Abstract

Background The clinical research data lifecycle, from data collection to analysis results, functions in silos that restrict traceability. Traceability is a requirement for regulated clinical research studies and an important attribute of nonregulated studies. Current clinical research software tools provide limited metadata traceability capabilities and are unable to query variables across all phases of the data lifecycle.

Objectives To develop a metadata traceability framework that can help query and visualize traceability metadata, identify traceability gaps, and validate metadata traceability to improve data lineage and reproducibility within clinical research studies.

Methods This research follows the design science research paradigm where the objective is to create and evaluate an information technology (IT) artifact that explicitly addresses an organizational problem or opportunity. The implementation and evaluation of the IT artifact demonstrate the feasibility of both the design process and the final designed product.

Results We present Trace-XML, a metadata traceability framework that extends standard clinical research metadata models and adapts graph traversal algorithms to provide clinical research study traceability queries, validation, and visualization. Trace-XML was evaluated using analytical and qualitative methods. The analytical methods show that Trace-XML accurately and completely assesses metadata traceability within a clinical research study. A qualitative study used thematic analysis of interview data to show that Trace-XML adds utility to a researcher's ability to evaluate metadata traceability within a study.

Conclusion Trace-XML benefits include features that (1) identify traceability gaps in clinical study metadata, (2) validate metadata traceability within a clinical study, and (3) query and visualize traceability metadata. The key themes that emerged from the qualitative evaluation affirm that Trace-XML adds utility to the task of creating and assessing end-to-end clinical research study traceability.

Ethical Approval

The [Dakota StateUniversity] University Institutional Review Board, chaired by Jack H. Walters, PhD, granted expedited approval (#2016–2016–114) to conduct this research with human subjects.

Publication History

Received: 27 March 2020

Accepted: 09 June 2020

Publication Date:
07 September 2020 (online)

© 2020. Thieme. All rights reserved.

Georg Thieme Verlag KG
Stuttgart · New York

中文翻译：

---

通过元数据框架增强临床研究数据的可追溯性。

摘要

背景 临床研究数据生命周期，从数据收集到分析结果，在限制可追溯性的孤岛中发挥作用。可追溯性是受监管临床研究的一项要求，也是不受监管研究的重要属性。当前的临床研究软件工具提供有限的元数据可追溯功能，并且无法跨数据生命周期的所有阶段查询变量。

目标 开发一个元数据可追溯性框架，可以帮助查询和可视化可追溯性元数据，识别可追溯性差距，并验证元数据可追溯性，以改善临床研究中的数据沿袭和再现性。

方法 本研究遵循设计科学研究范式，其目标是创建和评估明确解决组织问题或机会的信息技术 (IT) 工件。IT 工件的实施和评估证明了设计过程和最终设计产品的可行性。

结果 我们提出了 Trace-XML，这是一种元数据可追溯性框架，可扩展标准临床研究元数据模型并采用图遍历算法来提供临床研究可追溯性查询、验证和可视化。Trace-XML 使用分析和定性方法进行评估。分析方法表明，Trace-XML 可以准确、完整地评估临床研究中的元数据可追溯性。一项定性研究使用访谈数据的主题分析来表明 Trace-XML 增加了研究人员在研究中评估元数据可追溯性的能力。

结论 Trace-XML 的优点包括：(1) 识别临床研究元数据中的可追溯性差距，(2) 验证临床研究中的元数据可追溯性，以及 (3) 查询和可视化可追溯性元数据。定性评估中出现的关键主题肯定，Trace-XML 为创建和评估端到端临床研究可追溯性的任务增加了实用性。

伦理批准

[达科他州立大学] 大学机构审查委员会（由 Jack H. Walters 博士担任主席）批准加快 (#2016–2016–114) 对人类受试者进行这项研究。

出版历史

收稿日期：2020 年 3 月 27 日

接受日期：2020 年6 月 9 日

出版日期：
2020 年 9 月 7 日（在线）

© 2020. 蒂姆。版权所有。

Georg Thieme Verlag KG
斯图加特·纽约