Abstract
Healthcare organizations are under increasing pressure to improve productivity, gain competitive advantage and reduce costs. In many cases, despite management already gained some kind of qualitative intuition about inefficiencies and possible bottlenecks related to the enactment of patients' careflows, it does not have the right tools to extract knowledge from available data and make decisions based on a quantitative analysis. To tackle this issue, starting from a real case study conducted in San Carlo di Nancy hospital in Rome (Italy), this article presents the results of a process mining project in the healthcare domain. Process mining techniques are here used to infer meaningful knowledge about the patient careflows from raw event logs consisting of clinical data stored by the hospital information systems. These event logs are analyzed using the ProM framework from three different perspectives: the control flow perspective, the organizational perspective and the performance perspective. The results on the proposed case study show that process mining provided useful insights for the governance of the hospital. In particular, we were able to provide answers to the management of the hospital concerning the value of last investments, and the temporal distribution of abandonments from emergency room and exams without reservation.
Highlights
Nowadays, hospitals try to streamline their processes to deliver high quality care improving revenues and reducing costs
We present in this article a process mining project performed in the healthcare domain, in San Carlo di Nancy hospital located in Rome, Italy
The PM2 methodology consists of six stages, whose instantiation in the range of the case study will be thoroughly discussed in the subsections: (i) planning, FIGURE 2
Summary
Hospitals try to streamline their processes to deliver high quality care improving revenues and reducing costs. A complicating factor is that healthcare is characterized by highly complex and extremely variable patient care processes, referred to as careflows [1]. A wide range of careflows with different characteristics and requirements are daily managed and executed. The delivery of complex care may involve several departments and organizations, and requires an active collaboration between different professionals and practitioners having heterogeneous skills. Traces of executions of careflows are available in information systems in the form of raw system logs, which make it difficult for the management of a hospital
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
