| BackgroundAcupuncture,a treasure of Chinese culture,was listed on UNESCO's Representative List of the Intangible Cultural Heritage in November 2010.1n the past 10 years,more than 2,000 English articles on acupuncture and moxibustion have been published abroad,and not to mention countless Chinese articles on acupuncture and moxibustion.W ith the support of so many studies,still there are numerous doubts about acupuncture therapy.The national institutes of health(NIH)has canceled the recommendation of acupuncture for treatment of lumbago due to the results of a partial evidence-based medical system evaluation.In China,the infection of 115 people infected by acupuncture in Qixian,Kaifeng,Henan province,has triggered a wide discussion on acupuncture treatment.How should acupuncture researchers improve the level of evidence for acupuncture,when the efficacy and safety of acupuncture are questioned for different reasons?It is worth exploring ways that conduct standard clinical research on acupuncture and moxibustion.ObjectiveIn this study,by combing the literature of randomized controlled trials in Chinese and English for the treatment of migraine in recent 5 years,the design of randomized controlled trials was investigated by classification,the existing problems were found out and analyzed,and reasonable suggestions for improvement were put forward,so as to provide reference for the construction of reasonable design schemes for clinical studies of acupuncture and moxibustion.MethodsRetrieval on January 1,2014 to June 30,2019 Chinese and English literature.Retrieval database randomized controlled trials of acupuncture treatment of migraine,and all the documents will be retrieved by its inclusion and exclusion criteria.Extract the information in the library after the screening.The filtered contents are put into the table.Analysis content are as follows:1.The basic information of the literature:literature title,author,year of initiation and publication of the experiment,article source,and region distribution of the experiment;Research project fund support,etc;2.Information of experimental design project:(1)research purpose;(2)sample size estimation;(3)diagnostic criteria;(4)inclusion and exclusion criteria;(5)random grouping method and operation,blind method and operation;(6)Experimental group and control group Settings;(7)observed indicator,etcThe information extracted above is sorted out and summarized to find out the problems:(1)observation and research by logical analysis and specific design classification of each analysis item;(2)make bibliometric records and express them in the form of charts;(3)make an Excel sheet for t-test based on the overall proportion,test the data of the main experimental design of the Chinese and English articles,and observe the statistical difference in the proportion of the main data in the Chinese and English articles.Results1 Calculation of sample size(1)Among the 123 articles(109 in Chinese and 14 in English),14 mentioned the estimation of sample size,7 were in Chinese,accounting for 6.4%of all Chinese articles and 7 were in English,accounting for 50%of all English articles.There were 109 articles that did not mention the sample size estimation,of which 102 were in Chinese and 7 were in English.Use numerical variable to calculate the sample size in 9 cases,accounting for 64.2%of the articles mentioned in the sample size estimation.Use categorical variable to calculate the sample size in 2 cases,accounting for 14.2%of the articles mentioned in the sample size estimation.It is found that the researchers do not pay much attention to the sample size estimation,but the English articles are more concerned than the Chinese ones.(2)The sample size calculation problems in this study can be roughly divided into the following four categories:?there are 3 articles reported the sample size calculation method generally(1 Chinese,2 English);?4 articles' description of sample size calculation is repetitive(3 in Chinese and 1 in English);?Improper formula selection and errors in the setting of sampling error in 3(1 Chinese,2 English);?8(7 in Chinese and 1 in English)are not clear about the sample size of the safety test.2 Selection of diagnostic criteriaThis study involves the diagnostic criteria of 18 kinds,A total of 51 articles used the 2004 international headache association(ihs-2)standard,which was the most used.The guiding principles of clinical research on new Chinese medicine are widely used in TCM,with a total of 14 articles.46 did not specify diagnostic criteria,including 42 articles in Chinese,English 4 articles;There were 4 Chinese articles that did not indicate the source of diagnostic criteria.In the use of diagnostic criteria there were also a large number of outdated and unupdated diagnostic criteria:41 of them selected the appropriate and timely diagnostic criteria,35 Chinese articles(32.1%)and 6 English articles(42.9%).There were 32 literatures that selected the outdated and unupdated diagnostic criteria,including 30 Chinese articles(27.5%)and 2 English articles(14.3%).3 Setting of standards for inclusion,exclusion,elimination and sheddingIn this study,the setting records of inclusion,exclusion,elimination and shedding criteria in the literature were counted.The inclusion criteria were not included in 41 articles(36 in Chinese,accounting for 33.0%,5 in English,accounting for 35.7%).There were 41 articles with unreasonable inclusion criteria(34 Chinese articles,accounting for 31.2%;English 7,accounting for 50.0%)The inclusion criteria were set reasonably in 41 articles(39 Chinese articles,accounting for 35.8%;English 2,14.3%).There were 37 articles with no exclusion criteria,all of which were in Chinese(33.9%).Most of the problems were recorded and set for the elimination of falling standards,and 83(71 in Chinese,accounting for 65.1%)did not mention the elimination of falling standards.English 12(85.7%);29 articles(28 articles in Chinese,accounting for 25.7%)were reasonably set.One English article(7.1%);There are 9 missing entries,all of which are in Chinese(8.3%).4 Setting of random and blind methods(1)Randomization:60 of the articles(49%of the total number of articles)did not mention the specific randomization method or only said in the abstract and body of the article that the trial was a randomized controlled trial,with the same situation in both Chinese and English articles.The most widely used random number table method is the random number table method,with 28 papers,accounting for 44.4%of all random methods.(2)Blind setting:89%of the articles did not mention this problem,and only 11%of the articles were blind.There were 12 articles mentioning the three blind states of patients,analysts and statisticians,including 7(6.4%)in Chinese and 5(35.7%)in English.There is a big difference in the proportion of Chinese and English articles.5 Trial control intervention and type of observation purpose(1)Records:in the experimental group,there were 54 articles on the treatment of migraine with acupuncture alone,accounting for 43.9%of the total articles.There were 25 migraine treatments in the textbook of acupuncture and moxibustion,accounting for 20.3%of the total number of articles.In the control group,24 cases of false acupuncture accounted for 19.5%of the total number of articles.In the control group,single western medicine or western medicine group had the most articles,with a total of 60 articles,accounting for 49%of the total articles.(2)Specific analysis:in the setting of control intervention,51 articles(41.5%of the total number of articles)(48 in Chinese and 3 in English)were written for the purpose of "the difference in the effect of one therapy on migraine compared with other therapies".There were 51 articles,45 in Chinese and 6 in English,for the purpose of "observing the efficacy of acupuncture on migraine".There were 16 articles,including 15 in Chinese and 1 in English,for the purpose of "discussing the influence of a certain factor of acupuncture therapy(such as acupoint selection and manipulation)on the treatment of migraine".There were 10 articles,6 in Chinese and 4 in English,on the purpose of "discussing the mechanism of acupuncture in the treatment of migraine".Chinese articles tend to compare single or comprehensive therapies,techniques and acupuncture points.English articles tend to observe,prove and explain the curative effect of acupuncture.Most of the control intervention Settings fit the control purpose,but a small number of articles were set incorrectly.6 Observe the setting of indicators.(1)Records:observation of treatment status,a total of 9 observation indicators or conditions were recorded:VAS score was the most used for observation,with a total of 54 articles,including 44 Chinese literatures and 10 English literatures.The most commonly used observation index of traditional Chinese medicine is "guiding principles for clinical research of new Chinese medicine",of which 21 articles are written in Chinese and 1 in English.(2)Specific problems:the observation indicators did not accord with the nature of the disease itself(i.e.the absence of observation on pain).There were 6 articles,5 in Chinese and 1 in English.There were 8 missing safety observation indicators in the experiments for the purpose of safety observation,including 7 articles in Chinese and 1 article in English.ConclusionThis study include design problems of the literature found more,improper selection formula of sample size,sample size is not enough,choose obsolete diagnostic criteria,inclusion and exclusion criteria missing items,no standard is set for removing or falling off,random methods with high risk of bias were selected,patients,analysts and statisticians were not blinded,the experimental control intervention method is not reasonable,the observation index is inconsistent with the observation purpose,etc..In some specific problems,the design of English articles is slightly better than that of Chinese articles,but the overall difference is small.Most trials pay little attention to the quality of clinical studies on acupuncture and moxibustion,and the design of randomized controlled clinical trials on acupuncture and moxibustion needs to be improved urgently. |
PDF Full Text Request