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

• DOI: 10.1371/journal.pone.0279641
• Publication Date: 2023-01-03T18:49:56Z
• Authors: Kangah Park; Minsu Cho; Minseok Song; Sooyoung Yoo; Hyunyoung Baek; Seok Kim; Kidong Kim
• Keywords: Databases, Factual; Science; Q; R; 02 engineering and technology; 3. Good health; 0202 electrical engineering, electronic engineering, information engineering; Medicine; Humans; Electronic Health Records; Female; Health Facilities; Delivery of Health Care; Research Article; Retrospective Studies; Data Management

Recently, Electronic Health Records (EHR) are increasingly being converted to Common Data Models (CDMs), a database schema designed provide standardized vocabularies facilitate collaborative observational research. To date, however, rare attempts exist leverage CDM data for healthcare process mining, technique derive process-related knowledge (e.g., model) from event logs. This paper presents method extract, construct, and analyze logs the Observational Medical Outcomes Partnership (OMOP) mining demonstrates CDM-based with several real-life study cases while answering frequently posed questions in environment.We propose OMOP types including inpatient, outpatient, emergency room processes, patient journey. Using proposed method, we extract retrospective of surgical procedure (i.e., Total Laparoscopic Hysterectomy (TLH), Hip Replacement (THR), Coronary Bypass (CB), Transcatheter Aortic Valve Implantation (TAVI), Pancreaticoduodenectomy (PD)) Korean tertiary hospital. Patient extracted each operations analyzed using techniques.Using clinical pathways, outpatient models, journeys demonstrated The result shows CDM's usability as novel valuable source analysis, yet few considerations. We found that should be complemented by different internal external sources address administrative operational aspects particularly ER analyses.To best our knowledge, first exploit mining. Specifically, step-by-step guidance demonstrating analysis locating relevant tables visualizing results tools. can widely applicable across institutions. work contribute bringing perspective existing users changing Hospital Information Systems (HIS) environment also facilitating studies research community.