Morgan Price

Changing human behaviour is complex, especially in the context of chronic disease management; it can present challenges to patients, providers, and the entire healthcare system. This dissertation was guided by the sociotechnical systems approach. It adapted and applied multiple methodologies to explore behaviour change at different levels of chronic disease management, specifically: the Lead User Method, the Theory of Distributed Cognition, sociotechnical systems theory, the Theoretical Domains Framework, Personas, i* Framework, and the Behaviour Change Technique Taxonomy v1. Methods used within the dissertation included semi-structured co-creative interviews and a rapid prototyping and co-design process. A primary aim of the dissertation was to document existing practice and inform behaviour change challenges for physical activity prescription in primary care. A secondary aim of the dissertation was to explore a theoretical and methodological basis for behaviour change strategies for physical activity prescription in primary care.A novel method was developed and piloted offering a practical way of addressing behaviour change interventions in primary care settings and viewing behaviour change within primary care from a sociotechnical systems perspective. It also provided a way to document and inform improvements to the process for prescribing physical activity in primary care by using a visual approach to the discussion where lead users co-design potential solutions to identified problems. The system-level goal model and behaviour change technique clusters were created to inform the potential design of an information technology tool that would support behaviour change interventions in primary care aimed at increasing physical activity levels as a way to manage non-communicable diseases.Future work stemming from this research has the potential to address the complexity of human behaviour change from a sociotechnical systems perspective both within different settings of primary care and beyond the context of primary care.

View record

At present, the pervasive integration of genomics and other big data into routine clinical care has not been realized, particularly in primary care. One of the critical problems of personalized medicine is an effective and efficient presentation of large genomic data and evolving knowledge in a generalist clinical encounter setting. To address this issue, this study aimed to design and evaluate a user interface for a genomic clinical decision support system intended for primary care physicians. This was a serial, multiple-methods study. This study focused on frailty and the clinically actionable aspects of the frailty lifecycle, such as risk assessment. In phase one of this research, the Lead User method for the participatory design was used for the design of the user interface for genomically-enabled decision support. The concept ideation phase was followed by the design synthesis process in phase two. Phase two generated a set of system-agnostic and evidence-based requirement patterns and an integrated user interface design based on the patterns. In phase three, the integrated design was validated with Representative Users, and the patterns were refined. The key novel contributions of this work were user interface requirement patterns for genomically-enabled clinical decision support and a requirement integration method that supported the pattern development. The nineteen novel and validated requirement patterns are geared towards primary care providers as clinical users. The produced patterns addressed the presentation of CDSS notifications at the point-of-care and the display of detailed personalized risk information, including the risk factors and suggested interventions to address risk. These patterns are technology-agnostic and provide information to future implementers of clinical information systems. Producing theoretically-grounded and user-validated design patterns for presenting large evolving clinical data and knowledge, rather than a particular implementation, allows for this work to be relevant in various software-intensive clinical systems and contexts. Methodologically, the study contributed by developing a requirement integration method that is practical, reproducible, and applicable to a wide variety of design problems where it is necessary to synthesize multiple design perspectives. The method ensures traceability of requirement origin and evolution. It supports theory-informed design and triangulation of evidence.

View record