Designing Personal Health Technologies for Translating Population-Level Evidence into Personal Understanding

Thesis (Ph.D.)--University of Washington, 2020Growing interest in better tracking and managing health is leading a global market for self-monitoring technologies to reach an estimated $72 billion by 2022. This boom has contributed to a new potential to support people in collecting and interpreting data about their own health and well-being. However, there is often a mismatch between what technology currently delivers (e.g., step counts, sleep scores) versus what people expect from it (e.g., personalized health insights and recommendations). Current technologies often fall short of their potential to address individual health needs due to complex and interrelated challenges (e.g., in meeting individual needs, in data quality, in their integration into clinical practice). Such personalized understanding of one's own health is vital for people with various chronic conditions where symptoms and their triggers generally vary across individuals (e.g., irritable bowel syndrome, migraine). To scaffold people’s attempts to better understand their chronic conditions through collecting and interpreting personal health data, I first present a framework for self-experimentation. The framework provides guidelines on how tools can be designed to provide the necessary support in avoiding common pitfalls associated with tracking health data and in assisting people in answering health questions in a scientifically valid manner. Through the framework, I propose better design of tools aimed at assisting people in answering questions related to the individualized nature of their condition using personal health data. To demonstrate a practical application of the framework, I next present my work in the domain of Irritable Bowel Syndrome. I leveraged the self-experimentation framework to design TummyTrials. TummyTrials incorporates existing medical knowledge to assist people in conducting scientifically valid self-experiments in the context of their day-to-day life to determine which foods are triggering their symptoms. Through my research in TummyTrials, I demonstrate the feasibility of people conducting rigorous experiments when a system provides sufficient scaffolding to bridge associated expertise gaps. I next explored supporting people with chronic liver diseases in gaining an understanding of their cognitive function. I developed Beacon, a platform that enables cirrhotic patients to self-measure critical flicker frequency, a measure of cognitive function that has been considered promising in this population but has previously been impractical to collect. Through Beacon, I am enabling a new data stream for patients and providers to gain a easy-to-measure proxy of cognitive functioning in a high-risk population. Across these projects, the common theme is my pursuit of designing, developing, and promoting tools which leverage the existing population-level evidence in medicine to provide new value at a personal level. I synthesize design recommendations based on empirical findings from building the above tools to assist future tool builders in designing improved tools. I emphasize that realizing the full potential of new tracking technologies requires that designers and researchers scaffold the necessary domain expertise and the individual context to provide meaningful data experiences