大型流行病学调查资料的数据管理和质量控制  

Data management and quality control of the large-scale epidemiological survey study

作　　者：葛爱华[1] 魏永越[1] 李瑛 吴玉麟 彭志行[1] 潘丽 张肖敏[1] 陈峰[1] 

机构地区：[1]南京医科大学公共卫生学院流行病与卫生统计学系,江苏南京210029 [2]江苏省生殖健康检验中心,江苏南京210036

出　　处：《南京医科大学学报（自然科学版）》2011年第4期544-548,共5页Journal of Nanjing Medical University(Natural Sciences)

基　　金：江苏省科技支撑计划(社会发展)项目(BS2007080);国家"十一五"科技重大专项(2008ZX10001-003;2009ZX10004-904)

摘　　要：目的:以1项大型流行病学调查研究为例,探索该类研究资料的数据管理和质量控制。方法:从制度建立、人员培训、现场组织、实验室和数据录入等一系列环节进行质量控制,并通过数据核查评价数据质量。结果:将核查输出的异常数据与原始表格核对修正部分异常数据后,随机抽取的3个项目点的人次错误率都在10%左右,项目错误率都在0.5%以内。结论:医学科学研究中,影响数据质量的因素是多方面的,数据管理和质量控制虽然需要大量人力和物力,但根据实际制定合理的数据管理和质量控制方案是保证数据质量的关键。Objective：To explore the data management and quality control of the large-scale epidemiological survey study through an example.Methods：The quality control was applied in the whole process including quality control system,staff training,field organization,laboratory,data entry and so on,and the data quality was assessed by data checking.Results：After the original forms had been verified and the abnormal data had been revised,the error rates of people and the error rates of variable of 3 randomly selected projects were all about 10% and less than 0.5%,respectively.Conclusion：There are many factors that can affect data quality in the medical research.It is essential for data quality to develop the reasonable programs of data management and quality control according to the actual situation,although considerable manpower and material resources are required.

关 键 词：宫颈癌 数据核查 数据管理 质量控制 

分 类 号：R181.8[医药卫生—流行病学]