Construction of Empirical Care Pathways Process Models From Multiple Real-World Datasets

Area of Science:

• Health Services Research
• Process Mining
• Data Science in Healthcare

Background:

• Care pathways (CPWs) are vital for standardizing patient care.
• While their impact on outcomes is studied, real-world implementation analysis is lacking.
• Existing research often uses limited data, missing a comprehensive view.

Purpose of the Study:

• To develop and apply a process mining methodology for constructing empirical care pathway (CPW) process models.
• To leverage heterogeneous real-world data (RWD) for a detailed analysis of healthcare delivery.
• To enable evaluation of CPW conformance and impact on health outcomes.

Main Methods:

• Synthesized expert knowledge into an activity log representing healthcare delivery.
• Applied process mining techniques to construct CPW process models from the activity log.
• Linked heterogeneous real-world data (RWD) sets for comprehensive analysis.

Main Results:

• Successfully constructed an empirical process model for the Code Stroke (Acute Stroke CPW) care pathway in Aragon, Spain.
• Demonstrated the viability of using linked RWD for CPW process modeling.
• Established a foundation for evaluating real-world CPW conformance and impact.

Conclusions:

• Process mining applied to linked RWD offers a robust method for analyzing real-world care pathways.
• This approach provides valuable insights for health services research beyond traditional outcome studies.
• The methodology facilitates a deeper understanding of healthcare processes and potential areas for improvement.