Continuous unobtrusive assessment of meaningful change in older adults with cancer: The Pacific Aging & Cancer Study Collaborative—PACS collaborative.
Chao-Yi Wu,Deanne Tibbitts,Zachary Beattie,Hiroko H. Dodge,Tomasz M. Beer,Daniel Marks,Jackilen Shannon,Jeffrey Kaye,Kerri M. Winters-Stone
DOI: https://doi.org/10.1200/jco.2021.39.15_suppl.1569
IF: 45.3
2021-05-20
Journal of Clinical Oncology
Abstract:1569 Background: Knowledge of changes in health that precede a cancer diagnosis is challenging because of a lack of longitudinal, objective measurement techniques. Current approaches rely on periodic assessment via self-report which may miss when and how health changes, particularly when changes may appear subtly over time. Remote-monitoring technologies provide a mechanism to continuously, passively, and unobtrusively monitor changes in health so that trajectories of change in relation to a life event (i.e., cancer diagnosis, treatment) can be detected and described. We examined the changes in digital indicators of health and life events 1 year before and after a self-reported cancer diagnosis in community-dwelling adults aged 65 and older. Methods: This is a secondary, retrospective data analysis of older adults who self-reported a new cancer diagnosis in the Oregon Center for Aging & Technology (ORCATECH) cohort. Ten older adults (age = 71.8±4.9 years, 30% women) were included with various cancer types (esophageal, prostate, uterine, pancreatic, b-cell follicular lymphoma, multiple myeloma, basal cell melanoma, basal and/or squamous cell carcinoma). Daily physical activity was measured using step counts derived from an actigraph watch. Weekly health and life events (pain severity, loneliness, hospitalization/emergency room (ER) visits, days away from home overnight) were self-reported from weekly online surveys. Results: A total of 3,624 days of actigraphy data (210 ± 88 days pre-cancer; 153 ± 81 days post-cancer) and 750 weeks of self-reported online survey data (36 ± 12 weeks pre-cancer; 39 ± 14 weeks post-cancer) were collected. Longitudinal linear mixed-effects models revealed that the trajectory of step counts was different pre- and post-cancer (β = -1.52, p <.001), with a gradual decrease in step counts before a cancer diagnosis. The trajectory of pain severity was different pre- and post-cancer (β = 0.01, p <.001), with a gradual increase in pain severity before a cancer diagnosis. There was a gradual increase in the occurrence of hospitalization/ER visits (OR = 1.07, p = 0.02) and days away from home overnight (OR = 1.04, p = 0.01) before a cancer diagnosis. Feelings of loneliness increased over time, regardless of pre- or post-cancer (OR = 1.04, p <.001). Conclusions: Changes in health and life events 1-year before a cancer diagnosis in older adults with varying cancer types and severity were unobtrusively observed. This study suggests that a remote-monitoring technology platform deployed in homes can detect meaningful intra-individual changes before and after a cancer diagnosis. Future studies can employ this technology as a pathway for improving the timeliness of detection and more effective therapeutic follow-up for older adults.
oncology