In-Home Technology for Caregivers of People With Dementia and Mild Cognitive Impairment: Wearables

This study aims to develop, evaluate, and commercialize an in-home supportive technology that is designed to alleviate anxiety, burden, and loneliness in spousal and familial caregivers of individuals with Alzheimer's disease, other dementias, or mild cognitive impairment by integrating wearable devices (e.g., Apple Watches).

This study aims to develop, refine, evaluate, and commercialize a hardware/software system designed to integrate in-home sensors and devices, Internet-of-Things technologies (i.e., devices that can be controlled and communicated with via the internet), and social networking to create a more safe and supportive home environment for caregivers and people who have Alzheimer's disease, other dementias, or mild cognitive impairment. The system monitors troublesome behaviors in people with dementia or mild cognitive impairment (e.g., wandering), and targets mechanisms (e.g., worry, social isolation) thought to link behavioral symptoms in people with dementia or mild cognitive impairment with adverse caregiver outcomes (declines in health and well-being). The system is designed to minimize demands on caregivers' limited time and energy and to provide a platform for data collection that can be used by researchers and care professionals. Hypotheses: Caregivers in the active treatment condition will have better health and well-being (i.e., less caregiver depression, anxiety, loneliness, and burden) and greater user satisfaction compared to those in the control condition. The magnitude of the difference in health and well-being benefits for caregivers in the active treatment condition compared to those in the control condition will increase over time (reflecting additional bot learning and adjustment to changing caregiver needs). In the active treatment condition, greater utilization of features related to warnings (e.g., activating warnings, receiving warnings), social connection (e.g., adding Trusted Circle members), and information (e.g., accessing on-line resources through app) will be associated with greater decreases in depression and anxiety and greater increases in well-being.

Participants are randomly assigned to treatment arms by People Power. Identical systems are self-installed by caregivers in all homes regardless of treatment arm. Initiating the features of the system appropriate to the assigned treatment arm is done remotely by a member of the People Power staff following the installation. All participants complete the same questionnaires at the same intervals.

The full system [(a) 1 gateway that connects with home internet to communicate/control the equipment; (b) 5 indoor motion sensors; (c) 3 door/cabinet entry sensors; (d) 1 water leak sensor; (e) 1 "call for help" button; (f) 2 motion-activated LED night lights; (g) 1 Apple Watch to detect falls and activity outside the home] will be self-installed by caregivers (N=60) in their homes. Monitoring of sensors, provision of warnings, messaging, and social networking features will be activated remotely for those participants who have been randomly assigned to this arm. Participation will extend over a 6 month period with questionnaires (e.g., health and well-being) administered 3 times (at the time of installation and every 3 months thereafter).

The full system [(a) 1 gateway that connects with home internet to communicate/control the equipment; (b) 5 indoor motion sensors; (c) 3 door/cabinet entry sensors; (d) 1 water leak sensor; (e) 1 "call for help" button; (f) 2 motion-activated LED night lights], with the exception of the Apple Watch that those in the experimental condition receive will be self-installed by caregivers (N=60) in their homes. Only monitoring of the water leak and associated warnings will be activated remotely for those participants who have been randomly assigned to this limited (sham comparator) arm.

Intelligent bots monitor the in-home sensors, learn typical patterns, and provide caregivers with text messages and alerts via cell phone when worrisome behaviors occur. Caregivers are able to: (a) select services (e.g., warnings for falls, wandering, late night activity); (b) access daily reports (summaries of daily activities that can also be shared with health care providers); and (c) obtain support (e.g. Caregiver Support Groups that connect caregivers with knowledgeable experts and other caregivers, Caregiver Events that provide virtual meetings about relevant topics, and Trusted Circle task management to distribute the caregiving work load).

Intelligent bots monitor the in-home water leak sensor and provide caregivers with text messages and alerts via cell phone when worrisome conditions occur.

Questionnaire to measure depression (Radloff, 1977). 20 items are rated on a 0-3 scale and summed (range = 0-60). There are no subscales. Higher scores represent worse outcomes. The clinical cut-off is usually set at a score of 16.

Questionnaire to measure depression (Radloff, 1977). 20 items are rated on a 0-3 scale and summed (range = 0-60). There are no subscales. Higher scores represent worse outcomes. The clinical cut-off is usually set at a score of 16.

Questionnaire to measure caregiver burden (Zarit, Reever, & Bach-Peterson, 1980). 12 items are rated on 0-4 scale. Range: 0-48. No subscales. Higher scores represent worse outcomes.

Questionnaire to measure caregiver burden (Zarit, Reever, & Bach-Peterson, 1980). 12 items are rated on 0-4 scale. Range: 0-48. No subscales. Higher scores represent worse outcomes.

Questionnaire to measure anxiety (Beck, Epstein, Brown, & Steer, 1988). 20 items are rated on a 0-3 scale and summed (range= 0-60). Higher scores indicate worse outcomes. There are no subscales. A score greater than 36 is considered to be clinically significant.

Questionnaire to measure anxiety (Beck, Epstein, Brown, & Steer, 1988). 20 items are rated on a 0-3 scale and summed (range= 0-60). Higher scores indicate worse outcomes. There are no subscales. A score greater than 36 is considered to be clinically significant.

Questionnaire measuring overall life satisfaction and well-being (Diener, Emmons, Larsen, & Griffin, 1985). 5 items scored on a 1-7 scale and summed (Range = 5-35). Lower scores indicate worse outcomes. A score of 20 is considered neutral with higher scores considered increasingly more satisfied and lower scores considered increasingly more dissatisfied.

Questionnaire measuring overall life satisfaction and well-being (Diener, Emmons, Larsen, & Griffin, 1985). 5 items scored on a 1-7 scale and summed (Range = 5-35). Lower scores indicate worse outcomes. A score of 20 is considered neutral with higher scores considered increasingly more satisfied and lower scores considered increasingly more dissatisfied.

Questionnaire to measure functional health (Brim, Ryff, & Kessler, 2004). 13 items asking about limitations in two parts (yes/no and if they have the limitation rating its duration on a 1-3 score). Items are summed (range 0-39). Higher scores indicate worse outcomes.

Questionnaire to measure functional health (Brim, Ryff, & Kessler, 2004). 13 items asking about limitations in two parts (yes/no and if they have the limitation rating its duration on a 1-3 score). Items are summed (range 0-39). Higher scores indicate worse outcomes.

Inclusion Criteria: Caregivers are fluent/literate in English Caregivers currently reside in the United States with spouse/family member who has received a medical diagnosis of Alzheimer's disease, other dementia, or mild cognitive impairment Caregivers primarily use a smartphone (e.g., iPhone, Android) Caregivers have internet and WiFi service Exclusion Criteria: Caregivers providing care for individuals with known non-neurodegenerative conditions affecting behavior and cognition Caregivers providing care for individuals with longstanding Axis I psychiatric disorder Caregivers providing care for individuals with metabolic disorder or major organ dysfunction Caregivers providing care for individuals with alcohol abuse or dependence (within 5 years of dementia onset) Caregivers providing care for individuals with head trauma with loss of consciousness greater than 30 minutes Caregivers providing care for individuals with contraindications to MRI imaging Caregivers providing care for individuals with large confluent white matter lesions Caregivers providing care for individuals with significant systemic medical illness Caregivers providing care for individuals who use a medication likely to affect central nervous system functions adversely

Because health and other protected data are involved, we need to consult with our institutional review board and other collaborators concerning which data can be shared.

Chen KH, Wells JL, Otero MC, Lwi SJ, Haase CM, Levenson RW. Greater Experience of Negative Non-Target Emotions by Patients with Neurodegenerative Diseases Is Related to Lower Emotional Well-Being in Caregivers. Dement Geriatr Cogn Disord. 2017;44(5-6):245-255. doi: 10.1159/000481132. Epub 2017 Dec 8.

Lwi SJ, Ford BQ, Casey JJ, Miller BL, Levenson RW. Poor caregiver mental health predicts mortality of patients with neurodegenerative disease. Proc Natl Acad Sci U S A. 2017 Jul 11;114(28):7319-7324. doi: 10.1073/pnas.1701597114. Epub 2017 Jun 27.

Otero MC, Levenson RW. Lower Visual Avoidance in Dementia Patients Is Associated with Greater Psychological Distress in Caregivers. Dement Geriatr Cogn Disord. 2017;43(5-6):247-258. doi: 10.1159/000468146. Epub 2017 Apr 11.

Brown CL, Lwi SJ, Goodkind MS, Rankin KP, Merrilees J, Miller BL, Levenson RW. Empathic Accuracy Deficits in Patients with Neurodegenerative Disease: Association with Caregiver Depression. Am J Geriatr Psychiatry. 2018 Apr;26(4):484-493. doi: 10.1016/j.jagp.2017.10.012. Epub 2017 Dec 27.