I’ve had the Monday spot on this blog for quite a long time now. While there are many advantages to posting on Monday, not the least of which is having more time to put a post together (although that is also a disadvantage because it incentivizes my taking more time than I sometimes should), one distinct disadvantage is that all the Monday holidays are mine. That leaves me a choice on, for example, Memorial Day and Labor Day every year. It’s a choice with three options: The first is: Don’t post. (Do stop laughing, please. I know it’s me.) The second is: Use a guest post. I thought about this, and there are guest posts in the pipeline, but I don’t like to use a guest post just because I’m feeling lazy. It strikes me as an abuse of being the managing editor. Finally, I could post, which is what I decided to do.
I mainly decided to do a post because late last week there was something to blog about that was pretty important and interesting, courtesy of the National Center for Complementary and Integrative Health (NCCIH), which was formerly known as the National Center for Complementary and Alternative Medicine (NCCAM). Basically, on September 1, the NCCIH published a press release, Review Examines Clinical Trial Evidence on Complementary Approaches for Five Painful Conditions. Elsewhere, NCCIH Lead Epidemiologist and first author of the review, Richard Nahin, PhD, MPH, bragged on the NCCIH blog, New Review Offers Providers and Researchers Evidence-Based Information on Complementary Health Approaches for Pain. It didn’t take long for headlines to start appearing that said things like:
- NIH Study Gives Primary Care Docs Needed Evidence to Complement Chronic Pain Treatments
- Review Suggests Safe, Effective Ways to Relieve Pain Without Meds
- Yoga, Acupuncture Can Help Some Pain, Studies Find
You get the idea.
The NCCIH even held a Twitter chat on Thursday under the #talkpain hashtag to promote its new review, which had been carried by Nahin in conjunction with other investigators at the NCCIH. Co-authors included Robin Boineau, MD, MA, a medical officer at the NIH; Partap S. Khalsa, DC, PhD, a chiropractor and Deputy Director, Division of Extramural Research—yes, Deputy Director; Barbara J. Stussman, BA; and Wendy J. Weber, ND, PhD, MPH, a naturopath and Branch Chief, Clinical Research in Complementary and Integrative Health Branch, Division of Extramural Research. If you want to know why NCCIH supports so much pseudosciene, look no further than its having chiropractors and naturopaths in high ranking positions. Basically, this is a review by NCCIH for NCCIH believers.
I must admit that the Twitter chat was amusing. It was scheduled for 1 PM Thursday. Normally at that time I’m still in clinic, as I’m seldom able to finish my Thursday morning clinic before 1 PM, but I’ve noticed that my clinic is often slow on weeks before major holiday weekends, and this week was no exception. Not only was I finished before 1 PM, but I had had time to stop at the food court and get myself some grub to chow as I watched the festivities, which consisted of a whole lot of credulity, with a few skeptics interjecting some reality every now and then.
Did the review article—not an original study!—rate the promotion? Let’s find out.
Framing CAM as “non-pharmacological interventions”
The first thing I feel the need to point out here is the framing of this article, which borrows heavily from the framing used by advocates of “complementary and alternative medicine” (CAM), or, as it is now more frequently called “integrative medicine” (or, as I like to call it, integrating pseudoscience and quackery into medicine). That framing derives from a common device used by CAM advocates since the 1990s at least, in which potentially science-based interventions based on lifestyle changes, such as dietary interventions and exercise, were co-opted as being somehow “CAM,” even though there was nothing “alternative” about them. Even after CAM morphed into “complementary and integrative” approaches, there’s nothing “complementary” about these science-based methods, either, or at least there shouldn’t be. After all, what is the first science- and evidence-based intervention for type II diabetes? It’s to alter the diet, lose weight, and exercise, with the hope of reversing it, or at least preventing the need to begin pharmacological management. Ditto hypertension. I remember when I was first diagnosed with hypertension several years ago, partially likely thanks to the curse of genetics but also partially due to my sedentary lifestyle. My doctor insisted on at least three visits to see if I could bring it down without drugs, even though I thought it was too high for that. I couldn’t; so I ended up on an antihypertensive. Anecdotal evidence? Sure. However, it demonstrates that a lot of primary care doctors don’t go straight to the prescription pad, contrary to the stereotype promoted by CAM advocates.
So here’s where the framing gets more sophisticated, as it does in this review article. After co-opting sensible and potentially science-based lifestyle interventions as being “complementary” or “integrative,” CAM advocates, such as Nahin et al, for pain Nahin et al divide interventions into two categories, “pharmacological” and “non-pharmacological.” Naturally, CAM claims all the non-pharmacological interventions for its own, furthering the stereotype that science- and evidence-based medicine is about nothing more than drug therapy:
Back pain, joint pain, neck pain, and headaches are among the most common types of pain experienced by US adults (Table 1).3, 4, 5, 6 The prevalence rates for these conditions have remained relatively stable over time. Among the many pharmacological and nonpharmacological approaches that have been incorporated into pain management strategies are complementary health approaches. This broad category of care includes procedures by licensed practitioners such as acupuncturists, chiropractors, and massage therapists, as well as self-care approaches such as relaxation techniques (eg, meditation) and meditative movement-based approaches (eg, yoga and tai chi) and natural products such a glucosamine and herbal medicines. National surveys going back more than 25 years have consistently found that these complementary approaches are used by about 30% to 40% of the US public in a given year,7, 8, 9, 10, 11 although use of a given approach may wax and wane over time (Table 2).
There is no doubt that chronic pain has become a major public health problem. Flowing from this public health problem has been an epidemic of abuse of prescription opioids, an issue that has recently been prominently covered in the press and recognized by the CDC as a massive public health problem, causing record numbers of deaths in 2014. So there’s no doubt that finding science-based non-opioid strategies for patients with chronic pain to obtain relief is very important, arguably one of the top three or five priorities for public health in the US. Naturally, whenever there’s a public health problem, CAM advocates see an opportunity to promote pseudoscience, such as acupuncture. Of course, the CAM modalities examined range from quackery (acupuncture) to potentially helpful and science-based but co-opted by CAM (various exercise approaches like tai chi and yoga) to supplements, which, when they work, are basically pharmacological interventions disguised as nonpharmacological interventions.
The evidence base for CAM interventions for pain: Methods
In their review, Nahin et al examined existing evidence addressing the efficacy, effectiveness, and safety of seven common CAM approaches: acupuncture; spinal manipulation or osteopathic manipulation; massage therapy; tai chi; yoga; relaxation techniques including meditation; and selected natural product supplements, including chondroitin, glucosamine, methylsulfonylmethane (MSM), S-adenosyl-L-methionine (SAMe), and omega-3 fatty acids. They examined these interventions in five common painful conditions: back pain, osteoarthritis, neck pain, severe headaches and migraine, and fibromyalgia. Their methodology is described thusly:
The randomized, controlled clinical trial (RCT) is considered the strongest study design for investigating the efficacy and safety of pharmacological, behavioral, and physical interventions. To identify examples of RCTs for each complementary approach, we performed searches of the MEDLINE database for articles published from January 1, 1966, through March 31, 2016, using the search strategy outlined in Supplemental Appendix 2 (available online at http://ift.tt/10yXC6N). In order to make this review as relevant as possible to primary care physicians in the United States, we limited this review to RCTs either proformed in the United States or that included participants from the United States. This decision was based on 2 factors. First, the unique health care system in the United States vs other countries means that the standard care or usual care control groups used in the United States and other countries may vary substantially. Thus, whether a given complementary approach performs better than usual care in another country may not reflect how the approach would perform in US trials. Another factor is that the training and licensure of acupuncturists, chiropractors, and naturopathic doctors vary substantially among countries, as does the marketing, regulation, and use of dietary supplements. For instance, in Germany, the location of some of the largest acupuncture trials, acupuncture is only practiced by medical doctors, whereas the vast majority of acupuncture treatment in the United States is provided by licensed acupuncturists. Thus, the findings from German trials may not be directly comparable to acupuncture as practiced in the United States.
Notice something here? One sentence stood out: “Thus, whether a given complementary approach performs better than usual care in another country may not reflect how the approach would perform in US trials.” I can’t recall how many times I and other members of the SBM team have pointed out that a given intervention will nearly always perform better than usual care. In acupuncture trials, for instance, acupuncture virtually always produces better results than “usual care.” What matters is whether an intervention performs better than a sham control. In the case of acupuncture, the evidence is very clear. Acupuncture does not perform better than sham acupuncture using the “wrong” acupuncture points or retractable needles that disguise to both practitioner and patient whether the needles actually enter the skin. As I like to say, it doesn’t matter where you stick the acupuncture needles. It doesn’t even matter if you stick the needles in, my favorite example being a German study that used toothpicks twirled against the skin as its sham control and found no difference between “true” or “verum” acupuncture over sham. That’s because acupuncture is nothing more than theatrical placebo.
Be that as it may, basically a trial result on efficacy or effectiveness was termed “positive” if the CAM approach led to statistically significant improvements in pain severity, pain-related disability, and/or function, compared to the control group. A negative result meant that there was no difference between the intervention and control groups.
The evidence base for CAM interventions for pain: Results
The first thing I noticed reading this review is that, contrary to the headlines about the review, the actual results of the review were far more measured and—dare I say it?—wishy-washy. Of course, it is always thus for CAM, where the requirements of publishing in the peer-reviewed literature demand discussing all the caveats and weaknesses of a finding. Keeping that in mind, here are the conditions and CAM treatments for which Nahin et al reported “positive” results:
- Acupuncture and yoga for back pain
- Acupuncture and tai chi for osteoarthritis of the knee
- Massage therapy for neck pain—with adequate doses and for short-term benefit
- Relaxation techniques for severe headaches and migraine.
Let’s take a look at acupuncture for back pain, as reported by Nahin et al:
We found 4 RCTs (total participants, 1092)15, 16, 17, 18 that assessed the clinical benefit of acupuncture for treatment of low back pain (LBP) (age range, 28-60 years; most participants were white) and used primary study outcomes of self-report of pain intensity (numeric rating scale or visual analog scale [VAS]) and/or functional disability (Roland-Morris Disability Questionnaire, Oswestry Disability Index [ODI], or Disability Rating Index). Cherkin et al15, 16 reported modest improvement in pain intensity and function compared with usual care. In pregnant women using auricular acupuncture, Wang et al17 found a significant reduction in pain intensity and improved functional status compared with no treatment. Comparison of verum to sham acupuncture had mixed results, with 2 RCTS16, 18 finding no significant difference and 1 RCT17 finding a slight but significant difference. No significant adverse events were reported.
This is the very definition of a negative review. The only clearly positive results were found in studies that compared acupuncture versus usual care or no treatment and only one study comparing “verum” acupuncture to sham showed a positive result, that result being “slight but significant.” To me, “slight but significant” means small and statistically significant but almost certainly not clinically significant. This sort of result is very common in studies of treatments that are no better than placebo controls, and the proper way to report this would have been that there is no good evidence that acupuncture does better than sham/placebo for low back pain.
We see the same dance with phrasing with yoga and back pain:
We identified 6 RCTs of yoga for cLBP57, 58, 59, 60, 61, 62 (total participants, 596; all adults, predominantly female). Three named forms of yoga were studied: hatha,57, 58 viniyoga,59, 60 and iyengar61, 62; all were performed in group settings, with class durations from 60 to 90 minutes and the number of sessions ranging from 12 to 24, either once or twice per week, with recommendations for home practice. Compared with usual care, 2 studies59, 60 found that yoga provided improvements in pain and function, but the results were mixed when compared with exercise/stretching. A dose-response study57 compared once-weekly to twice-weekly classes and found that they produced equivalent improvements in pain intensity and function. Three smaller studies compared yoga with wait list58, 61 or education control62 and reported significant modest reductions in pain intensity and function/disability. No RCTs reported significant adverse events.
Note again that the only unequivocal differences were between yoga and usual care and that even then there was no detectable “dose-response” effect. Even then, the effects were modest, completely consistent with placebo effects. Like acupuncture, when a sham intervention (non-yoga stretching) was compared with acupuncture, the results were “mixed.” They’re always “mixed” with placebo medicine. Again, the correct way to report this result would have been that stretching exercises appear to help compared to doing nothing (although placebo effects can’t be ruled out) and that there’s no good evidence that yoga is any better than any other form of stretching exercises for low back pain.
Tai chi also falls into this category as a form of low impact exercise designed to emphasize body awareness and control; there is no good evidence that it is significantly better than other forms of low impact exercise for pretty much anything, including pain. That doesn’t stop actual high profile CAM advocates from somehow differentiating it from exercise. No less a figure than Dr. Brent Bauer at the Mayo Clinic even tries to separate tai chi from other forms of exercise:
I have many chronic pain patients who aren't able to exercise – I often have them try tai chi, as gentle way to get moving again #talkpain
— Brent Bauer (@babmd) September 1, 2016
I would have none of it:
— David Gorski (@gorskon) September 1, 2016
Seriously, one more time: What is tai chi but a form of slow, gentle exercise?
Then there’s acupuncture and osteoarthritis:
Four RCTs examined whether acupuncture could relieve symptoms associated with OA of the knee.87, 88, 89, 90 These studies used similar definitions of knee OA. Participants were predominantly female, had mean ages between 60 and 65 years, and had knee pain for an average of 9 to 11 years. All studies incorporated either the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) total score or the WOMAC pain subscore as the primary outcomes. In 2 of the trials,89, 90 no difference was seen between verum acupuncture and sham acupuncture for either the primary or secondary outcome measure. The third trial,87 which used a sham control, found significantly better improvement in both the WOMAC pain subscale and function subscale after 14 weeks of treatment. For the 3 trials that also incorporated either an attention control87 or standard care controls,88, 90 verum acupuncture produced significantly better improvement in primary and secondary outcomes than that observed in the control group.
I could just cut and paste the paragraph I wrote after the section on acupuncture for low back pain, substitute the word “osteoarthritis” for “low back pain,” and adjust the numbers to match the paragraph above, and it would fit just as well. This is the very definition of a negative review, just like the one for acupuncture and low back pain. Acupuncture, being the theatrical placebo it is, does better compared to no treatment, education, or wait list controls. Compared to sham acupuncture, the results are “mixed” with only one study being “positive,” as in statistically significantly positive, even though it’s almost certainly not clinically significant.
And so it goes with the other comparisons. It’s OK, though, for CAM aficionados, because they know that the popular press will strip their findings of all those pesky ambiguities and make it sound as though the results were far more positive than they were. Meanwhile, the quack press takes the reports and gives them headlines like Alternative Treatments Reduce Pain Safely, something this NCCIH review definitely does not demonstrate.
The selling of “integrative medicine” by NCCIH
I’ve sometimes noted that NCCIH is not as bad when it comes to ran pseudoscience as it was a decade ago, when it was still NCCAM. Since Dr. Josephine Briggs took over several years ago, consistent with the previous NCCAM strategic plan (which I like to characterize as “let’s do some real science for a change”) and the current plan (which I like to characterize as “meet the new plan, same as the old plan”) there have been fewer blatantly pseudoscientific “studies,” like studies of homeopathy, distance healing, and the like, funded and a shift towards varieties of CAM that seem less quacky on the surface. Indeed, I have not been able to find a study of homeopathy for which NCCIH provided grant support since 2008.
However, this shift towards co-opting sensible and potentially science-based modalities such as diet and exercise is no less harmful, particularly given the presence of outright quacks like chiropractors and naturopaths in high-ranking positions within NCCIH. Basically, reeling from all the negative studies it funded of truly “alternative” treatments, NCCIH has seen an opportunity to seem relevant by more tightly co-opting “nonpharmacologic treatments for pain,” rebranding them as somehow “integrative” or “complementary,” and then lumping them together with the remaining pseudoscience that has enough seemingly positive studies combined with enough popularity among naturopaths and other quacks that help oversee NCCIH, to imply that the whole “integrative package” is the answer to our opioid abuse crisis.
For instance, here, co-author and naturopath Wendy Weber claims that this reviews does address whether these therapies work, neglecting to point out that it shows that they almost certainly don’t:
She also basically admits, with a whole lot of hand waving, that we don’t know the mechanisms by which “complementary” therapies “work” for pain. Of course, that’s because the ones reviewed almost certainly do not, hoping that “future research” will answer those questions. Oh, and, Congress, keep funding us to do studies like the ones chiropractor and co-author Partap Khalsa wants to see:
In other words, let’s do more tooth fairy science and study differences in effects in different populations, mechanism, dosage, and the like without actually having conclusively demonstrated beforehand that the treatments under study actually do work.
As I noted before, NCCIH is now emphasizing as its key scientific priority pain and difficult-to-control symptoms, in particular nonpharmacologic methods of relieving pain. This review clearly is a result of that priority. Unfortunately, this review also demonstrates just how little promise there is in the specific “nonpharmacological approaches” rebranded as CAM examined. Basically, this review is most consistent with none of them working for chronic pain. However, NCCIH can’t admit that, which is why the NCCIH has discussed how to do “innovative clinical trials” by using “pragmatic clinical trials to address questions about the integration of complementary health approaches into health care systems, or to study the effectiveness of complementary or integrative approaches in comparison to standard care.” Unfortunately, pragmatic trials are trials that are done only after more rigorous trials actually show that an intervention works, with the intent to see how it works in the “real world.” Doing pragmatic trials of treatments not yet shown to work is putting the cart before the horse.
That’s pretty much CAM in a nutshell, at least CAM that doesn’t have infinitesimally small prior probability, like homeopathy. Basically, NCCIH is proposing to study how these CAM modalities work without having established that they work.“Non-pharmacological treatments for pain” ≠ CAM, no matter how much NCCIH wishes it so David Gorski