J. Nicholas Dionne-Odom
Distress Prediction in Advanced Cancer Family Caregivers and their Care Recipients using Digital Phenotyping
The purpose of this project is to explore the feasibility, acceptability, and potential utility of passively-collected smartphone data to assess distress in family caregivers and their care recipients with advanced cancer in an underserved context. We will passively collect smartphone behavioral data, including GPS, accelerometer, and anonymized call and text messaging use patterns, and develop time-varying statistical models to detect behavioral anomalies and correlates these anomalies with participant-reported distress and quality of life over 24 weeks. Our sample will include 50 family caregivers and 50 of their care recipients with newly-diagnosed advanced cancer who have a personal smartphone, over-recruiting underserved individuals who are either rural-dwelling and/or African-American. Participants will download an app called Beiwe to securely collect de-identified research-quality smartphone data. We will assess acceptability by conducting qualitative interviews with participants. The results could be exceptionally high impact as they can be potentially used to develop new clinical models of early palliative care to enhance outcomes and reduce disparities in cancer and other serious illnesses.