Case study: Machine learning and patient participation in home-based clinical trials

This research investigates the relations between data, learning, and care in the context of wearables, real-time data capture, and patient-reported outcomes in home-based clinical trials. Rather than requiring patients to go to a medical facility, such trials (also known as decentralised, distributed or site-less trials) make use of wearables combined with machine learning and platform technologies to measure and monitor patient outcomes on a 24/7/365 basis.  

More specifically, this research aims to understand how patient experience and clinical outcome measures are measured in the home and then translated through processes of blurring, annotation, smoothing, and analysis into metrics that can then be used to inform clinical trials. It involves tracing the flows of data across these different processes, the teams, people, and spaces involved, as well as the theories, values, and relations underlying these processes.  

In tracing and coming to understand these translations and quantifications, this work brings together several further streams of theory and thinking: explorations and understandings of ‘the home’; the intersections of space, affect, and the production of knowledge; forms of closeness and distance in care; rarity in medicine; and illness experiences.  

We ask, how do these developments reconfigure clinical knowledge and the ways in which patients and their families understand disease?  How do they shape experiences of research, care, the clinic, and the home?