I read a lot of the pseudo-medical websites. The writing is at best pedestrian, often turgid, and, at its worst, incoherent. It is rarely either engaging or clever.
Wit, the clever bon mot, the amusing turn of phrase or retort, is rare at best. So rare I cannot think of an example. It is ironic that those who engage in fantastical treatments are so often lacking in cleverness with language and thought. The closest you get to humor are the painfully-lame cartoons at the Natural News. I am sure that the readers will flood the comments with examples of all the clever writing I have missed in the world of pseudo-medicine just to prove me wrong. Not that the reality-based world is much better. It is the rare author on the internet whose style keeps me coming back for more.
But for some reason I found “Dear Science Based Medicine, Just a Few Questions About Acupuncture” funny and engaging, at odds with most of the purple quasi-paranoid articles I normally read. Just the right amount of chatty snarkiness to be enjoyable, at least for me. So refreshing given the style of the usual pro-acupuncture comments. Your millage may vary.
I thought when reading the article that I would enjoy discussing acupuncture over a beer with Mel. And may I call you Mel? It is how you signed the entry, but I do not want to be creepy, so I am a find-and-replace away from changing it to Ms. Koppelman. If there is any personality in an article I tend to form a totally-erroneous mental picture of the writer. Oh, and sorry Mel, the first few times I read the article I didn’t look at your bio or picture (it was on a phone) and assumed Mel was a male name. My bad. It says something about my biases, n’est pas?
Mel asks many questions of SBM and I will try to answer them as best I can, although I lack the time to reply to all of them. And I will take the opportunity to ask some of my own questions as well. Mel can have a part two to her article. The gift of blog fodder keeps on giving.
Before she gets to the questions, Mel makes a key statement:
I now understand that the positive results I see in practice are due to the subjective impression of improvement without actual improvement and the lack of a controlled setting and basically people in general being, how do you describe them? Kinda simple and a bit moronic? I mean, people actually think that they’re in less pain, sleeping better, feel better in themselves, pooping more when they were constipated, pooping less when things were moving too quickly, taking less medication, and taking fewer days off of work, etc etc. What a bunch of gullible dodo brains, am I right? That post hoc, ergo propter hoc gets them over and over (and over and over) again. Poor dears.
To be clear I never think of users of pseudo-medicine as ‘kinda simple’ or a ‘bit moronic’ or as having ‘gullible dodo brains’. Mel: can you give me a half-dozen examples from the bloggers (as opposed to the commenters). I would be curious of an example. Oh yeah. Food Babe. But as my kids would say, based on her content, objectively those adjectives may very well apply. So besides the Food Babe.
While I may think acupuncture is a bit moronic, I (and I would wager my colleagues as well) do not have the same opinion of most of its users or practitioners. I try hard in my dotage to disparage the message, not the messenger, but I have no doubt you can find examples where we have referred to those who use acupuncture as some version of an idjit.
Proponents of most pseudo-medicines are human and, like me, are at risk for the innumerable ways, the many cognitive biases, that make our interpretation of causality suspect. We are horrible at understanding reality. Being wrong does not make a person a gullible dodo brain. Only wrong. Otherwise we would all be dodo brains, perhaps a poor example.
So Mel, I would ask the question, what criteria would you use to judge whether a therapy works? Subjective endpoints? Objective endpoints? Personal experience? Careful clinical trials?
And what do you mean by “works”? When I say acupuncture does not work, I mean it does not alter any primary anatomical or physiologic processes causing the disease in question. Do patients perceive benefit from acupuncture or other pseudo-medical therapies? You bet. Like all effective placebos, it can have positive psychological effects. Interactions with other, caring humans have a perceived benefit, the same benefit that occurs when a child has a boo boo kissed or two apes groom each other.
And Mel, how would you control for the endless ability of people to see what they want to see, to have their biases confirmed? To quote the masters:
There’s an old saying in Tennessee — I know it’s in Texas, probably in Tennessee — that says, fool me once, shame on — shame on you. Fool me — you can’t get fooled again.
Or perhaps more coherently:
The first principle is that you must not fool yourself — and you are the easiest person to fool.
I always remember N-Rays: at the beginning of last century some physicists ‘saw’ a new form of radiation that made no sense based on their understanding of radioactivity, and published hundreds of articles in physics journals on the topic. The researchers continued to see N rays even after the machine was disabled without their knowledge; it turned out N rays were a figment of their imagination.
And that is in the hardest of the hard of sciences, experimental physics. Whenever I see an alleged therapeutic effect, due to a pseudo-medicine or a reality-based intervention, I always ask if they are publishing the equivalent of N rays. And a good deal of the time, unfortunately, they are.
I would note up front that Mel and I differ in our standards: I never, ever, trust my experience for determining if a therapy is effective. There are too many ways I can be fooled into thinking what I am doing is effective. It is why the three most dangerous words in medicine are “I lack insurance.” No. Sorry. “In my experience.” And I remember every day that I am the one most likely to be fooled by my experience.
It is why I generally trust independent studies where bias is removed and the endpoint is not dependent on the whims of the patient and researcher: the double-blind, placebo-controlled trial. One of the issues I look for in acupuncture trials is whether the patient and researcher were blind and if the blinding was successful, since there is:
empirical evidence of pronounced bias due to lack of patient blinding in complementary/alternative randomized clinical trials with patient-reported outcomes.
And:
Lack of blinded outcome assessors in randomized trials with subjective time-to-event outcomes causes high risk of observer bias. Nonblinded outcome assessors typically favour the experimental intervention, exaggerating the hazard ratio by an average of approximately 27%
Warping bias can even happen in animal studies:
Lack of blinding of outcome assessors in animal model experiments with subjective outcomes implies a considerable risk of observer bias.
Combine the above with the issues of placebo effect and bias for pain:
We did not find that placebo interventions have important clinical effects in general. However, in certain settings placebo interventions can influence patient-reported outcomes, especially pain and nausea, though it is difficult to distinguish patient-reported effects of placebo from biased reporting. The effect on pain varied, even among trials with low risk of bias, from negligible to clinically important.
And I conclude that it can be very difficult to know if therapeutic effects are real if blinding is not fastidious, especially if the endpoints are subjective. People see what they want to see.
So Mel, I ask, since we here at SBM are often called arrogant (although not by you I hasten to add), what is more arrogant given the sins to which memory is prone? Those who rely on their experience, or those who are skeptical of that experience?
Why do you trust your experience as valid given all the potential biases that could be coloring your perception of efficacy? In other words, Mel, how do you know that you are not seeing N rays when you see acupuncture efficacy?
Mel continues:
On your site, you define acupuncture as “the practice of placing very thin needles through the skin in specific locations of the body for the purpose of healing and relief of symptoms.”
Mel does not object to the definition; it describes acupuncture but is inadequate. So I would ask, as long we are asking questions: Mel, are all 30–40–50 styles of acupuncture (Chinese, Japanese, Korean, Ear, Wonli etc. etc. etc.) equally legitimate?
I ask because as I understand symptoms, such as ‘pain, sleeping better, feel better in themselves, pooping more when they were constipated, pooping less when things were moving too quickly’ as due to the result of different physiologic processes and anatomy.
The causes of insomnia are different from the causes of diarrhea, and each can have multiple different potential etiologies. The insomnia of existential angst and guilt (am I over sharing?) is not the same as too much coffee. The constipation of hypothroidism is not the constipation of narcotic abuse.
From my reductionist reality-based approach to disease, each process would require a different intervention. Different mechanisms result in different treatments of the underlying process. And yes, I try to treat underlying processes, not symptoms. If all forms of acupuncture are equally effective for all forms of healing and relief of symptoms, then what is the underling mechanism? Because I cannot see needles in the skin having an effect of such a wide variety of fundamentally-diverse processes. Or any process for that matter, at least based on known anatomy and physiology.
According to the World Health Organization, the following are amenable in some way to acupuncture (Mel, any you disagree with?):
Diseases, symptoms or conditions for which acupuncture has been proved— through controlled trials—to be an effective treatment:
Adverse reactions to radiotherapy and/or chemotherapy, Allergic rhinitis (including hay fever), Biliary colic, Depression (including depressive neurosis and depression following, stroke), Dysentery, acute bacillary, Dysmenorrhoea, primary, Epigastralgia, acute (in peptic ulcer, acute and chronic gastritis, and, gastrospasm), Facial pain (including craniomandibular disorders), Headache, Hypertension, essential, Hypotension, primary, Induction of labour, Knee pain, Leukopenia, Low back pain, Malposition of fetus, correction of, Morning sickness, Nausea and vomiting, Neck pain, Pain in dentistry (including dental pain and temporomandibular, dysfunction), Periarthritis of shoulder, Postoperative pain, Renal colic, Rheumatoid arthritis, Sciatica, Sprain, Stroke, Tennis elbow,
Diseases, symptoms or conditions for which the therapeutic effect of acupuncture has been shown but for which further proof is needed:
Abdominal pain (in acute gastroenteritis or due to gastrointestinal spasm), Acne vulgaris, Alcohol dependence and detoxification, Bell’s palsy, Bronchial asthma, Cancer pain, Cardiac neurosis, Cholecystitis, chronic, with acute exacerbation, Cholelithiasis, Competition stress syndrome, Craniocerebral injury, closed, Diabetes mellitus, non-insulin-dependent, Earache, Epidemic haemorrhagic fever, Epistaxis, simple (without generalized or local disease), Eye pain due to subconjunctival injection, Female infertility, Facial spasm, Female urethral syndrome, Fibromyalgia and fasciitis, Gastrokinetic disturbance, Gouty arthritis, Hepatitis B virus carrier status, Herpes zoster (human (alpha) herpesvirus 3), Hyperlipaemia, Hypo-ovarianism, Insomnia, Labour pain, Lactation, deficiency, Male sexual dysfunction, non-organic, Ménière disease, Neuralgia, post-herpetic, Neurodermatitis, Obesity, Opium, cocaine and heroin dependence, Osteoarthritis, Pain due to endoscopic examination, Pain in thromboangiitis obliterans, Polycystic ovary syndrome (Stein–Leventhal syndrome), Postextubation in children, Postoperative convalescence, Premenstrual syndrome, Prostatitis, chronic, Pruritus, Radicular and pseudoradicular pain syndrome, Raynaud syndrome, primary, Recurrent lower urinary-tract infection, Reflex sympathetic dystrophy, Retention of urine, traumatic, Schizophrenia, Sialism, drug-induced, Sjögren syndrome, Sore throat (including tonsillitis), Spine pain, acute, Stiff neck, Temporomandibular joint dysfunction, Tietze syndrome, Tobacco dependence, Tourette syndrome, Ulcerative colitis, chronic, Urolithiasis, Vascular dementia, Whooping cough (pertussis)
That is one hell of a list Mel. How can needles in the skin have so many effects? What is the mechanism that ties them all together, the one ring to bind them, the commonality underneath acne, schizophrenia, Whooping cough and excess saliva? Or is there a different mechanism for each disease? To my way of thinking either these all have the same underlying mechanism to provide healing from acupuncture OR it is all N-Rays. Which brings us to the quote:
But when you say that acupuncture doesn’t work, the main argument you repeat over and over (and over and over again) is that acupuncture doesn’t work because there’s no such thing as qi.
It is, by the way, a side comment but when writing a blog entry do you presume that the reader has the background of all the prior blog entries or should the entry be self-contained? I choose self-contained, and that does lead to lots of repetition. As this entry no doubt confirms.
Mel: if there is no qi, how does so many one process (needles in the skin) accomplished dozens of different ways (Chinese, Japanese, etc) effect so many radically different processes? Seems far-fetched to me.
Mel continues with a good point:
Whether or not inserting and stimulating small needles for therapeutic benefit is effective for symptom reduction or disease resolution really doesn’t rely on the existence of qi as an explanatory model.
That is true. Acupunctures could work for both infertility and pain as well everything else on the list from an as-yet unknown underlying unified mechanism. Because I can’t see how acupuncture could work for all those processes based on what we do know about anatomy and physiology.
But the plausibility of the mechanism is important if you are Bayesian kind of gal. If you have an implausible intervention then any efficacy is likely to be a false positive and due to bias/poor methodology
Needles in the skin to alter constipation and fertility and acne and schizophrenia seems highly implausible. And as much as Bayes makes my head hurt, it appears to be the way the world works, so that it is highly probable that any effect of acupuncture is likely bias.
Smarter people than I have written about Bayes and its application of pseudo-medicine. There is some arbitrariness to deciding prior plausibility. But given the above discussion of the issues of bias and the difficulty in ascribing one therapy with such a multitude of effects, you can see how I could rate the prior plausibility of acupuncture as close to zero as you can get without reaching it. And in your essay you do not offer compelling reasons for me to change my mind:
You’re basically saying that in order for sticking needles into someone to have a therapeutic effect, then qi must exist and if qi doesn’t exist, then acupuncture has no therapeutic benefit.
Yes and no. I am saying that if qi does not exist than you have no mechanism to account for the effects of acupunctures on a multitude of processes that are all fundamentally different. You have no reason for a specific effect on a specific process.
That is different from a therapeutic benefit. To repeat my favorite metaphor, acupunctures are beer goggles: it gives the appearance of improvement with no actual change. Kind of a Zen koan, huh? Although the only thing I got out of my time in the Zen temple in Kyoto all those years ago was a welt on the back from being hit by a board during meditation to shock me into enlightenment. Fail.
Penicillin will cure S. mitis endocarditis every time, whether you believe in germs or not. Acupuncture works best when the patient believes they are getting acupuncture and believes that acupuncture is effective. When a therapeutic effect depends primarily on belief it suggests that any effect on the conditions WHO lists is illusory given the lack of a reasonable mechanism.
But Mel, you remain vague throughout, and vagueness is a wee bit harder to discuss than specifics. What then is acupuncture, and which style and how does it work and for what process?
Mel continues:
One of your main arguments against acupuncture is that it “lacks a plausible mechanism.”
But you also argue that the copious amount of research into acupuncture’s mechanisms is “not relevant to the argument” of whether or not acupuncture works. I find this really confusing because you say that “basic science considerations are, in essence, ignored in determining whether there is
sufficient prior scientific plausibility of acupuncture to treat, for instance, infertility (1) or depression (2), and equivocal, bias-prone clinical trials are ranked much higher than the basic science considerations that make the hypothesis that acupuncture can do anything for infertility so implausible as to border on impossible, barring new evidence speaking to its plausibility.”
So, if I understand you correctly, you’re saying that acupuncture as a treatment modality is implausible but you’re not going to look at the research that explores its plausibility because it’s irrelevant to the argument? In order to take a “science-based” approach to the question of acupuncture’s biological plausibility, you are going to make a conscious decision not to evaluate the research into its effects on
the brain using fMRI studies, purinergic signalling, autonomic tone using heart rate variability, effects on gene expression using high throughput metabolimics, and mechono-transduction studies compared to sham. That’s an interesting stance to take.
Assuming of course that these results are valid. To paraphrase another:
According to Professor John Ioannidis’ very well-respected and highly cited article on research methods, most published research findings are actually false.
Yep, you read that correctly.
For a variety of reasons, including funding sources, poor research design, and good old bog standard bias, most published research findings end up being unrepeatable and/or overturned. And it is on these very results that the entire institution of medical acupuncture is poised. Shaky foundations, indeed. When your acupuncturist prescribes you a needle or recommends an herb, assuming that these decisions are even based on the most up to date and highest quality research, there’s a good chance that those conclusions will be shown to be flat out wrong in due course.
Sobering, ain’t it?
Yep. Sobering. Most of the positive research on acupuncture is likely wrong.
An you note later that:
Wow, that’s a very … nuanced position. I feel relieved that the good people of Science Based Medicine are equipped to tease out these subtleties
Nuance is important in these issues. The devil, and the fascination, is in the details of a topic. Precision of language hopefully reflects precision of thought, so here’s hoping I can be precise and clear. Wish me luck.
Sticking needles in animals, especially as part of healing ritual in humans, is going to have local effects and effects in the brain. So will stubbing your toe. The question is whether it is more than the effects of the needle in the skin and results in altering the panoply of processes the WHO says are amenable to treatment from acupuncture.
I see the literature about the alleged mechanism(s) of acupunctures as than the literature of the effects of poking needles in the skin. It has no wider applicability to treating any disease or symptom, the raison d’être of acupuncture
So Mel, how do those fMRI studies, purinergic signalling, autonomic tone using heart rate variability, effects on gene expression using high throughput metabolimics, and mechono-transduction studies compared to sham shed light on mechanism for using acupunctures for stroke, leukopenia and depression? Or the GI issues noted in your opening statement?
And I wonder. As I understand it, in TCM the diagnoses for which acupuncture is used is based on tongue and pulse characteristics. Mel, just how do you translate the tongue/pulse diagnosis to reality-based diagnosis and fMRI and purinergic signalling as examples?
Any mechanistic studies about acupuncture that you have managed to read that do show a plausible mechanism don’t actually address acupuncture, just the insertion of tiny acupuncture needles in an acupuncture treatment-reminiscent fashion. Got it.
Correct. By Jove, I believe she’s got it! The nee(dle) near knee does nothing to the qi. The pin near the shin does nothing… The point near the joint… I really should edit that out… Trying way too hard but so close…
And think back to N rays. They had all sorts of research and theories to explain how N rays worked. Or cold fusion. If the effect you are seeing is an illusion, then the research explaining it is either an illusion or, more likely, irrelevant.
Harriet coined the term tooth fairy science:
“Tooth Fairy science” is an expression coined by Harriet Hall, M.D., (aka the SkepDoc) to refer to doing research on a phenomenon before establishing that the phenomenon exists…Fairy Tale scientists mistakenly think that if they have collected data that is consistent with their hypothesis, then they have collected data that confirms their hypothesis. Tooth Fairy science seeks explanations for things before establishing that those things actually exist. For example:
You could measure how much money the Tooth Fairy leaves under the pillow, whether she leaves more cash for the first or last tooth, whether the payoff is greater if you leave the tooth in a plastic baggie versus wrapped in Kleenex. You can get all kinds of good data that is reproducible and statistically significant. Yes, you have learned something. But you haven’t learned what you think you’ve learned, because you haven’t bothered to establish whether the Tooth Fairy really exists.
Given issues of prior plausibility and issues of bias in studies, I am more inclined to see such explanatory studies through the lens of Tooth Fairy science.
And:
By assigning a low prior plausibility score to acupuncture, any positive studies now magically have very little positive predictive value – in other words the chances that a positive study is a true, rather than false positive, are inversely proportional to the prior plausibility value that you made up off the top of your head without reading the literature! Bravo!!
You got it. Perfect. You do understand. Except for the reading the literature part. I did. You just cannot apply it to your own practice.
Mel continues with:
Pragmatic studies – Better to be a skeptic with a migraine than a pain free idiot, am I right? (Up high!!)
Mel notes:
If we take a condition like migraines, for example, we see in double blind RCTs that needling acupuncture points is roughly equal in effectiveness as needling non-acupuncture points and that both of these are very very effective, indeed.
The classic interpretation is that if an intervention is no better than placebo, it does nothing specific for the process being treated and you are seeing the sum total of the nonspecific and usually beneficial effects of a medical intervention a.k.a. the placebo effect.
My favorite example is internal mammary artery ligation, which was used for angina in the 1960’s until it was shown that fake surgery had the same effects as the procedure.
Mel: How would you interpret the internal mammary artery ligation studies and, more importantly, how would you apply it? By your interpretation should we still being doing internal mammary artery ligation?
Or arthroscopic partial meniscectomy? Both are equal to a sham procedure. Look at the graphs: real and sham surgery had an equal decrease in their knee pain, about 50%. About the same degree of effect you credit to both real and sham acupuncture. As a Gedankenexperiment let’s ignore the specifics and say acupuncture and arthroscopic partial meniscectomy cause an equal (50%) decline in knee pain and ignore the cost/risk/morbidity of surgery. So would you say arthroscopic partial meniscectomy works for knee pain as well as acupuncture and should be used? And how about sham arthroscopic partial meniscectomy? I mean, if you wanna be all ‘pragmatic-study’ about it?
Although not done, I would wager both sham and arthroscopic partial meniscectomy are better than wait list or medical therapy. The more complicated the placebo, the larger the effect. I see all of them as not working: not affecting a specific anatomical or physiologic process. They do have beneficial effects. They are placebos. Like acupuncture.
It is one of the ongoing ethical/interesting questions. Are placebos ethical? I think no, since to use them requires lying to the patient.
And if you’re saying that these studies show that acupuncture doesn’t work because acupuncture and minimal acupuncture are similarly effective, what do they say about the pharmaceuticals?
Some pharmaceuticals are often overrated in their effects? Water is also wet and fire is hot. The numerous issues with modern medicine and the perversion of clinical trials are a different question. It is the “there are issues with airlines, so let’s use flying carpets” argument.
Don’t you like people and want to learn more about what’s effective in making them better?), doesn’t it at least make you ask a couple of questions about the so called “science-based medical treatment” and why it’s not more effective than “SCAM” (as you so cleverly call it) in the real world? I mean, if being nice to someone for an hour a week while duping them into buying expensive woo is more effective than science-based medical drugs, with all the risks that they entail, then shouldn’t that be further explored? I mean, if you wanna be all ‘science-based’ about it?
I suspect you are using the term ‘duping’ ironically as is the rest of the sentence, but at a fundamental level that is how I un-ironically see virtually all the pseudo-medicines discussed on this site, except the duping part, which implies willfully fooling people.
Which again raises interesting questions of approach to health care. Here is mine.
My job as a physician is to accurately diagnose my patient, tell the patient what is going on, what their options are and make them better if I can. They have entrusted me with their health, their life, their hope, their money and their time, all of which are precious and I take that responsibility very seriously. Although an atheist all my life, I curiously see being in health care a calling (although a calling to what I will be damned if I know). Pseudo-medicines like acupuncture can and do waste health, waste time, waste money, waste hope because at their core they are N rays.
So no, it shouldn’t that be further explored. The answer to bad medicine is to improve it, not to explore magic systems divorced from reality.
Don’t you like people and want to learn more about what’s effective in making them better?
Mel. Mel. Mel. Such a wonderful snarkfest ruined by disparaging the motives of the messenger rather than addressing the message, albeit a minor example. I have to admit it fries my bacon, gripes my cookies or whatever angry food analogy that whips your smoothie. I am proud of the work I have done for the last 33 years and have always strived to do the best I can for my patients. I do this gig, and medicine, because I hope I am a caring, committed healthcare provider whose primary motivation is to help my patients. Certainly that is true of almost every HCW I have known, reality-based or otherwise. But that is another difference between pseudo-medicine providers and those of us at SBM; we usually do not presume malign intent. I think you are wrong about acupuncture, but I presume your motives are beneficent. But as someone said
some haters always gonna hate. That’s ok, we’ll let them.
I know. Boo hoo, poor poor pitiful me. I have an issue, where is my tissue?
Mel has more questions and I lack the time to address them further. I have a finite amount of time for SBM and need some time with my family and other parts of my growing multi-media empire. I usually do the final draft Thursday nights but with my eldest off for his final semester of college I will be taking him out for dinner instead of answering the rest of your questions. Damn. Time does fly.
Mel, I hope that helps answers at least some of your questions. I also learn a lot by writing these articles and I hope you did too.
I do not think we will ever agree, but I enjoy your style on the blog even if I do not agree with the content. I look forward to your answers and, I hope, at least one laugh at my expense.
All the best to you as well,
Mark