Collaborators: Kira Birditt (PI), Marco Angrisani (MPI), Florian Keusch, Ritika Chaturvedi (Co-I), Fred Conrad (Co-I), Richard Gonzalez (Co-I), Joan Monin (Co-I)
Funding: National Institute on Aging
Grant: R01 (AG083097-01)
Project Summary: Up to 36 million Americans are caregivers to disabled adults and approximately 16.1 million provide informal caregiving to people with Alzheimer’s disease and related dementias (ADRD).1,2 Compared to other types of caregiving, ADRD caregivers often report greater burden, poorer psychological and physical health, lost productivity, and financial strain. Given the rising number of older adults with disability and ADRD and the accompanying shortage and cost of paid caregivers,3 family and other unpaid caregivers are increasingly vital in the long-term care of older adults. Caregiving is often shared by multiple social partners across the life course (e.g., family, stepfamily, and non-family) who may experience heterogeneous burden. Yet, most research focuses on spouse and child caregivers and often lacks information regarding non-White and young adult caregivers who may experience caregiving differently due to health, socioeconomic, or cultural factors. Digital technology-enabled studies of daily stress exposure (e.g., interpersonal tensions, work stress) and reactivity (i.e., the link between exposure and daily well-being) may provide crucial information regarding the mechanisms by which caregiving is linked with greater burden or worse psychological and physical health outcomes. Digital technologies (collected from smartphones and wearables) offer passive as well as active, non-invasive tools for measuring everyday life and health data. Fitbits, for example, passively collect relatively accurate, validated, and continuous person-generated health data (PGHD). Smartphone-based ecological momentary assessments (EMA) involve actively assessing individuals’ daily lives without imposing excessive participant burden or access barriers. The present study proposes to include new survey instruments in the Understanding America Study (UAS), a well-established, probability-based Internet panel currently comprising 10,000 individuals ages 18 and older, to identify caregivers and implement new EMA assessments in conjunction with wearable devices to capture caregivers’ daily experiences and their links with daily emotional and physiological well-being. We will obtain a representative sample of ADRD and non-ADRD family and non-family unpaid caregivers, across multiple age groups that allows for comparison among often under-represented groups of informal caregivers and accounts for multiple caregiving roles (i.e., providing care for more than one recipient). Consistent with other caregiving studies, we define caregiving as assisting a person with basic activities of daily living because the recipient is not able to do the activities without help. We address three aims: 1) Build a representative, life-course sample of unpaid caregivers to follow longitudinally and collect real-time daily experience and physiological data, called UAS-Caregiving Lifecourse Experiences Assessed in Real-time (UAS-CLEAR). We will collect two waves of data, 1.5 years apart, each lasting 1 year and with four components. Specifically, we will 1A) survey all UAS respondents to identify caregivers and assess their experiences (e.g., relationship with care recipient, conditions of care recipient, caregiving intensity/task, burden, future expectations for providing/receiving care); 1B) deploy a previously developed smartphone application to collect EMAs of daily stress (6 a day for 7 days); 1C) conduct short monthly surveys about caregiving status, physical and psychological health; and 1D) use Fitbits to continuously measure biometrics including heart rate (HR), heart rate variability (HRV), sleep, and physical activity throughout the 1year data collection phase. A unique public use dataset will be created for the larger research community. 2) Compare daily experiences (stress exposure, stress reactivity) among ADRD and non-ADRD family and non-family caregivers from young adulthood to old age. Statistical and machine learning (ML) methods will assess whether daily experiences, self-reported and objective well-being outcomes and links between experiences and well-being (i.e., stress reactivity) vary by relationship type, ADRD vs non-ADRD status, and age group. (H2.1) ADRD caregivers report greater daily stress exposure and exhibit greater stress reactivity than non-ADRD caregivers, and (H2.2) controlling for caregiving intensity, family caregivers report greater stress exposure and exhibit greater reactivity compared to non-family caregivers. 3) Identify individuals who are more or less resilient to daily stress and examine whether resilience factors vary between ADRD and non-ADRD family and non-family caregivers. We will examine daily appraisals, behaviors (coping strategies, health behaviors) and psychosocial resources (e.g., efficacy, relationship quality) that either buffer or exacerbate the effects of daily experiences on self-reported well-being (emotional, physical) and physiological health. We will explore whether the associations vary by relationship type, ADRD status, and age group. (H3.1) Reactivity to stress will be reduced among individuals who report better quality social relationships (i.e., more supportive, lower negativity). (H3.2) Maladaptive health behaviors (e.g., smoking, poor eating) will exacerbate the effects of daily stress.