Preventable falls are a leading cause of death among nursing home and senior living patients, so using new technologies to predict and prevent them is crucial. One new tool uses a smartwatch to detect signs of falls as much as 32 days before they happen.
The new tool is being developed in the UK with Age NI, a charity based in Northern Ireland that advocates for services and mobility aids for people 65 and older. It claims to detect signs of falls up to 32 days before they occur by using a smartwatch to monitor activity levels, sleep, heart rate and oxygen saturation. It’s currently being tested on people 65 and over in Northern Ireland.
Patients wear a smartwatch linked to a monitoring platform that includes benchmarks for each person’s normal activity patterns and movement levels, heart rate and amount of sleep each day. Using predictive analytics, the smartwatch’s software detects changes to forecast an increased risk of falling over the 32-day period. Alerts are then sent to caregivers so that they can check on the patient, offer assistance or change care regimens.
“It is not just someone wearing a watch and relying on the technology,” Vicki Caddy, head of communications at Age NI, told Interesting Engineering. “There is very much a human interface behind this as well.”
Other fall prevention technologies are being tested that could have an impact on senior living providers in the US. In one new partnership, Lilli, a UK-based machine learning company for the home care industry, teamed up with digital care management provider Nourish Care to help caregivers implement fall prevention services. Using machine learning, the sensor-based monitoring platforms and apps help senior living caregivers keep residents safe.
Additionally, an Apple Watch app, BoundaryCare, lets caregivers improve resident safety by helping them detect potential falls and wandering behaviors through geo-fencing. Another device, Cherish Serenity, uses AI radar technology to detect biometrics, body movements, falls and other safety risks through sensors.
