This article is a part of the ‘Technology, transformation and health care” series created in partnership with AMS Healthcare. These solutions-focused articles will focus on emerging technologies and their potential for transformational change in our health-care system.
Throughout his nearly 30-year career working with seniors, physician and researcher Frank Knoefel says he’s been concerned by an increasingly common problem – how to safely discharge his elderly patients back home.
“The patients that were typically admitted to my unit usually came in with a hip fracture,” Knoefel says. “After the hip was repaired in an acute care hospital, they still couldn’t just get up and walk home” since many also had other medical conditions such as cognitive and memory changes.
As of 2020, there were an estimated 597,000 people living with dementia in Canada, a number that’s expected to double by 2031. Night-time wandering is a common symptom of dementia, as well as one of the major predictors of caregiver burnout and subsequent institutionalization.
To make matters more challenging, many elderly patients don’t have family to take care of them living nearby.
“We need to discharge them home, but is this safe? And how long is it going to last? I can’t possibly follow up with everyone that I discharge just to make sure they’re doing OK. The resources just aren’t there,” says Knoefel.
When he first began searching for answers in the early 2000s while working in geriatric rehabilitation, Knoefel was inspired by the advanced technology his engineering friends were using in their winter vacation homes.
“They’d have a cottage where they’d go skiing and could control the heating from their cell phones,” meaning they could do things like warm up their cabins remotely before they arrived, he says. “That’s lovely, but it has no social value whatsoever.
“But I thought, how can we use this technology and do social good? Can we apply it to frail older adults that want to go back into the community when it might not normally be very safe?”
Fifteen years later, Knoefel is now chair of the University of Ottawa Brain and Mind – Bruyère Research Institute in Primary Health Care Dementia Research. He has co-authored several studies on how greatly improved, commercially available smart-home technology can be used to support aging populations.
“How can we use this technology and do social good?”
In 2020, one such study done in collaboration with researcher, Rafik Goubran, the Dementia Society of Ottawa and the Champlain Local Health Integration Network, used commercially available equipment to design supportive smart homes for people living with dementia. “We literally went to Home Depot and (then) ordered a couple things off Amazon. We put together a little sensor system that works really well for the person with dementia, but even more for the person caring for the person with dementia.”
To understand how the system works, Knoefel gives the example of a person living with dementia named “Colin.” This particular system starts with a pressure sensor under Colin’s mattress. When Colin gets out of bed, the pressure sensor sends a message to the cloud-computing system, which then turns on the smart lights lining the hall, illuminating the way to the washroom. Once Colin is safely back in bed, the pressure sensor tells the cloud to turn the lights back off.
But if Colin wanders too far and ends up in the kitchen, the motion sensors send a signal to a smart speaker that then plays a pre-recorded message of Colin’s wife, telling him that it’s the middle of the night and he should come back to bed.
Knoefel says the system is largely designed with the caregiver in mind. “What we’re doing is we’re helping the person with dementia find their way around the house. At the same time, we’re giving the caregiver a chance to sleep through the night.” Should something dangerous happen, for example, a loved one wandering away from home in the middle of the night, an alarm would sound. “But only then do we wake the caregiver,” Knoefel says.
Designs like these were outfitted in homes around Ottawa and monitored for 12 weeks. The study found that caregiving participants reported lower levels of anxiety and depression. The project was such a success that once it was completed, Knoefel says a number of the participants asked to keep the equipment.
Once the team knew the smart-home setups could be used effectively in private homes, the question then became whether these designs could be scaled and adapted for other care settings.
“We installed the system in all 12 rooms in a retirement home. Once we wired up all the rooms, we gave a cell phone to the night staff to give them updates. For example, Mrs. Jones has just left her room or Mrs. Smith is in the bathroom. We can customize it so that it only gives you important information.”
Richmond Care Home was outfitted for the study nearly two years ago, and Knoefel says it is still using this system today.
His research team has since handed over the personalization and installation of the smart-home systems to a start-up called Esprit-ai. The company manufactures its own sensors and does installations for a mix of private homes and private for-profit retirement homes and has dozens of these systems running in real time around Ottawa.
The next major step was installing smart-home systems in hospitals where, Knoefel says, getting new technology approved can be particularly complex.
“The challenge was that we needed the top-down approach to get permission and to get the infrastructure to allow this to happen. But we still wanted it to be bottom up so we could brainstorm with nighttime bedside nurses as to what patient information was most important to them.”
As a first step, the research team installed the devices at the Greystone Transitional Care Unit in Bruyère, an alternative level of care (ALC) facility serving the acute care hospitals of Ottawa. Two weeks after installing the equipment, the team followed up with health-care providers to see how the system was running.
“[The nurses then relayed] ‘Oh, this was crazy. We got 12 alarms a night for five nights in a row; I was ready to throw the phone out the window,’” Knoefel explains. “This is a real problem in health care. If you jump every time something is slightly out of the ordinary, you’re going to be jumping all the time. But the question is: What’s the right kind and amount of information? What do you really need to know to do your job?”
The team then tailored the amount of data to balance what information is critical for which patient, but without overwhelming nurses with alerts. For example, if a patient is at high risk of falling, it would be valuable for the nurses to know as soon as that patient leaves their bed. But for someone who is more mobile, nurses may only want to receive an alert if that patient leaves their room.
“Everyone always starts out thinking they want to know everything,” Knoefel says. But tailoring systems to only deliver what’s essential is just one of the critical issues facing mass implementation.
Knoefel co-authored an evidence to impact report with Bruce Wallace that highlights several of the issues facing supportive smart homes. Some are seemingly straightforward kinks to iron out: How often batteries may need to be changed; the sophistication of the technology; making devices last longer; and improving the quality of both the sensors and cloud analytics. But Knoefel adds that there are a number of sticky issues that also will have to be resolved.
“We need to sort out the ethics of people’s home data getting thrown into the internet. We know that hackers can hack, but do we want people hacking into your home and turning on the stove when you’re not there? There are real security issues to consider.”
Knoefel says there are ethical issues to contend with as well, such as questions around who can consent to the devices being installed; whether children should be able to consent even if their parent isn’t on board or doesn’t have capacity to consent; how to protect and filter what kind of information gets relayed and to whom when there’s, for example, video monitoring in someone’s bedroom?
“If this is really going to make a difference in health care, then how do we get the government to pay for it?”
There’s also the issue of who should pay. Knoefel says that for the study, his team was able to outfit homes for about $2,000 for the equipment, then the engineers would program and personalize the systems. With private companies loaning the equipment and providing this service on a subscription basis, the question of accessibility and whether or not this should be billed to the public payer system is a legitimate one.
“If this is really going to make a difference in health care, then how do we get the government to pay for it?” Knoefel says, “At the end of the day, it’s the government who’s saving money on labour. But right now, it’s on the backs of family members who are caring for their loved ones.”
Another kink to iron out is that in Canada, the classification used for “health and medical” devices is different legally than what would be used for something deemed a “wellness” device. Knoefel says that figuring out that line needs to be clear since registration and compliance as a medical device is more expensive and carries a higher standard for accuracy.
And that in itself raises another issue: liability.
“How you define what you’re doing is going to be very relevant here.” says Knoefel, “Neither Google nor Amazon right now wants the liability of somebody saying, ‘My mum was on the floor for two days and you never notified me.’ They don’t want that risk.”
Knoefel analyzes these issues in a new book he’s co-authored on supportive smart homes, which he says he expects to be published in April 2023.
When Knoefel first began this work, he says the health-care system didn’t have adequate resources to take care of older adults but with “boomers coming at us like a steamroller,” solutions and innovations are even more desperately needed today.
Though there is still “decades worth of work to do” on supportive smart homes, Knoefel and his team are off to a very promising start.
“People get that this is not only an opportunity, but a necessity. There’s no way we’re going to be able to hire enough nurses and doctors to support the elderly, particularly if they want to age in place. We’re going to need technology to help bridge that gap.”