Quality assurance and quality control in longitudinal studies
Quality assurance and quality control in longitudinal studies
复制标题
DOI：
10.1093/oxfordjournals.epirev.a017973
复制
发表时间：
1998-01-01
影响因子：
5.5
通讯作者：
Wahl, PW
中科院分区：
文献类型：
--
作者：
Whitney, CW;Lind, BK;Wahl, PW
队列研究的质量控制的许多要素与其他类型的研究相似，即方案(研究程序)的标准化、所有研究人员之间的良好沟通、对要求的明确期望以及确保满足要求的监测。然而，队列研究增加了纵向数据收集的维度，这增加了与随时间漂移、设备变化(包括因磨损而退化和因新技术而升级)以及工作人员更替有关的问题。如果这项研究使用中央实验室或阅读中心来处理数据，这些网站尤其容易受到与时间变化相关的质量问题的影响。涉及多个中心的队列研究在不同地点收集的数据的可比性方面存在额外的问题。本报告将提供许多类型医学研究共同的质量控制要素的概述，但主要集中在与队列研究相关的问题上。重点将放在多中心队列研究上，因为与单中心队列研究相比，它们面临着额外的质量控制问题。许多关于质量保证和质量控制的文献都是在临床试验的背景下出现的，重点是通过标准化的研究范围和局部方案、研究人员的培训和数据管理系统来最大化数据的质量(1-9)。质量保证和控制程序一般在特定研究的背景下描述，例如多风险因素干预试验(10-12)、高血压初级预防试验(3、13)或视神经炎治疗试验(14、15)，一些文章提供一般指南(4、16-18)。在非临床试验研究方面的文献较少，尽管一些研究特定的方法(8，19-22)和一般质量保证/质量控制指南
Many elements of quality control for cohort studies are similar to those for other types of studies, ie, standardization of protocol (study procedures), good communication among all study staff, clear expectations of requirements, and monitoring to ensure that requirements are met. However, cohort studies have the added dimension of longitudinal data collection, which adds issues related to drift over time, changes in equipment (including both degradation due to wear and upgrades due to new technology), and staff turnover. If the study uses central laboratories or reading centers to process data, these sites are particularly subject to quality issues related to changes over time. Cohort studies involving multiple centers have extra issues related to comparability of data collected at different locations. This presentation will provide a general overview of quality control elements common to many types of medical research studies, but primarily focuses on issues related to cohort studies. Emphasis will be placed on multicenter cohort studies, as they face additional quality control issues compared with single-center cohort studies. Much of the literature on quality assurance and quality control has arisen in the context of clinical trials, focusing on maximizing the quality of the data through standardized study-wide and local protocols, training of study personnel, and data management systems (1-9). Quality assurance and control procedures are generally described in the context of a specific study, such as the Multiple Risk Factor Intervention Trial (10-12), the Hypertension Primary Prevention Trial (3, 13), or the Optic Neuritis Treatment Trial (14, 15), with some articles providing general guidelines (4, 16-18). There is less literature in the context of nonclinical trial studies, although some study-specific approaches (8, 19-22) and general quality assurance/quality control guidelines have been