SHRINE: enabling nationally scalable multi-site disease studies.

Area of Science:

• Health Informatics
• Clinical Research
• Biostatistics

Background:

• Medical research findings are often inconsistent or irreproducible.
• Limited patient subject availability and population variability hinder study generalizability.
• Differences in medical treatments and record systems across institutions complicate multi-site data analysis.

Purpose of the Study:

• To develop a system for aggregating patient observations from multiple hospitals uniformly.
• To create open-source software facilitating multi-site clinical studies.
• To report on the early applications and potential of the developed system.

Main Methods:

• Developed the Shared Health Research Information Network (SHRINE) system.
• Reused electronic health data from clinical care for research.
• Ensured patient privacy and hospital autonomy while harmonizing data across institutions.

Main Results:

• SHRINE enables the aggregation of statistically significant patient populations across numerous hospitals.
• Data harmonization allows for parallel querying across multiple institutions.
• Early implementations have supported multi-site studies in diverse areas like autism, arthritis, and cancer.

Conclusions:

• SHRINE addresses challenges in multi-site clinical research by enabling data aggregation and harmonization.
• The system facilitates statistically robust and independently validated research findings.
• The broad applicability suggests SHRINE's potential beyond current research uses and participating institutions.