Current Cochrane acupuncture reviews might be biased without searching Chinese, Korean and Japanese databases

Article type
Authors
Lu ZX1, Lao YF2, Yao L3, Yan PJ4, Hui X4, Han X2, Feng XW2, Fu YW2, Ma YX2, Li J4, Yang KH5
1First Clinical Medical College, Lanzhou University, Lanzhou
2Second Clinical Medical College, Lanzhou University, Lanzhou
3Health Research Methodology, Department of Health Research Methods, Evidence and Impact, McMaster University
4Evidence-Based Social Science Research Center, School of Public Health, Lanzhou University, Lanzhou
5Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou
Abstract
Background: Acupuncture is wildly used as an alternative therapy and evaluated by different Cochrane reviews. Some acupuncture reviews searched Chinese, Korean and Japanese databases to include more potential studies published in local language, whereas some other reviews did not search that. In this study we want to investigate whether Chinese, Korean and Japanese databases should be considered in Cochrane acupuncture reviews.
Objectives: To explore the value of Chinese, Korean and Japanese databases to Cochrane acupuncture reviews.
Methods: Search and include all Cochrane reviews and protocols related to acupuncture, and extract the information of countries, databases searched, and included studies. Local databases are defined as Chinese, Korean and Japanese databases, and local studies are defined as studies only could be searched from local databases.
Results: 134 acupuncture reviews and 20 protocols were included, the authors are mainly from China, Australia, USA, UK, Korea and Canada, among which, 86 reviews did not search any local databases, and 48 reviews searched local databases (median=4, range: 1-13). We found 85% reviews searched local databases included at least one Chinese, Korean or Japanese authors. I also found recruiting local authors was associated with searching local databases in the review (OR=35.79, 14.28 to 89.67). Cochrane acupuncture reviews searched local databases included 61% non-English studies, and 60% could only be obtained from local databases. In addition, those reviews that did not search any local databases included 7% non-English studies from PubMed, EMBASE, CENTRAL and other databases.
Conclusions: There is no consensus of searching local databases in current Cochrane acupuncture reviews. Our study shows it’s valuable to do that, otherwise, the results might be biased due to selection bias. the possible solution is to included Chinese, Korean and Japanese authors in the review.
Patient or healthcare consumer involvement: There was no patient or healthcare consumer involved in this project.

Table 1: The characteristics of Cochrane acupuncture reviews

# The result shows that Cochrane acupuncture reviews recruiting local author with those not have statistic differences on searching local databases.

Figure 1: The characteristics of included Cochrane acupuncture reviews

Figure 2: Countries cooperation network of Cochrane acupuncture reviews