Descriptions of sham acupuncture in randomised controlled trials: a critical review of the literature

Our study comprehensively reviewed the reporting quality of sham acupuncture in relevant RCTs based on STRICTA 2010 and TIDieR-Placebo. The overall reporting quality of sham acupuncture was suboptimal. Items for which the reporting rate was above 50% included a brief name that described the sham acupuncture, needle type, names of points used, number, frequency, and duration of treatment sessions, needle retention time, and number of needle insertions per subject per session. Items for which the reporting rate was between 30% and 50% included the rationale for choosing sham acupuncture, needle stimulation, depth of insertion, and the manufacturer of the needles. Other items for which the reporting rate was below 30% included practitioner-related information, response sought, measurement of blinding, intervention mode and environment, assisting tools, and the extent to which treatment varied. Items including “Modifications”, “How well (planned)” and “How well (actual)” were not found to be reported in any of the literature obtained.

Our results were consistent with those of a previous study. A cross-sectional evaluation [21] of sham acupuncture descriptions revealed that between 2009 and 2018, the reporting qualities of sham acupuncture remained low and did not improve over time, only 7 items out of 25 were found to be reported in more than 50% of the studies obtained. Another similarity is that both studies found that the number of sham needles, needle retention time, locations of points used, and treatment sessions were highly reported items, whereas items concerning the categories of “practitioner” and “protocol and settings” [21] were seldom reported. However, there are subtle differences between the previous study and our study. For example, the item “depth of insertion” was found to be reported in over 50% of studies in the previous study, but in less than 50% of the available studies in our study. Meanwhile, the item “needle stimulation” was reported in less than 50% of studies in the previous study but in more than 50% of available studies concerning sham laser acupuncture and TENS/TEAS (not all studies included) in our study. In addition, the previous study did not capture the quality of descriptions of important elements of sham acupuncture, such as needle model, assisting tools, manufacturer of needles, brief name of sham acupuncture, intervention mode and environment, and practitioner background. Differences in sample size and in the guidelines of the previous study and ours might partially explain these distinct results. The previous study included 117 studies for analysis since it only involed a search of the PubMed database for articles published from 2009 to 2018, and excluded trials using sham interventions such as sham electroacupuncture, sham TENS/TEAS, and sham laser acupuncture. In our study, three databases were searched for articles published from inception to 2022 and included trials using multiple kinds of sham acupuncture, thus having a larger sample size. Besides, items for evaluation in the previous study were determined according to STRICTA, CONSORT and TIDieR [22], however, in our study, items were chosen based on STRICTA 2010 and TIDieR-Placebo which was specially designed for sham interventions including but not limited to sham acupuncture, and other considerations.

We found that, for most items extracted in this study, compared with the overall descriptions of sham acupuncture, the reporting rates of items of sham laser acupuncture and sham TENS/TEAS did not seem to differ markedly. However, some items, showed subtle differences. Lower reporting rates were shown in the item of practitioner-related information, including work experience, educational background, training, and profession. We infer that the inadequacy of descriptions of practitioners might be mainly due to the lower dependence on practitioners’ skills in sham laser acupuncture and TENS/TEAS than in sham manual acupuncture and sham electroacupuncture. Therefore, researchers may not pay sufficient attention to describing this information. Sub-items such as “Needle type”, blinding evaluation, and model of electrical equipment also had lower reporting rates than those of all trials. This information needs more attention from researchers. Meanwhile, the sub-item “Needle stimulation”, which refers to the settings of electrical stimulation including intensity, frequency, waveform, and width, was found to have a higher reporting rate than the overall level. Parameters of electric stimulation are key factors of sham laser acupuncture and sham TENS/TEAS, so reporters paid more attention to describing these contents.

Considering the complexity of sham acupuncture components, the effects of sham acupuncture are largely influenced by specific design details. Contextual factors [23], such as treatment settings, intervention mode and environment, practitioner experience, practitioner-patient relationship, and patients’ understanding of or expectations toward acupuncture intervention [24,25,26], are considered to be among the main contributors to the effectiveness of acupuncture. Altogether, contextual factors along with specific details of the sham acupuncture design can have a nonspecific (potentially specific) yet statistically and clinically significant influences on outcomes [27,28,29]. However, STRICTA 2010 recommendations showed the significance of specific components of needling while neglecting many of the contextual factors. Regarding contextual factors in sham controls, several recommendations listed in TIDieR-Placebo, although not specifically designed for sham acupuncture, could be combined with sham acupuncture settings to guide the reporting of contextual factors in sham acupuncture. For example, considering the possible influence of practitioner-related characteristics on outcomes, STRICTA 2010 recommended that researchers report the relevant background of practitioners who implement the interventions, such as their qualifications or profession years spent practicing acupuncture and other relevant experience. When guiding the reporting of sham acupuncture, relevant information from practitioners in the sham acupuncture group was not required to be reported. Regardless of its low reporting quality, practitioner background was required in item 5 “Who provided” of TIDieR-Placebo. In addition, more contextual factors such as communication with participants, information offered in informed consent, and intervention information offered to patients might affect patients’ psychological state and expectations [30], influencing research results and necessitating transparent descriptions for better interpretation.

Besides contextual factors, to better comprehend the effects of sham acupuncture, other important factors such as modifications of the sham acupuncture regimen, practitioner adherence, and blinding evaluation need more consideration. Owing to the complex components and manipulations of sham acupuncture, unforeseen modifications might occur during actual clinical trials. Modifications of sham acupuncture might lead to altered procedures or circumstances of implementation [20]. The reporting of any modification of the sham acupuncture procedure or implementation is warranted for trial transparency and better interpretation and replication. Another factor that has long been neglected in reporting is practitioner adherence, which refers to whether the practitioner performs each step of the sham acupuncture as planned. It is crucial to accurately assess and monitor whether practitioners’ practices meet the predefined requirements and whether planned sham acupuncture protocols and procedures are consistently performed as different effects may occur due to differences in stimulation parameters, such as stimulated points and insertion depth. Regarding blinding evaluation, comprehensively assessing and reporting the results of the blinding status can provide readers with more evidence about the validity and credibility of sham acupuncture in the trial. Although few relevant studies were found to report these items, they should still be emphasized as they can significantly alter outcomes.

After STRICTA 2010 was published, researchers suggested that some important elements, such as the angle and direction of insertion [31], subjects’ prior experiences of acupuncture, as well as standards and methods on acupoint selection and location [32], should be considered in STRICTA 2010, considering their potential to impact the effects of acupuncture. Items beyond STRICTA 2010 and TIDieR-Placebo, such as insertion method and patient’s body position, are also important factors that might influence the effects of acupuncture and are seldom adequately reported. Lack of this critical information may lead to ambiguity for readers to comprehend the real effects of sham acupuncture, thus leading to overestimation or underestimation of the specific effects of acupuncture. With regard to their impact on the effects of sham acupuncture, we strongly argue that these items should be reported in the RCT literature with regard to sham acupuncture.

It was assumed that the low reporting rate of sham acupuncture items might be partly due to the lack of specific guidelines for promoting the reporting of sham acupuncture. Currently, STRICTA 2010 and TIDieR-Placebo are adopted in combination to examine the quality of reporting of sham acupuncture. However, neither of them was specifically developed for this purpose. In general, STRICTA 2010 tends to be more relevant to acupuncture intervention, while TIDieR-Placebo generally covers guidance toward a wider range of sham interventions. For example, item 4 “What (Procedure)” of the TIDieR-Placebo refers to a requirement “Describe each of the procedures, activities, and/or processes used in the placebo/sham intervention, including any enabling or support activities.” Correspondingly, items in STRICTA 2010 include “Number of needle insertions per subject per session”, “Names of points used”, “Depth of insertion”, “Response sought”, “Needle stimulation” and “Needle retention time”. Moreover, when used as a guiding/evaluation tool, both of them tend to fail in terms of adequacy and accuracy; i.e., both checklists lack adequate guidance toward the specific reporting of sham acupuncture. Our previous study also found that both checklists could provide certain references for descriptions of sham acupuncture, but neither seemed to be fully applicable after performing an in-depth comparison of the STRICTA 2010 and TIDieR-Placebo [33]. Hence, we strongly recommend the development of a specialized set of guidelines for guiding the reporting of sham acupuncture in sham-controlled trials of acupuncture based on STRICTA 2010 and TIDieR-Placebo. As the Acupuncture Controls gUideline for Reporting humAn Trials and Experiments (ACURATE) checklist [34] got published, efforts of clear descriptions of sham acupuncture in accordance with ACURATE checklist should be made to facilitate interpretation and replication of trials and precisely assess specific effects of acupuncture.

Our study has several strengths. First, this work involved a comprehensive literature review with a large sample size to assess the quality of reporting of sham acupuncture in RCTs. To identify as many eligible trials as possible, we comprehensively searched the three most commonly used databases (PubMed, MEDLINE and Embase) from inception to 2022 to obtain a substantial amount of literature. This is also the first study to analyze the quality of reporting of sham acupuncture based on STRICTA 2010 and TIDieR-Placebo, as well as other important items related to sham acupuncture settings. Our study provides validated evidence for the quality of reporting of sham acupuncture and is critical to encouraging researchers to pay more attention to improving the transparency of sham acupuncture. However, this study has some limitations. First, we only included studies reported in English based on the consideration that most sham-controlled RCTs were reported in English. Besides, we may have made a subjective judgment when determining if items recommended in STRICTA 2010 and TIDieR-Placebo were reported in available literature because one item may contain information from several aspects, and partially reported items were categorized as “N”. We also did not evaluate the overall quality of the included studies, which might have influenced the strength of our results.

留言 (0)

沒有登入
gif