Frailty in late life is a public health challenge which contributes to increased risk of multiple adverse outcomes, including falls. Based on recent estimates, 10% of community-dwelling older adults are frail and another 41.6% are pre-frail. Detection of frailty at the pre-frail stage paves the way towards alleviating the associated risks by employing intervention strategies. Additionally, since frail older adults have a high risk of falling, detection of frailty can help prediction of prospective falls, which in turn serve as a basis for fall prevention efforts. However, the ability to establish a diagnosis of frailty and prediction of future falls remains a major challenge in the geriatric population.

The aim of this project is to develop a wearable sensor-based platform for identification of frailty status and prediction of future falls in older adults. This wrist-worn, smartwatch-based platform will be used to automatically calculate measures of functional performance including walking speed, power of sit-to-stand transitions and activities of daily living. Using these measures, models will be developed for the target population to classify their frailty status into non-frail, pre-frail and frail and predict their risk of falling in the future.

The outcome will be used in identifying the transition from non-frail to pre-frail stage, which is increasingly recognized as the stage to target with different therapeutic modalities. The Canadian healthcare system is facing a serious challenge in the form of an aging population, prevalence of chronic diseases and the accompanying rising costs. In response to these challenges, the proposed novel technology, capable of identifying frailty status of older adults at the pre-frail stage and predicting prospective falls, could improve the quality of care for older adults while reducing the cost of care through early detection/intervention.