Evaluation of multidisciplinary collaboration in pediatric trauma care using EHR data.

Area of Science:

• Health Informatics
• Medical Informatics
• Trauma Care Research

Background:

• Electronic Health Records (EHRs) are crucial in modern healthcare, but understanding collaborative usage patterns remains a challenge.
• Optimizing care for pediatric trauma patients requires insights into team communication and workflow efficiency within EHR systems.

Purpose of the Study:

• To identify distinct collaborative electronic health record (EHR) usage patterns among pediatric trauma patients.
• To determine the relationship between identified EHR usage patterns and patient outcomes, specifically emergency department length of stay (ED LOS).

Main Methods:

• Process mining and network analysis were applied to EHR metadata and trauma registry data.
• Event logs were segmented, and usage patterns were constructed by analyzing functional role interactions within temporal segments.
• Graph kernel and spectral clustering methods were used to classify usage patterns, followed by comparative analysis of patient demographics, clinical data, and ED LOS.

Main Results:

• Three distinct EHR usage patterns were identified: fully connected (clique), partially connected, and disconnected (isolated).
• The fully connected pattern was associated with a significantly shorter median ED LOS compared to the partially connected pattern.
• Partially connected patterns were more common during day shifts and involved specific services like otolaryngology and ophthalmology.

Conclusions:

• A higher degree of multidisciplinary collaboration, represented by a clique-like EHR usage pattern, is associated with reduced ED LOS in pediatric trauma patients.
• Further research is needed to understand the causal factors influencing these collaborative patterns to improve patient care pathways.