临床试验数据质量的评估

Measuring data quality of clinical trials

临床试验数据的采集、报告与处理的每一步骤都可能伴随错误 ,临床试验数据质量的评估是临床试验质量控制的重要措施。数据质量评估技术如 :数据精度评估技术 ,研究者信度评估技术 ,汇总统计 (AggregateStatistics)和对试验过程的统计监控图 (StatisticalProcessControlCharts)技术 ,数据记录一致性的交叉核对技术 ,实验室报告常规格式核对技术 ,关联文件与关联数据的核查技术 ,源数据与CRF以及CRF与数据库对照检查技术 ,数据库数据证实(DataValidation)技术 ,数据处理过程的差错率评估技术 ,实验室数据标准化技术 (laboratorynormalizationtechniques) ,数据清理 (dataclean)

Each step about data which are collected, reported and manipulated has an error rate associated with it. Therefore, measuring data quality is a key element of quality control for clinical trial data. The techniques of measuring data quality includes: evaluating data precision, establishing inter-rater reliability,aggregating statistics and statistical process control charts, cross checking between referential data, checking the format of lab reports, checking among relevant documents and data, comparing between source data and CRF, and between CRF and database, data validating, calculating error rate in the processes of data processing, laboratory normalizing and data cleaning.