Towards Understanding the Usage Pattern of Web-based Electronic Medical Record Systems

Area of Science:

• Health Informatics
• Computer Science

Background:

• Electronic Medical Record (EMR) systems are vital in healthcare, but widespread adoption is hindered by challenges in secure data access and patient privacy.
• Understanding EMR usage patterns is essential for overcoming these adoption barriers.

Purpose of the Study:

• To conduct an in-depth trace analysis of a large-scale EMR system to understand user-initiated session patterns.
• To identify key characteristics of EMR system usage relevant to security and performance optimization.

Main Methods:

• Analysis of user-initiated sessions from a decade-long EMR system at Vanderbilt Medical Center.
• Examination of workload stability, user behavior consistency, and access patterns between users and medical records.

Main Results:

• EMR system workload exhibits high stability with a consistent weekly pattern.
• Individual user behavior within the EMR system is consistent over time, with slow migration across sessions.
• Access to medical records is sparse, reflecting real-world patient-caregiver relationship limitations.

Conclusions:

• The observed usage patterns provide valuable insights for developing EMR-specific security measures, including anomaly detection systems.
• These findings can guide system performance optimization efforts for large-scale EMR implementations.
• Understanding user behavior is key to improving secure and efficient EMR system design.