Research Reproducibility in Longitudinal Multi-Center Studies Using Data from Electronic Health Records

Area of Science:

• Biomedical Informatics
• Clinical Research
• Data Science

Background:

• Reproducibility is a cornerstone of scientific integrity.
• Specific guidelines for ensuring reproducibility in electronic health record (EHR) research are lacking.
• Researchers need clear criteria to assess and implement reproducibility in EHR-based studies.

Purpose of the Study:

• To articulate specific requirements for the reproducibility of research utilizing electronic health record (EHR) data.
• To develop a framework for assessing and ensuring reproducibility in EHR research.
• To guide researchers in identifying necessary provisions for reproducible EHR studies.

Main Methods:

• Analysis of three distinct clinical research projects employing EHR data.
• Identification of key project features influencing reproducibility requirements.
• Development of a structured framework based on empirical analysis.

Main Results:

• Defined a set of essential requirements for reproducing research conducted with EHR data.
• Identified specific project characteristics that dictate these reproducibility needs.
• Established a foundational framework to guide reproducible EHR research.

Conclusions:

• A clear framework is needed to ensure reproducibility in electronic health record research.
• Understanding project-specific features is vital for implementing reproducibility measures.
• This work supports the development of strategies for trustworthy EHR-based scientific findings.