GRADE方法中，随机试验起评即为高质量证据，观察性研究起评即为低质量证据；但若证据本身存在高发表偏倚风险，则两者证据质量级别都应降低。即使最佳证据汇总表纳入的各项研究仅有低发表偏倚风险，发表偏倚仍会极大高估效应值。当可得证据来自小样本研究、且多数由厂商资助时，作者应怀疑存在发表偏倚。若干基于检验数据类型的方法可用于评价发表偏倚，其中最常用的为漏斗图，但这些方法都有较大局限。发表偏倚可能较常见，必须特别关注早期结果、对样本量与事件数都很小的早期试验结果尤需小心。
The financial conflict of interests is drawing a wide attention, while several problems such as the unclarity of definition and blurred boundary of classification exist when it comes to the nonfinancial conflict of interests. Taking clinical practice guidelines as the example, this study introduce how nonfinancial conflict of interests happen, their classifications and impact based on present situation of relevant studies.
ObjectivesTo evaluate the reporting status of conflict of interest and economic evidence in Chinese clinical practice guidelines published in 2017 and to provide implications for the guideline development in China.MethodsCNKI, WanFang Data and Google Scholar were electronically searched to collect clinical practice guidelines published in 2017 from January 1st to December 31st, 2017. Two authors independently screened literature, extracted information of interest, such as conflict of interest disclosure and economic evidences, then, the analysis was performed by using the Microsoft Excel 20013 software.ResultsA total of 53 guidelines were included; in which, 14 (26.42%) disclosed conflict of interest, 10 (18.87%) stated sources of funding, and 6 (11.32%) mentioned economics evidence.ConclusionIn China, inadequate attention is addressed to the conflict of interest disclosure and economics evidence during the development of clinical practice guidelines. We propose the inclusion of conflict of interest and economics evidence disclosure in clinical practice guidelines, which will improve their objectivity, independence, and transparency.
ObjectiveTo develop a transparent evaluation tool for the transparency of clinical practice guidelines.MethodsLiterature, books and policy documents related to the transparency of guidelines were retrieved from PubMed, EMbase, Web of Science, WanFang Data and CNKI databases to form a pool of items. After two rounds of Delphi expert consultation, the items were modified and scored to determine whether to be included in the final evaluation system. Expert consensus would determine the organization and evaluation method of the final tool.ResultsA total of 77 references were included, and 399 items related to transparency of guidelines were extracted. The items pool containing 53 items was formed after normalization and de-duplication. After two rounds of Delphi consultation, the evaluation tool for transparency of clinical practice guidelines covering 32 items in 6 areas was formed. The established tool consisted of planning (5 items), personnel selection (6 items), conflict of interest management (6 items), recommendation formation process (8 items), external review process (3 items), and guide reporting process (4 items).ConclusionsThe developed evaluation tool for transparency of clinical practice guidelines can provide a standard for the formulation, evaluation and utilization of guidelines from the perspective of transparency.
ObjectiveTo develop a transparency evaluation tool 2.0 of clinical practice guidelines (CPGs) oriented by public trust. MethodsThe Delphi method was employed to score and select the importance and operability of evaluation indicators. The analytic hierarchy process was used to determine the weights of the indicators. And the final evaluation indicator system was determined through expert consensus meeting. ResultsIt constructed two first-level indicators including conflict of interest and formulation process, and six second-level indicators, including guideline developers, reviewers, protocol and registration, evidence production, recommendation formation, and external review. Based on the second-level indicators, a total of 21 third-level indicators were constructed from the perspectives of disclosure and management. The logical structure of the tool is rigorous and harmonious. ConclusionThe CPGs transparency assessment tool 2.0 developed in this study provides measurement standards and an evaluation framework for assessing transparency in CPGs.