Cutting-edge technology could help solve one of the most difficult things about dementia
An estimated 5.3 million Americans have Alzheimer's disease, the most common form of dementia, and that number is expected to double by 2050, according to the Centers for Disease Control and Prevention.
We'll have to care for all those people somehow, but dementia is a challenging problem. One of the most difficult things about dementia is that patients often require round-the-clock care on top of frequent appointments with specialists. Technology may help meet these growing needs.
In particular, the ability to connect basically any object to the internet so it can send and receive data, the premise of the so-called Internet of Things, could help provide high-quality care to a wider number of patients - especially those living at home.
A group of researchers at the University of California San Francisco Memory and Aging Center has developed something they call a "care ecosystem," which uses the Internet of Things and machine learning, a type of artificial intelligence, as tools to provide care that's "more continuous and proactive and personalized" than what people with dementia commonly get today, Dr. Katherine Possin, one of the study directors, told Tech Insider.
The technology that powers the "care ecosystem" has been developed, and the researchers are now enrolling patients in a clinical trial to try it out.
The hope is that the tech savvy approach will not only improve the quality of life of people with dementia and their caregivers, but also reduce the amount of money spent on their health care.
What the technology does
The care ecosystem for each patient will be overseen by a "care team navigator" who checks in with the patient and their family by phone at least once a month and whom the family can call at any time as problems come up. The care team navigator uses an internet dashboard to keep track of how patients are doing and what their particular needs are.
The Internet of Things comes in with activity sensors that are placed strategically throughout a patient's home. Additional information about a patient's movements and location are collected from a smartwatch or smartphone.
As it's collected, all the data will be fed to a machine learning algorithm, which will first establish a baseline of what a patient's normal activity looks like, then alert the patient's care team navigator if their activity looks abnormal. The algorithm can pick up sudden changes (like if the patient doesn't get up at their normal time one morning or strays outside a certain geographical area) as well as gradual changes (if the patient gets less active over a couple weeks).
The sensors gather information about how the person with dementia is using appliances (if they left the stove on, when they open the refrigerator, etc.), how often they flush the toilet (going to the bathroom a lot can be a sign of a urinary tract infection), if they are leaving doors ajar, or any other activity marker that's relevant for a particular patient. Sensors could also track when patients open pillboxes to take medicine, but this isn't currently part of the care ecosystem, Possin says.
When alerted by the machine learning algorithm that something about a patient's movement or activity is off, the care team navigator can check in with the patient and their family to see what's going on and if anything needs to be done.
A company called Lively is already selling a similar product. Lively's system includes a watch with a pedometer and activity sensors like those in the UCSF system which sends data to a web page or app a caregiver can check. That way, a caregiver who doesn't live with the person they're caring for can still see if there's a concerning change in activity.
In the UCSF care ecosystem, the information gathered from sensors and analyzed through machine learning is integrated with the patient's medical and care history to provide personalized care.
The goal
In designing the care ecosystem, the scientists wanted to "use innovative technologies to help make the system very efficient but at the same time personalized," Possin says.
The idea is that personal attention from the care team navigator who can connect patients and their families to resources, as well as social workers, pharmacists, and nurses who specialize in dementia care, can provide high quality care to people who don't live close to specialty care centers.
A problem with current dementia care is that when patients don't see a doctor frequently, problems that start small can come up between visits. If they're left untreated, those small problems can snowball into emergencies, Possin says. But when patients and their families can talk to their care team navigator about concerns at any time, they can address issues early and avoid emergency room visits and spending time in the hospital.
The team behind the care ecosystem also hopes it will keep patients out of nursing homes longer, all of which would reduce health care costs (In 2010, Americans spent more money on dementia care than on health care for either heart disease or cancer).
The care ecosystem is being tested in a clinical trial to see how it affects patient quality of life, caregiver burden and depression, how much additional health care patients use, and how much is spent on that health care. If it proves effective and scalable, millions of Americans and their families could have access to good care that's not available to them now.