Whitney C W, Lind B K, Wahl P W
Department of Biostatistics, School of Public Health, University of Washington, Seattle 98195, USA.
Epidemiol Rev. 1998;20(1):71-80. doi: 10.1093/oxfordjournals.epirev.a017973.
As we have presented, it is evident that cohort studies are confronted with their own special, non-trivial issues of quality assurance and quality control. Such studies are typically large-scale designs and involve an extensive amount of data to be collected and processed, the quality of which depends on a variety of factors related to study personnel and equipment. The fact that data are collected over an extended period of time and at several centers greatly increases the magnitude of the data processing task, significantly increasing the likelihood of discrepancies and measurement error in the data. As presented in tables 1 and 2, the quality assurance and quality control procedures span the entire course of the study and include a multitude of tasks. Such tasks are delegated to various committees and/or are undertaken by participating centers, all of which must take responsibility for understanding, implementing, and following through on all procedures that maximize data quality. The quality of the quality assurance/quality control process is highly correlated with the quality of the communication within and between centers and all researchers. Maintaining standardization of procedures across centers and long-term stability of equipment and analytic procedures are integral components of quality control. In conclusion, the magnitude of the quality control process in a multicenter longitudinal study should not be underestimated, requiring a significant commitment of study resources. The quality control process is key to the integrity of the study, and an integral part of the design of the study. In a well-designed study, with a good quality control process and dedication to the process by the research team, the validity of the conclusions of the cohort study can be established.
如我们所阐述的，队列研究显然面临着自身特殊且并非微不足道的质量保证和质量控制问题。此类研究通常是大规模设计，涉及大量需要收集和处理的数据，其质量取决于与研究人员和设备相关的多种因素。数据在较长时间内且在多个中心收集这一事实，极大地增加了数据处理任务的规模，显著提高了数据中出现差异和测量误差的可能性。如表1和表2所示，质量保证和质量控制程序贯穿研究的全过程，包括众多任务。此类任务被委托给各个委员会和/或由参与中心承担，所有这些都必须负责理解、实施并贯彻所有能使数据质量最大化的程序。质量保证/质量控制过程的质量与各中心内部以及所有研究人员之间的沟通质量高度相关。保持各中心程序的标准化以及设备和分析程序的长期稳定性是质量控制的重要组成部分。总之，多中心纵向研究中质量控制过程的规模不应被低估，这需要投入大量的研究资源。质量控制过程是研究完整性的关键，也是研究设计的一个组成部分。在一个设计良好且有良好质量控制过程以及研究团队致力于该过程的研究中，队列研究结论的有效性才能得以确立。
