 9, 2010 at 15:26 #

    “Excellent post”….”great post”….

    Meanwhile outside of bahh-bahh-land, your bias is rather conspicuous. As with those in the camp that oppose you, you show plenty intelligence and industry in finding the faults that suit your own presumptions, but remain blind to those that challenge your presumptions.

    In the real world both the Hewitson and the Hertz-Picciotto are very weak to extent of worthlessness, in my view, as I shall explain.

    The Hewitson depends on a bizarre assumption that they can validly speed up the schedule of vaccinations fourfold. But are we seriously to believe that everything happens four times as fast in a monkey’s brain? The pharmacokinetics four times a fast. The blood flow, the rate of diffusion. Absurd.

    Meanwhile I pointed out the fallacies of Ms Hertz-Picciotto et al’s recent study at http://www.autismcauses.info/2010/07/hertz-picciotto-et-al-2010-re-blood.html. As detailed more fully there, fish consumption does not constitute a genuine source of mercury toxicity/burden, and blood mercury has been long recognised as a pretty useless indicator of anything much beyond how much mercury there is in the blood at that time. The outcome is that H-P 2010 was an utter waste of time, evidencing nothing.

    I also find your preoccupation with “lower impact” rather pathetic. Does it not occur to you that viewpoints that challenge the big biz corporate agenda are vastly less likely to find a fair welcome in big biz’s lavishly funded and promoted trade journals (pretending to be neutral science)? Do you really expect your readership here to credulously suck up such simplistic falsehoods?
    See http://www.the-scientist.com/2010/8/1/30/1/
    “These studies were published in lower-profile journals (all with current impact factors of 6 or below), suggesting they should have had less of an impact. But these papers eventually accumulated at least 1,000 citations.”

    “Excellent” indeed, pah. Anyway, I award you some Gold Stars for trying though! (At least you’re open for discussion unlike certain quack sites I won’t name here.)

  5. Joseph August 9, 2010 at 20:31 #

    Part of what made the Hertz-Picciotto study strong was the fact that they controlled for fish consumption. Correct me if I am wrong, but I believe this is something that Ip did not do, nor did DeSoto and Hitlan in their re-analysis.

    That is correct. None of the studies to date, however, control for urbanicity (which I’d treat as a continuous variable, not a discrete one.)

    The urbanicity confound is something that I believe escapes Dr. DeSoto (or it used to) judging by a paper of hers on EPA Superfund sites in Minnesota.

    It’s well known that people in urban areas have a higher blood concentration of mercury than people who live in rural areas.

  6. passionlessDrone August 9, 2010 at 21:29 #

    Hi Joseph –

    The urbanicity confound is something that I believe escapes Dr. DeSoto (or it used to) judging by a paper of hers on EPA Superfund sites in Minnesota.

    It looks as if you never bothered to read DeSoto’s paper on Superfund sites.

    An additional weakness is the correlational nature of the relationship that is being reported. Essentially, there appear to be more cases of ASD diagnosed near highly polluted areas. However, such an association could result from a third variable, such as proximity to urban centers. For example, it might be that families affected by autism might choose to relocate to an urban area that might have a wider range of services to offer. Although such an explanation cannot explain the overall pattern of results reported in the literature (Kamer et al., 2004; Yurong et al., 2001; Hoshino et al., 1982), it could play a role in the association here reported between ASD prevalence and proximity to EPA Superfund locations.

    - pD

  7. RPClarke August 9, 2010 at 22:23 #

    Passionless’s point about urban I find quite persuasive, albeit speculative at present.
    Re the notion that H-P was strong because they controlled for fish consumption, I have just above pointed to the fact that fish consumption is irrelevant to these studies, and also that blood mercury is very well known to be a fairly useless indicator of mercury toxicity/burden, such that these negative results amount to timewasting and unnecessary abuse of children. Even my mere GP back in 2004 was able to correctly tell me that a blood test for mercury would be worthless. So why are these supposedly hot researchers so ignorant? The trick of deliberately designing a study to be dud is far from novel, not that I have any basis for attributing bad faith rather than limits of learning here.

    • Sullivan August 10, 2010 at 15:02 #

      RPClarke,

      and also that blood mercury is very well known to be a fairly useless indicator of mercury toxicity/burden, such that these negative results amount to timewasting and unnecessary abuse of children

      By your statement all the work, Ip, DeSoto and Hitlan, and Hertz-Picciotto et al. worthless and designed to fail, timewasting and an unnecessary abuse of children. However, you seem to only apply your logic to the Hertz-Picciotto study. By your own words, “these negative results amount to…”

      If we start from the basics: autism and mercury poisoning do not have the same symptoms, we can find that millions of dollars, uncounted hours of researcher time and the health of many children who have suffered adverse reactions to chleation, has been wasted on the mercury hypothesis.

  8. Joseph August 9, 2010 at 23:54 #

    @pD: I probably did not read it in its entirety, and it was a while back either way. This part does seem familiar, though:

    Although such an explanation cannot explain the overall pattern of results reported in the literature (Kamer et al., 2004; Yurong et al., 2001; Hoshino et al., 1982), it could play a role in the association here reported between ASD prevalence and proximity to EPA Superfund locations.

    Note that she simply asserts that it cannot explain the pattern of results from past literature. It’s not clear what that assertion is based on.

    What I did remember is that she made no effort to control for urbanicity. This is not very difficult to do. You don’t even have to come up with a multiple regression model. She could simply check to see if pollution can explain the residuals of a simple model of population density vs. autism. (I recall that back then I did check this using California county-level data, and the association goes away.)

  9. RPClarke August 10, 2010 at 18:22 #

    Dear Sullivan, For very good reason, negative results tend to have much lower status than positive results. The H-P’s negative can be rightly dismissed for using the infamously poor criterion of blood mercury which tends to give false negatives. But if some studies happen to find significantly positive even when using blood mercury as a criterion, then the significance levels of those studies cannot be equally dismissed on that account. On the contrary they are all the more telling.

    As for your dismissal of a notion that there is evidence linking autism with mercury, you have so much to learn there that this blogsite is not really the place for my trying to achieve it. But I can point you to the Geier/Kern/Geier review in the same vol of Acta Neurobiol Exp http://www.ane.pl/showarticle.php?art=7025, particularly page 215, albeit I don’t find their case against Thimerosal more than marginally persuasive.

    Adverse reactions to chelation? Yes chelation can be rather nasty if done incompetently (as Andrew Cutler explained yonks ago), providing there is actually a toxic burden already there to be stirred up and swished around (so the adverse reactions tend to prove the toxin hypothesis rather than disprove it). But how many have ever died from chelation? None from DMSA or ALA, perhaps one from incompetent misuse of EDTA.
    In contrast millions are killed by approved medical pharmaceuticals being “competently” prescribed by ~qualified~ doctors; I look forward to hearing you raging against that sometime here.

    Wasting time and money on dud research? That stupomega-hyped report in Nature the other month takes some beating on that score!:
    http://www.autismcauses.info/2010/06/copy-number-variations-are-only.html

  10. RPClarke August 10, 2010 at 18:27 #

    (Continuing) So what sort of case is there against the involvement of mercury in autism? I note that Michael Fitzpatrick’s book was greatly enthused about by Paul Offit no less. And yet his book can be seen to be drivel of the highest level, as per http://www.autismcauses.info/2009/12/shallow-critiques-of-holmes-and.html. If that’s what passes for excellence in the no-mercury-involved case, then I can’t imagine what the also-ran arguments must be like.

  11. Joseph August 10, 2010 at 21:23 #

    For very good reason, negative results tend to have much lower status than positive results.

    This assertion has no basis in fact. What could be argued is that a negative statistical result from a small sample size has much less weight than a positive result with a big sample size. (This is because statistical power is lost with a smaller sample size, so a negative could simply mean that there wasn’t enough data to achieve statistical significance.)

    However, this does not apply in this case, because the H-P study is the bigger one of the 3 we’re talking about. Furthermore, it takes more confounders into account. Clearly, it’s the superior study among the 3, and a negative result here is easily explained by its superior methods.

    That’s the purpose of reproduction in science: To see if results can be reproduced, but also to consider other possibilities and check if more mundane explanations might be in order.

    Additionally, if it’s true that chronic exposure to mercury is a major risk factor for autism (e.g. it’s clustered geographically), then this chronic exposure should obviously be detectable in the blood at the time of testing. (And I fully expect that it would be, prior to control for confounding.)

  12. RPClarke August 10, 2010 at 23:41 #

    “This assertion [that negative results tend to have much lower status than positive results] has no basis in fact.”

    I never said it did. Its basis is in reason, namely the reasonable point that it is a very much easier (less improbable, less level of significance) to have something not work than it is to have it work.

    I have already explained that the H-P used the defective criterion of blood mercury, as a result of which its negative result is worthless. End of story. You can rabbit on with rationalisations till the proverbials come home but it doesn’t change the truth of the matter.

    Furthermore, it takes more confounders into account.

    An irrelevant pseudo-confounder such as fish consumption. And taking more confounders into account strengthens only positive results not negative ones (as it is basically about reducing the risk of false positives caused by confounders).

  13. RPClarke August 11, 2010 at 00:28 #

    @Joseph: “if it’s true that chronic exposure to mercury is a major risk factor for autism (e.g. it’s clustered geographically), then this chronic exposure should obviously be detectable in the blood”

    Not “obviously” at all. But genuinely obviously you failed to take in the point which I twice wrote above, namely:
    “blood mercury has been long recognised as a pretty useless indicator of anything much beyond how much mercury there is in the blood at that time.”
    and
    “Even my mere GP back in 2004 was able to correctly tell me that a blood test for mercury would be worthless.”

    I.e, even the lowest rung of even the medical establishment itself was already familiar with this most basic fact of this field way back six years ago. So what’s holding you up from learning it yourself? Just doesn’t suit your pre-set convictions I guess.

  14. Tom August 11, 2010 at 19:23 #

    “Just doesn’t suit your pre-set convictions I guess.”

    This from the mouth that claims his dental amalgams produced an onset of autism at age 15.

  15. Joseph August 11, 2010 at 19:25 #

    I never said it did. Its basis is in reason, namely the reasonable point that it is a very much easier (less improbable, less level of significance) to have something not work than it is to have it work.

    Again, in this case this is incorrect. If study A finds a positive result with N=100 and study B finds a negative result with N=400, then it’s probable that study A’s positive was nothing more than a chance finding (or a false positive due to confounds.)

  16. Joseph August 11, 2010 at 19:31 #

    I have already explained that the H-P used the defective criterion of blood mercury, as a result of which its negative result is worthless.

    If we were to accept this, then the results of all blood-mercury studies need to be thrown out, not just the ones you don’t like.

    There’s no rule in science that says a data collection method can be considered good in positive studies but poor in negative ones. That’s just nonsense.

    Additionally, the fish-consumption confound is clearly not bogus in the particular population studied by H-P et al., as it resulted in a statistical effect.

  17. RPClarke August 11, 2010 at 21:58 #

    Well, in respect of all the responses above I have to conclude that I am not in the presence of an audience that can tell sense from nonsense, such is its bias to its pre-set conclusions. So sadly I see little point in continuing here. I’ll just reply to one particularly huge piece of nonsense above, namely

    “Just doesn’t suit your pre-set convictions I guess.”

    This from the mouth that claims his dental amalgams produced an onset of autism at age 15.

    But this mouth never claimed anything of the sort. I do not have any sort of autism etc, nor ever said so. Rather at age 15 I suddenly started to develop the quite un-autistic devastating symptoms of adult amalgam poisoning. The effects were with utmost starkness documented by my school reports and subsequent documentation of the time and long after. Oh, but of course dental amalgams don’t really make a person ill let alone a mental wreck. And the falsifications in the SCENIHR report are mere hallucinations, sure. And the FDA citing Clarkson & Magos but, ooh, “forgetting” to cite Mutter et al’s demolition job reply in the same journal, couldn’t really have happened.

    I think you may be driving me crackers as I am having a vision of a website run by a group of its own trolls!
    Best wishes anyway.

  18. RPClarke August 12, 2010 at 02:33 #

    Ok, one last bit of explanation (mainly for any open minds that drop by) before I really do leave the residents here to their self-justifying rationalisation paradise!

    You appear to view the H-P et al 2010 as a useful piece of evidence of no mercury involvement, whereas supposedly the Ip et al (DeSoto-corrected) you reject. And yet I have indicated here at least one serious objection that rules the H-P out of that role. Even if H-P et al had a strong point as you reckon, that would not compensate for its defect of using blood mercury (the so, so clever strategy of using a faulty measurement to oh so sadly fail to get the required result, not that I have any reason to think they were deliberately there rather than just bungling).

    Meanwhile you seem to think that I consider the other studies that used blood mercury to be useful evidence of mercury involvement, and that makes me some sort of irrational with double standards. But that is partly a misunderstanding of my position, albeit I may have failed to properly indicate it above.

    If I were presenting a case for mercury involvement in autism causation, I would not appeal to any blood mercury evidence, for the simple reason that there is whopping other vastly more powerful evidence which very much more than suffices. Indeed I already have just such a draft on my desk right now, prepared for publication in the real world. Ip, corrected by DeSoto was not needed in my case, and yet it does have evidential value, and that is not because it is a positive blood result (as explained above, your non-appreciations notwithstanding), but crucially because Ip et al was primarily a study of HAIR mercury.

    Which you failed to mention. I note that all the others on this site all consistently ignore anything I mention here that fails to fit within their dogmas (and for which they can’t find glib enough rationalisations). For instance my link above to the utter balderdash in Michael Fitzpatrick’s “excellent” book. And on a previous occasion the graph of stark change of ratios of ages of onset.

    How sad such self-delusion is. Byebye!

  19. David N. Andrews M. Ed., C. P. S. E. August 31, 2011 at 22:14 #

    … in which R P Clarke shows exactly how far up his own arse his head has gotten stuck!

  20. RPClarke August 31, 2011 at 23:08 #

    It’s curious that the collection of rare geniuses here-above are rejoicing in the H-P et al 2010 blood mercury paper as evidence/proof that mercury is not causal of autism. And yet…
    The first author has been well aware that the paper does not really disprove causation by mercury. In a February 2010 interview she stated as follows.
    AHEARN: “So if autistic kids have the same blood mercury levels as typically developing kids, can we rule out mercury as a potential cause of autism?”
    HERTZ-PICCIOTTO: “Not from this study. That was not the design of this study.”
    The paper itself admitted only on the 20th of its 24 text pages that: “Notably, since half-lives of methyl Hg in blood and whole body inorganic Hg range from 60-90 days, (Clarkson et al. 2003) these measurements cannot address whether Hg exposures in either the prenatal or early postnatal period play an etiologic role in autism. [….]. As only 5% of body burdens of Hg are estimated to be in circulation, (Burbacher et al. 2005; Stinson et al. 1989) reliable conclusions about distribution are not possible from one-time observational measurements in blood.”
    Only hidden away in its second-last sentence the paper stated: “This report did not address the role of prenatal or early life Hg exposures in etiology of autism.” But why was this crucial clarification not placed right at the beginning instead?

    • Sullivan August 31, 2011 at 23:44 #

      RPClarke,

      this would beg the question: do the Ip data (the dataset that DeSoto and Hitlan mis-analyzed, even without consideration for the lack of control for fish consumption) have anything to say about autism etiology?

      The answer: no.

      leaving us with: the DeSoto/Hitlan paper is basically worthless. You’d have to agree, based on your own statements. Well, an intellectually honest person would.

  21. RPClarke August 31, 2011 at 23:19 #

    The top article here states:
    “Part of what made the Hertz-Picciotto study strong was the fact that they controlled for fish consumption.”

    Question. In what way was the H-P study “strong”? In the way that (as my preceding above indicates) even H-P and the paper itself agreed it was completely UN-strong (i.e. as evidence re mercury causation of autism)?

    • Sullivan August 31, 2011 at 23:55 #

      The Hertz-Picciotto paper is strong in the sense that the methodology was strong. That is clear in what was written above. The interpretation you are trying to make is not mine. You do see that, don’t you? Or did you misunderstand?

      It is not I who claims that papers on blood mercury levels have anything to do with autism etiology. That would be Prof. DeSoto.

      I’m with EpiWonk on this one (in his discussion of DeSoto 2007): “We can conclude absolutely nothing about the association of ethylmercury in vaccines to autism from these data.”

  22. RPClarke September 1, 2011 at 00:58 #

    The Hertz-Picciotto paper is strong in the sense that the methodology was strong. That is clear in what was written above. The interpretation you are trying to make is not mine. You do see that, don’t you? Or did you misunderstand?

    So your interest here in the H-P paper is in its strong methodology of showing quite what? The utterly trivial strong finding of no blood mercury association with autism? Oh come off it. It’s been paraded everywhere in the corporatosphere as proof of no mercury causation of autism. Tell me, why did its authors not mention its irrelevance anywhere in the first 20 pages, let alone not mention it in the abstract? Why has such a trivial result been so prominently publicised by certain factions?

    Re your first reply above – I thought I’d already explained that I am not in the business of trying to present here all the case for mercury being causal in autism. And there’s no point in presenting just bits of it either. And if I hadn’t made that clear before, I do so now. And sad to see you all continuing to blink away the elephants I mentioned that don’t fit in your particular vision of this living room. Cheers

    • Sullivan September 1, 2011 at 01:15 #

      “The utterly trivial strong finding of no blood mercury association with autism?”

      Right. This is utterly trivial. That’s why the DeSoto re-analysis of Ip was ignored by those promoting vaccine causation? (yes, that’s sarcasm)

      “It’s been paraded everywhere in the corporatosphere as proof of no mercury causation of autism.”

      I guess since you can’t find where I was “parading” this around, you have to wave your hands at some mythical conspiracy. Well, take your complaints to whomever “paraded” it around. Hence the fact that I am not defending against your strawman argument.

      • Sullivan September 1, 2011 at 01:30 #

        “So your interest here in the H-P paper is in its strong methodology of showing quite what?”

        The DeSoto re-analysis of the Ip data was promoted strongly as evidence that autism is caused by mercury, including mercury in vaccines. This went so far as to have a congressman write to the special masters involved in the Omibus Autism Proceeding about the re-analysis. The Age of Autism blog, for example, had multiple posts on this. Do you want links? I’d hate to be in the position you are in of referencing some vague “corporatoshphere”.

        For those who rely on such weak explanations as the DeSoto paper, there are multiple responses:

        1) the analysis was poor. This has been discussed in many places. EpiWonk did a great one. Other discussions are out there.

        2) The concept of using blood mercury levels is, as you point out, not valid

        3) the data DeSoto relied upon is (a) not reproduced and (b) weak in that it did not control for diet.

        People who rely on papers such as DeSoto’s, such as the writers at the Age of Autism blog, are intellectually dishonest and purposely misleading their readership.

        That is the point. Are we clear?

  23. RPClarke September 1, 2011 at 01:14 #

    Just for a little clarification (though far from the whole picture of my analysis)….
    I don’t consider the reanalysed Ip et al or reanalysed Soden or anything else in DeSoto/Hitlan, or the H-P 2010 deceiving rubbish, as the evidence FOR mercury involvement. I just find them to be three out of three cases of claptrap studies falsely made out as the proof AGAINST merc involvment. As in other areas of pseudoscience, one can form a reasonable judgement from the sham foundations on which they rest their laurels.

  24. RPClarke September 1, 2011 at 01:27 #

    Right. This is utterly trivial. That’s why the DeSoto re-analysis of Ip was ignored by those promoting vaccine causation? (yes, that’s sarcasm)

    I mean it is recognised as utterly trivial by all scientifically-competents, who understand the irrelevance of blood merc levels. The fact that the scientifically-illiterate vacc>>autism true believers make a big thing of it is not my business.

    I guess since you can’t find where I was “parading” this around,

    You were going on about it being a strong finding just above. And I was anyway implying mainly substantially bigger fish than yourself, such as the website of the IACC and so on. Do a search for the web coverage of the HP paper and see how it gets predictably spun by your authorised authorities. Strong, wow!

    • Sullivan September 1, 2011 at 01:34 #

      “I mean it is recognised as utterly trivial by all scientifically-competents, who understand the irrelevance of blood merc levels. ”

      So, the writers for, say, the Age of Autism blog are not “scientifically competent”? Your term, not mine.

      “You were going on about it being a strong finding just above”

      Wow. I make one brief statement months ago, and I’m “going on about it”? What do you consider your perseveration on this post months later?

      “And I was anyway implying mainly substantially bigger fish than yourself, such as the website of the IACC and so on. ”

      What? You were “implying” this? Sure. That came through.

Trackbacks/Pingbacks

  1. Tweets that mention Autism Blog - Prof. DeSoto discusses mercury and autism « Left Brain/Right Brain -- Topsy.com - August 4, 2010

    [...] This post was mentioned on Twitter by Kev and Squillo, Emily Willingham. Emily Willingham said: RT @kevleitch: Prof. DeSoto discusses mercury and autism: A recent issue of the journal Acta Neurobiologiae Experime… http://bit.ly/aTjYZy [...]

  2. Hand me down clothes for the “mercury causes autism” mannequin « Losing In The Lucky Country - August 31, 2011

    [...] up three references. The fourth is DeSoto and Hitlan who court ample controversy not least for citing Geier and Geier [...]

Comments are closed.

Follow

Get every new post delivered to your Inbox.

Join 1,028 other followers

%d bloggers like this: