Can A Small Sensor Protect The Elderly From Dangerous Falls? A CMU Research Team Thinks So

Jun 15, 2018

Baptist Homes is a sprawling senior care campus, located on a lush, quiet hillside in Mt. Lebanon. About 50 people live in the personal care part of the facility, where they can get help with everyday tasks. 

Administrator Janet Wasko said falls are a big problem--about five patients fall on a regular basis.

"They may slide off the chair, they may slide off the bed, their shoes may get caught on the carpet," she said. 

When a patient falls, they undergo a physical exam to make sure they aren't concussed or injured. The nursing staff does an assessment of the room to see if things like rugs or clutter can be rearranged to make things safer. 

"And even after all of that, some of these residents just fall for whatever reason," Wasko said.

It’s estimated that almost one-third of adults over 65 years old fall every year, resulting in injuries that ultimately diminish that person’s independence. Jane Cauley, an epidemiologist at the University of Pittsburgh, said a recent study from the American Geriatrics Society found falls cost the healthcare system about $50 billion in 2015. The price tag, she said, hurt both the health care system and the patient's quality of life.    A sensor in development aims to prevent falls before they happen. 

Imagine the scene in Jurassic Park, where the dinosaur’s massive footsteps cause ripples to appear in a cup of water. Human footsteps create vibrations as well.

“So when people walk around we sense the vibration caused by each footstep, to identify and localize and get different gait patterns out of it,” said Hae Young Noh, a researcher at Carnegie Mellon University who is developing the sensor. 

Noh said walking patterns are a bit like fingerprints, because they’re different for each person.  When someone is about to fall, their walking pattern changes; they might take two quick steps in a row.

For a year and a half, Noh’s team measured the unique walking patterns of patients at Baptist Homes. 

“If we can look at the pattern in the past and predict what’s going to happen, then you can prevent before it happens,” she said.

Noh said these small gadgets can be attached to the floor or a piece of furniture, so it’s also possible to tell when somebody is trying to get out of bed, which in some facilities requires caregiver assistance.

“But late at night, or sometimes you don’t want to bother other people, so they’re like ‘Oh, I’ll just get out of the bed on my own and just drop by the bathroom quickly,' and then they try to get out without help and that’s when they may fall,” Noh said.

If a sensor picks up irregular movement, it can alert the appropriate person, such as a nurse, with a message on their phone or computer that says a patient could be at risk of a fall.

Baptist Homes’ Wasko said she’s thrilled by the idea that the sensors could be deployed in hospitals. That way, when a new resident arrives at the facility, the staff can measure their normal gait pattern, which would help them identify the abnormalities that precede a fall. And, she said, the population of people most at risk for falls is growing every year.

“Especially with the baby boomers coming up, we’re all looking for technology," Wasko said. "What can we get to help us in our daily living?”

Noh’s team is still fine-tuning the sensor and the algorithm to analyze the data it gathers. Eventually, Noh said she’d like to see the sensors deployed not only in hospitals and nursing homes, but also in private residences, so people can be alerted if their loved one is at risk of a fall.