Perception and reality in acupuncture

Last week, I wrote about an overhyped acupuncture study that purported to show (but didn’t, really) that acupuncture is more effective than “conventional” therapy in the treatment of low back pain. This story reverberated through the Internet and blogosphere as “proof” that acupuncture “works” when in reality the study was very weak evidence of any effect due to needles being placed into the skin and only showed that meridians are a meaningless concept (i.e., the finding that sham acupuncture was just as effective “real” acupuncture). If acupuncture does do anything, this study was pretty thin gruel to demonstrate it.

In my field, cancer, one common claim for acupuncture is that it is a useful adjunct that alleviates chemotherapy-induced nausea. It turns out that a study was recently presented at the European Cancer Conference in Barcelona last week that addresses this very question, and in doing so it sheds some light on the appeal of acupuncture:

Barcelona, Spain: Despite widespread belief among cancer patients and health care professionals that acupuncture helps relieve nausea caused by cancer treatment, new research in radiotherapy has found it does not.

The study, presented today (Wednesday) at the European Cancer Conference (ECCO 14) in Barcelona, evaluated the effectiveness of acupuncture in 215 patients with various types of cancer who got either active acupuncture or a sham treatment that involved an identical looking and feeling needle that retracted into the handle on contact with the skin.

The treatment, in the forearm, was given two or three times a week during the whole period of radiotherapy, which is a common schedule. The patients documented their nausea and vomiting in diaries and completed questionnaires during and after the radiotherapy course.

“Both groups of patients reported they believed the treatment had been invasive and effective in reducing nausea. However, 68 percent of patients who got the acupuncture experienced nausea for an average of 19 days during radiotherapy and 61 percent of the patients who got the sham treatment suffered nausea for an average of 17 days,” said the study’s lead researcher Anna Enblom, a physiotherapist and doctoral student at the Faculty of Health Sciences, Linköping University in Sweden.

Vomiting was experienced by 24 percent of the patients getting acupuncture and 28 percent of patients receiving the sham treatment, Enblom added.

Fifty-eight of the patients received chemotherapy in combination with radiotherapy. Among them, 82 percent of those in the acupuncture group developed nausea, compared with 80 percent of those treated with the sham needles.

“There was no statistically significant difference between the groups in the number of days with nausea or vomiting or in the intensity of the nausea, neither in the patients receiving radiotherapy alone, nor in those receiving a combination of chemotherapy and radiotherapy,” said Enblom. “Our study may indicate that attitudes and expectations play a major role in the experience of the effect of the treatment.”

Acupuncture is increasingly used in cancer care as an alternative to drugs to treat therapy side effects such as pain and nausea, despite unclear evidence that it has any real effect. The percentage of cancer patients using acupuncture for various kinds of symptoms, including nausea and pain, varies from two percent to 30 percent depending on the population.

My first thought was to marvel that this study actually used the sort of retractable needles suggested by Steve Novella as controls. This aspect of the study design alone, more than anything else, made it possible to actually blind the caregivers as well as the patients as to which treatment group they are in or dealing with. My second thought was that I’m not particularly surprised. Although it is theoretically possible that acupuncture works for pain control or nausea, a beneficial effect on chemotherapy-induced nausea due to acupuncture is less scientifically plausible than an effect on pain.

What is really interesting about this study is that it suggests a reason why so many people believe that alternative medicine, be it acupuncture, homeopathy, or whatever, works for them, even in the absence of any real objective evidence that it does anything. Here we have two groups of patients, one of which received “real” acupuncture and one did not, but who were truly blinded to which group they were in. Even though there was no objectively-measurable difference in the intensity or duration of chemotherapy-induced nausea and vomiting among the groups, the overwhelming majority of both groups thought acupuncture helped them and would like to use it again. My only quibble about this study is that it would have been strengthened by a third group receiving no treatment, although the addition of such a group could have altered the study by destroying its double-blind nature. Be that as it may, just the expectation that the treatment will work clearly led both groups to believe that it did, even though there was no objective evidence that either group did any better.

Certainly, this study won’t be the final word, and it will be interesting to see what the final manuscript shows, when all the details of the study design are presented. It turns out that the literature is all over the place regarding the use of acupuncture for chemotherapy-induced nausea, with the vast majority of studies tending to be of low quality and mostly negative or only weakly positive (like most alternative medicine studies). There is also a tendency to confuse electroacupuncture (acupuncture with electrical stimulation), for which there is some, albeit weak, evidence of efficacy in chemotherapy-induced nausea but which, I would argue, is a very different thing than acupuncture, given that the originators of acupuncture didn’t have electricity to run through their needles. (Maybe applying current is a better way of manipulating the flow of qi.)

One of the reasons that evaluating so many of these alternative therapies scientifically can be so difficult is because of this rose-colored view many patients have of their treatments, such that they look for every evidence of their efficacy and, even if there really isn’t any, still view the treatment as helpful, a phenomenon known as confirmation bias. This is a very difficult form of bias to overcome, both in patients and their caregivers. As the press release concludes:

In the study presented at ECCO, 66 percent of patients who got acupuncture and 71 percent who got the sham treatment said they would be highly interested in having acupuncture again if it turned out they needed another course of radiotherapy.

In Sweden, the cost of acupuncture is covered by the state healthcare system when given for indications where it is proven to work.

As long as the best possible available drug treatment for nausea is given to cancer patients receiving radiotherapy, the patient may be satisfied with a gentle and pain free acupuncture treatment and believe it helps. It is not recommended to give acupuncture if it produces pain or harm.

If I lived in Sweden, I’d say that I have no objection–as long as the patient pays for the acupuncture himself. After all, if there’s no scientific evidence that it objectively decreases the severity, duration, or intensity of nausea and vomiting, why should taxpayers be expected to pay for it? In a state-funded health care system, any money diverted to ineffective woo is money that doesn’t go to pay for effective treatments for other people. It’s exactly the same reason that, if I lived in the U.K., I’d be very upset that my tax money was being spent on homeopathy. It’s a zero sum game.


  1. #1 Marcus Ranum
    October 2, 2007

    The only value woo woo like accupuncture offers is that it serves as a measure of how comfortable/disaffected patients are with science-based medicine. I watched Dawkins’ “Enemies of Reason” series and was struck by the point he made, namely that “alternative” medical practitioners spend a full hour with their “patients” and palpate them, ask them lots of questions, pay them attention and generally make them feel special and important. Contrast that with a visit to a real doctor, and it looks like medical costs, dealing with insurance paperwork, and poor bedside manner has a lot to do with what makes woo attractive.

    If I were puking with chemo, I’d hire a hot 19-year-old to wear a cheerleader outfit and massage my feet while showing me lots of cleavage. My guess is it would cost about the same and it’d be an equally effective course of therapy. Perhaps services like that could be offered under the alternative therapy banner. Having a crowd of people keening, wailing, and rending their garments outside my death-chamber would be kinda cool, too. If I still have a memory when I’m getting close to going, I’ll see if I can do that one. That might be fun. “What’s going on?” “the prophet is dying!!! *wail*” “Huh?”

  2. #2 coturnix
    October 2, 2007

    But does it work in monkeys and other animals?

  3. #3 Ezekiel Buchheit
    October 2, 2007

    Again, this is what makes woo so very evil. It prays on the most desperate. If you’ve got cancer I imagine it’s maybe, just a bit, somewhat difficult to keep a clear perception of reality.

    I’m with Marcus. I get cancer and I’m stoned from sun-up to sun-down. I haven’t used a recreational drug in years – hell, I don’t even drink. But that cancer comes in, I’m a shaman.

    Makes me wonder if anybody has ever thought and gone through the effort to set up some kind of volunteer comisceration service. Sometimes it’s nice to have attention focused on you from otherwise complete strangers.

  4. #4 Chemgeek
    October 2, 2007

    Marcus, I don’t disagree with you. However, I bet real doctors need to see more patients that the altie doctors. I’m sure most (not all) doctors would love to spend a full hour with each patient. But, seeing only 10 of the 30 patients you need to is not going to help the other 20. Maybe I’m wrong.

    As for the cheerleader idea, now you are on to something. *brilliant*

  5. #5 Graculus
    October 2, 2007

    Having a crowd of people keening, wailing, and rending their garments outside my death-chamber would be kinda cool, too.

    I am interested in your ideas and would like to subscribe to your newsletter.

    Hey, Orac, how about some discussions of any studies of the benefits of marijuana for allieviating chemotherapy nausea? I haven’t seen any that used placebos (what do you use for placebo, crabgrass?).

  6. #6 Clare
    October 2, 2007

    The patient’s expectations and the ritual employed contribute to woo’s appeal, I’m sure. But as Marcus says, and as the anthropological literature on cross-cultural woo shows, all that goes on *around* the ritual (which can be much more time-consuming than the ritual itself) makes a difference too. But that’s just my conjecture; I don’t know how you’d test this experimentally.

  7. #7 sailor
    October 2, 2007

    Something we may be missing in our discussion of placebo, needles and other things, is that we have to recognize two separate things; the symptom and how the symptom is perceived. In other words you could have two patients showing the same symptoms, (failure in a test) but for one it is a minor inconvenience, for the other it demands a shaded room and bed rest. Shifting someone from the second case to the first is a very desirable outcome. No measures of this were taken, but I would think this would be an easier shift for placebo than changing the symptoms.

  8. #8 Sid Schwab
    October 2, 2007

    I recall seeing a video in med school of an operation done, supposedly, with only acupuncture as anesthesia. It was, in fact, a thoracotomy, which is a very painful operation. Having been produced in China, I couldn’t say now that it was legit. But I was impressed at the time, and remember thinking that in 5000 years those guys may have come up with points in the neural pathways, input from which might affect perception of pain. Now, with a few actual studies like this being reported, I put it in with all the other woo. And I wonder what really went on back then.

  9. #9 Ezekiel Buchheit
    October 2, 2007

    Sid: I think Steve Novella talked about that video a while back on his (f-ing excellent) podcast The Skeptics Guide to the Universe (get it on iTunes). If I remember correctly (given that the video is the same), the man getting the surgery is repeating over and over something like “pain, pain, pain” or something to that affect while the surgeon tells him, in essence, to just shut up.

  10. #10 qetzal
    October 2, 2007


    I don’t know about placebos for smoked pot, but there’s already an approved drug in the US that consists of delta-9-THC dissolved in sesame oil. It’s supplied in a gelatin capsule for oral adminstration, and it has been studied in double-blind, placebo controlled trials. See here.

    Sid Schwab,

    Consider. If someone really could do that under controlled circumstances, why would the only evidence consist of some obscure Chinese video? Even if it really were some mystical process, and only one guy in China could do it successfully, that one guy would only have to do it one time in a conventional hospital setting, under observation by trained surgeons and anesthetists. He’d be famous, and acupuncture would be proven beyond doubt.

    So why hasn’t that happened? I can think of three possibilities:

    1) The effect is real, but everyone who really knows how to do it wants to keep it a secret.

    2) The effect is real, but the vast conspiracy of conventional anesthesiologists and surgeons have been wielding their enormous power to keep such a demonstration from happening. (Shame on you, Orac!)

    3) It’s a load of BS.

    Put me down for #3.

  11. #11 Theo Bromine
    October 2, 2007

    Dammit, Jim, I’m an engineer, not a doctor, so this may be a stupid question, but I am wondering about “electric acupuncture”. I have seen a physiotherapist treat neck pain (not mine, but I was there) by sticking needles around the painful area and running an electric current through them. No chi or meridians were cited, and as far as I could tell the physiotherapist was completely committed to conventional medicine. I’m curious as to where this would register on the woo-meters of Orac and knowledgable readers.

  12. #12 Sastra
    October 2, 2007

    I was once talking to someone who was studying to be some sort of alternative “energy healer” and, to my surprise, under persistent questioning she admitted that it wasn’t important if the effects were only placebo, imagined, or erroneously thought to be helpful when they really weren’t. Maybe there was no such thing as healing energy. So? All that counted was whether someone was happier because of something she did.

    Truth, lies, accuracy, misrepresentation — everything fell away before the only thing that mattered to her: making people feel better, if only a little. She saw no moral conflict here at all. In fact, I was the one with the problem, because I cared more about “being right” than I cared about people. Shallow of me.

  13. #13 boojieboy
    October 2, 2007

    Forgive me if I missed something, but I am confused about the decription of the placebo:

    The placebo needles made contact with pressure, but did not penetrate/puncture the skin, is that correct?

    The problem this creates is that it may not fairy distinguish between the two treatment conditions. It seems to me that some acupuncturists might claim that the pressure is in and of itself sufficient to cause the effect, and puncture isn’t necessary.

    Or did I miss something?

  14. #14 Obdulantist
    October 2, 2007

    “Or did I miss something?”

    No, that is a valid criticism of the methodology.

  15. #15 Obdulantist
    October 2, 2007

    Something we may be missing in our discussion of placebo, needles and other things, is that we have to recognize two separate things; the symptom and how the symptom is perceived. In other words you could have two patients showing the same symptoms, (failure in a test) but for one it is a minor inconvenience, for the other it demands a shaded room and bed rest. Shifting someone from the second case to the first is a very desirable outcome.
    sailor | October 2, 2007 11:03 AM

    There is a serious problem with this. It relies heavily on the assumption that the reported symptom accurately reflects a common underlying physiological state in the two patients.

    But just because two patients exhibit apparently the same symptom, it doesn’t follow that they have no major underlying physiological difference, and that the different experiences of that symptom can be simply attributed to psychological factors (if that is what you are arguing, it is not entirely clear).

    When people start talking of patient perception, there is usually little discussion of the often arbitrary and untested (and sometimes objectively untestable) perception/attribution of the clinician as well. This sword definitely cuts both ways.

  16. #16 Shiritai
    October 2, 2007


    In addition to not puncturing the skin, the needles were not placed on the meridians that acupuncture dictates. So, we have random pressure as the equivalent to precisely placed needles, which refutes acupuncture in this case.

  17. #17 Graculus
    October 3, 2007

    It’s supplied in a gelatin capsule for oral adminstration

    I’m sorry, but if something is needed to treat nausea, oral administration is a non-starter.

  18. #18 RS
    October 3, 2007

    Graculus, lots of anti-emetics have an oral route of administration. It is only when things get really bad that you have to switch to alternative routes.

  19. #19 Orac
    October 3, 2007

    Quite true. Oral anti-emetics are usually the first treatment, unless the patient can’t keep anything down.

  20. #20 Graculus
    October 3, 2007

    Oral anti-emetics are usually the first treatment, unless the patient can’t keep anything down.

    IIRC, the studies done on marijauna were on people who had not responded to regular anti-emetics, anyway. It seems to me that for any nausea severe enough to prevent eating in the first place it might be good medical practice to skip the orals.

    Also, I believe that there are studies that show that bare THC is not as effective as the whole herb, perhaps some synergistic effect with one of the other compounds.

  21. #21 Marcus Ranum
    October 3, 2007

    Ezekiel writes:
    Makes me wonder if anybody has ever thought and gone through the effort to set up some kind of volunteer comisceration service. Sometimes it’s nice to have attention focused on you from otherwise complete strangers.

    I bet that you’d find it would be amazingly therapeutic (mentally) for patients, especially terminal ones. I had a distant acquaintance 2 years ago who was in the final stages of MS. She knew she was going to die, I knew she was going to die, and I happened to be in town for a conference and invited her out. She said it was absolutely fantastic to be able to hang out with someone who wasn’t walking around her on eggshells or trying to tell her “it’s going to be OK…” and blowing upbeat lies in her face. It’s not like I was a bummer, though – I took her out to one of Toronto’s better topless clubs (carried her wheelchair and her up a flight of stairs!) and bought her lapdances (the dancers really exerted themselves for her; it was great…) and listened to her bitch about how much it sucked to be young and dying. Afterwards she said that she’d held that in for years because she felt she had to be strong for her friends and couldn’t complain – but it was OK to complain to a near-stranger. I never heard from her/of her again; I suspect she’s dead by now. Christine – it was nice knowing you.

    I was completely wrung out, afterwards, but I found the whole experience really refreshed my appreciation for being alive, being able to drink a couple shots of tequila, and get a good lap-dance while I can.

    Could someone offer such a thing as a volunteer service? Absolutely. Would it be valuable? Absolutely. My guess is it would be intensely draining on the volunteers – basically you’re taking a heavy load on; I couldn’t do that more than once or twice a year without getting seriously depressed.

  22. #22 Ezekiel Buchheit
    October 3, 2007

    Marcus: I wonder how one goes about setting something like this up. There’s got to be all sorts of logistics to deal with.

    I am genuinely interested in starting a volunteer service like this (how the hell I would fit that into my schedule, I don’t know.)


New comments have been temporarily disabled. Please check back soon.