(ĐTĐ) - Acupuncture is superior to both sham acupuncture and standard care for the treatment of different types of chronic pain, suggesting that the effects of acupuncture are more than just placebo effect, a new meta-analysis shows.
The analysis found that about 50% of patients who got acupuncture had improvement in pain compared with 30% who didn't get acupuncture and 42.5% who had sham acupuncture.
"In other words, 20% of patients were feeling better because they had acupuncture; about one third of those would only feel better if the right needles were put in the right points to the right depth, and two thirds of them would feel better getting any kind of acupuncture," lead study author Andrew J. Vickers, DPhil, attending research methodologist, Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, New York, told Medscape Medical News.
So in contrast to other interventions, for which the placebo effect is typically about one third of the effect of the treatment, "in acupuncture, it looks like it's two thirds," said Dr. Vickers. "That's quite a large benefit and that's what the patient will actually experience in real clinical practice," where the decision is not whether to have true or sham acupuncture but whether to get a referral for acupuncture or not.
The analysis is published online September 10 in Archives of Internal Medicine.
Individual Patient Data
The researchers searched MEDLINE, the Cochrane Collaboration Central Register of Controlled Trials, and the citation lists of systematic reviews for randomized, controlled trials (RCTs) that included at least 1 group receiving acupuncture needling and 1 group receiving sham acupuncture or no-acupuncture control.
Unlike previous investigators, researchers carried out an individual patient data meta-analysis of only high-quality randomized clinical trials. Such meta-analyses are superior to summary-data meta analyses because they enhance data quality, enable different forms of outcomes to be combined, and allow use of statistical techniques of increased precision, according to the authors.
This analysis included 29 RCTs with a total of 17,922 patients from the United States, the United Kingdom, Germany, Spain, and Sweden who had 1 of 4 indications: back or neck pain, shoulder pain, chronic headache, or osteoarthritis. Of the 29 studies, 18 with 14,597 patients compared acupuncture and no acupuncture, and 20 with 5230 patients compared acupuncture and sham acupuncture. All patients had access to analgesics and other standard treatments for pain.
Sham acupuncture included needles inserted superficially; needles that retract into the handle rather than penetrate the skin; and non-needle approaches, such as deactivated electrical stimulation or detuned laser. The usual care in no-acupuncture control groups also varied; for example, in 1 RCT, control group patients were merely advised to "avoid acupuncture." In some trials, both acupuncture and sham groups received a course of additional treatment, such as exercise.
The meta-analysis showed that acupuncture was statistically superior to control for all analyses (P < .001). The effect sizes among patients receiving acupuncture were 0.23, 0.16, and 0.15 standard deviations (SDs) lower than those among sham controls for back and neck pain, osteoarthritis, and chronic headache, respectively. Effect sizes were larger for the comparison between acupuncture and no-acupuncture control than for the comparison between acupuncture and sham.
Table. Effects of Acupuncture vs Sham or No Acupuncture in Chronic Pain
|Indication||Studies (n)||Fixed Effects||P Value|
|Acupuncture vs sham|
|Neck and back pain||8||0.37||<.001|
|Acupuncture vs no acupuncture|
|Neck and back pain||7||0.55||<.001|
|Shoulder pain||0||No trials|
The authors provided an example of what the effect sizes might mean in "real terms." A baseline pain score on a 0-100 scale for a typical RCT might be 60, and, given an SD of 25, follow-up scores might be 43 in a no-acupuncture group, 35 in a sham acupuncture group, and 30 in patients receiving true acupuncture.
"If responses were defined in terms of a pain reduction of 50% or more, response rates would be approximately 30%, 42.5%, and 50%, respectively," they write.
Sensitivity analyses showed that neither restricting the sham RCTs to those with low likelihood of unblinding nor adjusting for missing data had any substantive effect on the main estimates. Inclusion of summary data from RCTs for which raw data were not obtained or that were published recently also had little effect.
Repeating the meta-analyses excluding RCTs with a sample size of less than 100 had essentially no effect on the results, nor did an analysis examining the effects of pooling different end points measured at different periods of follow-up.
The type of acupuncture didn't seem to make a difference to the results, said Dr. Vickers. "Some acupuncturists will tell you not to go to such and such a person because that person doesn't put the needles in the right way, or they don't use the right theories, or they're not as well trained, but the particular type of acupuncture you get doesn't seem to make a large difference."
Dr. Vickers and his colleagues are comparing 2 accepted approaches to acupuncture, Chinese and western. "At this point, we can only say that if there's a modest difference between doing real acupuncture and something that no one would think of as being acupuncture, then the difference between 2 bona fide types of acupuncture are unlikely to be large."
Some 3 million Americans receive acupuncture treatments every year, most for relief of chronic pain. A "good proportion" of those are referrals from doctors, said Dr. Vickers.
"Many doctors would be quite reasonably concerned about referring a patient for acupuncture for chronic pain due to a lack of evidence that it would benefit that patient," said Dr. Vickers. "A proportion of those doctors may read this paper and conclude that this evidence is sufficiently robust that they now believe that such a referral would in fact help."
Dr. Vickers noted that many patients with chronic pain are very well managed on medication. Patients who don't get sufficient relief from medication or have side effects from those drugs can choose from a wide variety of other treatments, including physical therapy, manipulation, and behavior therapy. "How exactly acupuncture should be used amongst those other options is actually not clearly understood right now."
Although this analysis didn't tackle the topic of cost-effectiveness, other research has generally found that acupuncture provides "bang for your buck," said Dr. Vickers. "Those studies have typically found that the health gain per dollar spent is well under the typical threshold."
In an accompanying editorial, Andrew L. Avins, MD, MPH, research scientist, Division of Research, Northern California Kaiser Permanente, and professor, Departments of Medicine and Epidemiology & Biostatistics, University of California, San Francisco, said the study authors have provided "robust evidence" that acupuncture provides modest benefits over usual care for patient with chronic pain.
After carrying out an "exhaustive" literature search, using a prespecified algorithm, establishing "clear and justifiable" study eligibility criteria, incorporating appropriate analytical techniques, and conducting a wide range of sensitivity analyses, the authors have provided a "fresh" contribution to the debate surrounding acupuncture, said Dr. Avins.
The authors were able to obtain the raw data from 29 of 31 eligible trials, permitting data to be pooled and analyzed at the individual level, which, said Dr. Avins, is "an impressive demonstration of the great value of collaboration and data sharing."
Dr. Avins did note that the choice of outcome measure — standardized effect size, or the magnitude of change expressed in SD units — is reasonable yet "problematic." The authors' assertion that the overall observed estimate of about 0.5 SD is of "clear clinical relevance" is difficult to substantiate because the clinical relevance probably varies with the measure used and the outcome being assessed, said Dr. Avins.
"Nevertheless, for patients with chronic pain, whose therapeutic options are limited, even a modest benefit from a safe intervention is welcome."
Although it's ideal to understand the mechanism of action for acupuncture, the ultimate question is whether this intervention actually works, said Dr. Avins.
The Acupuncture Trialists' Collaboration is funded by a grant from the National Center for Complementary and Alternative Medicine at the National Institutes of Health to an author, and by a grant from the Samueli Institute. Dr. Vickers and, the other study authors or for Dr. Avins have disclosed no relevant financial relationships.
Arch Intern Med. Published online September 10, 2012.