Challenges of systematic reviews in the field of traditional Chinese medicine

Session Type
Capacity building in evidence synthesis
Cai Y1, Liu J1, Liu X1
1Centre for Evidenceā€Based Chinese Medicine, Beijing University of Chinese Medicine, Beijing, China, China

Nowadays, systematic reviews and meta-analyses (SRs) are widely applied in the field of traditional Chinese medicine (TCM), with about 14754 SRs have been published. However, there are some challenges in the construction of clinical questions, topic selection, methodology design, and reporting of SRs in the field of TCM, include: 1) the western medicine diagnostic system was mostly used in the inclusion criteria, instead of the TCM diagnostic system; 2) lack of represent of Syndrome Differentiation and Treatment of TCM; 3) the interventions were set broadly, for example, when the TCM was set for the intervention, the studies of Chinese medicine treatment and non-pharmaceutical therapy were generally combined, that led to great heterogeneity, made it difficult to clarify the effect of the single therapy and translate evidence into clinical practice; 4) it may difficult to clarify the effect come from the acupuncture treatment or acupoint selection when combined the acupuncture trials; 5) some did not state whether the intervention was used directly or loaded in addition to usual care when the comparison was usual care; 6) the selection of outcomes was not comprehensive, lacking of safety outcomes, economy outcomes, gold standard outcomes, and outcomes with TCM characteristic; 7) lack of registration; 8) there were some irregularities in the formulation of search terms, such as missing aliases and processed products of Chinese medicine or Chinese patent medicine, and missing international codes of acupoints; 9) some only searched Chinese databases without English databases or specialized databases, inviting incomplete literature retrieval; 10) grey literature was ignored, because some studies did not search protocol registration platforms or include unpublished studies; 11) some did not strictly follow the inclusion and exclusion criteria for literature screening; 12) improper meta-analyses, for example, the combination of randomized controlled trials (RCT) and non-RCTs or the combination of high-quality trials and low-quality trials may cause biased results; 13) lack of publication bias assessment; 14) lack of reports on conflicts of interest.
Patient, public and/or healthcare consumer involvement: No involvement.