Management of data quality--development of a computer-mediated guideline

Area of Science:

• Health Informatics
• Data Management
• Epidemiological Research

Background:

• Electronic health data quality is critical for research and healthcare.
• Current data quality management relies on source data verification (SDV) and feedback, which can be resource-intensive.
• A need exists to adapt these procedures for efficiency, particularly within German research networks.

Purpose of the Study:

• To develop a guideline for managing data quality in electronic health data.
• To adapt SDV and feedback procedures to current quality levels, reducing data management costs.
• To support research networks in Germany conducting epidemiological studies and managing registers.

Main Methods:

• Conducted a thorough literature review to identify quality indicators.
• Assigned quality indicators to levels: plausibility, organization, and trueness.
• Defined operational methods for automatic calculation of quality indicators and SDV sample size.
• Developed a guideline implemented in a software tool with adaptation and improvement cycles.

Main Results:

• Identified numerous potential quality indicators but noted a lack of systematic assessment and concepts for SDV and feedback.
• Developed an SDV sample size calculation that adjusts based on data quality.
• Created a guideline with a software tool incorporating study-specific adaptation and a Plan-Do-Check-Act (PDCA) cycle for continuous quality improvement.

Conclusions:

• The developed guideline offers a systematic approach to managing electronic health data quality.
• The adaptive SDV and feedback mechanisms can reduce data management costs.
• The guideline is applicable across various medical documentation contexts, including clinical studies, registers, and surveillance databases.