The DIPH.DEM study
Digital phenotyping for changes in activity at the end of life in people with dementia: an observational trial based on sensing technology (DIPH.DEM) is a study investigating activities and symptom changes in persons with dementia, with the goal of describing the activity trajectory toward the end of life. DIPH.DEM is funded by the Regional Health Authorities of Western Norway (Helse Vest) and functions as a pilot to the larger 5-D study.
Main content
Almost 90% of people with dementia develop behavioral and psychological symptoms (BPSD). Recent research shows that data acquired from mapping the physical, mental, and social activities of a person can serve as a marker for some clinical conditions, including BPSD. The use of digital phenotyping (analysis of characteristics and behaviors through sensor technology data) in people with dementia is still mostly unexplored, therefore there is value in investigating whether these measures can enhance the objectivity of tracking activity changes during the last period of life in nursing home residents with dementia.
Objectives
The primary objective of DIPH.DEM is to evaluate the physical activities of people with dementia living in the nursing home, as many experience restlessness, pain and poor sleep during their stay. For this, the study will combine digital measures with validated assessment tools to describe the activity trajectory. The secondary objective is to discover which sensor types and data resolution are required to successfully monitor activity in people with dementia. The study functions as a pilot to the larger 5-D study.
Sensors and devices
Sensing technology used in the study includes the Garmin Vivoactive5 smartwatch (movement, heart rate) and the contactless wall-mounted radar-based sensor Somnofy (respiration, light, sound, air quality, movement, sleep). Data is collected over 7-day periods every 6 months up to 1 year, with continuous measurement between 8–12 weeks after a significant event causing a change in health status. All data will be stored using the secure server infrastructure at the University of Bergen.
Team
Coordinating investigator is PhD candidate Lydia D. Boyle, who is working with an international multi-disciplinary team including Bettina S. Husebø (professor and palliative care physician, main supervisor of the candidate), Monica Patrascu (systems scientist, project manager and co-supervisor), and Brice Marty (neuroscientist).
The collaborators of DIPH.DEM are Kristoffer Haugarvoll (neurologist, co-supervisor), Ole Martin Steihaug (geriatrician, co-supervisor), Rune Samdal (user representative), Ipsit V. Vahia (psychiatrist, USA), Rui Nouchi (psychologist, Japan), and Wilco P. Achterberg (elderly care physician, The Netherlands).
Current status
The study was initiated in May 2023, and received ethical approval in October 2023 (REK). The DIPH.DEM team met with partner institutions in the U.S. in the fall of 2023, allowing for enhanced collaboration about the project between the University of Bergen, Yale University, Harvard University, and MIT. Active recruiting efforts at the Bergen Red Cross Nursing Home began in January 2024. The primary contacts and participants have been provided with education, training, and support for the project. DIPH.DEM has so far included 11 total participants, with baseline measurements completed and data analysis underway.
The DIPH.DEM team is very thankful to the families, participants, and healthcare team at the Bergen Red Cross Nursing Home for making this study possible.