Developing an ethical framework-guided instrument for assessing bias in EHR-based Big Data studies: a research...

Area of Science:

• Big Data Health Research
• Medical Ethics
• Public Health Informatics

Background:

• Big Data in health research offers advancements but presents ethical challenges.
• Potential biases in electronic health records (EHR) data and data handling processes are under-researched ethical concerns.
• Addressing bias is crucial for the integrity of Big Data health studies.

Purpose of the Study:

• To develop, refine, and pilot test an ethical framework-guided instrument for assessing bias in Big Data research.
• The instrument specifically targets bias within electronic health records (EHR) data sets.
• To enhance the ethical integrity of health research utilizing large-scale data.

Main Methods:

• An iterative process involving literature/policy review, content analysis, and interdisciplinary dialogues.
• Development of an ethical framework and an EHR bias assessment guideline.
• Pilot testing the guideline within a National Institutes of Health (NIH)-funded Big Data HIV project.

Main Results:

• An ethical framework and a practical EHR bias assessment guideline were developed through an iterative, stakeholder-engaged process.
• The guideline was refined through content analysis and interdisciplinary discussions.
• The developed instrument is ready for pilot testing in a real-world Big Data research project.

Conclusions:

• The developed EHR bias assessment guideline provides a structured approach to identifying and mitigating bias in Big Data health research.
• Engaging diverse stakeholders is key to creating effective ethical tools for health data research.
• This framework and guideline will contribute to more equitable and reliable health research outcomes.