To meet challenges in healthcare, care services require development of approaches and methods for coordination and quality improvement, including new forms of system approaches and methods that support process-based organisation within health care departments as well as between clinics, between various public health and caregiving organisations.
Logistics is as an effective solution in the organization of working time to care staff by offering them the opportunity to concentrate on their core activities and improve patient care conditions. The management of logistics activities goes beyond traditional physical flows, and it considers other flows such as patients throughout the care chain.
In this thematic area, we consider a data-driven approach to the organization of workflows in care chains. Use cases cover Akademiska Uppsala Hospital, Karolinska University Hospital (Solna and Huddinge), and Barn och Ungdom Psychiatry Stockholm (the organization providing psychiatric care to children and young people), all in Sweden. The datasets are large and complex in nature, ranging from 50k records to over 40 million records. The data describes the events that occur along various care chains through these complex systems.
The goal is to establish an empirical, data-driven base of evidence for improving workflows in hospitals. Using traditional data mining and machine learning techniques, we use the data to uncover bottlenecks in care chains, patient groups and variations, inefficiencies in resource utlization and so on to improve workflow efficacy, resource utilization and work conditions for staff.