Articles

195 thoughts on “Pro-CAM Wikipedia – Skeptics Need Not Apply

  1. clgood says:

    I really appreciate the patience of the good Doctors Novella, Gorski, et all but, with all due respect, isn’t arguing with pec just hurting the S/N ratio? It seems to be at odds with Michelle’s attempts to just ignore her.

    I mean, the irony-loving part of me is all tingly waiting for all this promised “proof” that pec is going to dump on us any day now, but that joke gets old really fast.

    Shouldn’t pec be asked nicely to just go get her own blog somewhere where she can preach to the credulous? My vote would be for ignoring her, as in really ignoring her. Maybe she’ll go away.

    Back on topic: I always take Wikipedia with a grain of salt. It’s nice to know it’s been seen as unfriendly by the CAM crowd. Kinda takes my opinion of it up a small notch.

  2. clgood says:

    Oh, dear. I do know there’s only one ‘l’ in ‘et al’. I claim fatigue.

  3. matalford says:

    Give someone an inch……

  4. pec says:

    “There were negative ones too. And when you consider them together with accepted statistics the net effect is negative.”

    That would have to be mentioned in the abstract. You can say whatever you want because we can’t get the article. You did NOT provide the context.

  5. pec says:

    “one study will not establish a new phenomenon. Only a research program demonstrating a repeatable and replicable effect, looking at it from multiple angles, consistently showing a real effect will be compelling.”

    I have NEVER said I would accept anything based on one study! I said I was gathering references to research supporting the existence of biological energy or fields. I said it would take time and I did not want to post isolated studies for exactly this reason! Then I was challenged to find even one, because they thought I was lying when I said there were many studies supporting this kind of energy. So I posted just one, which happens to be by a mainstream biologist with many years of research experience. She is not considered an idiot or fake who would use deceptive devious tactics (which is what you accuse her of).

    I am NOT accepting energy healing based on one experiment. I have MANY MANY reasons for suspecting it is real.

    I said I did not want to post any references until I had a complete lit review — which they wanted to be done in one day — because I knew the kind of devious tactics used here. You are saying the study was negative overall and that is utter BS, and you know it. But you don’t care because this is political warfare.

    And you “skeptics” depend on one small unblinded, underpowered, poorly designed, negative study to prove that energy healing is all fake. Just one study is good enough if it confirms your ideological preconceptions about what is plausible.

  6. pec says:

    “No significant differences were found in DNA synthesis, assayed by [3H]-thymidine incorporation at 1 or 2 weeks for SaOs-2 or 1 week for HOBs. However, after four TT treatments in 2 weeks, TT significantly (p = 0.03) increased HOB DNA synthesis compared to controls. Immunocytochemistry for Proliferating Cell Nuclear Antigen (PCNA) confirmed these data. At 2 weeks in differentiation medium, TT significantly increased mineralization in HOBs (p = 0.016) and decreased mineralization in SaOs-2 (p = 0.0007), compared to controls. Additionally, Northern blot analysis indicated a TT-induced increase in mRNA expression for Type I collagen, bone sialoprotein, and alkaline phosphatase in HOBs and a decrease of these bone markers in SaOs-2 cells. ”

    The planned comparisons came out supporting energy healing. There is no mention of unplanned ad hoc comparisons for this study — that would have to be stated in the abstract. Novella is lying. And experienced and respected researcher would never resort to that kind of deception and a publisher would not accept it.

    And why can’t Novella email me a copy of the article?

  7. pec says:

    And furthermore, there were 2 studies, not one.

  8. pec says:

    And Novella is NOT any kind of expert on statistics. But the “skeptics” here will mindlessly believe whatever he says.

  9. matalford says:

    zzzzzzzzzzzzzzzzzzzzzz

  10. weing says:

    Sounds like you are rabidly anti-science, and as far as I can see, base your beliefs only on abstracts. Science is self-correcting. You nor I don’t have to accept anyone on authority. Just follow your reason. You do know what that is, don’t you?

  11. pec says:

    I am not anti-science and you know it. I don’t have a subscription. If there were a negative effect that cancelled out the positive effect, this would be reported in the abstract, and you know it. Novella is either deliberately or inadvertently misinforming us. And he is ignoring one of the studies.

    I told you many times I do not draw a conclusion based on one experiment. You or qetzel or pretzel or whoever it was challenged me to find even one study supporting energy healing. Well I gave you 2, and there will be many more. And you won’t be able to discount them all, even using deception.

  12. pec says:

    You’re fighting a war against CAM? Well I’m fighting for truth and fairness and objective science. I am fighting against your authoritarian anti-science ideological close-mindedness.

  13. This was not two studies – this was one study published twice – or divided into two papers. Right from the press release: “Her findings were published in the Journal of Orthopaedic Research and The Journal of Alternative and Complementary Medicine.”

    This has not been replicated.

    Someone with an ounce of intellectual honesty would at least read the paper before accusing someone else of lying about it.

    It is naive to say that the abstract must be accurate and complete. Why, in this very blog our own Mark Crislip recently wrote: “Seems impressive, especially the conclusion. Chiropractic is not to blame for VBA strokes. But abstracts are like movie trailers. They give a flavor of the movie, but often leave out many important plot devices and characters.”

  14. pec says:

    You know very well that if there were a negative result that cancelled out the positive result IT WOULD BE REPORTED IN THE ABSTRACT. That would not be buried somewhere in the results section. You know it, but you would rather deceive your readers into thinking the experiments found nothing. You are pretending that a respected researcher would try to mislead people into believing something that she is not selling or promoting. Why???

    You simply cannot concede that this research looks promising for energy healing. You will say anything at all to avoid admitting that.

    But there is so much more evidence for the existence of some kind of life energy. You are desperately fighting a losing war.

  15. pec says:

    And by the way, why would the editors of the Journal of Orthopaedic Research allow a researcher to use deliberate deception and make it sound, in the abstract, as if there were a positive effect when in fact there was no effect? It would never happen, and you know it. You don’t understand statistics well enough to interpret the results.

  16. pec says:

    ” The abstract is only presenting the positive correlations. There were negative ones too. And when you consider them together with accepted statistics the net effect is negative.”

    Why don’t you post the part of the results section that shows this? Then those of us who know something about statistics can judge for ourselves.

  17. pmoran says:

    Pec, what are you trying to prove? Even if you find an in vitro study with no obvious defects, it would STILL be outweighed by all the other evidence that leads us to regard TT as a pseudoscience-based placebo treatment. Formulating (or dismissing) a proper scientific hypothesis entails considering all the evidence from all possible sources and it uses other scientific knowledge and experience when weighing one fom of evidence against another. It is a complicated process.

    Thus, not all of numerous possible faults in trial design, performance, and interpretation will be obvious from the published material. For example, conscious or unconscious fraud by enthusiasts wanting to broadcast “the truth” will not be obvious from any reading of the papers.

    The authors may not even know that there was a problem with the controls: a lab assistant may have left some of the Petri dishes lying around for a while in a different temperature or atmosphere.

    Remember also these are still subtle statistical results similar to those found in clinical trials and they have been well shown to be fallible and easily influenced by observer biases.

    Perhaps we skeptics need to try and explain to you why we seem so biased against TT. But that might take many years.

  18. pec says:

    It is discouraging when people care more about politics and ideology than about reason and evidence.

    Harriet wrote a long negative review of Gary Schwartz’s energy healing book. She never described the experiments, never explained why she does not believe the results. Instead she went on at length about the one negative experiment, which was performed by “skeptics,” was unblinded and under-powered, and lacked ecological validity. Bu Harriet loves that experiment, because it agrees with her ideology. What about all the energy healing studies with positive results? She fails to mention them, intentionally, so people will assume they were meaningless.

    I am in the process of learning all I can about scientific research on energy healing. I am appalled at the complete lack of concern for science I see at blogs like this, and I do not intend to ignore it.

    I have studied both mainstream and alternative science all my life. I am not ignorant and credulous like most of this blog’s readers. I am not awed by authorities and experts and I do not belong to any ideological group. It makes me angry when reason and evidence are cast aside because of politics and money.

    However I do think the authors and readers of this blog are sincere and deeply believe in scientific materialism. I think you are fighting for what you believe, and do not realize how brainwashed you are. I am here to help and to educate.

  19. TsuDhoNimh says:

    Steven … So you are saying they did the “experiment” with the bone cells, then analysed many different possible characteristics, threw out the ones that didn’t show a positive result, and then published the positive results?

    What characteristics did they examine and reject? Are they intellectually honest enough to list them?

  20. pmoran says:

    The original TT claim was that the practitioner could detect defects in the subjects’ energy fields and correct them. These claims were made without any attempt to show that the defects in the energy field could be reliably detected by different practitioners, or that anything at all was being felt, or that the fields were improved by their “unruffling”, or that the defects and their undoing was correlated with the presence or absence of any clinical state. On top of this the main clinical claims (relief of anxiety and pain) were adequately explained by placebo influences.

    And you are here now complaining that we don’t take therapeutic touch seriously– that we don’t regard them as serious and trustworthy scientists? For Heaven’s sake, you claim to have a PhD! That should indicate some capacity for critical thought..

  21. pec says:

    “Perhaps we skeptics need to try and explain to you why we seem so biased against TT”

    Oh pmoran, you wish I were an ignoramus but I hate to disappoint you, I am not. I designed and analyzed experiments for 4 years while getting a PhD, so you don’t have to explain the scientific process to me.

    There is one experiment that I know of that “disproves” TT, and of course that’s the one experiment you “skeptics” know about and constantly cite. In that experiment the TT practitioners did not practice healing but tried to sense an energy field when their hands were not moving. They failed, but that does not mean they would have failed in a normal healing context. Furthermore, this was an unblinded experiment with low power — but I never hear you “skeptics” complain about that.

    There have been many studies showing that TT can be effective. Of course the quality of the studies must vary widely and I would not accept TT based on only one of them. And I am not saying I do accept TT — I am saying there is evidence for it, and it would be unscientific to discard and ignore evidence.

    You materialist “skeptics” think TT is implausible because you do not believe in what is variously referred to as life energy, qi, chi, prana, bioenergy, biofields, etc., etc. And you do not believe in this because vitalism is not the current fashion in biology. There were no scientific experiments that disproved vitalism; it simply faded out of fashion.

    You are taught in biology classes that vitalism is wrong, and you obediently accept whatever you were taught. No amount of evidence can change your mind, because the authorities and the experts all agree, and you trust them.

    Rejecting vitalism helped modern scientific medicine to distance itself from all sorts of folk medicine, which helped to elevate the status of the medical profession, to differentiate it from ancient “superstition” and “magic.”

    Never mind that energy healing and belief in life energy is common to all cultures at all times and places — with the exception of modern materialist science. All the others were wrong, ignorant, superstitious. Only the materialists are right.

    So that is where we are today. Mainstream materialist medical science has run up against some brick walls and interest in CAM is increasing. Scientific research is very expensive, but CAM is being funded now, researching theories that mainstream medicine despises. No wonder you’re so angry, no wonder your minds are slammed shut — your exalted pedestal is getting shaky.

    And I am NOT saying all alternative therapies have value, or that every ancient belief is true. Or that every mainstream idea is wrong. I am saying that materialist science is dead set against certain ideas and will not even consider the evidence, even when the researchers are highly qualified, respected, experienced scientists.

    Because there are some scientists whose curiosity about nature is stronger than their need to conform.

  22. pec says:

    “And you are here now complaining that we don’t take therapeutic touch seriously– that we don’t regard them as serious and trustworthy scientists? ”

    There was no placebo effect in this study, because the subjects were cells. This was a controlled experiment. Novella’s claim that there was no effect is untrue. There was a positive effect that clearly supported the hypothesis.

    If you don’t give a damn about scientific evidence, then what is the point of doing research? Why go to all that trouble if the results mean nothing at all?

  23. pmoran says:

    Pec. “There was no placebo effect in this study, because the subjects were cells. This was a controlled experiment.”

    I don’t believe that you have designed or participated in such experiements, as you don’t seem to even understand what you are being told here. Let’s see your name on some published papers.

    We should not be needing to explain to any experienced scientist how easily contaminants, artefact and loss of effective controls can creep into nearly all kinds of biological experimentation. Researchers can also easily unconsciously manipulate results by disregarding results that “don’t look right”. This goes on all the time as the result of pressure to publish and to get the “right” results.

    And don’t manipulate my words. My reference to placebo was clearly to the supposed effects upon anxiety and pain that TT practitioners were claiming. The resemblance of TT to placebo is enhanced by some of the studies for example the well-known burn study which was supposed to “finally prove TT”. In that the patients said that the procedure made them less anxious and in less pain, but they did not get out of hospital earlier or use less pain mediciation.

    So the hypothesis that TT may confer some limited benefits for patients is not seriously disputed (by me, at least). What is disputed is that unique processes or forces lie behind it, that it can help in any major clinical capacity, and that it can perform better than any other other kindly-intended hands-on placebo treatment..

    I will be travelling for five weeks and may have little ability to participate for a while.

  24. I won’t have access to the study again until Monday when I am at work.

    There was no deception or discarding of results. They presented all the results. They even admit in their discussion that for some things they measured there was a difference for others there were not – and when they analyze all the results together there is no statistically significant effect. They concluded that this was because that statistical analysis was too conservative and their study was underpowered to see an effect with this conservative analysis.

    So – they presented everything. It was just their abstract and their ultimate conclusions that were misleading – which is extremely common. I see this all the time in non-controversial studies. It simply makes this data very weak. It certainly is not enough to warrant rewriting the physics books.

  25. Jim1138 says:

    Given it’s claims, energy therapy would easily be verifiable with a double blind study. The intent here is obfuscation of the fundamental issue that this hocus pocus energy has never been demonstrated in a scientifically sound manner. It probably has not as the practitioners would be fools to participate.

    People here are wasting their time chasing down papers with fundamentally flawed assumptions. I can see why Michelle B reposted HCN’s comments. The troll’s bait has been swallowed hook, line, and sinker.

  26. pec says:

    ” for some things they measured there was a difference for others there were not – and when they analyze all the results together there is no statistically significant effect.”

    It doesn’t make any sense that this would not be mentioned in the abstract. I would like to see the results section. I have never seen an article that claimed to have an overall effect in the abstract, but really had none. I think you misunderstood.

    If the overall planned comparison showed no difference, this would be stated in the abstract.

    And there were 2 studies, both claiming the predicted effect — so which one are you talking about?

  27. pec says:

    “We should not be needing to explain to any experienced scientist how easily contaminants, artefact and loss of effective controls can creep into nearly all kinds of biological experimentation.”

    I have never said otherwise! I also said, repeatedly, that I would not accept a hypothesis based on a small number of studies. Of course there can be all kinds of fraud and error in any experiment — that is exactly why I thought Harriet, and many other “skeptics,” were wrong for accepting the anti-TT study with no hesitation.

    The anti-TT study was by a complete novice, a 9-year-old, while the ones I cited here were by a highly qualified, experienced biologist. You automatically assume the pro-TT study is full of errors and bias, but I have not seen any “skeptic” make those accusations about the anti-TT study.

    And, as I keep saying over and over and over, I did not intend to cite one or 2 studies. I am doing whatever lit review I can without being a subscriber to medical journals. I was challenged to provide even one quality reference, because some of the “skeptics” here were absolutely certain I could not find even one.

  28. pec says:

    ” the study may have been unpowered to support use of the
    conservative, Bonferroni approach to performance of
    multiple, pairwise statistical tests.”

    “when they analyze all the results together there is no statistically significant effect. They concluded that this was because that statistical analysis was too conservative and their study was underpowered to see an effect with this conservative analysis.”

    You don’t understand what the Bonferroni test is. In no way does their statement imply that there was no overall effect.

  29. pec says:

    Since Bonferroni is overly-conservative or inappropriate for many analyses, it would not be considered important enough to mention in the abstract. But publishers may require it as a formality. You were wrong to conclude from that statement that there were opposing effects that cancelled each other out. I will have to see more of the results section.

  30. weing says:

    Let me get this straight. You were conditioned to believe in energy healing and find confirmation in your reading of abstracts that it is real. You do not read the studies themselves and analyze them critically? I assume you also believe in cold fusion. Boy have I got a bridge to sell you in Brooklyn.

  31. TsuDhoNimh says:

    PEC –
    Unless something can be reliably repeated – if not 100%, at least with better than chance odds – by anyone who has the tools and follows the procedure, it’s useless in a medicalsetting.

    Instead of multiple experiments, each claiming to show something is happening … you need to pick one that had a decent protocol and REPLICATE it in another lab with different staff, following the published procedure.

    1 – How do you select the TT practicioners for a study? What is the screening test for this ability? How is the energy adjusting ability acertained? How do you know they aren’t just faking it?

    2 – Without seeing the experimental protocol, I can’t tell if the staff caring for the cultures knew which were the “TT” cells, which were the “sham”, and which were the “ignored” containers. Cell cultures are, in their own way, as sensitive to handling as lab rats.

    3 – If I were designing this, if possible, I would have a “jukebox” cell culture handler that would select the appropriate cell culture/s when the human entered their code number, present it to be therapied or shammed or ignored inside a climate-controlled area for X minutes and put it back.

    *********
    Also, and we’ll have to wait for Steve on this, did the TT make any parameters worse?

  32. urology-resident says:

    Name that logical fallacy in PECs comments…

    Argument from authority:
    “I am saying that materialist science is dead set against certain ideas and will not even consider the evidence, even when the researchers are highly qualified, respected, experienced scientists.”

    Ad hominem attacks:
    “No wonder you’re so angry, no wonder your minds are slammed shut — your exalted pedestal is getting shaky. You complete idiot. You are such a complete moron.”

    Appeal to ancient cultures:
    “Never mind that energy healing and belief in life energy is common to all cultures at all times and places”

    Argument from Personal Incredulity:
    “It doesn’t make any sense that this would not be mentioned in the abstract. I would like to see the results section. I have never seen an article that claimed to have an overall effect in the abstract, but really had none.”

    I bet someone else can find some of the other 20 logical fallacies mentioned here: http://www.theskepticsguide.org/logicalfallacies.asp.

  33. pec says:

    That’s ridiculous. We have absolutely no reason to think an article published in a mainstream journal by a respected researcher is completely worthless. It is NOT a logical fallacy to mention that this researcher is qualified to do experiments. It is NOT a logical fallacy to say it’s a mainstream journal. If it were not a mainstream journal you would say it’s worthless for that reason.

    I don’t have access to the journal and I am not thinking about spending a fortune subscribing to medical journals. This is only a hobby for me. Dr. Novella can post the results section if he likes — I would appreciate that.

    Meanwhile, I can guarantee that there were no negative effects that cancelled out the positive effects. That would be mentioned in the abstract, because that would be extremely relevant. Novella is mistaken. The test he mentioned would lower the cutoff for p values when a large number of comparisons are being made. You could have extremely reliable effects for the important comparisons, and less reliable effects for unimportant comparisons, and this test would not account for that. For example, cancer cells did not respond to TT — well that actually confirms the hypothesis, although it was not predicted.

    The abstract should explain the relevant and important findings. Statistics done merely as a formality would be omitted from the abstract. If, as Novella claimed, it was a complete wash out, that would absolutely be reported in the abstract.

  34. urology-resident says:

    Assuming that all relevant information is in the abstract is not always true.

    I just reviewed a paper from one of the Urology Journals for our monthly Journal Club in which they claim that Alfuzosin (a drug that blocks the muscle in the ureter) decreases time to expulsion of a distal ureteral stone (with p<0.05 and so forth) in the abstract.

    Then you read the results section and the confidence intervals for the time to expulsion in the placebo and the drug were very similar, although the group in the drug was in average “statistically better”.

    The study would have convinced me if the confidence intervals were not as similar, for example: 1-4 days compared to 5-10 days, as opposed what they found on the study: 0.8-12 days in the placebo versus 1-15 days in the drug (I don’t remeber the exact numbers at the moment).

    This is just to say that its hard to make any conclusions out of an abstract and is naive to assume that all people are 100% completely honest and will never omit information from an abstract.

  35. weing says:

    “Meanwhile, I can guarantee that there were no negative effects that cancelled out the positive effects. That would be mentioned in the abstract, because that would be extremely relevant.” Only guarantee if you yourself verify it. I have read enough articles to know better. You would think an increased risk of MIs in patients on vioxx would be extremely relevant, wouldn’t you?

  36. pec says:

    “This is just to say that its hard to make any conclusions out of an abstract and is naive to assume that all people are 100% completely honest and will never omit information from an abstract.”

    I never said I had complete faith that the abstract is accurate and honest. I said that I think Novella’s claim must be false. I do not think he read the results carefully and I think he jumped to a wrong conclusion because that is what he wants to believe — that energy healing is all fake or delusion.

    All we’re doing now is speculating since we don’t have the article. I will bet you, though, that Novella was wrong in implying that the abstract was utterly misleading. As long as I don’t have the article, he can say anything at all and the “skeptics” here will believe it. Because you don’t want to think that an energy healing experiment can have meaningful positive results.

  37. pec says:

    weing,

    What Novella is claiming — that these experiments had no overall effect — would always be reported in the abstract. It would be crazy to leave out something that important. He obviously misunderstood the statistics. If he were right, the “skeptics” would be all over it. I can pretty much guarantee that he is mistaken. But we’ll see.

  38. weing says:

    That has not been my experience. Sorry.

  39. pec says:

    “That has not been my experience. Sorry.”

    Well maybe we will find out. But there are plenty of other energy experiments. I’m sure you will assume all are the result of fakery, incompetence and self-deception. We could make that assumption about any research though — and then what is the point of doing research?

  40. weing says:

    Would you say all research published on the existence of the Easter Bunny are the results of fakery, incompetence, and self-deception?

  41. qetzal says:

    Hiya, pec.

    You wrote:

    I have noticed, however, that they guys who challenged me to find even one experiment supporting energy healing have quietly disappeared.

    Not disappeared, just too busy to respond. I seem to recall it took you well over a month to find this study, so I’m sure you’ll grant me a few days to make time to look at it. Especially since I’ll have to stop by the library in person. (I can access that journal only for articles older than 3 months.)

    I can, however, access the JOrthopRes article on line. May not have time to look at it until tonight, though.

    I’ll just echo what others have said – abstracts are meant to cast an article in the most positive light so people will read them. They’re expected to be truthful and not egregiously misleading. But it’s perfectly common for an abstract to tout some seemingly interesting result, with statistical significance, yet admit quite candidly in the discussion that the data’s not adequate to conclude that the effect is real.

    I’ll have to wait to see for myself if that’s the case here.

  42. qetzal says:

    pec’s quote ends with the word “disappeared” above. The rest of the words are mine.

    Sorry for the missed tag and any resulting confusion.

  43. Below are the paragraphs from the paper on the statistics used. Orac has also written on these two papers (http://scienceblogs.com/insolence/2008/08/maybe_we_should_use_therapeutic_touch.php)
    and came to the same conclusion as me.

    He also noted that the JACM version of the paper did not have the more elaborate statistical analysis that the JOR version did. The methods below are from JOR.

    Essentially, the Bonferroni method is used to adjust for the fact that multiple comparisons are being made. This is to avoid cherry picking positive results out of the negative ones. When this was done the results were generally not statistically significant. This was not reflected in the abstract.

    Again from the discussion:

    “However, once again the study may have been unpowered to support use of the conservative, Bonferroni approach to performance of multiple, pairwise statistical tests. The p-values at 4 and 6 weeks were both equal to 0.029. Although these pvalues fell below the nominal 0.05 cutoff for significance, they did not reach the more extreme threshold of 0.0167 required by use of the Bonferroni method.”

    Actually, reading through again I see that there was one result that was significant even after Bonferroni: calcium content in TT compared to placebo at 2 and 4 weeks. What is odd, though, is that this was not significant when compared to control, which overall showed a smaller effect that placebo. The pattern is not consistent. There is also no clear dose-response effect.

    So most of the effects vanished with proper statistical analysis, and there was no pattern in the results that makes sense if there were a real effect at play. You see the kind of scatter that is evident with studies of null effects.

    And there were problems with the methods – such as no mention of how the handling of any of the plates were blinded.

    Here is the statistics section of the methods:

    Statistics
    Generally, most replications of an experiment involved three sets of six-well plates. DNA synthesis, mineralization, and a
    Northern blots were performed in one experiment. HOBs and SaOs-2 cells were plated in different dishes, treated separately,
    but in the same TT session. Both HOBs and SaOs-2 were analyzed from one experiment in a blinded, nonbiased manner with coding and with the same biochemical assay at the same time.

    Data analysis focused on comparing the distribution of levels of proliferation and mineralization across study conditions, for
    example, ‘‘therapeutic touch versus control’’ or ‘‘therapeutic touch versus control versus placebo.’’ All comparisons used
    ‘‘exact’’ nonparametric statistical tests. Nonparametric tests were selected because study measures typically did not follow
    normal distributions and sometimes exhibited clear evidence of heterogeneity of variance between groups. ‘‘Exact’’ versions of the tests were performed to avoid reliance on ‘‘large-sample’’ approximations in the calculation of p-values.

    For two-group comparisons (therapeutic touch versus control) that involved combination of observations across replicated
    experiments, the stratified Wilcoxon rank sum test proposed by Lehmann was used.21 This technique involves ranking observations within experiments to account for ‘‘block effects’’ that may vary from one experiment to the next. The
    significance level for these comparisons was set at 5%. For three-group comparisons (therapeutic touch vs. control versus
    placebo) that involved combination of observations across replicated experiments, the Lehmann technique was applied
    to each of the three possible pairwise comparisons of groups, and the Bonferroni method was used to account for potential
    inflation of the Type I error through multiple testing.22 Thus, in these circumstances, a 1.67% significance level was applied to
    each pairwise comparison to keep the overall probability of a Type I error in any comparison at or below 5%.

    The Bonferroni method is known to be ‘‘conservative’’ in that the ‘‘effective’’ probability of any Type I error is often less than the nominal, desired level of 5%.22 In one case, a three-group comparison was conducted that did not involve combination of data from different experiments. In this circumstance, the Kruskal-Wallis method provided the basis for initial testing. When that test provided a statistically significant result, the Wilcoxon rank sum test was used in conjunction with the Bonferroni correction to evaluate all pairwise comparisons. In another case, two groups (therapeutic touch vs. control) were compared at three time points (2, 4, and 6 weeks) using data from a single experiment. The difference in distribution between groups was assessed at each individual time point using the Wilcoxon test and the Bonferroni correction was applied to account for multiple testing across time points.

  44. pec says:

    You left out the actual results. Thanks a lot.

    If all the results were in the predicted directions, and all p values were below .05, than what you are saying is bs. But I need to actually see the results, obviously.

    “The p-values at 4 and 6 weeks were both equal to 0.029. Although these pvalues fell below the nominal 0.05 cutoff for significance, they did not reach the more extreme threshold of 0.0167 required by use of the Bonferroni method.”

    This does NOT mean researchers would generally consider this to be a chance result! You know that very well, or you should.

    And I can’t imagine why you posted selectively so I can’t see any of the data.

  45. pec says:

    “The p-values at 4 and 6 weeks were both equal to 0.029″

    For the overall comparison? And how many comparisons were there? One in twenty are expected to be chance. If there were 20 comparisons and all were significant at p < .05 then anyone would agree the results were positive.

    But you left all that out, so readers here will assume there was no overall effect.

  46. pec says:

    Well Orac sure is desperate to discredit this study.

    Anyway, now I know that Novella was hoping we would mindlessly accept his untruthful statement, that the positive effects were cancelled out by negative effects, amounting to nothing overall. There were p values well below .05 for the overall comparisons. Any reasonable researcher would acknowledge that being a tiny bit over some arbitrary — and in this case, high conservative and not necessarily appropriate — cutoff does not mean there was no difference between experimental and control groups. There were p values well under the usual .05 cutoff.

    So Orac goes on and on and on about how this can’t be real, this can’t be valid research, TT is woo, it can’t be true, oh no what will we do, what if science gets overrun by woo.

  47. pec says:

    Dangerous comments must be moderated. Our materialist worldview must not be threatened.

  48. pec says:

    So uh maybe my comments will show up some day.

  49. weing says:

    I always thought the Bonferroni approach meant you have to divide the alpha ie 0.05 by the number of variables you are comparing to determine significance. Otherwise you’ll see connections where none exist. Am I wrong?

  50. Weing – you are correct.

    I cannot post the entire study for copyright reasons. Also – I have it in pdf, which takes considerable time to copy and then fix the formatting to make it readable. Register and get the full study if you want. Or take a look at the link I provided to Orac’s discussion – he reproduces some of the data with graphs.

    The bottom line is that this data is WEAK. This study is not convincing at all and is entirely compatible with there being no real effect.

  51. pec says:

    “The bottom line is that this data is WEAK.”

    And, very conveniently, you can’t show it to us!!

    Wait wait wait for moderation. Dangerous comment.

  52. pec says:

    If all the planned comparisons were in the expected direction, and most of them had p values well below .05, then the data is NOT WEAK.

  53. daedalus2u says:

    Is it legitimate to treat a single plate of 6 wells as six independent experiments? If there are minor differences in temperature, atmosphere composition, light exposure, plating technique, etc, each group of 6 wells that is treated “the same” is going to be more similar to themselves than to a separate group of 6 wells that is necessarily treated slightly differently.

    Any differential treatment of different plates requires that they be separated in space for some period of time.

    I would really like to see multiple control plates go through the protocol to see how much variation there is between plates.

  54. qetzal says:

    pec, if you want to see the whole paper, you can purchase access for $39.00 here.

    Or, you can go to your local university biomed library if they have a subscription. (That’s what I’m planning to do, BTW. I don’t have on-line access to the most recent pubs for this journal, and I’m not about to spend $39 on it.)

    If you’re not willing to do either of those, then that’s your problem. No one here can legally send you a copy, nor can they legally transcribe all the results and post them here. That’s due to copyright laws, not to any nefarious intent of anyone on this blog.

    In other words, you should probably shut up and stop accusing people of arbitrarily withholding things from you. And maybe see someone about that persecution complex you’ve developed.

  55. pec says:

    “If you’re not willing to do either of those, then that’s your problem”

    I am studying the subject of energy healing, as a hobby, partly out of interest and partly to show that you are uninformed and misinformed on the subject. If I paid for every article, or drove to the university library, it would become a major sacrifice of money and time. And you know that, so I should not have to say it.

    What you are saying, essentially, is that people who are outside the profession and do not have subscriptions at work should have blind trust in the authorities and experts. We should just shut up and let you lead the way.

    Novella, and Orac, deliberately left out most of the data we needed to see. And they did this because the results were not weak or ambiguous or doubtful. If the data had been weak, they would have displayed it proudly.

    This article is a serious threat to your worldview. But there are many others, and now they are starting to become mainstream. I am studying the subject as well as I can without access to medical journals — there is a lot of free information and I am making use of it.

    If Novella wanted to be honest and straightforward he would have posted the relevant data. Instead, he told us, untruthfully, that the results were a wash out. If that were true, Orac would have said it also, but he didn’t. Orac did his damnedest to point out faults, but he never claimed a flat or negative result — just not quite up to some exceedingly conservative and unrealistic probability level.

    So Novella showed himself to be either extremely careless or deliberately dishonest. And I am supposed to blindly trust his evaluation of this research?

  56. pec says:

    Ah, wait wait wait for moderation. Dangerous comment warning.

  57. pec says:

    If my comments ever make it past censorship, you will see them.

  58. pec says:

    I answered you qetzal, but it still has to get past the authorities.

  59. pec says:

    And I don’t see why these results would be considered surprising, since electromagnetism can accelerate bone healing, and organisms generate electromagnetic fields. W

  60. pec says:

    I took out the swear word, but it’s still in the censor trap. Too bad, can’t answer you qetzal. So much for scientific debate.

  61. weing says:

    “I am studying the subject of energy healing, as a hobby, partly out of interest and partly to show that you are uninformed and misinformed on the subject. If I paid for every article, or drove to the university library, it would become a major sacrifice of money and time. ”

    No pain, no gain.

  62. qetzal says:

    Sorry, pec, don’t know what to tell you.

    From the sounds of it, perhaps your IP has been flagged by the blog’s spam filter for some reason. I assume it wasn’t by intent of any of the blog authors.

    (If it was an intentional action, I think it was misguided. OTOH, you’ve established a long history of misquoting, mischaracterizing, and miscontruing what the blog authors and others say, so I don’t exactly hold you blameless.)

    That aside, I’ve looked over the J Orthopaed Res paper (Jhaveri et al., 2008, Therapeutic Touch Affects DNA Synthesis and Mineralization of Human Osteoblasts in Culture). That one I could access on line. (The other I can only get in person at the library. That will have to wait a few days.)

    Here are my thoughts on Jhaveri et al.

    They did quite a few comparisons of TT-treated & control cells. Most showed no significant effect. For example, they say they tried treating HOB and SaOs-2 cells with a range of different TT doses (1/wk – 5/wk) over a 1-week period & saw no specific effect.

    But they also reported significant effects for a few conditions. For example, HOB treated twice per wk for 2 wks showed apparently significant increases in 3H-thymidine uptake (a measure of DNA synthesis). Interestingly, the same treatment regimen had no significant effect on SaOs-2 cells.

    They did PCNA staining of the affected HOB cells, & also reported a significant effect. If those HOB cells really did have increased DNA synthesis, you’d expect to see increased 3H-thy uptake and increased PCNA staining. In other words, those two observations suggest the HOB cells really were showing a difference from control, but if so, they’re really only describing a single effect. (Meaning that increased 3H-thy and increased PCNA are two aspects of the same effect; they’re not independent examples of the purported effect of TT.)

    They also report that TT significantly increases</i. mineralization in HOB cells, but decreases it in SaOs-2 cells. That’s rather odd.

    Finally, they report differences in certain mRNA levels.

    Based on everything they report, I think it’s reasonable to conclude there were some real differences in some of the treated cultures. The question is whether those differences are due to the TT treatment, or some uncontrolled variable. Frankly, I can’t tell. They talk about their efforts to avoid bias, and some of the steps are good. But it’s hard to tell if they covered all their bases. It doesn’t help that the paper reads as if written by someone for whom English is not a first language. (Not meant as a slight to anyone, BTW.)

    Key things that aren’t clear to me:

    1) Were control and TT-treated cultures always processed in parallel? I think so, but it’s not made clear.

    2) Were possible environmental differences adequately controlled? The authors talk about clamping treated and control dishes at opposite ends of an L-shaped lab. Did they test whether temperatures might differ in the two locations? If so, they don’t say. That alone could be more than enough to cause significant differences. For example, if a TT-treated culture was clamped at one end of the lab, and the matched control culture was clamped at the other end, if the control side was cooler, DNA synthesis in those cells would likely be inhibited.

    3) Were the reported significant results shown to be reproducible? This is a big one, and it’s fairly maddening that you can’t tell. The authors frequently note that results represent 3 or 4 or sometimes more experiments. But they also make clear that in some cases, at least, they combined all the data from multiple experiments before testing for statistical significance.

    So, for example, they say:

    TT significantly increased
    HOB tritiated thymidine incorporation by 316% after
    2 weeks of treatment compared to controls (Fig. 1A) (p = 0.03, determined by the stratified Wilcoxon rank sum test referenced in Materials and Methods, N = 3 experiments).

    and,

    No effect was found with SaOs-2 cultures at 2 weeks ( p = 0.24, N = 5 experiments).

    Did 3 independent experiments with HOB all showed significant effects? If so, that increases confidence in their reported results. But their wording suggests to me that they pooled the results of 3 HOB experiments and then analyzed for significant, and similarly pooled 5 experiments for SaOs-2. (If that’s not what they did, why would they cite a single p value for multiple experiments?)

    If the latter is the case, then they’re really reporting a single, unverified results that could be simply due to chance.

    There are a few other bothersome inconsistencies. For example, at least twice they cite p values in the text that don’t appear to be consistent with the relevant graphs. Referring to their figure 2C, they state in the text that:

    Control compared to the TT groups obtained a p-value of 0.019, control compared to placebo reached a p-value of 0.75, and TT compared to placebo was significantly different, after Bonferroni correction, at p = 0.003.

    Yet visual inspection of Fig 2C shows that the difference between placebo and TT is less than the difference between control and TT, while the error bar for placebo appears larger than for control. If that’s correct, the p value for TT versus placebo should be higher than for TT versus control, not lower (as stated in the text). Perhaps that’s just a simple misstatement or mislabeling in the figure, but there are at least a couple of things like that, which is disturbing.

    Bottom line, if they repeatedly observed that a certain treatment regimen gave a reproducibly significant effect in certain cells, that would be interesting. Obviously, it would need to be repeated by other investigators to really rule out any confounding factors the authors may have missed, just like any other interesting result.

    But as best I can tell, they have NOT reported a reproducible effect. They seem to have seen a true difference between certain groups, but they apparently haven’t shown that those differences are reproducible across experiments. They made quite a few comparisons, most of which even they admit were not significantly different. If they haven’t reproduced the supposedly significant ones, there’s no reason to think they weren’t significant just by chance. And of course, that’s exactly what the Bonferroni-corrected analysis suggests.

  63. qetzal says:

    pec wrote:

    Novella, and Orac, deliberately left out most of the data we needed to see. And they did this because the results were not weak or ambiguous or doubtful.

    and

    If Novella wanted to be honest and straightforward he would have posted the relevant data. Instead, he told us, untruthfully, that the results were a wash out.

    In other words, you’re saying Novella and Orac deliberately lied about what the article says, even though you admit you haven’t read it!

    That’s low, even for you pec. It’s contemptible. You should be ashamed, but I doubt you have that capacity.

  64. pec says:

    “you’re saying Novella and Orac deliberately lied about what the article says,”

    NO, I said Novella either lied or was mistaken — he said negative effects cancelled out the positive effects, and the overall result was nothing. That was not true. Orac didn’t say that, and he certainly would have said it if it were true!

  65. pec says:

    Darn it, I can’t even answer you Qetzal, it’s censored again.

  66. pec says:

    Orac seemed to give accurate, but incomplete, data.

  67. pec says:

    This is too frustrating. Qetzal you said I said something I did not. But I can’t answer!!!!

  68. pec says:

    One of the bloggers said something that was not true. Only one of them.

  69. pec says:

    Dr. N misrepresented the data. Dr. O did not. Although neither gave us enough.

  70. pec says:

    Maybe that’s the secret! I can’t include their names!

  71. pec says:

    “you’re saying N and O deliberately lied about what the article says,”
    NO, I said N either lied or was mistaken — he said negative effects cancelled out the positive effects, and the overall result was nothing. That was not true. O didn’t say that, and he certainly would have said it if it were true!

  72. pec says:

    Ok, the above is my comment. I took out the blogger’s names and it went.

    Damned stupid blog software.

  73. pec says:

    “They made quite a few comparisons, most of which even they admit were not significantly different.”

    Ok, but were they in the predicted direction? And were they significant by normal criteria? And do not count the cancer cell treatments — those should not be positive anyway.

  74. pec says:

    ” if a TT-treated culture was clamped at one end of the lab, and the matched control culture was clamped at the other end, if the control side was cooler, DNA synthesis in those cells would likely be inhibited.”

    A researcher would have to be brain-dead to not think of something like that.

  75. pec says:

    I did not post this study as proof of energy healing. This one suggests that there is something going on, and there are many others.

    I am reading Gary Schwartz’s energy healing book right now. Harriet reviewed it and said nothing about the many experiments he did for over a decade. Nothing at all. But she went on and on about the one “skeptic” energy experiment that supposedly proved energy fields can’t be detected by healers. Only one study was enough for her. But Gary Schwartz has done many, as have other researchers. None of that counts. You try to poke little holes in any positive experiment, but a negative experiment, just one, disproves them all.

    This study is not perfect — none ever are.

  76. Fifi says:

    Wow, I popped by and apparently it’s all pec all the time now! No wonder the spam filter thinks she’s a spambot!

  77. qetzal says:

    pec whines:

    This is too frustrating. Qetzal you said I said something I did not. But I can’t answer!!!!

    The interested reader (assuming there are any!) is invited to judge for themselves:

    Novella, and Orac, deliberately left out most of the data we needed to see. And they did this because the results were not weak or ambiguous or doubtful. If the data had been weak, they would have displayed it proudly.

    That is an unambiguous accusation that Novella and Orac of lied by omission.

    Two paragraphs later from the same comment:

    If Novella wanted to be honest and straightforward he would have posted the relevant data. Instead, he told us, untruthfully, that the results were a wash out.

    Again, an unambiguous claim of deliberate dishonesty.

    pec, you’re a liar, a punk, and a troll. P!ss off.

  78. David Gorski says:

    Pec,

    You really have to stop flooding SBM comments threads with so many comments. It’s not fair to other commenters and gives the distinct impression that you’re trying to filibuster and in essence drown out those who disagree with you using a tsunami of verbiage in the form of multiple short comments. Please consolidate your observations into far fewer comments.

    Don’t think that we bloggers at SBM haven’t noticed your behavior and that we don’t consider your behavior at present to be incredibly obnoxious and annoying. We do. We have tolerated it thus far because our dedication to free speech and letting anyone have their say has (so far) won out over the need to keep the discussion forums from degenerating into mindless free-for-alls. Behind the scenes, however, even the normally very mild-mannered Steve Novella is starting to become exasperated with your antics, and it takes a lot to annoy him–far more than it takes to annoy me. In fact, if the decision were up to me alone I wouldn’t hesistate to ban you for a month for flooding comment threads with your nonsense over the last few days and then, after the ban is ultimately lifted, put you on probation with a very low threshold for banning you again. It has nothing to do with “suppressing” your free speech and everything to do with keeping you from harming the community that has developed on this blog and destroying the value of the discussions here.

    Certainly waking up this morning, perusing the blog, and finding “all pec all the time” (as Fifi put it) in the comments made me think even more that you are not contributing anything but irritation to our regular commenters (and us), not to mention that you’re dampening their enthusiasm for substantive discussion by hijacking discussions. Your casting aspersions on people’s honesty is also most definitely unjustified and unappreciated. Stop it. Now.

  79. weing says:

    pec,

    If you are reading his book, then do it. Don’t just take his word for it. Review his actual experiments and the methodology and then make up your mind regarding the validity of his claims, not now. Then come back armed with the relevant data to back up your claims.

  80. pec says:

    I was challenged to find even one positive study on energy healing. I have found many, but of course have not read them all. I posted one that is mainstream and suggests that there is something real going on.

    I had an extremely hard time posting, and that is the main reason there are so many short comments from me. I would be accused of something and try to reply only to have most of my replies blocked. So I posted a lot of short comments. It is extremely frustrating to be accused and not able to defend yourself.

    I think the main reason Dr. G is so annoyed with me now is that I am obviously right about some important things. No one could honestly say that experiment is a pile of trash — of course Dr. N tried (can’t post their names for some reason).

    What you hate most about me is that I am logical and scientific, yet open-minded about CAM. You have convinced yourself that anyone who is open-minded about CAM has to be a self-deceiving, gullible idiot. Well you are wrong.

  81. Fifi says:

    Actually, pec was challenged to find one well conducted study on energy healing that supports her beliefs (not one “positive” study that she hasn’t even bothered looking over).

    I for one don’t “hate” you pec. I actually feel pity for you since you’re so desperate about posting here, paranoid enough to think a spam filter (you know, an algorithm aka a mathematical equation!) is personally out to get you because you’re unable to grasp even rudimentary aspects of how algorithmic programs work (which kind of points to the fact that you lie about your education in an attempt at false authority), and you clearly feel lonely and victimized by life, the universe and everything from your family, to doctors to strangers on the internet. Since you’ve made this blog all about you all the time it seems only fair to discuss you and why you act so irrationally and so much like a troll. (I think it’s quite possible you’re not a professional troll but just a rather sad, lonely, old lady who life has passed by and who is looking for someone to blame for her misery. I suspect this is the most social interact and connection to others you get, which is why you’re so attached and need constant attention.)

  82. weing says:

    You have convinced yourself that anyone who is open-minded about CAM has to be a self-deceiving, gullible idiot. Well you are wrong.

    Well then prove it. Do the work that’s required to prove you are not self-deceived and gullible by critically evaluating studies and by approaching CAM skeptically and not accepting its claims at face value.

  83. David Gorski says:

    I had an extremely hard time posting, and that is the main reason there are so many short comments from me. I would be accused of something and try to reply only to have most of my replies blocked. So I posted a lot of short comments. It is extremely frustrating to be accused and not able to defend yourself.

    Not a single comment of yours has been “blocked” by me. I check the spam filters periodically, and I have personally approved many of your comments. I’m sure our fearless leader Steve has done the same. Your comments are approved as soon as one of us gets around to moderating them. You are simply too impatient to wait for that to happen. Also, it wouldn’t surprise me if your recent flooding of the comments has led to the spam filter to view your behavior (flooding) in the comments as sufficiently spam-like that it doesn’t take much in addition in the content of your comments for it to flag them as spam.

  84. Fifi says:

    Now, back to the actual blog topic….

    It’s worth mentioning that while the internet is set up in quite a democratic manner it’s the way it’s not innately “free” or democratic for a number of reasons (access to technology being one, the other being that it can be easily controlled by changing structures and implementing various kinds of structures – which is why huge corporations throw so much money at the legal battles around controlling the internet and how people use it).

  85. Fifi says:

    It’s pretty clear that pec’s short on personal responsibility and understanding of cause and effect, and long on blame.

  86. pec says:

    “just a rather sad, lonely, old lady who life has passed by and who is looking for someone to blame for her misery”

    Shove it Fifi you are an ugly asshole. I don’t care if I get banned for this. You are sick and disgusting and evil and brainless.

  87. pec says:

    You’re an asshole Fifi. Crawl back in your ugly cave.

  88. pec says:

    You are a hideous brainless asshole. I have never had to insult anyone on a blog before. You are intolerable.

  89. pec says:

    oh fifififif you are so brilliant. I am sure your understanding of algorithms, and everything else in the universe, is profound. I just know that everyone here is deeply impressed by your wisdom and understanding. I’m sure no one thinks you are an ugly evil brain-damaged pathetic insulting no-nothing.

    I am sorry, censors, I have never tried to insult anyone else on a blog in my entire life.

  90. Forgive me if someone has already cited this elsewhere, but here is an article that discusses, in detail, what’s wrong with some of Gary Schwartz’s research:

    “How Not to Test Mediums
    Critiquing the Afterlife Experiments”

    Ray Hyman

    Abstract:

    “Professor Gary Schwartz makes revolutionary claims that he has provided competent scientific evidence for survival of consciousness and–even more extraordinary–that mediums can actually communicate with the dead. He is badly mistaken. The research he presents is flawed, and in numerous ways. Probably no other extended program in psychical research deviates so much from accepted norms of scientific methodology as this one.”

    Available at:

    http://www.csicop.org/si/2003-01/medium.html

  91. David Gorski says:

    oh fifififif you are so brilliant. I am sure your understanding of algorithms, and everything else in the universe, is profound. I just know that everyone here is deeply impressed by your wisdom and understanding. I’m sure no one thinks you are an ugly evil brain-damaged pathetic insulting no-nothing.

    What did I just say about ad hominem attacks and insults, pec? I’ll tell you. I said: Stop it. Now.

    I don’t want to hear who started it. I no longer even care. Just stop it, and stop acting like a four year old child seeing how much she can get away with.

    While I’m at it, I would point out that my admonitions to remain civil go for everyone else who has indulged in nastiness, not just pec, although only pec has recently reached such a consistent level of obnoxiousness as to warrant a stern ultimatum. I understand that pec makes it easy to sink to her level, but, please, please, please, do try to avoid the temptation.

  92. Editors Note: pec has been banned from commenting on this blog for 30 days for unacceptable behavior including over-posting, harassment of other commenters, and rude behavior. She was given what we consider to be more than fair warning but continued to behave in a way that was disruptive to productive discussion on the issues raised in this blog.

    In the name of open discussion we have had an extremely high threshold for banning readers from commenting. We regret having to take this action, but we reserve the right to maintain the integrity and functionality of this blog.

    Also note that any attempt to subvert this temporary banning will result in a permanent ban.

  93. Fifi says:

    Thanks for the link Dr Atwood.

    And a general apology for addressing pec and her accusations/assumptions and speculating. It’s just a bit hard not to view someone as a specimen and want to poke them to see what they are when they’ve totally dominated the comments section with lies and slander and are such tenacious believers in and promoters of CAM on a science-based medicine blog.

Comments are closed.