Seite - 18 - in Applied Interdisciplinary Theory in Health Informatics - Knowledge Base for Practitioners

Exactly as described in GST. In Learning Health Systems (LHS), these ideas are extended to include developing new medical learning and there is a strong emphasis on the use of health informatics solutions as both the provider of the data that will be used for evidence-based medicine and the vehicle for delivering knowledge to the clinical teams through automated decision support and workflow management. The Heimdall Framework [10] provides a taxonomy of types of learning health system where new clinical insight and patient process improvements are driven by the analysis of data from the electronic health record (EHR) and other health information systems. In GST terms, clinical and management control is informed by feedback about processes and outcomes and is implemented as interventions to the inputs and process. More data, faster data flows and improved analytical abilities improve control and the organization's long-term ability to continuously learn and adapt to its changing environment. A key insight from GST is that of systems-within-systems, each contributing to overall success. An LHS approach can therefore be applied to a surgical team, a ward, a department or clinical specialty as well as the organizational, regional and national systems in Friedman’s vision. Following GST carefully would suggest that LHS should indeed be implemented at all levels of the organizational hierarchy including the individual human as reflective practitioner. Adoption of integrated informatics solutions, interoperability standards and improved methods for mining health data are essential for LHS but the long term vision is of systems that self-learn through embedded AI and a new generation of digital-native clinicians who are part of, but remain firmly in control of, their health system. LHS is seen as a driver for health informatics but to succeed it requires the deeper understanding of the relationships between organizational structure, people, processes and technology that comes from applying GST. 2.4. Applications of GST in Process Mining of Care Pathways The care pathway is a commonly used concept for considering how the processes of delivering healthcare should best be organized around the needs of the patient [11]. A care pathway is a design template for a healthcare process – it describes the sequence of care that is recommended for patients with similar conditions requiring similar treatment. Comparing the actual care that patients received as recorded in the EHR against the intended care pathway should help healthcare organizations understand the gap between what they think they are doing and what they are actually doing, a key requirement for learning. Coiera [12] suggests that LHS should use process mining to develop automated process-level metrics and identify common multi-variate process patterns to help better understand how healthcare delivery is structured. Process mining is a set of big data analytics tools and techniques that use time-series event data to specifically address process characteristics and there is growing interest in process mining in healthcare [13]. Ronnie Mans and Wil van der Aalst [14] provide a comprehensive guide to process mining in healthcare including health reference models and pathways. Process mining has been combined with process simulation to create a mixed methods approach to support the development of LHS [11]. In the following example we illustrate how process mining of a care pathway fits with GST and an LHS vision. O.Johnson /GeneralSystemTheoryand theUseofProcessMining to ImproveCarePathways18