Abstract
Efverman’s recently published study on the placebo and nocebo effects in both genuine and sham acupuncture provides new insights into the effects of acupuncture through a reanalysis of data from a previously conducted randomized clinical trial (RCT), and reading this article makes us reflect on the research we have conducted so far. 1 The question of whether sham acupuncture is an appropriate control for acupuncture has been raised repeatedly, and we believe the findings of this study provide additional evidence supporting such concerns. 2 Following is our elaboration of our reasons for saying this:
When calculating Bang’s Blinding Index to assess the success of blinding based on previous published original RCT data of this study, the index for the acupuncture group was 0.84 (95% CI: 0.73-0.94), and for the sham acupuncture group, it was −0.63 (95% CI: −0.78 to −0.47). 3 This indicates that both groups similarly believed they were receiving real acupuncture. For blinding to be considered successful in clinical trials, an index between −0.2 and 0.2 is more desirable, as it would indicate that participants cannot reliably determine their group assignment. When most participants believe they received the genuine treatment intervention, psychological factors such as placebo or nocebo effects are likely to influence the outcomes.
According to the author, the most frequently reported negative or positive side effects in both groups were relaxation and tiredness. It is well-known that vagus nerve activation through tactile stimulation, such as acupuncture or massage, can improve sleep quality and reduce anxiety.3,4 Even the Park Sham method cannot avoid skin stimulation, in practice, it causes various forms of stimulation during the process. 5 Sham acupuncture often induces relaxation, likely because the intensity of stimulation is weaker than acupuncture, though the manner of applying the stimulation appears similar to the real group. While participants in the acupuncture group experienced “irradiating sensations” more frequently, reports of such sensations in the sham acupuncture group suggest that these phenomena might overlap. Interestingly, sensations such as soreness or numbness were unique to the acupuncture group, leading one to speculate whether they are related to specific physiological responses distinct to real acupuncture. These sensations might result from needle penetration into the fascia, or muscle, or stimulation of nerves or perineural areas. Conversely, certain adverse effects—such as sweating, headache, and pleasant drowsiness—were reported exclusively in the sham acupuncture group, albeit by sham acupuncture-specific effects. Can we attribute all of these effects to psychological factors like placebo or nocebo responses when there are overlapping but different physiological effects beyond placebo in each group?
In addition to these observations, there are further questions for this study. In the original study, out of 522 participants, 237 were randomized, with 120 assigned to the acupuncture group and 117 to the control group. Among these, 97 in the treatment group and 100 in the control group completed the final follow-up. 3 However, this study reports data from 215 participants, with 109 in the acupuncture group and 106 in the control group. It is unclear how the data were selected. Furthermore, in Efverman 1 Table 3, percentages are supposedly indicated in parentheses, but many figures seem uninterpretable. Additionally, the original study classified 2 participants (one from each group) as “did not answer” in the blinding statement, but this study reports 17 participants (12 in the treatment group and 5 in the control group) in this category, raising questions about whether the data were correctly extracted. Lastly, we are curious whether prior approval from an Institutional Review Board (IRB) was obtained for this secondary analysis of existing data.
The controversies surrounding acupuncture—the role and size of placebo effects, or whether sham acupuncture serves as an appropriate control—remain unresolved. The authors’ conclusion that “This implies that acupuncture-delivering therapists striving to maximize placebo-responses and minimize nocebo-responses may consider strengthening the patient’s treatment expectations, and offer a pleasant, pain-free, acupuncture treatment” is intriguing from a clinical perspective. 1 However, it risks the misconception that acupuncture’s effects are entirely psychological or nonspecific, like placebo or nocebo effects. Using placebo controls (eg, sham acupuncture) needs greater attention in the acupuncture clinical trials. Not controlling for the unintended and unavoidable physiological effects of the sham intervention is unacceptable, at least pre-clinical physiological studies are needed to ensure that the physiological effects of the sham beyond placebo lack any specific effects on the symptom under investigation. Considering these, it might not be possible to estimate placebo effects or even “control” for placebo in the clinical trial. More advanced discussions about sham acupuncture and further research into the physiological mechanisms of acupuncture are urgently needed.
