Digital phenotyping for changes in activity at the end of life in people with dementia

Abstract

L.D. Boyle, B.S. Husebo, B. Marty, K. Haugarvoll,O.M. Steihaug, M. Patrascu

Purpose: Almost 90% of people with dementia develop behavioral and psychological symptoms (BPSD) such as apathy, agitation, and sleep disturbances.¹ Recent research shows that data acquired from mapping the physical, mental, and functional activities of a person can serve as a marker for conditions including BPSD.^2-4 The application of digital phenotyping to people with dementia is still mostly unexplored, with issues of ethical nature at the forefront. There is therefore value in investigating whether digital phenotyping can enhance the objectivity of measuring activity and symptom changes during the last period of life. We hypothesize that use of sensing technology will enable better estimation of time of death, facilitating improvement of end-of-life interventions and directives for people with dementia. Method: DIgital PHenotyping in DEMentia (DIPH.DEM), a 3-year prospective cohort study (N=25), will monitor participants living in a nursing home at baseline, and every 6-months (7-days), up to 1-year. The objective of the study is to use sensing-based digital phenotyping (Somnofy and Garmin Vivoactive 5), combined with validated assessment tools, to describe the activity trajectory and associated processes of the end of life in persons with dementia. Presumed consent will be obtained for all participants and traditional outcomes measures completed by proxy. Results & Discussion: The use of digital phenotyping can provide knowledge on living, and dying, with dementia. This can improve directives and provide guidance for timely, appropriate interventions, including referral to palliative services and pharmacological interventions. Preliminary results from the first 11 participants are currently being analyzed (MATLAB and STATA). Analyses will include development of individual models based upon descriptive statistics, traditional outcome measures (proxy-rated), actigraphy (movement) and sleep quality data collected at baseline and at 6-months.  

References:

1. Sandvik, R. K., Selbaek, G., Bergh, S., Aarsland, D., & Husebo, B. S. (2016). Signs of imminent dying and change in symptom intensity during pharmacological treatment in dying nursing home patients: A prospective trajectory study. *Journal of the American Medical Directors Association, 17*(9), 821-827.
2. Husebo, B. S., Heintz, H. L., Berge, L. I., Owoyemi, P., Rahman, A. T., & Vahia, I. V. (2019). Sensing technology to monitor behavioral and psychological symptoms and to assess treatment response in people with dementia: A systematic review. *Frontiers in Pharmacology, 10*, Article 1699.
3. Vahia, I. V., & Forester, B. P. (2019). Motion mapping in humans as a biomarker for psychiatric disorders. *Neuropsychopharmacology, 44*(1), 231-232.
4. Au-Yeung, W. M., Miller, L., Beattie, Z., May, R., Cray, H. V., Kabelac, Z., et al. (2022). Monitoring behaviors of patients with late-stage dementia using passive environmental sensing approaches: A case series. *American Journal of Geriatric Psychiatry, 30*(1), 1-11.

Keywords: dementia, technology, end-of-life, phenotyping, sensors

Affiliation: Centre for Elderly and Nursing Home Medicine, Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway

Corresponding Author Email: lydia.boyle@uib.no

Ethics: This study was approved by the Regional Committee for Medical and Healthcare Research Ethics (REK) on October 23, 2023: 634938.

Acknowledgements: This research is supported by the Western Norway Regional Health Authority (Helse Vest RHF); sponsor protocol code: F-12829-D10484.