数据管理质量对临床研究证据评价的影响

Effect of Data Management Quality on Clinical Research Evidence Evaluation

目前对临床研究文献质量的评价有多种方式,多侧重于研究设计质量的评价,对研究实施过程质量的评价尚不完善,特别是试验实施过程中的数据管理环节未得到重视。临床研究的数据是评价临床研究结果的基础,尽管在研究方案中可能对数据管理有严格的规范与要求,但在目前的临床研究证据评价体系中却未能得到相应的体现。数据管理是试验实施过程的重要组成部分,是数据收集、录入、清理和管理过程的综合体现,其目的是获得高质量的研究数据用于统计分析,从而得到真实可信的结论。为全面评估临床研究设计及实施情况,笔者建议完善现有的临床研究证据的质量评价指标,增加对研究实施过程中数据管理质量的评价。将数据管理计划、数据检查及核查标准和要求、违背方案数据及异常数据的处理规定加入临床研究证据的质量评价指标中,重视数据管理质量对临床研究证据评价的影响。

Current clinical evaluation of literature quality has various ways. Most of them lay spe- cial emphasis on the evaluation of the design quality, but the evaluation of the implementation process quality is not perfect. Especially data management is not fully emphasized during the enforcement of clinical trials. Data from clinical research were bases for evaluating clinical findings. Although strict specifications and requirements for data management might be strictly written clearly in research protocols, they were not embodied in current clinical research evidence evaluation system. Data management is an im- portant part of implementing the whole clinical trial process, which is a comprehensive reflection of data collecting, logging, sorting, and managing. Its objective is to obtain high quality research data for statisti- cal analysis, thereby coming to a true and reliable conclusion. In order to overall evaluating clinical design and implement, we suggest that present quality evaluation indicators of clinical trails should be completed, and add data management quality evaluation during the whole implement process. Data management plans, standards and requirements for data checking, and management regulations for disobeying data and exception data should be added in quality evaluation indicators for clinical research evidence. The effect of data management quality on clinical research evidence evaluation should be emphasized.