Anticipating the consequences of technical change in healthcare: a conceptual modelling solution

Abstract

Introducing new Information Technology (IT) into healthcare environments usually leads to both expected and unexpected outcomes. Unfortunately, the expected benefits of technology use in healthcare are largely unmet, and the costs associated with unexpected consequences continue to rise. Research is needed to ensure that new IT improves, rather than hinders, the quality of healthcare provided and the ease with which healthcare work is coordinated.To understand the impact of technology, analysts need to understand the underlying work system and organisational processes into which the technology will be introduced. At present, business process analysts often document organisational processes in a graphical way (e.g., using a notation), and they use these representations to assess the impact of technical change on activity flow, resources, and outcomes. Although this practice has been successful for less complex domains, it is uncertain if the existing notations are adequate for representing healthcare processes. In healthcare processes, workers adapt the way they perform their duties in response to fluctuating needs of the clinical situation and changing conditions in the environment. The specifications of existing notations are insufficient to represent either the dynamic human interactions within the clinical workflows or the variability with which healthcare workers perform activities. The limitations of existing notations may be preventing analysts from effectively realising the consequences of implementing new technology into their organisations.The purpose of this thesis is to develop a notation that is tailored to the complex healthcare environment so that analysts might readily understand, and be able to promote, effective organisational change. The notation developed in the thesis, Health Process Notation (HPN), is intended to facilitate the representation of the interactions that occur between people, technological, and clinical aspects of the domain using pictorial elements and annotated connector objects. It is hoped that HPN might help analysts overcome the difficulties of implementing new forms of IT within a health organisation, to envision more easily the impact of new technical support on the workplace, and to identify areas of concern. Cognitive engineers endeavour to understand the tradeoffs within complex systems while also addressing the needs of both the workers and analysts responsible for implementing new forms of IT within a health organisation. This thesis is a novel extension of Cognitive System Engineering research because it involves the application of a paradigm from the Information Systems discipline known as Design Science Research (DSR) to a problem commonly addressed by cognitive engineers. DSR guides the development of artefacts to be used in practice. DSR artefacts, such as a notation, are developed through a series of research cycles that are supported by an understanding of the environment. The DSR cycles were applied in the following way. First, the DSR’s Relevance Cycle required a thesis of a healthcare environment, called a domain study. The domain study provided the foundation for identifying the concepts and relations within healthcare processes that would need to be captured by the notation. The domain study ensured that any subsequently developed notation was developed with consideration of the specific requirements of the healthcare domain, such as conditions of accessing information from different sources, coping with changes in the environment, and the demands of coordinating care among multiple providers.Second, through the DSR’s Design Cycles, the notation was developed and tested through iterations to ensure that it supported people’s understanding and reasoning about the healthcare domain. Three tests were conducted: one after the notation’s development (Test 1), one after initial refinement (Test 2), and one after additional refinement (Test 3). Test 1 identified opportunities to improve the notation, after it became clear that a subset of symbols was not understood. Following refinement to the notation, Test 2 and Test 3 compared the properties of HPN to those of the industry standard, Business Process Modelling Notation (BPMN). Test 2 demonstrated some practical benefits of HPN; specifically, participants using HPN exhibited a better understanding of which activities happened simultaneously and which activities involved multiple people. Test 3 verified the usefulness of both HPN and BPMN for understanding complex processes and reasoning about a technical change. Performance metrics showed that HPN was no less effective than the industry standard, BPMN.In summary, the thesis presents a novel process modelling notation for representing complex clinical workflows, and demonstrates how to use processes and concepts from Cognitive Systems Engineering and DSR to achieve a well-formed design solution that supports analysts’ work. The thesis is one of the first known instances of applying DSR to the problem of envisioning the impact of technical change in healthcare. In the future, it is hoped that HPN—and the methods by which it was developed—will prove useful to analysts and help them better identify and successfully manage the range of potential impacts that IT implementations can have within healthcare environments.