Abstract:
Objective: The objective of research was to understand and support quality use of medications in older adults living independently in an aged care facility. Considering ageing of populations, shortage of skilled caregivers, increasing chronic disease burden, use of multiple potent medications and high medication-related risk of morbidity and mortality in older people remain important problems. Improving adherence to and continuity of medications while balancing the risk of medication-related complications and morbidity with the use of information technology remains a desirable but underachieved goal. Instead of being limited to a range of existing solutions, the researcher intended to understand the reality from the perspective of older users. It was considered important to unravel the relationship between personal use of medications and a dynamic social environment in order to understand the implicit needs and demands of potential users from a technology-driven medication management system. The thesis explored the subjective reality of older people while using their medications and also while interacting with a collaboratively designed solution. The research intended to develop and test an interactive dialogue system on the touch screen interface of a social robot focussing on two important research questions: 1. What is the theory underpinning appropriate design of technology to enable older people to better manage their medications? 2. Can an automated medication management solution be developed successfully on a robot while being informed by the theory of automated medication assistance? Method: It was an interpretive and formative research attempting to discover theory rather than testing it. Based on the methodologies of Grounded Theory (GT) and Participatory Design (PD) within four Action Research (AR) cycles, the research elicited design implications and tested the design configuration addressing the unique task requirements. Within the overarching methodological framework, a variety of qualitative and quantitative methods were used that were appropriate to each stage. The beginning of the AR cycles was informed by an ethnographic analysis of the context of older residents of Selwyn Village (an aged care facility in Auckland). The study participants were elderly residents and concerned healthcare providers, caregivers and family members. The apparatus was designed and developed as part of a multi-disciplinary ‘Healthbots’ project at New Zealand Korea Centre for healthcare robotics, hosted within University of Auckland. Results: An initial ethnographic study mapped medication-related practices and processes, identified actors and scenarios defining the context for Action Research. The first AR cycle developed and tested a paper prototype and identified implications for software architecture and interface design. The second AR cycle observed residents interacting with a prototype and found them to be generally satisfied with it. The results informed further refinement of the prototype. A refined system in the third AR cycle led participants through a series of daily interactions, discovering a pattern of task mastery. The fourth and final AR cycle allowed older participants to independently use a robot within their apartments, discovering a successful interaction and safe medication use. The research confirmed that an ideal dialogue system aimed at older users could successfully meet the requirements if it was delivered on a touch screen (mounted on a robot as a computing device). A closed-loop system (as opposed to an open-loop standalone reminder) was designed, where the robot/device could work in synchronization with a web-based electronic medication record to enable real-time dynamic interaction with healthcare providers, caregivers and family members. A dynamic and complex set of variables around medication use, possible error situations, need for personalisation, need for patient education, monitoring of therapeutic efficacy and safety was unravelled. In each AR cycle qualitative data was collected and analysed using open, axial and selective coding along the principles of Grounded Theory, to arrive at four theoretical elements to the process of medication management, namely: Empowerment, Engagement, Collaboration and Safety. Conclusion: The research journey uncovered needs of older people beyond a simple reminder alarm and pill box. It successfully developed and tested a medication management module and proposed a theory of automating medication assistance for older people. It showed that older people can independently use a robot to help them manage self-care tasks. Qualitative methods such as the hybrid GT-PD-AR approach may be particularly helpful for innovating and articulating design requirements in challenging situations. A successful automated medication assistant for the elderly should effectively engage with and empower its users to self-manage their medications safely, while collaborating effectively with others who care for them. The research discovered new components to the medication management process in the context of older people and self-care task automation and opened a direction of research into the concept of ‘empowering technologies’.