Signal from the Noise: A Mixed Methods Process Mining Approach to Evaluate Care Pathways

Abstract

ObjectiveMapping real-world practice patterns vs. deviations from intended guidelines and protocols is necessary to identify and improve the quality of care for emergent medical conditions like stroke. We propose a process mining algorithm applied to Electronic Health Record (EHR) event log data as a unique opportunity to more easily identify and compare real-world care processes.MaterialsData was obtained from the event log of a major EHR vendor (Epic) for Stanford Health Care Hospital patients aged 18 years and older presenting to the ED from January 1, 2010 through December 31, 2018 and receiving tPA within 4.5 hours of presentation. Our algorithm was built using the Python programming language.MethodsAn unsupervised process-mining algorithm was developed and used to create a process map for our cohort. This map was then used to identify the most common path as well as individual and average conformity to this path across all encounters.ResultsOur automatically generated process mining graph, specifically its most common path, mimicked our institution’s recommended “code stroke” clinical pathway. The average conformity score for our cohort was 0.36 with a range from high of 0.64 and low of 0.20.DiscussionThis method allows for greater detail into common process measures to be more easily illustrated to evaluate the quality of care at a given institution. It may be extended to other, similarly well-defined processes or those which currently lack standardized clinical pathways.ConclusionOur mixed methods approach represents an essential data analysis step to improve complex care processes by automatically generating a qualitative and quantitative process measures from existing event log data which can then be used to target quality improvement initiatives.