Exploring the potential of OMOP common data model for process mining in healthcare

Abstract

Background and objective: Recently, Electronic Health Records (EHR) are increasingly being converted to Common Data Models (CDMs), a database schema designed to provide standardized vocabularies to facilitate collaborative observational research. To date, however, rare attempts exist to leverage CDM data for healthcare process mining, a technique to derive process-related knowledge (e.g., process model) from event logs. This paper presents a method to extract, construct, and analyze event logs from the Observational Medical Outcomes Partnership (OMOP) CDM for process mining and demonstrates CDM-based healthcare process mining with several real-life study cases while answering frequently posed questions in process mining, in the CDM environment.

Methods: We propose a method to extract, construct, and analyze event logs from the OMOP CDM for process types including inpatient, outpatient, emergency room processes, and patient journey. Using the proposed method, we extract the retrospective data of several surgical procedure cases (i.e., Total Laparoscopic Hysterectomy (TLH), Total Hip Replacement (THR), Coronary Bypass (CB), Transcatheter Aortic Valve Implantation (TAVI), Pancreaticoduodenectomy (PD)) from the CDM of a Korean tertiary hospital. Patient data are extracted for each of the operations and analyzed using several process mining techniques.

Results: Using process mining, the clinical pathways, outpatient process models, emergency room process models, and patient journeys are demonstrated using the extracted logs. The result shows CDM's usability as a novel and valuable data source for healthcare process analysis, yet with a few considerations. We found that CDM should be complemented by different internal and external data sources to address the administrative and operational aspects of healthcare processes, particularly for outpatient and ER process analyses.

Conclusion: To the best of our knowledge, we are the first to exploit CDM for healthcare process mining. Specifically, we provide a step-by-step guidance by demonstrating process analysis from locating relevant CDM tables to visualizing results using process mining tools. The proposed method can be widely applicable across different institutions. This work can contribute to bringing a process mining perspective to the existing CDM users in the changing Hospital Information Systems (HIS) environment and also to facilitating CDM-based studies in the process mining research community.

Fig 1. Process mining in the changing HIS environment (adapted based on [21, 26]).

Fig 2. Relevant CDM tables for event log construction.

Code names are sourced from [50].

Fig 3. Finding the optimal number of clinical orders in TLH CP.

Fig 4. CP discovery for selected surgeries.

(From left to right) (a): CP before THR for mainstream behaviors only with all paths, CP before THR with mainstream behaviors and major paths only; (b): CP after CB for death cases, CP for 24 hours after CB for death cases; (c) CP before TAVI, CP for 24 hours after TAVI; (d): CP for 24 hours after PD with all activities, CP for 24 hours after PD with 30% activities.

Fig 5. Outpatient process models.

(a) CDM-based model in frequency view; (b) CDM-based model in performance view; (c) EHR-based model in performance view [42].

Fig 6. Dotted chart analysis.

(a) dotted chart for all activities by case, occurred between 2003 and 2019; (b) dotted chart for activities by case, occurred in 2013; (c) dotted chart for all activities occurred by case, sorted by case duration; (d) dotted chart for all activities occurred by case, sorted by case duration, excerpted from [42].

Fig 7. Emergency room process.

(a) Process model with all activities included; (b) Process model with mainstream behaviors only; (c) Process model with mainstream behaviors only, based on median time duration between events.

Fig 8. Patient journey map using visit-level events.

(a) entire patient journey with all activities included (left) and the enlarged screenshot of the early encounters only (right); (b) patient journey with mainstream behaviors only.

Fig 9. Patient journey map using descendent-level events.

(a) process model for the entire visits (left); enlarged view of the first two outpatient visits (top right); enlarged view of the TLH inpatient visit (middle right) (b) process model with mainstream behaviors only.

Fig 10. Process mining data spectrum and representation ability of CDM, based on [52].