Aligning Event Logs to Task-Time Matrix Clinical Pathways in BPMN for Variance Analysis.

Area of Science:

• Health Informatics
• Healthcare Management
• Business Process Management

Background:

• Clinical pathways (CPs) are essential for standardizing healthcare and ensuring quality.
• Analyzing CP compliance and variances aids in training and pathway redesign.
• Business Process Model and Notation (BPMN) offers flexible semantics for complex protocol modeling.

Purpose of the Study:

• To present a novel method for parsing complex BPMN models and heuristically aligning patient data traces.
• To enable automated variance analysis between task-time based CPs and real-world electronic medical record (EMR) data.
• To demonstrate the feasibility of analyzing CP compliance using EMR data.

Main Methods:

• Developed a new method to parse complex BPMN models, focusing on task-time matrix representations of CPs.
• Implemented a heuristic alignment approach to match patient data traces from EMR databases to the parsed BPMN models.
• Conducted a case study using a practical CP and two large EMR datasets for variance analysis.

Main Results:

• Automated variance analysis between BPMN task-time models and real-life EMR data was demonstrated as feasible.
• The proposed method successfully analyzed variances in a clinical pathway using actual patient data.
• Existing analysis techniques were insufficient for this type of automated variance analysis.

Conclusions:

• Automated variance analysis of clinical pathways using BPMN and EMR data is achievable with the presented method.
• This approach offers a viable solution for identifying deviations and improving healthcare delivery.
• The findings provide valuable insights for enhancing clinical pathway design and implementation.