| BackgroundOutcome is one of PICO(population,intervention,comparison,outcomes)in clinical trials,which is the key issues that should be concerned in clinical trials of Traditional Chinese Medicine(TCM).At present,there are some problems in the clinical outcomes of TCM clinical trials,such as the diversity of outcomes in similar clinical trials,short outcome measurement time and lack of endpoint outcomes,the lack of major outcomes or without complete major outocmes reporting;adverse reactions/side effects are not adequately reported,composite outcomes reporting,lack of patients report outcomes,insufficient TCM syndrome reporting,as well as diversity outcome measurement instruments,or no outcome measurement instruments reporting.These problems will reduce the value of clinical trials and result in investment waste,because clinical trials can not be included in the systematic review or can not be conduct meta-analysis,so that the clinical trials can not provide a higher level of evidence for clinical practice to some extent.These problems also signify the potential selective reporting bias,of which the resulting may mislead clinical practice.Also,the lack of TCM syndromes outcome reporting can not reflect the advantages of TCM.One of the potential ways to address these issues is to develop the smallest outcome set,which should be reported in all of clinical trials in a specific disease.Of which is known as core outcome set(COS).In the late of 1970s,World Health Organization recommended the use of standardized outcome measures in the clinical trials of oncology.Then,there are much more researchers developed COS of other disease.However,the development of COS is slow,as well as the methodology is diversity.In 2010,the COMET(Core Outcome Measures in Effectiveness Trials)initiative was established to promote international consensus of COS,and promote methodological research of COS.From then on,the development of COS has gradually increased.Of the 259 projects that are complete,236 are ongoing now.However,these studies are developed or participated by European and American countries,the viewpoint of participants in low-income countries and developing countries are less.Therefore,the degree of consensus is not enough,and it is difficult to widely use.In 2013,Chinese researchers introduced the concept of COS and proposed the combination of clinical trials of TCM.Due to the lack of TCM syndrome outcomes and the lack of Chinese researchers and patients participating in the completed studies,it is extremely urgent to develop COS of TCM.Some domestic researchers have paid attention in the field,but mainly focused on exploring the significance and methodology of COS.Because the concepts of TCM are vague and the syndromes are not standardized,it is difficult to quantify the information of the Chinese medical examinations,as well as the methodology of the COS is still not perfect,which poses difficulties for developing COS of TCM.Researchers should pay attention to TCM syndromes when developing of TCM.However,there is no in-depth discussion on how to deal with TCM syndromes in COS,and there is no relevant COS to provide reference.There is no gold standard for developing COS,and some aspects,such as the number of panels,the group size,the methods of scoring,nature of feedback presented between rounds,the methods of assessing attrition(response bias)between rounds,ObjectivesThere are three objectives of this study.1.This study refers to international experience of developing COS,as well as combined with the characteristics of TCM,exploring the methodology of developing COS for clinical trials of TCM,2.Developing a COS for clinical trials of TCM in treating non-valvular atrial fibrillation to verify the feasibility of the method.3.Different methods are used to evaluate and improve the factors that may influence the results of the COS in Delphi survey to enrich the methodology of COS.MethodsThis study present that developing COS in clinical trials of TCM can be divided into two parts:the universal COS of TCM and the core TCM syndromes set.Developing a universal COS of TCM should develop an initial list of outcomes based on systematic reviews and interviews,then conduct the Delphi survey to acquire the views of different stakeholder groups.A consensus conference can promote the estabilishment of final COS.then the measurement instrument and measurement time of each outcome in COS can be established.The methods of developing core TCM syndromes set is similar.First,it is necessary to obtain the names of TCM syndromes of specific diseases through systematic review and retrospective study,then TCM syndrome name database can be established from them.Second,researchers can get TCM four diagnostic information through cross-sectional survey,then cluster analysis or factor analysis can help researchers identifying TCM syndromese classification.Third,the researchers can extract the syndrome elements and the target of the syndrome elements of each type of syndrome,compares them with the database of the names of syndromes of TCM,so that the names of candidate syndromes acquired.Forth,the Delphi survey and consensus meeting can help developing the name of the syndrome and the core symptoms/signs of the TCM syndromes,of which is called core TCM syndromes set.Results1.This study developed a COS of non-valvular atrial fibrillation in clinical trials of TCM,which is an example to verify the feasibility of the methodology.2.There are 12 outcome in 9 outcome domains included in the COS of non-valvular atrial fibrillation(paroxysmal atrial fibrillation),such as clinical efificacy of atrial fibrillation(time of first recurrence of atrial fibrillation,rate of sinus rhythm conversion),cardiovascular mortality,cerebrovascular events(hemorrhagic stroke,ischemic stroke,transient ischemic attack),thromboembolic events(systemic embolism),quality of life,adverse reactions(heavy bleeding),healthcare resource use(emergency visits),treatment compliance(lost visits),TCM syndromes,etc.Among them,the core TCM syndromes outcome is eventually developing consensus in four syndromes,such as qi stagnation and blood stasis syndrome(core symptoms/signs:palpitations,chest distress or pain,dark purple tongue,or tougue with ecchymosis or petechiae,irregular pulse or uneven pulse),heart-kidney yang deficiency syndrome(core symptoms/signs:palpitation,chest distress,fatigue,weakness,chills,pale complexion,frequent urination,wheezing,edema on the face or both lower extremities,oliguria,slippery pulse or slender pulse or deep pulse),qi and yin deficiency(Core symptoms/signs:palpitation,chest distress,fatigue,shortness of breath,fine pulse,spontaneous perspiration,night sweats,forgetfulness,lassitude,red tongue,little or no moss,and fine pulse),heart and spleen deficiency(Core symptoms/signs:palpitation,chest tightness,spontaneous perspiration,abdominal distension after eating,loose stools,pale tongue,weak pulse).Because the majority of the outcomes in the COS of nonvalvular atrial fibrillation(paroxysmal atrial fibrillation),so the consensus meeting experts did not recommend outcome instrument except quality of life(SF-36).Due to the lack of evidence for the measurement time of each outcome,the timing of the outcome measure is not recommended.3.The factors whichi may influence the result of Delphi survey(1)In the Delphi survey,when there are several panels from different stakeholders comparing with one panel,the results reflect the perspectives of stakeholder which has the maximum participants.If there are several panels,it can reflect different stakeholders'perspectives and the results would not be affected by the group size.(2)Compared the results of the first round of questionnaire scoring in the Delphi survey with different methods.Feedback in the form of average scores and feedback in the form of histograms had a significant impact on whether the experts modified the questionnaire scores(?2=7.822,P<0.05),and the histogram form was more likely to promote consensus,the average score form may reduce the degree of consensus.(3)Adopting two different methods to evaluate the bias of attrition:1)Average round-1 scores were calculated for each participant then plotted according to whether participants completed round 2 or not.2)Comparing average scores of individual outcomes amongst those who do and not complete later rounds.The results showed that there was no attrition bias.However,the second method had a higher probability of false positives.(4)Comparing the relations of scoring changes and the importance/consensus degree of the outcome between two rounds of Delphi survey,the results show that although the majority of participants changed their score,but the score changes did not change the importance/consensus degree of the outcomes.We conclude that the 9-point Likert scoring system is the reason of which results in false positive in the evaluation of attrition bias,and it is difficulties for the the experts to select score.3-point method maybe a better choice.ConclusionsThis study developed a COS for non-valvular atrial fibrillation in clinical trials of TCM,which vertified that the method of developing COS of TCM is feasible.At present,the methodology of the COS is not perfect,and it is difficult to judge the quality of COS.Based on COS-STAD recommendations,this study evaluated the standard of the COS of non-valvular atrial fibrillation in TCM clinical trials.The results showed that this study was basically consistent with COS-STAD standard.The COS of Non-valvular atrial fibrillation in TCM clinical trials also has some limitations:1.The number of participants are different in different stakeholders,2.It is not complete mergering outcomes or outcome domains.3.There is no core outcome domain survey.4.The survey method and scoring system are not perfect(lack of "uncertain" options).5.The structure of questionnaires is single,wichi is not suitable for all stakeholders.6.Disease classification is not clear.7.Outcome measurement instruments and measurement time are not adequately studied.The results of this study show that,in the Delphi survey,if the group size is different between different stakeholders,it can be divided into different panels;histogram is a good method to feedback between rounds,it can reflect different experts' different in evaluating the bias of attrition with comparing average scores of individual outcomes amongst those who do and not complete later rounds,false positive result should be assessed;3-point approach maybe a good choice. |
PDF Full Text Request