A data flow process for confidential data and its application in a health research project

Area of Science:

• Health Informatics
• Data Science
• Medical Research

Background:

• Linked healthcare data offers significant potential for research and service improvement.
• Secondary use of health data raises critical ethical and privacy concerns.
• A standardized, lawful approach to data linkage and anonymization is needed.

Purpose of the Study:

• To define a clear, end-to-end data flow protocol for creating anonymous research datasets.
• To describe the implementation and utilization of this protocol in a real-world case study.

Main Methods:

• Developed a data flow protocol generating pseudonymous, reversibly linkable datasets.
• Protocol allows for irreversible linkage to form a fully anonymous research dataset.
• Designed and implemented within the Comprehensive Patient Records (CPR) study in Leeds, UK.

Main Results:

• A clear, ethico-legally approved approach for creating anonymous research datasets was defined.
• Individual-level linkage was achieved through a non-computer-intensive, irreversible mechanism.
• Successfully applied to electronic health records and patient-reported outcomes for 365,193 patients.

Conclusions:

• The defined approach provides a framework exceeding current anonymization standards.
• The methodology ensures lawful and ethical processing of sensitive health information.
• The resulting anonymous dataset is available for cancer research via DATA-CAN.