Improving Delphi Process in Acupuncture Decision Making: Overall Descriptions and Quality Assessment of Delphi Reports

Yi-yang Sun, Peng-fei Wang, Gui-rong Yang, Dong-qing Du, Chun-jing Li, Zi-jun Mu, Yu-xia Ma, Na Zhang Department of Acupuncture-Moxibustion and Tuina, Shandong University of Traditional Chinese Medicine, Jinan, People’s Republic of ChinaCorrespondence: Yu-xia Ma; Na Zhang, Email phdmayuxia@126.com; zhangna2008zhentui@163.comBackground: Clinical acupuncture decisions are highly operator-dependent and require physician-patient interactions. The Delphi method allows subjective factors such as expert experience and preference of patients to be taken into account in clinical decision making, which is particularly applicable to acupuncture. Currently, the Delphi method is widely used to support clinical decisions in acupuncture. Therefore, it is necessary to provide high-quality and complete descriptions of the Delphi process when making clinical decisions. This study aims to evaluate the quality of the Delphi process in acupuncture, facilitate its standardization and rigor for further clinical decision making in acupuncture.Methods: Articles sourced from six databases were searched systematically to assess the quality of the Delphi consensus process based on the standards for conducting and reporting Delphi studies (CREDES). Descriptive statistics and analysis were presented according to the percentage of each item. Five-score Likert scale was used to evaluate the reporting quality of four domains as well as each item in CREDES by two independent researchers, combined with ICC-value to assess the consistency.Results: A total of 37 qualified articles were included according to eligibility criteria. As for the low reporting rate, the item “External validation” was reported as the lowest positive rate at 32.43% and the item “Prevention of bias” was 48.65%. The item “Adequacy of conclusions”, “Definition and attainment of consensus”, and “Discussion of limitations” were reported at a positive ratio of 62.16%, 64.86%, and 67.57% individually. The average scores of the four domains based on CREDES from highest to lowest were, respectively, as follows: planning and design (68.75%), reporting (66.07%), rationale for the choice of the Delphi technique (65.54%), study conduct (45.10%).Conclusion: The reporting quality of the Delphi consensus process in acupuncture is acceptable currently, but the reporting rate on some items is still low. Further standardization, including either clearer checklists or study reports, should be developed and strengthened to guide clinical decisions in acupuncture.Keywords: delphi, consensus, acupuncture, CREDES, quality assessment