As I drink a glass of water


So I’ve recently been bothered by the out-of-hand dismissal of homeopathy as a fancy and placebo effect, much like “drinking a glass of water”:http://oracknows.blogspot.com/2005/12/more-evidence-that-alternative.html, and I’ve decided to do a little more digging on my own.

First, let’s start with a “definition”:http://skepdic.com/placebo.html

bq. The placebo effect is the measurable, observable, or felt improvement in health not attributable to treatment.

If you are to believe the source and the quotations within, the placebo effect requires the belief that a treatment will make you better. Taken further, this suggests the care of a trusted practitioner, although some theories suggest natural improvement over time or “regression to the mean”:http://www.visi.com/~thornley/david/philosophy/thinking/mean.html.

If, for the sake of argument only, we assume that homeopathy acts mainly through the placebo effect (and I will question this assumption in a minute), then most theories of the placebo effect suggest that even then homeopathy is not the same as “drinking a glass of water.” Drinking a glass of water doesn’t involve the care of a trusted practitioner, and in most prevailing theories, does not equate to the placebo effect.

Let me state this up front: I am for a fair, balanced scrutiny of homeopathic treatments. I do not think that homeopaths or other CAM do themselves any favors by claiming that RCTs are “inappropriate for homeopathy”:http://www.open2.net/othermedicine/programme2_transcript2.html [or other CAM practice]. I feel that any practice of medicine, except for faith healing, are open to scientific investigation. (And faith healing, which is based solely on one’s beliefs and so is not so easy to examine scientifically, should be used in conjunction with some form of medicine given by a sympathetic practitioner.)

I also understand the argument that RCTs are hard to conduct in the case of homeopathy because of the nature of the process of the practice. Let me explain this a little more.

A RCT for a pharmaceutical compound assumes that either a compound works similarly for everybody, or that factors that mitigate the response can be explained _a priori_. For example, weight often affects the impact of a compound on symptoms. This makes sense, because weight often affects maximum concentration and ability to eliminate a compound. (This is one of the many reason pediatric studies are so important.) However, weight can be identified up front as a mitigating factor. In hypertension, race is an important factor. Black people, for whatever reason, don’t respond as well to most blood pressure medications. This is a factor that is identified up front and considered in analysis.

In homeopathy, these factors are not necessarily identified up front, but come up during the process of care. In an example “here”:http://www.hpathy.com/research/bhatia-lancet-homeopathy.asp, Dr. Bhatia notes that there are over 217 remedies for injury (one of which is Arnica), and the homeopath determines which one is appropriate for a specific injury through a set of over 100 questions. (I believe we answered about 150 questions for our child’s ear infection!)

I agree that knee-jerk RCTs aren’t warranted hereSensationalistic articles entitled “The End of Homeopathy” are probably also not warranted, but I do think that homeopathy can be subjected to an honest scrutiny through a well-planned RCT. For example, I’d like to see the following noninferiority/superiority trial run:

A sample of patients (sample size determined through an appropriate sample size calculation taking into account an appropriate margin of inferiority over conventional methods, and appropriate superiority over placebo) is randomized into four groups:
# A “conventional” placebo group, where a patient sees a “conventional” doctor and receives a “conventional” placebo designed to look like a standard-of-care pharmaceutical therapy.
# A “homeopathic” placebo group, where a patient sees a “homeopath” and goes through the homeopathic process, but then receives a sugar pill designed to resemble a homeopathic remedy.
# A “conventional” treatment group, which is as the conventional placebo group but the patient receives an active standard-of-care, and
# A “homeopathic” treatment group, which is as the homeopathic placebo group but the patient receives the remedy according to homeopathic best practices and the practioner’s experience.

Then, the rate of recovery and time to recovery are analyzed as a factorial design, and safety data is reported as always. It’s worth noting here that exacerbation of symptoms is often expected in homeopathy before recovery, so that issue (and I’m sure many others I haven’t mentioned as well) need to be controlled in the protocol.

Such a study, properly controlled, documented, and analyzed should be able to tease out whether homeopathic medicine works, or whether it’s the individualized care that works, or simply does not work. Comparisons can be drawn between placebo and homeopathy as well as standard-of-care according to Western medicine standards. Granted, I’ve left out a lot of details, and some honest negotiations can be made as to what an appropriate endpoint is, and whether laboratory results are appropriate measures of efficacy. However, that is beyond my scope here and beyond my expertise. (I don’t make up the endpoints, I just analyze them.)

There are other two types of scrutiny of homeopathic practice, which really are two sides of the same selection bias coin. One is the scrutiny with an intent to prove, and the other is the scrutiny with the intent to disprove. I find most of the discussion on the interwebs to fall into one of these two categories.

There are also two approaches to scrutinizing homeopathy. The first, and the one I detailed with a hypothetical study design above, is a clinical approach. The second is a mechanistic approach: how does homeopathy, if it does, work? And it is this type of scrutiny that the “drinking a glass of water” objection falls into.

Specifically, the glass of water objection makes use of the assumption that diluting a substance to a concentration of less than 1 in “Avogadro’s number”:http://www.google.com/search?q=avogadro%27s+number&start=0 means that not a single molecule of the original substance remains in the dilution. Homeopathic remedies work by successively diluting the remedy. For example, a dilution of 6X (meaning that the dilution is cut down to one-tenth successively six times) is one part per million. 30C means that the remedy goes through a dilution of one-hundredth 30 times, giving a dilution of 1 in 10^60^, far smaller than Avogadro’s number. Therefore, skeptics of the system claim, a homeopathic remedy is simply a placebo.

Advocates of homeopathy claim that the water holds a “memory” of the remedy and cite the “prion hypothesis”:http://en.wikipedia.org/wiki/Prion (note the article as of 12/30/05 might have errors, but that doesn’t detract from my point) as an comparison. This article and the referenced ??New Scientist?? “article”:http://www.newscientist.com/article.ns?id=dn3817 article start to call into question the Avogadro’s number limit and the accompanying claim that homeopathy=placebo. While the work is certainly controversial and by no stretch of the imagination definitive and conclusive, it does invite us to call our own certainties and assumptions into doubt. In short, it invites us to be skeptical.

In short, I espouse an honest examination of homeopathy. This honest examination is not one of hostility, but one of impartiality and curiosity. It requires an understanding of how the homeopathic process works, as well as a curiosity of the mechanisms underpinning it. It requires an understanding of good science and how scientific evidence fits together to form a picture, or, if you will, theory.

Until I find that honest examination in the literature, I remained unconvinced about what science has to say about it, and I’ll have to go on my own observation and experience. And,
from the play-the-winner strategy, I’ll be trying homeopathy again.

*Update:* As I look even further, I’m finding “more assessments”:http://nhscrd.york.ac.uk/online/dare/990167.htm like mine. Wikipedia presents a “balanced”:http://en.wikipedia.org/wiki/Homeopathy, if somewhat multiple-personality-disordered, view of the subject. Although I do find it rather odd that their list of 10 sources “critical” of homeopathy involve the work of Stephen Barrett 3 times, skepticreport.com 2 times, and skepdic.com.

Advertisements

16 Responses

  1. You say “It’s worth noting here that exacerbation of symptoms is often expected in homeopathy before recovery, so that issue (and I’m sure many others I haven’t mentioned as well) need to be controlled in the protocol”

    Um…. MOST things get better eventually. Whether they’re viral, bacterial, caused by blunt injury, or by underfeeding, things heal.

    What makes any treatment effective — what makes it BETTER than ‘doing nothing’ — is that you:

    1) get better faster, e.g. recover more quickly than you would if you left things alone,

    or

    2) get ‘more better’ as my daughter would say; e.g. recover more fully than you would if you left things alone,

    or

    3) both.

    A treatment which commonly allows symptoms to worsen isn’t particularly wonderful, in most people’s view…

  2. Erik,

    Thank you for your lecture on what constitutes treatment effectiveness, complete with bullet points. You seem to have covered all the bases with your mentions of viral and bacterial infections as well as injury and malnutrition.

    You should have no trouble at all backing up your assertions and opinions about what “most people” think with a properly designed survey and study. Please let me know when you publish the results. Be sure to include such issues as comparisons with adverse drug reactions, recurrent illnesses, and chronic illnesses.

  3. Hmm. Are you disputing that definition of what constitutes an effective treatment? Obviously it’s somewhat simplified — only so much time to post — but I’m curious to know if you have a problem with those ‘bullet points’. And, if so, what are YOUR alternative suggestions?

    Surely you’re not suggesting that getting better isn’t the purpose of a treatment? Or that what makes a treatment work is that it’s better than, and distinguishable from, “no treatment”?

    If you define treatment broadly enough, you’ve widened the goal to ensure a score. But that’s not very useful. Here’s an example: I claim I can make it rain through waving a wand and shouting. I claim it is highly effective. My caveats are 1) It can take a while to work, up to 6 months; 2) If I do it soon after rain clouds have left, it may get dryer before rain comes again; 3) Some areas (Death Valley, for example) are not receptive to my act; and 4) The rain sometimes manifests as snow, sleet, or hail.

    Now, as I have defined my ‘rain cure’, it will be almost impossible (if not completely impossible) to show that it DOESN’T work. Hell, here in New England I’d bet $1,000,000 it will work every time. And it may make drought-strapped farmers feel better, but outside that it is not useful.

    I’m also confused about your demand for a published study on what ‘most people think’, as you seem to be asking. It would not be a noteworthy “discovery” to find that — for example — fewer people think homeopathy is effective for a headache than think Tylenol is effective. Simply looking at sales amounts can answer that fairly easily, and it’s not very a publishable subjct.

    But all in all, your demand for published studies seems odd, coupled with your disparagement of the ones which have been done. There do seem to have been studies, mostly by scientists, which have not shown homeopathy to be effective. Even if I grant you the benefit of the doubt and assume that your objection to the studies has absolutely no relation to your personal belief in homeopathy, it appears you’re not going to like — or accept — any negative studies.

    But fame, fortune, and the saving of homeopathy is in your hands. Prove that a solution of arnica which has been diluted far past 1 molecule/liter can still contain an essence of arnica and you’ll make it into Science and have the grant money rolling in. Or put together a good double blind study which will pass peer review and show that your remedies exceed those of a placebo. You will easily be published if you produce reliable results.

    Oh yeah, 2 other issues:
    You say: “If, for the sake of argument only, we assume that homeopathy acts mainly through the placebo effect (and I will question this assumption in a minute), then most theories of the placebo effect suggest that even then homeopathy is not the same as “drinking a glass of water.” Drinking a glass of water doesn’t involve the care of a trusted practitioner, and in most prevailing theories, does not equate to the placebo effect.”

    This is an incorrect example. Equating the homeopathic remedies to the glass of water means that if a patient who didn’t know better were to be told by a homeopathic practitioner “drink this water with a remedy in it, it will heal you”, the patient would improve by the same amount irrespective of whether there was actually any remedy in the water. There is NO general disagreement, in the alternative or conventional worlds of medicine, that having someone tell you “this will heal you” helps. Water, alone, would work (some studies suggest that sugar pills work even better, I think). The question regarding placebo is trying to determine whether you get ‘more better’ BECAUSE OF the added effect, if any, of the homeopathy.

  4. Erik, you have succeeded in the following:

    • taking my words out of context — _e.g._, the glass of water example came from a quote by Dr. Edzard Ernst, the subject of the article I’m commenting on:

      You may as well take a glass of water than a homeopathic medicine.

      and by Orac:

      Dr. Ernst is correct; homeopathy is no more better than drinking a glass of water because it is no more than drinking a glass of water.

    • assuming (incorrectly) my intentions for this article (e.g. “saving homeopathy,” whatever that means), especially since I clearly state my intentions here
    • rambling on about what constitutes an “effective treatment,” which I didn’t dispute (and I don’t need you to tell me anyway) — what I did dispute was the “exacerbation of symptoms” aspect and the relevance of your bullet points to the issues I raised in the entry, and which is much more complex that you state anyway (answer this: what constitutes an “effective treatment” of cancer — “more better”? resolution of symptoms? reduction of symptoms?)
    • failed to consider the issues I raised, namely whether “most people” think that exacerbation of symptoms is tolerable in the course of an effective treatment (as opposed to, say, a longer treatment duration, the case of recurrent illnesses, or other adverse drug reactions) (I’ll admit I probably wasn’t too clear on that)
    • suggesting a study that even homeopaths will tell you will fail (the arnica example, which if you read the references I linked to in the article, has some reasonable criticisms of such a study design), and
    • putting words into my mouth about why I am “disparaging” studies and falsely claiming that I will not accept “negative” studies when I clearly state what I think about homeopathy studies:

      I agree that knee-jerk RCTs aren’t warranted here^1^, but I do think that homeopathy can be subjected to an honest scrutiny through a well-planned RCT. For example, I’d like to see the following noninferiority/superiority trial run …

      Try reading through the post again before reading my mind on my opinion about the study of homeopathy.

    • Ignoring the fact that I address the placebo effect in my study design (and if anyone has seen/performed such a study, I’d love to see the write-up)

    I’d say you have a hack job of a comment. Try better next time, preferably after you get some actual information on the subject from a source that understands homeopathy. If you don’t like the homeopathy advocacy sites, you can try NCCAM or even the FDA, though the FDA link is a rather old. The Wikipedia article I referenced in the entry is decent, too, though you have to read the sometimes hilarious discussion on the neutrality of the article.

    Finally, if you don’t think a study on what “most people think” is worthwhile, or any evidence at all for that matter, then don’t make an assertion about what “most people think.” Your Tylenol example doesn’t apply at all to a study about whether exacerbation of symptoms is tolerable in the course of a treatment to resolve symptoms.

    If you want to be skeptical or disbelieving of homeopathy, that’s fine, and there are fine reasons for being so (I used to hold that view). What I do not find fine, however, is espousing that skepticism by bringing up irrelevant examples and issues, changing the subject, pretending to read minds, and making assertions without any evidence whatsoever to back them up.

  5. You may have been quoting Dr. Ernst. But your summary of the ‘glass of water’ argument, and your selective quotatino, calls your bias into question. Your first few paragraphs here, in particular the one which starts “If, for the sake of argument only….” appear on their face to refute the ‘glass of water’ concept, for example. Yet as a self-described biostatistician, you’d have to be being deliberately obtuse not to recognize that the quoted comment implied a course of treatment.

    You next say that “…I did dispute was the “exacerbation of symptomsâ€? aspect and the relevance of your bullet points to the issues I raised in the entry, and which is much more complex that you state anyway (answer this: what constitutes an “effective treatmentâ€? of cancer—“more betterâ€?? resolution of symptoms? reduction of symptoms?)”

    OK: 1) An effective treatment of cancer is always ‘more better’. 2) Defining that depends on the cancer involved, but can (though need not) include increases in quality of life, length of life, and general health; or decreases in pain, other discomfort, and need for additional surgery. While the patient may get ‘worse’ in one or more of those particular area due to treatment, the OVERALL status of the patient will improve as a result of any valid treatment. (e.g. even though you are in a painful semi-coma immediately after the successful removal of a cancerous breast, your overall health–no cancerous masses in your body!–is vastly improved).

    You next say: “you failed to consider the issues I raised, namely whether “most peopleâ€? think that exacerbation of symptoms is tolerable in the course of an effective treatment (as opposed to, say, a longer treatment duration, the case of recurrent illnesses, or other adverse drug reactions) (I’ll admit I probably wasn’t too clear on that)”

    I’ll agree you weren’t clear, but if that’s what you were asking, I think I just addressed that above. Though of course, what people THINK isn’t especially relevant, unless their thoughts are based on scientific evidence and study. (if I think that eating bear paw will cure my baldness, that does not make it so).

    What I think this really comes down to is simple:

    Generally, science operates on a “prove it” principle.

    -Do you believe we’re all made up of atoms? Prove it.
    -Do you think mold will kill germs? Prove it.
    -Do you think lithium salts will reduce manic depressive episodes? Prove it.
    -Do you think beaming radiation into people’s body can increase their life expectancy? Prove it.
    -And so on.

    Now, what is seems you’re asking for (though maybe I’m wrong) is the REVERSE…. you seem to be saying “You think homeopathy doesn’t work? Prove it!”

    Is this true? And if so, why should homeopathy be granted that sort of “free ticket”?

  6. 1. Why did I select the glass of water comment to comment on? Because I wanted to. Because a doctor made a quote concerning homeopathy being like “drinking a glass of water,” and I disagreed. Every blogger does this on some issue or another. It’s not obtuseness, it’s a discussion on point somebody made. If you don’t like it, feel free to skip it. If you feel the glass of water comment isn’t warranted, as I do, then dispute the original source of the quote, not my reaction to it.

    2. If you don’t think that what “most people THINK” isn’t relevant, then don’t talk about what “most people think”:http://www.randomjohn.info/wordpress/2005/12/30/as-i-drink-a-glass-of-water/#comment-14 (or, in your words there, “most people’s view”). You’re the one that brought it up, so at one time you apparently thought it was relevant.

    3. If someone is going to say “homeopathy doesn’t work,” I expect some proof of the assertion. There are major differences in meaning among the statments “homeopathy is unproven,” “studies on homeopathy have returned mixed results,” and “homeopathy doesn’t work.” If you read and comprehend the blog entry, you’ll see that I’m not giving anything a “free pass.” Instead, you’ll some problems I’ve seen with the studies currently in the literature, and some ways I feel those problems can be addressed, given the fact that I’ve designed and analyzed clinical studies.

    4. Your incessant hammering of ‘more better,’ whatever that means, doesn’t address exacerbation of symptoms. You do finally seem to address the exacerbation issue (which was apparently the point of your first comment, and you seemed to make my point quite well), but your ‘more better’ only detracts from what you are saying.

    5. Please “put spaces around your double hyphens”:http://www.randomjohn.info/wordpress/2006/01/03/comment-formatting/ to avoid the Textile plugin’s triggering del tags.

  7. 1) First, would you mind answering my last question:
    Why should homeopathy be granted a “free ticket�?

    I am curious to hear your response.
    ——–

    2) Next, you are also entirely misreading my point regarding what people “think”, to a degree that I have to wonder if it’s intentional.

    Most people don’t believe in homeopathy. That is what being an alternative medicine treatment MEANS. If the majority accepts it, it’s not alternative. But I have no particular vested interest in that argument and hell, if it makes you happy I’ll even pretend to think that 100% of all people, MDs included, think homeopathy is the bee’s knees.

    Similarly, if it makes you happy I can also pretend to think that most laymen believe that it’s just fine and dandy to get worse before you get better, and that even if they could avoid getting worse they’d be OK with it.

    Fortunately, these concessions wouldn’t matter anyway:

    Whether or not someone believes in homeopathy, or the effectiveness of antibiotics, or that men are taller than women, or the existence of quarks, has NO EFFECT on whether such things are _objectively true_. It MAY have an effect on that individual’s reponse to placebo based on their belief. But that’s about it.

    Similarly, if you accept that there are some objective measures of treatment effectiveness (discussed in my post above), whether some people think that getting worse is necessary to get better, or that a certain course of treatment is more effective, or not, isn’t relevant except MAYBE to that person.

    ———–

    3) I am not sure that you and I agree on ‘exacerbation of symptoms’. Perhaps you could restate your point yourself?

    —-

    4) As you might suspect, I _DO_ believe that homeopathy is tantamout to drinking water. That is why I disagree with your poorly phrased attack, and not with the original quote.

    This belief is based on my acceptance of the past few centuries of science, and the “no free pass” rule. As such I am attempting to conform my belief to the best available evidence regarding what is known to be true.

    I have read the preliminary studies to which you refer. And if those studies are repeated, and it turns out I’m wrong, then heck, I’ll change my mind. I’ve a science background and I adapt to new research all the time.

  8. 1. If you’re curious about my response to your loaded question, try reading my above comments again and even the blog post again. I stated the answer clearly several times in several ways, including quotes from the original blog entry. If there’s a point you don’t understand, you can ask for a clarification and I will try to clarify. If you wish to debate my response, I might even engage in that. I’m not going to respond to your loaded question again. However, I will point out that your repeated use of the question makes it clear that you have not comprehended my entry.

    As an aside, you may find that my viewpoint isn’t unique to me. Try scrolling down and searching for the Barrett vs. King Bio “here”:http://www.randomjohn.info/wordpress/2005/12/29/data-please-holding-quackbusters-to-their-own-standards-part-ii-orac-you-know/ (and follow the link therein and ignore the irritating marquees for a larger excerpt of the judgment).

    2. If you don’t care what most people think, let’s drop that idiotic point of discussion. I don’t either. Again, you’re the one that brought up what most people think:

    bq. A treatment which commonly allows symptoms to worsen isn’t particularly wonderful, in most people’s view…

    Somehow, you morphed it into what “most people think about homeopathy.” You’re the one that lectured me on what most people think, and I asked for proof, which you never gave.

    3. The exacerbation of symptoms question was related to what “most people think:”

    bq. A treatment which commonly allows symptoms to worsen isn’t particularly wonderful, in most people’s view…

    Is this the point you want to discuss?

    4. Disagreeing with me on the original quote is fine. There are people whose perspectives and opinions I respect who disagree with me. However, there is a big difference between your disagreeing with me on Dr. Ernst’s quote and your saying

    bq. This is an incorrect example. Equating the homeopathic remedies to the glass of water means that if a patient who didn’t know better were to be told by a homeopathic practitioner “drink this water with a remedy in it, it will heal you�, the patient would improve by the same amount irrespective of whether there was actually any remedy in the water.

    The quote was not, “You may as well take a glass of water and be told it has a remedy in it than a homeopathic medicine.” It wasn’t, “Homeopathy relies exclusively on the placebo effect.” It was, “You may as well take a glass of water than a homeopathic medicine.” I’m simply separating the two issues of “taking a glass of water” and the placebo effect, and you seem to have a problem with it. Personally, I think it’s a minor point, except it adds to the drama of the quote.

    By the way, if you accept the “past few centuries of science,” why not look at the history of the AMA(American Medical Association), and its relationship with the practice of homeopathy.

    Finally, your use of “mind reading”:http://ucsfhr.ucsf.edu/assist/articles/info.html?x=1059 and telling me why I’m arguing a certain way terribly weakens your argument.

  9. 1) I feel like I’m getting the runaround; why are you avoiding the question? The question seems “loaded” as you put it, because it is RELEVANT. That is why I am asking it. And I sure can’t find your “answer” here; neither have you given me any help other than “read my post”. Of course, maybe I’m just missing it…

    SO hey–assume I’m an idiot, if it makes you happy. Just restate your answer, in a sentence or two, clearly, so I can get it more easily.

    Then we’ll either agree, or have something extremely important to discuss, don’t you think?

    I’ll ask the question (really, two questions) again, so any other readers don’t have to go search:

    Generally, science operates on a “prove itâ€? principle…. Now, what is seems you’re asking for (though maybe I’m wrong) is the REVERSE. You seem to be saying “You think homeopathy _doesn’t_ work? Prove it!â€?

    1) Is this your belief?

    2) And if so, why should homeopathy be granted that sort of “free ticket�?

  10. […] An annoying conversation I’ve had with a commentator about whether “homeopathy works” got me to thinking about our biases when we conduct science (i.e. carry out the scientific process on some conjecture or hypothesis). At least when statistical reasoning is involved, there are generally two competing hypotheses: the null, which states that the conjecture is false and the status quo is true, and the alternative, which states that the conjecture is true and the status quo is false. Then, in an ideal situation, we construct our study in such a way that the probability of the success of the study is very small if our conjecture is really false (this is the famous “alpha” and is usually set at 5%, though other values are used—the FDA usually requires 2.5%). Furthermore, we often calculate a study size (e.g. number of patients in a clinical study) such that the probability of failure of the study is modest (10% or 20%) if our conjecture is really true. If the study is successful, we say that we “reject the null hypothesis” and the study backs up the alternative.1 […]

  11. Erik: none of this is personal. I could care less what your intelligence is, except for the fact that I’m going to assume that you can find the “Control-F” or “Command-F” key combination in your browser.

    Try three search phrases:

    * “i am for”
    * “i do think”
    * “i espouse”

    Please report the results.

  12. “please report the results” sounds a bit snippy, I must say, but hmm, let’s see…

    You supposedly want a “fair, balanced scrutiny”
    and you apparently think one could acheive “honest scrutiny through a well-planned RCT”
    and perhaps you would also like an “honest examination”

    Those terms are meaningless, as you know. For some people all those criteria have already been satisfied. (for an easy example of why these are meaningless, go ask a democrat and a republican whether the swift boat ads were ‘fair’, ‘balanced’, and ‘honest’)

    But I’m all for real honesty. I even read your entire referenced post regarding this “annoying” conversation here.

    Lots of generalistic statements there a well, and a lot of that “honesty” stuff again.

    SO let’s aim for honesty here, shall we?

    You seem to be claiming (feel free to correct me if I’m wrong)

    1) Failure to prove efficacy does not automatically imply nonefficacy.

    I have no argument with that limited statement. I certainly understand the stats and have run my fair share of studies.

    2) Nonefficacy is harder to prove.

    This is certainly true; sometimes it’s almost impossible to prove something cannot work. In my opinion it’s one of the reasons that certain folks like to assume they’re right, and ask others to DISprove their position. (see, for example, the work of ID advocates. See also, for example, the work of some alt med advocates).

    In logic, this is known as an “a priori” argument, and is widely accepted as a fallacy. In science, it’s equally bad form.

    3) Things whose effect has not been proven to work, nor proven not to work, exist in a sort of “unproven limbo”. “Honesty” (your term, not mine) sometimes requires that we give some of these effects the benefit of the doubt.

    Ahhhh, HERE we get into the root of the disagreement. This is the root of your error.

    The reasons that science does not bother with much DISproof are many, though I will list only the primary three.
    First, there’s little incentive to disprove anything, as there are frequently a variety of potential treatments who have NOT failed an initial efficacy test; if you’re trying to find a cure, why waste time working on your failures?
    Second, it’s extremely difficult, and expensive, to do a “disproving” right, and even if you don’t DISprove effect it doesn’t prove effect…. it makes much more sense to simply test again for effectiveness.
    Third, if you plan your study right you can have an excellent chance of catching any effect that is there. A well planned study, or series of studies (all human treatments go through multiple studies) will have a very high probability of catching any effect. (never perfect, but we don’t deal in perfection). You can calculate your chances of missing an effect fairly accurately.

    As a result, in MANY cases where researchers are trying to prove efficacy, their eventual failure to do so also serves as extremely good proof of LACK of efficacy. Run through enough subjects testing a failed pain reliever and you can reasonably conclude that it not only doesn’t reduce pain, but that it doesn’t increase it or have any significant effect at all.

    Now, when we don’t find an effect and something lands in ‘unproven limbo’ the reason that we assume things DON’T work should be obvious, but in case any readers, or you, are missing it:

    If we start assuming that some things in ‘limbo’ work and require disproof before acceptance (that’s a priori arguing),

    and yet we accept that other things in ‘limbo’ DON’T work, and require proof before acceptance,

    Who decides which category things get put into?

    You?

    I don’t think so.

    Remember, after all, that putting things into the “try to disprove this” category gives then a HUGE step up.

    And given your purported background, I have trouble believing that you would engage in a wholesale rejection of the scientific method. Though you mockinlgy refer to the “status quo” in describing the null hypothesis, you probably aren’t disputing that what we know now is based on research and knowledge, which was itself tested.

    So that brings me back to my original questions, though slightly modified to include your latest post. Let’s go for honesty here on your part. Even better, lets go for CLARITY. These are all yes/no questions, so you shouldn’t have to type much:

    Do you believe that all compounds and/or hypotheses should be assumed to be INeffective and only adopted once proven?

    Do you believe that all compounds and/or hypotheses should be assumed to be effective and only discarded once DISproven?

    Do you believe any aspects of homeopathy should be assumed to be effective and only discarded once DISproven?

  13. If it seems a bit “snippy” to you, it’s because I’m done with your twisting of my words, your oversimplification of the issues, and your pretending to understand my intentions of these blog entries when it’s clear that you either don’t understand or won’t understand them.

    But, since at least this time you made a better attempt at addressing my points, I’ll try one last time:

    [on “Nonefficacy is harder to prove”] This is certainly true; sometimes it’s almost impossible to prove something cannot work. In my opinion it’s one of the reasons that certain folks like to assume they’re right, and ask others to DISprove their position. (see, for example, the work of ID advocates. See also, for example, the work of some alt med advocates).

    In logic, this is known as an “a priori� argument, and is widely accepted as a fallacy. In science, it’s equally bad form.

    Why is it that certain skeptics, apparently you included, like to connect ID and alternative medicine? Wow. Given that ID has produced no scientific hypothesis and relies primarily on argument from incredulity (not to mention is trying to make it’s entry into the world of science by suing to get into the high school science curriculum), I’d say there is no connection or analogy here, be it for descriptive purposes or for evidence to back up your statement.

    Seeing as how you engage the correct characterization of my point later on, I’ll assume you retract your nonsense on whether I believe we should assume something is right and disprove it.

    Please point to where I say

    bq. “Honesty� (your term, not mine) sometimes requires that we give some of these effects the benefit of the doubt.

    I haven’t been able to find that anywhere. If you’re going to use my words, please use the whole context. Perhaps your framework for understanding hypothesis testing is radically different from mine. Perhaps your framework doesn’t accept limbo. Perhaps your framework doesn’t include neither accepting nor non-accepting. In which case, you are never going to understand my position, and you are never going to understand why I’m not giving *anything* a “huge step up.”

    Finally, I’ll take issue with a few other statements:

    bq. And given your purported background, I have trouble believing that you would engage in a wholesale rejection of the scientific method.

    Good, because I’m embracing the scientific method as a very useful tool. The scientific method is not on trial. If you notice in this entry and in the other and the other, what I attack is hostility. What I attack is editorials of negative studies in alternative medicine that make far-reaching statements such as “The End of Homeopathy” and “[Echinacea] bites the dust,” especially when those studies have valid criticisms, and when, by design, negative studies are weaker evidence of no effect than positive studies are of positive effect.

    bq. hough you mockinlgy refer to the “status quo� in describing the null hypothesis,

    You’re clearly reading something much more into my words that is there. Wow. Maybe Valium will help, or a nice cup of chamomile tea and a hot bath.

    bq. If we start assuming that some things in ‘limbo’ work and require disproof before acceptance (that’s a priori arguing),

    This is a ridiculous clause. You can assume that if you wish, or you can assume that I’m saying that. In both cases, you’d be wrong.

    Who decides which category things get put into?

    You?

    I don’t think so.

    Seriously, dude. Look into chilling out. It’ll be more effective than anti-hypertensives.

    Let me put it real simple for you:

    # If you want to claim that homeopathy works, I expect to see evidence of that claim.
    # If you want to claim “The End of Homeopathy,” I expect to see evidence of that claim.

    Given the results of studies are mixed, recent ??The Lancet?? meta-analysis and accompanying over-reaching editorial notwithstanding (and this is a meta-analysis that is not without criticism), I think it fair to expect evidence of a claim in either direction, especially extravagant claims of the end of homeopathy and, in the case of the echinacea trial editorial, a call for the end of NCCAM.

    If you have a problem with this, have fun with your beta-blockers.

  14. Well, i can skip over the insults. Though I’d note for someone who accuses me of twisting your words, you’ve… oh hell, not worth it.

    Back to the fun stuff.

    I asked:
    Do you believe that all compounds and/or hypotheses should be assumed to be INeffective and only adopted once proven?

    Do you believe that all compounds and/or hypotheses should be assumed to be effective and only discarded once DISproven?

    Do you believe any aspects of homeopathy should be assumed to be effective and only discarded once DISproven?

    The closest thing I can find to an answer is this:
    ” 1. If you want to claim that homeopathy works, I expect to see evidence of that claim.
    2. If you want to claim “The End of Homeopathy,â€? I expect to see evidence of that claim.”

    Now, the followup question:
    Imagine I replace # 2 with “If you want to claim “The End of ________________,â€? I expect to see evidence of that claim.”

    And I fill in the blank with:
    -Astrology
    -Divination
    -Faith healing
    -Pyramids which make “super water”
    -Ear candling
    -Bleeding as a treatment for disease
    -Beatings as a treatment for schizophrenia
    -Sacrifices as a way to make it rain
    -Young Earth theory

    Would you also say the statement was true? Could I put ANYTHING in there, and would you agree it was true?

    And if not, why oh why do you treat homeopathy differently? COuld it be that dread “bias” you talk about?

    For example, given your citations of “many studies” in your other post, and given the many centuries of studies done on water which taken together suggest it is not affected by superdilution, you must have some trouble believing that the recent ‘water is affected’ study is accurate.

  15. bq. Well, i can skip over the insults. Though I’d note for someone who accuses me of twisting your words, you’ve …

    Provided quotes side-by-side showing how.

    Now, for your followup question:

    If someone wants to “end”
    * astrology
    * divination
    * faith healing (I qualify this in posts elsewhere, basically as an adjunctive therapy)
    * Pyramids which make “super water”

    If we are going to declare an “end” to these practices, I’d like to see a good reason why. These practices are what people do in the comfort of their own homes. Conversely, if people want me to believe in those practices or hypotheses, I expect some evidence. If people want to teach these practices in the science classroom, I expect some science.

    * ear candling

    I’m not sure what practice this is, but apparently it is part of a setup to make me look ridiculous.

    * Bleeding as a treatment for disease

    Never say never. The FDA has “approved leeches”:http://www.fda.gov/bbs/topics/ANSWERS/2004/ANS01294.html for some conditions.

    At any rate, this practice was, ironically enough, one of those in response to which homeopathy was developed. While no double-blind placebo-controlled studies have been run on the subject, I think other evidence (e.g. the success of other types of treatment and theories on the human body) sufficed to remove the medieval practice of bleeding as a serious contender of medicine.

    * Beatings as a treatment for schizophrenia
    * Sacrifices as a way to make it rain

    Given that these are considered unethical and, at least for schizophrenia, more humane alternatives are available, I don’t think we need to bring science to bear on these.

    * Young Earth theory

    Get real. Most scientific evidence is against it, but are we going to ban it? Are we going to go around to Sunday School classrooms with a bullhorn and declare an end to this theory?

    At any rate, these examples are not analogous to homeopathy, unless there are studies (well-controlled or at least with some attempt at good design) with mixed results on faith healing, bleeding, or beatings.

    bq. you must have some trouble believing that the recent ‘water is affected’ study is accurate.

    Sigh. Let’s try this again: you cannot argue both my side and yours. I’ll present mine again:

    bq. While the work is certainly controversial and by no stretch of the imagination definitive and conclusive, it does invite us to call our own certainties and assumptions into doubt. In short, it invites us to be skeptical.

    I’ll even agree with the following statement:

    bq. … and if those studies are repeated, and it turns out I’m wrong, then heck, I’ll change my mind. I’ve a science background and I adapt to new research all the time.

    I’ll end with this:

    bq. In short, I espouse an honest examination of homeopathy. This honest examination is not one of hostility, but one of impartiality and curiosity.

    All I’ve seen out of your arguments is hostility (toward homeopathy).

    So, let me turn the tables here. I’ll ask you one question, which doesn’t require you to accept anything:

    p{border: thin}. Should homeopathy be investigated further, or should investigation of homeopathy end now?

  16. I’ve closed comments on this thread because we seem to have migrated.

    Please go “here”:http://www.randomjohn.info/wordpress/2006/01/09/prove-it/ to leave comments.

Comments are closed.

%d bloggers like this: