Overview
The aging of the U.S. population, combined with health care workforce shortages and insufficient access to effective fall prevention and exercise programs, has contributed to a high burden of fall-related injuries in older adults, especially in those with Mild Cognitive Impairment (MCI), a precursor of Alzheimer's dementia (AD) and related dementias (ADRD). There is an urgent need for scalable, evidence-based interventions that can be delivered efficiently in the home setting. This study aims to evaluate the efficacy and impact of a comprehensive digital fall prevention program (Digital STEADI) delivered through a technology platform with two conversational agents that leverage generative artificial intelligence (gen-AI) with human oversight to increase program access, enhance adherence to clinical fall prevention protocols and best practices, and reduce fall risk among community-dwelling older adults.
Description
The Centers for Disease Control and Prevention's (CDC) Stopping Elderly Accidents, Deaths \& Injuries (STEADI) Initiative integrates fall risk screening, assessment, and intervention according to established clinical guidelines from the American and British Geriatrics Societies. Although STEADI is a proven multifactorial intervention, traditional dissemination models relying on in-person visits, paper-based materials, and intensive human monitoring and care coordination present significant barriers to widespread implementation for impact at the population level.
The sponsor in collaboration with clinical and academic partners, will test a digital STEADI intervention program incorporating an evidence-based exercise program (Otago, OEP) to provide strength and balance training and continuous coaching. The digital program is implemented via two conversational agents, a tablet-based avatar or mobile chat via SMS, supported by gen-AI and trained human staff who supervise the agents as needed. Both agents provide 24x7 companionship with protocolized health coaching and escalation to care providers as needed. The technology has been rigorously validated in cognitively normal (CN) as well as frail older adults with multiple chronic conditions including serious mental illness (SMI), MCI and ADRD. It is optimized for ongoing fall risk monitoring, personalized exercise prescription, and coordinated care. The full digital STEADI program was successfully tested for feasibility and safety in two pilot studies with older adults with and without MCI, one in a simulated home environment and another in private homes of community-dwelling older adults. Upon the successful completion of the randomized control trial (RCT), the Digital STEADI program will be prepared for national implementation and commercialization.
A randomized control trial (RCT, N= 300) will be conducted to evaluate the efficacy of the digital STEADI program including OEP compared to standard STEADI (STEADI-Standard) and attentional control (STEADI-Control) conditions. A subset of participants with MCI will be assigned to the STEADI-Avatar group to evaluate the digital program in this high-risk population. All participants will be community-dwelling older adults aged ≥50 years at risk for falls.
Eligibility
Inclusion Criteria:
- Montreal Cognitive Assessment (MoCA) score between 26 and 30
- Able to walk independently with or without an assistive device.
- Able to rise independently from a standard desk or dining room chair without arms.
- Live independently in a non-institutional or senior community setting.
- Able to speak, hear and understand English
- Have 20/40 vision from 10 ft with or without correction (e.g., glasses)
- Willing to participate in the requirements of the research program.
- Able to meet in a physical study location on Emory campus.
- Screen positive for fall risk per the Stay Independent screener.
Exclusion Criteria:
- acute medical illness requiring hospitalization.
- hospitalized in the past 1 month.
- diagnosed with dementia (ADRD) or a neurodegenerative disease other than MCI (e.g., Parkinson's, multiple sclerosis).
- uncontrolled congestive heart failure (requiring hospitalization in the past year and/or unable to the equivalent of one city block).
- history of hospitalization for stroke in the past 1 year.
- inability to perform study procedures.
