Change agents: Will virtual nurses and pocket health coaches improve patient engagement?
BOSTON - Healthcare organizations have a lot on their plates nowadays, and the challenges are only growing. Meaningful use is a big one, of course - not least Stage 2's new focus on patient engagement.
Another is a lack of new physicians, with the Association of American Medical Colleges projecting a shortage of some 150,000 docs in the coming years - even as ACA is set to add millions of newly-insured patients to the rolls. Still another is low health literacy, especially on the part of many of those same new patients.
Could virtual reality be the answer? Timothy Bickmore, associate professor at Northeastern University's College of Computer and Information Science, thinks so.
Bickmore works with relational agents, which he describes as "computational artifacts designed to build and maintain long-term, social-emotional relationships with their users."
Avatars like them, sometimes called "chatbots," have found footholds in the airline industry, at IKEA, at AT&T and beyond, offering help in lieu of actual humans. Health insurers, such as Aetna - whose online "Ann" avatar, unveiled in 2010, talks to plan members and takes their questions as she helps them navigate their benefits on the site - use them. But these intelligent virtual assistants (IVAs) could soon be poised for a much more significant clinical role.
Gartner has predicted that within the next eight years perhaps 85 percent of interactions between customers and companies could take place without a human on the other end. Healthcare could soon be headed for a similar shift, according to Tim Bickmore, associate professor in the College of Computer and Information Science at Northeastern University.
Working toward his Ph.D at the MIT Media Lab years ago, Bickmore's research included the study of "face-to-face conversation between people as a kind of user interface." That meant "studying hand gestures, facial displays of emotion, body posture shifts and head nods, and how these are used to convey information in face-to-face conversation."
His dissertation showed how that type of virtual interface could be brought to bear on healthcare counseling: "studying how doctors talk to patients, how nurses talk to patients and how we can take best practices from face-to-face counseling and build that into automated systems for educating patients and doing longitudinal health behavior change interventions."
Now he develops these virtual agents, which he says are well suited for healthcare, especially in areas around education and behavior change. He does so after real-life research of provider-patient interactions.
"We'll pick a particular area of healthcare - say, patient education at point of discharge and medication adherence - and we'll go with our video cameras and tape providers talking to patients, come up with models of dialogue and the language they use, both for the therapeutic part, and how they build trust and rapport and therapeutic alliance over time," says Bickmore.
He and his colleagues have been working in this area for a decade or so, but "the past few years have seen the technology getting more sophisticated," with regard to the kinds of tools and dialogue structures available, says Bickmore. "We can build systems more quickly, bring our architectures to new sorts of health problems more quickly."
One thing that's important in order to get patients to want to engage with the technology is to keep the facial design expressive, but very simple - "animated characters you have a conversation with," he says.
"We don't try to push into something that's very anthropomorphic or very human looking, explicitly because we want to avoid the uncanny valley," says Bickmore. "We find that the simpler figures are just as effective. And they work well.
Growing relationships
So far, the relational agents have shown promise when it comes to education.
"Everyone knows how to do face-to-face conversation, so we do a lot of our interventions with safety net hospitals and populations that don't have high levels of computer and health literacy," he says. "We find that they not only find these conversations easy to use but they actually are happier with them. They give us higher levels of satisfaction than other patients."
Their promise for longer-term patient engagement, "where you're doing a longitudinal intervention where you want patients to keep coming back, [or] a counseling session where you want people to stay on their medications or stick with their appointments," also seems promising. "Having a person that somebody builds a relationship with is a way of achieving that," says Bickmore.
Another argument in their favor is that patients have shown to be more honest with virtual clinicians, were more willing to accurately report their symptoms and habits, and some reportedly more at ease because they didn't feel rushed or condescended to, according to Bickmore's research.
Sculpting behavior
Max Celko, a researcher of the Hybrid Reality Institute, a research group that focuses on "human-technology co-evolution," wrote in a recent white paper that, "Interacting with "humanized" technology in a therapeutic and coaching context will turn mobile devices into what he called "identity accessories" - tools that "actively sculpt our behavior and identity" and thus will "heighten our emotional attachment to technology."
Victor Morrison, vice president of sales, healthcare markets, at Spokane, Wash.-based Next IT - which, in addition to creating IVAs for the airline industry, also developed Aetna's "Ann" avatar - sees a bright future for healthcare applications of the technology.
But he also sees a need for capabilities that are much more clinically rigorous than they are now if IVAs are really going to make a marked difference. They'll need to collect patient data, engage those patients via mobile devices to spur appropriate interventions, interpret EHR input and remote monitoring device output, collect data and deliver it to EHRs, Morrison says.
The IVAs Next IT is working on now are more than conversational interfaces build on a natural language model. They "can figure out customer intent, what that customer is looking for, and then access the correct database," he says. "The technology is actually very advanced."
Still, they're poised for even more promise. With a 150,000 doctors shortfall, "there's just not enough physicians around to do what needs to be done, and move the needle on compliance, adherence and ultimately outcomes."
Of course, an animated avatar - no matter how artificially intelligent, will never replace a physician. But they can help ease some of the burden when it comes to patient engagement, populating personal health records, medication adherence and more. And "if it's done right, it can do it in a very personalized way," says Morrison.
"If you don't find a sophisticated way to extend the reach, we're in big trouble in healthcare," he adds.
It may seem a ways off, but things change fast. Natural language technology has been around since the 1950s, and look at all the advancements that have been made in recent years. No one had heard of IBM's Watson or Apple's Siri three years ago - now they're everywhere.
"We're at a point, when you look at some of the [challenges] in healthcare, and the technology has come along far enough that you can start thinking about ways to deploying it in a way that creates efficiencies," says Morrison. "I think you're going to see this accelerate."