Migraine is a complicated neurologic disorder that manifests with unilateral, throbbing pain that can persist for 4–72 h during each attack [1]. It has a larger impact on global public health and is estimated to be responsible for 1 billion people [[2], [3], [4]]. As complications are common involved, particularly psychiatric and sleep disorders, migraine can result in not only physical pain but also mental burden and decreased quality of life [5,6].

Although options for treating patients with migraine have substantially expanded [4] over the past decade, only 13% of patients with migraine currently seek preventive drugs [7]. About 10%–15% patients have drug contraindications or limited efficacy [8]. Exploring non-drug interventions are of utmost necessity. Under this circumstance, acupuncture is gradually attracting attention. Especially in recent years, the efficacy of acupuncture in treating migraine has been widely recognized and even validated [[9], [10], [11]]. More important, the National Institute for Health and Clinical Excellence guideline has recommended acupuncture for the prophylactic treatment of migraine [12]. Previous studies have showed that acupuncture could typically produce a more significant effect than no-treatment or drug treatment on migraine [10,[13], [14], [15]]. However, compared with sham acupuncture, the results were generally indistinguishable or modest [[16], [17], [18], [19]]. A possible explanation is that as an elaborate and ritualistic treatment, acupuncture may lead to larger placebo effects, particularly for painful conditions [[20], [21], [22], [23], [24], [25]]. Moreover, during migraine treatment, a high placebo effect has been found by several studies [14,26,27]. For example, by comparing various interventions for migraine, Meissner et al. revealed that sham acupuncture had a more pronounced effect than oral pharmacological placebos [14]. Besides pain-related outcomes, the placebo efficacy of acupuncture also varied for other different outcomes [28,29]. Moreover, despite many studies having been conducted on the placebo effect of sham acupuncture in migraine, no studies to date have explored the factors that might influence this effect.

The current study is based on the belief that investigating the placebo effect of sham acupuncture through different outcomes and evaluating the factors affecting these outcomes could better guide recommendations for acupuncture clinical studies design. As commented above, based on the limitedly published literatures, placebo effect of acupuncture could be exist and affect efficacy of acupuncture. Nevertheless, only few studies to date have explored this placebo effect, and evidence regarding the factors influencing this effect remains deficient. Hence, additional research is warranted to better understand sham acupuncture by means of randomized clinical trials (RCTs) [30]. However, given the absence of an additional no-treatment group in most sham-acupuncture–controlled trials, investigating the placebo effect of sham acupuncture may be difficult. In 2018, the Society for Interdisciplinary Placebo Studies has proposed the notion of “placebo response,” which comprises all health changes resulting from the administration of an inactive treatment [31]. Unlike the placebo effect, the placebo response includes regression toward the mean and natural course of the disease [32]. The concept of placebo response has been acknowledged by leading experts worldwide. In 2015, the BMJ journal published an explanatory article presenting the concept and composition of placebo response [33]. Moreover, a recent study by Kaptchuk outlined in detail the research status of placebo response in chronic pain [34]. A meta-analysis published in a subjournal of JAMA also examined the magnitude of the placebo response across treatment modalities [35]. The introduction of this concept has elicited a novel insight into investigating the effect of sham acupuncture. The present systematic review and meta-analysis aimed to evaluate the placebo response of sham acupuncture on diverse outcomes in migraine RCTs. Subsequently, we identify the factors associated with this response in RCTs using a sham-controlled design.