Opportunities for collaborative clinical work: Predicting relapse onset in bipolar disorder from online behavior

Bipolar disorder (BD) can negatively impact the lives of individuals. Symptoms of BD can manifest not only in their offline behaviors, but online as well. Being able to identify manic and depressive mood episodes early on can lead to more effective interventions. In this work, we focus on understanding the feasibility and acceptance of an early warning system for patients with BD that leverages online behavioral data to infer mood episode onset. For this, we interview three participants with BD to probe how they envision this type of intervention system and might use it to manage BD. Our goal is to uncover the opportunities and constraints of the future of work in BD healthcare that connects intelligent tools and objective data to provide an effective partnership between patients, caregivers, and clinicians. Toward this goal, in this paper, we focused on understanding concerns and gathering design ideas from patients with BD. We present this study as a case for a new type of work, incorporating clinical perspectives from start to finish-both as collaborators and active participants - -to enhance clinical work experiences and provide better care.