Guiding interoperable electronic health records through patient-sharing networks

Area of Science:

• Health Informatics
• Network Analysis
• Public Health

Background:

• Effective clinical information sharing is vital for healthcare safety and efficiency.
• National data-sharing systems face challenges like cost, political hurdles, and lack of local relevance.
• Understanding patient movement between hospitals is key to optimizing data sharing.

Purpose of the Study:

• To analyze patient sharing patterns among English hospitals between 2013-2015.
• To identify and group hospitals into data-sharing communities.
• To assess the impact of these communities on clinical data continuity.

Main Methods:

• Utilized a dataset of 130 million hospital attendances for 19.6 million patients across 155 English hospitals (2013-2015).
• Constructed a hospital network based on patient sharing frequency.
• Applied the Louvain algorithm to partition the network into ten distinct data-sharing communities.

Main Results:

• Identified that 12% (14.7 million occasions) of patient interactions involved attending a different hospital than their previous one.
• The developed data-sharing communities improved data sharing continuity from 0% to 65-95% in instances of inter-hospital patient transfer.
• Demonstrated a significant enhancement in localized clinical information accessibility.

Conclusions:

• Local data-sharing communities can effectively improve clinical information accessibility.
• This approach offers a viable alternative to large-scale, national interoperable health information systems.
• Implementing localized strategies can enhance healthcare system efficiency and patient care continuity.