Other Titles

Harnessing AI-Driven Technology to Empower Dementia Caregivers Across Home, Clinic, and Community

Other Titles

Symposium Presentation

Abstract

The prevalence of behavioral symptoms, such as wandering and agitation, in over 90% of persons with Alzheimer's disease and related dementias (Pw-ADRD), with approximately 83% residing in home-based settings [1-3], underscores the urgent need for effective interventions to support informal caregivers. These caregivers are essential for maintaining quality care for Pw-ADRD, yet they frequently encounter substantial challenges in managing behavioral symptoms [4], revealing a significant gap in available resources and support systems.

To address this gap, we developed a prototype for an AI-based Dementia Care Voice Assistant (AI-DECAVA), powered by a large language model (LLM). The system is designed to aid informal caregivers in managing behavioral symptoms associated with ADRD. However, general-purpose LLMs are often inadequate for closed-domain question answering (QA) tasks, as they lack the specialized knowledge required for nuanced subdomains. Broadly pre-trained models typically do not capture the domain-specific terminology and contextual precision needed in healthcare applications—a limitation compounded by sparse conversational data and computational constraints in specialized fields.

Our methodological approach employs a Retrieval-Augmented Generation (RAG) framework, which combines targeted data retrieval with fine-tuning using domain-specific question-answer pairs. A core component of this process is Question-Answer Generation (QAG), used to create tailored training datasets. This strategy enables fine-tuned models to assimilate specialized terminology and contextual knowledge while remaining computationally tractable for domain-specific deployment. A principal advantage of this framework is its replicability across diverse domains and scalability for integration into varied user groups, positioning it as a versatile and generalizable solution for enhancing conversational assistants in specialized care contexts. Initial validation of AI-DECAVA was conducted with five informal caregivers, demonstrating usability and functional efficacy.

These preliminary results provide critical insights to inform the design of a larger feasibility study and serve as proof-of-concept for the application's potential to strengthen caregiver support and improve dementia care outcomes. This innovation contributes to nursing science by empowering informal caregivers to deliver quality dementia care to persons living with ADRD.

Notes

Presenter notes available in attached slide deck.

References:

[1] Alzheimer’s Association. (2025). 2025 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 21(4), 1598–1695. https://www.alz.org/getmedia/ef8f48f9-ad36-48ea-87f9-b74034635c1e/alzheimers-facts-and-figures.pdf

[2]Meline, M. (2025, May 16). Home health care use rose among Medicare beneficiaries with dementia until the pandemic. Penn LDI Research Updates. Retrieved from https://ldi.upenn.edu/our-work/research-updates/chart-of-the-day-home-health-care-use-rose-among-medicare-beneficiaries-with-dementia-until-the-pandemi

[3] Centers for Disease Control and Prevention. (2024). Helping dementia caregivers. https://www.cdc.gov/caregiving/resources/helping-alzheimers-caregivers.html

[4] N. M. Huang, L. Z. Wong, S. S. Ho, and B. Timothy, “Understanding Challenges and Emotions of Informal Caregivers of General Older Adults and People With Alzheimer Disease and Related Dementia: Comparative Study,” Journal of Medical Internet Research, vol. 27, Feb. 2025. Available: https://doi.org/10.2196/54847

Description

Overall Symposium Summary: This symposium presents preliminary findings on AI-driven interventions to enhance Alzheimer’s and dementia caregiving across home, clinical, and community contexts. Projects include: a voice assistant app guiding informal caregivers in managing behavioral symptoms; a gamified VR training tool cultivating dementia care competencies in healthcare students; and a data-driven app matching caregivers with tailored financial resources. These innovations aim to empower caregivers and improve outcomes.

Note: The attached slide deck is a combined symposium presentation containing the slides of all featured symposium speakers.  

To locate the other presentations in this symposium, search the repository by the Symposium Title shown in the Other Title field of this item record. 

Author Details

Presenter: Modupe Akintomide, PhD, MSN/MPH. PHNA-BC. CHES

Co-presenters: Xinyue Zhang, PhD, Divya Iyer, Madison Ellison, & Sumaya Tamenne

Sigma Membership

Non-member

Type

Presentation

Format Type

Text-based Document

Study Design/Type

Other

Research Approach

Other

Keywords:

Health Equity or Social Determinants of Health, Public and Community Health, Competence

Conference Name

37th International Nursing Research Congress

Conference Host

Sigma Theta Tau International

Conference Location

Toronto, Ontario, Canada

Conference Year

2026

Rights Holder

All rights reserved by the author(s) and/or publisher(s) listed in this item record unless relinquished in whole or part by a rights notation or a Creative Commons License present in this item record. All permission requests should be directed accordingly and not to the Sigma Repository. All submitting authors or publishers have affirmed that when using material in their work where they do not own copyright, they have obtained permission of the copyright holder prior to submission and the rights holder has been acknowledged as necessary.

Review Type

Abstract Review Only: Reviewed by Event Host

Acquisition

Proxy-submission

Date of Issue

2026-07-19

Click above link to access the slide deck.

Share

COinS
 

From Burden to Support: Validating an AI-Based Dementia Care Voice Assistant App

Toronto, Ontario, Canada

The prevalence of behavioral symptoms, such as wandering and agitation, in over 90% of persons with Alzheimer's disease and related dementias (Pw-ADRD), with approximately 83% residing in home-based settings [1-3], underscores the urgent need for effective interventions to support informal caregivers. These caregivers are essential for maintaining quality care for Pw-ADRD, yet they frequently encounter substantial challenges in managing behavioral symptoms [4], revealing a significant gap in available resources and support systems.

To address this gap, we developed a prototype for an AI-based Dementia Care Voice Assistant (AI-DECAVA), powered by a large language model (LLM). The system is designed to aid informal caregivers in managing behavioral symptoms associated with ADRD. However, general-purpose LLMs are often inadequate for closed-domain question answering (QA) tasks, as they lack the specialized knowledge required for nuanced subdomains. Broadly pre-trained models typically do not capture the domain-specific terminology and contextual precision needed in healthcare applications—a limitation compounded by sparse conversational data and computational constraints in specialized fields.

Our methodological approach employs a Retrieval-Augmented Generation (RAG) framework, which combines targeted data retrieval with fine-tuning using domain-specific question-answer pairs. A core component of this process is Question-Answer Generation (QAG), used to create tailored training datasets. This strategy enables fine-tuned models to assimilate specialized terminology and contextual knowledge while remaining computationally tractable for domain-specific deployment. A principal advantage of this framework is its replicability across diverse domains and scalability for integration into varied user groups, positioning it as a versatile and generalizable solution for enhancing conversational assistants in specialized care contexts. Initial validation of AI-DECAVA was conducted with five informal caregivers, demonstrating usability and functional efficacy.

These preliminary results provide critical insights to inform the design of a larger feasibility study and serve as proof-of-concept for the application's potential to strengthen caregiver support and improve dementia care outcomes. This innovation contributes to nursing science by empowering informal caregivers to deliver quality dementia care to persons living with ADRD.