Hi, and welcome to the presentation of the ER processes case study. In this case study, we will illustrate the use of process mining techniques for the analysis of the processes that daily take place in the emergency room of one of the main university hospitals in Chile. I am Marcos Sepulveda, professor of the computer science department at the Pontificia Universidad Católica de Chile and one of the members of the team that worked in this case study. This case study took place in the Red de Salud UC Christus, a medical network with more than 11 medical centres, 1,000 doctors, and 4,000 professionals.
The emergency room service of this network is one of the first emergency programmes created in the country and is committed to provide advanced medicine in an environment of efficiency and care for the patients. The Emergency Room, ER, has become one of the most significant first contact points with the healthcare system. ER must provide the required services to screen, examine, and provide care to patients in the most effective way. This has led to the increased efforts to improve the service levels, to reduce overcrowding, and to provide prompt and efficient care. These tasks are very challenging, because ER processes are intrinsically flexible, so they must adapt to the particular characteristics of each patient.
This flexible nature is evidenced through the presence of both mainstream and unusual behaviour in the ER. To address this challenge, there has been attempts to establish certain guidelines on how to treat patients, for example, by creating guidelines to address a specific diagnosis. Data from ER processes are stored in hospital information systems, HIS, information systems designed to manage all aspects of a hospital’s operation, including its medical, administrative, financial, and legal issues, and the corresponding processing of services. In our case, we worked with data extracted from a hospital information system called ALERT ADW Phase 1, which is a HIS used in the ER of the hospital. The data collected corresponds to July, 2014.
Different data were extracted by means of a specific report in CSV format from the database of the HIS. In particular, data related to problems described by the patients, vital signs, allergies, referrals, transportation, responsibility transference, diagnosis, professional activities, medication, final discharge, and triage. In total, 309,796 activities were released into the system, composing 7,160 different episodes. There are 64 different activity names registered, which illustrate the flexible nature of these processes. Based on the use of historical information related to process execution within the ER, it’s possible to provide answers to a number of questions frequently posed by experts in the field. They can be classified in both general questions and episode-oriented questions.
General questions follow the main process mining approaches: process discovery, conformance checking, performance analysis, and organisational analysis. Episode-oriented questions are based on certain clinical characteristics of data obtained when executing the process activities that are specific to the ER, for example the colour of the triage, the stay duration, or the discharge destination of the patient. The aim of this case study was to demonstrate the usefulness and applicability of process mining techniques to provide answers to several questions that were initially posed according to the specific needs of ER experts. For example, what process is executed for treating patients who are triaged red, are there certain diagnoses that are always triaged yellow and last more than 10 hours?
What are the activities carried out for providing attention to ER patients diagnosed with pneumonia. In particular, the specific question
that was addressed in our case study, was: what activities are carried out and what processes follow in providing care to ER patients diagnoses with appendicitis. Thank you for your attention.