M-04: A Review on Methodological Quality of Traditional Chinese Medicine’s Clinical Trials’ Design in 2016
Poster Presenter
Zhi Cui
Student
University of Macau China
Objectives
To figure out the current development of the traditional Chinese medicine’s clinical trials in 2016 and to analyze the main problems of the traditional Chinese medicine’s clinical trials.
Method
The keywords were used in two databases called CNKI and PubMed to search the clinical trials on TCM, published from January 1st to December 31st in 2016. The meta-analyses and the systematic reviews got excluded. The quality assessment was conducted according to the CONSORT statement.
Results
A total of 210 TCM clinical trials were identified and 158 (88.78%) RCTs were included, of which 42 were from CNKI and 116 were from PubMed. On average, 77.85% of the items on the checklist were provided for all the trials which were included. This is an obvious improvement compared with the reported ratio 62.9% in 2006 (checked using a 10-item checklist). Item participants, item interventions, item objectives, item outcomes were fully met while some items, item randomization sequence generation, item implementation and item statistical methods, remained partially met. In particular, item sample size, item allocation concealment and item blinding/masking showed in an unfavorite situation. For instance, only 44.94% of RCTs followed blinding principle. Among all the 158 RCTs included, their total average score was 69.18, which was a good level. As for the quality of different classifications, the average score of the researches from CNKI was 58.43 which was just a pass level. The average score of researches from PubMed was 73.08 which was a good level. Besides, the average score of the researches on drug was 67.00 which was a good level, while the average score of the researches on non-drug was 70.66, also a good level. Among all the 158 researches on traditional Chinese RCTs, 10 of them were the fail level, 64 of them were the pass level, 53 of them were the good level and 31 of them were the excellent level. As a result, the quality of the whole traditional Chinese medicine’s randomized controlled trials was just a good level, but the proportion of the excellent level was still too low, the quality of English traditional Chinese medicine’s randomized controlled trials from PubMed is much higher than the Chinese traditional Chinese medicine’s randomized controlled trials from CNKI and the quality of traditional Chinese medicine’s randomized controlled trials on drug was higher than the traditional Chinese medicine’s randomized controlled trials on non-drugs.
Conclusion
Although the quality is being improved rapidly in the recent ten years, some methodology issues warrant closer examination. There are mainly three problems in the way of improving the quality of the traditional Chinese medicine randomized controlled trials, the first one is there are many researches on traditional Chinese medicine randomized controlled trials lacking the necessary content the CONSORT checklist requires, which is also the biggest problem. The second one is that the descriptions of every item are not detailed enough for the researches which contain the necessary content. The last one is that there are also some structures problems for some researches, for example, they would mix different items together. The quality of traditional Chinese medicine clinical trials would not be improved unless the trials are designed under the guide of the randomized controlled trials. To improve the quality of traditional Chinese medicine’s randomized controlled trials, the items of the CONSORT statement should be paid great attention to and applied as much as possible. In particular, item sample size calculation, allocation concealment and blinding should be carefully carried out to ensure the validity of traditional Chinese medicine clinical trials.