Home-based dual-task gait assessment for older adults at risk of dementia

The National Institute on Aging awarded Hebrew Rehabilitation Center $318,000 on August 18, 2026, under the Aging Research program (CFDA 93.866) to conduct a pilot study of home-based dual-task gait assessment in older adults at risk of dementia.

The project will recruit 60 ambulatory adults aged 65 and older—30 with mild cognitive impairment and 30 age- and sex-matched controls—to use a smartphone-based gait application for daily assessments in their homes over a two-week baseline period, followed by weekly assessments for one year. Participants will complete comprehensive lab-based functional assessments evaluating health history, cognitive function, mobility, and factors influencing gait, and will report falls and injuries via the application. Follow-up phone interviews will occur when falls are reported. The study addresses the elevated fall risk in older adults with mild cognitive impairment, which accelerates cognitive decline and increases risk of Alzheimer's disease and related dementias. Existing clinical gait assessments conducted annually in clinical settings fail to capture real-time changes signaling acute increases in fall likelihood. Two specific aims will quantify day-to-day variation in normal and dual-task gait performance (walking while performing verbalized serial subtractions) during the baseline period and then track gait performance changes over the following year to identify patterns associated with acute fall risk.

Work is performed in Roslindale, Massachusetts. The period of performance runs from August 18, 2026, through July 31, 2028.

Abstract

Older adults with mild cognitive impairment (MCI) are twice as likely to fall as compared to age-matched controls. Falls in this vulnerable population not only disrupt daily activities, but also accelerate cognitive decline and thus increase the risk of alzheimer's disease and related dementias (ad/ADRD) in later life. Unfortunately, existing clinical assessments of fall risk, including gait assessments performed annually within clinical settings, are too infrequent to capture changes in gait over time that may signal an acute increase in the likelihood of suffering a fall in the near future. The proposed research has been designed to, for the first time, study changes in gait associated with an acute increase in the risk of falling in older adults with and without MCI. This pilot study will leverage a validated smartphone-based "gait APP" that enables the user to regularly complete brief, safe assessments of gait in the comfort of their home. We will recruit 60 ambulatory men and women aged 65 years and older: 30 with MCI and 30 age- and SEX-matched controls. Each participant will undergo a comprehensive lab-based functional assessment to evaluate health history, cognitive function, mobility, and other factors that influence gait. They will then use the gait APP to complete daily gait assessments in their homes over a two-week baseline period, followed by weekly assessments for one year. Participants will also report falls and injuries via the gait APP and follow-up phone interviews will be conducted in the event of a fall. These data will be used to complete two specific aims. Aim 1 will quantify the expected day-to-day variation in performance within self-administered tests of normal and "dual-task" gait (i.e., walking while completing verbalized serial subtractions) over a two-week baseline. With this knowledge in hand, aim 2 will then track gait performance once per week for one year, together with the incidence of falls, to obtain preliminary evidence of changes in gait preceding falls. Our primary hypotheses are that H1) older adults with MCI, compared to controls, will exhibit elevated day-to-day variation in dual-task gait performance over a two-week baseline period, and that H2) regular self-administered home gait assessments-particularly under dual task conditions-are capable of identifying changes in gait performance associated with an acute increase in the risk of falls. Together, these discoveries will provide preliminary insights into whether changes in home-based gait performance are associated with an increased risk of falling in older adults with MCI. The proposed research will directly inform a larger R01 study designed to use smartphone technology to develop predictive models that utilize baseline data and regular, prescribed gait assessments thereafter to identify when an individual's risk of falling acutely increases, so that fall prevention strategies can be implemented in more timely fashion.