Description

In 2024, approximately 6.9 million diagnosed cases of Alzheimer's disease and related dementias (ADRD) were reported in the United States, with 83% of the caregiving burden shouldered by unpaid family members or friends. The duration of ADRD varies between 4-20 years, during which individuals often transition into a state of complete dependency. Without family caregivers, the long-term services and support system would be unsustainable. The dearth of accessible support for ADRD caregivers constitutes a significant public health emergency. Caregivers are frequently uncertain of which services are needed, available, and how to gain access - often leaving those most vulnerable without essential support. When queried about service underuse, study respondents who participated in the K99 phase of this project highlighted an information deficit tailored to specific cultures, demographics, and dementia types. They described the excess of online information as overwhelming, often irrelevant, impractical, or unaffordable. Existing technology-based solutions targeted toward enhancing personalized caregiver support are limited. Leveraging continued innovations in technology to inform the training and testing of machine learning algorithms, which can match and update resources while accounting for individual needs, preferences (in-person, virtual), and barriers (e.g., employment, lack of respite care), holds great potential to enhance the precision of service linkage for ADRD caregivers. This R00 project aims to develop, refine, and pilot test CarePair (formerly the Caregiver Resource Room), a mobile application assessment and service referral platform for dementia caregivers. CarePair will leverage innovative machine learning algorithms to holistically evaluate caregivers' evolving needs, their barriers, and preferences to generate personalized service referrals relevant to their areas of identified need. Thus, the specific aims of the R00 are to 1) Use mixed-method and focus group data from the K99 phase to inform the iterative development of the CarePair, which includes a digital self-assessment tool employing machine learning to identify needs, categorize them, and generate targeted service recommendations; 2) Evaluate front- and back-end usability (e.g., via task analysis, heuristic evaluation) of the tool's content, design, features, functionality, and accuracy of service output. Feedback will inform modifications and iterative refinement of the CarePair Version 2.0; 3) Conduct a pilot randomized controlled trial to assess the feasibility, acceptability, and preliminary efficacy of CarePair in enhancing service awareness, addressing unmet needs, and improving mental health. The proposed research aligns with the NIA's strategic initiative to foster research scientists in aging and to develop promising interventions to better engage and support the well-being of ADRD family caregivers.