The senior living industry in general has been criticized for its lack of tech savvy and ability to implement solutions that are widely available in other sectors. But some tech providers are implementing new solutions that aim to remove that stigma completely.
One of those technologies that has emerged in several forms industry-wide is monitoring devices: wearable devices and remote devices, and among them devices that cannot even be sensed or detected by the subject which they are monitoring.
Taken a step further, these devices can provide artificial intelligence in an effort to improve the quality of care senior living organizations are able to provide. Whether addressing fall reductions, a lower incidence of heart attacks or fewer infections among residents, this intelligence—which empowers machines to copy intelligent human behavior—can be used in more ways than one, says one provider of such devices.
“Care is emerging across sensor-based technology in association with artificial intelligence to allow us to predict and prevent negative health events,” says Andy Belval, founder and CEO of Keystone Technologies.
A startup in existence for about 15 months, Keystone offers among its solutions a monitoring-based artificial intelligence platform, which it says is changing the nature of care itself.
Here are 4 ways artificial intelligence is doing this through remote monitoring devices.
Predict and prevent negative health events
The Keystone model relies on six sensors: four motion sensors, one bed sensor and one depth sensor. Contrary to wearable devices that have risen in popularity among the senior population, the sensors are placed in the senior’s apartment and collect data passively rather than having to be worn on the person. The bed sensor alone comprises 1,700 measuring points sending data points 50 times per second. By proactively measuring the activities of a particular resident, this artificial intelligence learns over time and can both recognize when there are changes, and can anticipate when there will be changes. It is designed to then provide alerts that will allow providers to prevent negative health events such as falls, pressure ulcers and heart attacks.
For example, say a resident bumps his knee on a chair. As a result of the bump, the resident’s gait changes.
“This resident went from low risk to high risk for a fall,” Belval says. “The provider can then go back to the resident and find out about the knee.”
The time it takes for the system to learn the behaviors of a particular resident: 10 days.
“It can learn the difference between a husband and wife in the same room, it can see when grandchildren showed up,” Belval says.
Eliminating negative health events means less time in hospitals and other higher acuity settings.
In a pilot program launched by Keystone at TigerPlace Independent Living in Columbia, Missouri in 2010, the company’s second generation technology was implemented for 52 residents versus a control group of 81 residents without the solution.
The average length of stay was 1.72 years longer for seniors living with the solution.
While an exact measure of return-on-investment is difficult to estimate, the length of stay increase can translate directly into revenue as well as savings, Keystone says, by keeping residents out of hospitals.
“We will see the opportunity to change the way the ACO model works,” Belval says. “You can have a person in the hospital with the sensors in place and can keep measuring the same results … from SNF to home. Not only can we keep people out of the hospital, but we can prove that the care works.”
The flexibility to provide real time information through non-human intervention and tracking what’s going on will revolutionize care, and will give senior living providers a place in the ACO conversation, he says.
Eliminate false positives
“There are other sensors out there that tell you they alert you to falls, and they do,” Belval says. “But a concern is false positives.”
With a system based on artificial intelligence, there are no false positives, he says. What providers see based on the data gathered is statistically significant, and creates a situation where care providers don’t need to enter rooms as frequently to check for false results.
“False positives result in being ignored by care providers,” he says. “That can be worse than not having system in place at all.”
Written by Elizabeth Ecker