FAIR+R: Making Clinical Data Reliable Through Qualitative Metadata.

Area of Science:

• Data Science
• Information Science
• Clinical Informatics

Background:

• Metadata is essential for researchers accessing data repositories for secondary use.
• The volume of metadata is increasing due to automatic generation and harvesting.
• Current metadata quality in repositories often falls short of its potential, impacting data reliability.

Purpose of the Study:

• To identify measures for assessing metadata quality in clinical care repositories.
• To propose an extension to the FAIR Guiding Principles to include data reliability.
• To lay the groundwork for assessing metadata quality in Medical Data Integration Centers (MeDICs).

Main Methods:

• Conducted an extensive literature review to identify metadata assessment measures.
• Synthesized findings to propose nine measures for metadata assessment.
• Developed three new principles for data reliability, forming a fifth block for the FAIR Guiding Principles.

Main Results:

• Identified nine key measures for assessing metadata quality in clinical repositories.
• Proposed an enhanced FAIR Guiding Principles framework with an added block for Reliability.
• Established a foundation for future metadata quality assessments in MeDICs.

Conclusions:

• Metadata quality is a critical component for ensuring reliable data in secondary use.
• The proposed reliability principles enhance the FAIR framework, promoting more trustworthy data.
• This work provides a basis for improving metadata quality and data accessibility in clinical research settings.