The United States, in particular, and the world, in general, faces a healthcare crisis of enormous proportions. The cost of care is already rising, but the combination of an aging population and changes to diet and lifestyle leaves a growing percentage of the population susceptible to chronic health conditions, such as heart disease and diabetes, which further drive healthcare costs. By 2020, the United States is expected to spend $4 trillion annually on healthcare —more than the GDP of all but a handful of nations.
Healthcare, then, is ripe for innovation. Much comes from academia, which lets data scientists, economists and medical professionals collaborate more freely than they might in a corporate setting, but new ideas do come from the private sector as well.
Feature: 13 Healthcare IT Trends and Predictions for 2013
Commentary: Why Tech Entrepreneurs Will Transform Healthcare
The Future of Health and Wellness Conference held earlier this month at the Massachusetts Institute of Technology highlighted the results of some of these collaborations. Collectively, they won’t address the 30 percent of wasted healthcare spending—some $800 billion per year—that the Obama administration hopes healthcare reform can eliminate, but they do demonstrate progress in understanding how patients age, cope with stress, change our behavior and interpret the information that doctors give them.
1. Reality Mining: Using Data to Influence Healthy Behavior
Using smartphones to collect information about what people are doing and how they are behaving, which Alex “Sandy” Pentland, director of the MIT Human Dynamics Laboratory, describes as “passive monitoring from the things you carry around every day,” results in a data set that’s “hugely richer than anything you’ve ever seen before.” It’s an extension of data mining known as reality mining, and its predictive capabilities seem to know few limits.
Gartner: Big Data, EHR Driving Healthcare IT Innovation
Related: 12 Tips to Prevent a Healthcare Data Breach
For security purposes, information is shared using an “answer architecture” that makes yes-or-no queries of the open personal data store (openPDS) on a user’s smartphone, Pentland says, much like the SWIFT platform banks use to exchange information. In this manner, and in accordance with the U.S. Consumer Privacy Bill of Rights and the European Union Directive on Privacy and Electronic Communications, the data belongs to the individual.
Societal patterns and habits inherent in these data sets can predict behavior, such as the likelihood of residents of a certain neighborhood developing diabetes or alcoholism. (Predicting behavior from verbal and visual cues, it turns out, is rather easy; technology that Pentland and his team have developed is used by two large health insurers to screen callers for signs of depression.)
However, if exposure to external forces drives behavior changes, Pentland says, then getting to the root of the problem means changing exposure. Through its research, Pentland’s lab reports that social influence—knowing that others are being rewarded for good behavior such as riding a bicycle to the office—is more than three times as effective as simply receiving that reward on an individual basis.
This influence has helped veterans coping with post-traumatic stress disorder, who see that fellow veterans are more active and social and decide to do something about it, but further uses within the highly individualized U.S. healthcare system are only emerging slowly.
2. Social Networking: For Best Results, Group Like-Minded People
An obvious source of social influence, of course, is social networking, the rapid rise of which has demonstrated the ability of anonymous peer-to-peer interaction to influence behavior. This occurs because social networks forge what Damon Centola, assistant professor of behavior and policy science at MIT, calls “weak ties” that, despite their moniker, actually increase the diffusion rate of behaviors and ideas. You may not have had a friend living in the Middle East during the Arab Spring, for example, but you likely had a friend of a friend. Put another way, it’s a small world after all.
News: Johns Hopkins Researchers Use Twitter to Track Flu Outbreak
More News: 5 Online Tools to Track the Flu
Centola says that changing behaviors as a result of social network influence doesn’t work so simply, though. The reason is the difference between what are called simple and complex contagions. It only takes a single contact to tell you that smoking is bad for you (simple contagion), but it will take multiple contacts to get you to quit smoking (complex contagion), Centola says. Information can spread among the weak ties of a randomized network, but when it comes to behavior change, it’s only a matter of time before there are no longer any ties that bind.
The answer, Centola found by leveraging the infrastructure of MIT’s online fitness program, is to connect users to “neighbors” based on the concept of homophily, meaning “love of the same.”
Users of the online fitness site were more likely to adopt new functionality—in this case, a calorie-and-exercise tracker that offered a real-time look at a user’s weight—if fellow users of the same gender or similar age and body mass index were also using it.
Policy implications of these findings, says Centola, who is helping the site PatientsLikeMe analyze data on its 130,000 users, who connect with people with similar medical conditions, include providing incentives for social interactions, conducting sentiment analysis of the information that’s shared to revise future messages and repeatedly messaging the people you want to join.
The last point stems from another experiment Centola conducted. In a closed, invitation-only network, users were more likely to come back the more “signals” they received indicating that others were joining the service; the “eager beavers” who joined early on, it turns out, were least likely to keep using the service.
3. Usability: Give Users Something Familiar
Getting people to use health and wellness services isn’t always about gathering data every 60 seconds and providing real-time feedback on what it all means. What patients often want, notes Dr. C. Anthony Jones, chief marketing officer for Philips Healthcare’s patient care and clinical informatics business group, is a seemingly “mundane” application that, for example, lets them make a doctor’s appointment on their smartphone.
In Pictures: 10 Mobile Apps That Promote Health and Wellness
Study: One in 10 Say Health-Related Websites Saved Their Lives
The challenge, of course, is making sure it’s not mundane—booking an appointment, Jones says, should be no more difficult than making a dinner reservation using OpenTable. Demographics are a key factor here. As Jones sees it, the wellness movement essentially began in the mid-1990s, when baby boomers turned 50 and marketers, seeing an opportunity, appealed to boomers’ interests like never before. As a result, wellness apps typically target a demographic less tech savvy than the people making them.
For wellness application developers, considerations extend beyond ease of use. Even typeface matters. In studying driving’s impact on seniors, Lisa D’Ambrosio, a research scientist in the MIT AgeLab, discovered that “humanist” fonts, with rounded letters and numbers that look handwritten, are easier for older drivers to read than fonts with squared edges.
4. Home Care: Make It Easy, Involve Everyone
Driving is immensely important for seniors—if they can’t get around, they become isolated. It’s part of a larger wish to remain independent and avoid entering institutionalized care. This isn’t surprising, D’Ambrosio says, noting that today’s seniors are wealthier, healthier, better-educated and more diverse than ever before.
To date, technology’s impact on keeping seniors in their homes remains small. Though 60 percent of seniors (ages 65 and above) told the AARP in 2008 they’d be willing to use a device such as an activity monitor if it meant remaining at home longer, adoption rates for services such as LifeLine are much lower than that, D’Ambrosio points out. (This is also true in the United Kingdom and Western Europe, where the government subsidizes such purchases.) Concerns are many and include privacy, affordability, accessibility, availability of technical support and lack of confidence in using the technology; moreover, these concerns are shared by patients as well as their caregivers, albeit less so if that caregiver is an adult child as opposed to a spouse.
Through the AgeLab’s e-home project, D’Ambrosio and her team studied the effectiveness of a desktop setup that aimed to improve seniors’ adherence to medication regimens. (Fewer than half of seniors keep prescriptions in an open, easy-to-reach area.) The setup included a monitor where patients and caregivers could leave notes, an “information globe” that amounted to a “you’ve got mail” icon and an RFID-enabled medication table.
Study: In-Home Health Monitoring to Leap Six-Fold By 2017
Case Study: Cisco Telepresence Lets Swiss Doctors Conduct Virtual Consultations
The ehome study concluded that patients were more likely to adhere to their schedules if they and their caregivers (in this case, their children) were notified about a missed dose, as opposed to only the patient receiving a reminder. More reminders also meant more communication between parents and children. Equally telling, though, was the participant feedback: The setup would have worked even better had it taken up less counter space.
5. Emotion Sensors: For the Willing, Anything Can Be Monitored
When you’re hard at work, deep in thought or emotionally stimulated, your brain sends signals to your skin, especially your hands and feet. This is called electrodermal activity, and it’s triggered by the sympathetic nervous system as part of our innate “fight or flight” response system. Measuring this activity used to require wires and sensors, which had its shortcomings—namely, wearers couldn’t wash their hands. Advances in sensor technology now allow for noninvasive monitoring all day and night.
In Pictures: Mobile Health Gadgets for Better, Longer Living
News: Researchers Build Featherweight Bio Chips That Dissolve in Water
Much of the work done by Rosalind Picard, founder and director of the MIT Affective Computing Research Group, concerns children with autism. Emotion sensors such as the Q Sensor (made by Affectiva, which Picard co-founded) can indicate when a child is about to act out or has calmed down.
The results can dramatically alter treatment plans. In a specific example, Picard was with a former student with autism wearing the sensors and waiting to deliver a delayed speech; the student was pacing as a means of calming herself, but the student’s friend told her to stop, saying it didn’t help. In analyzing her electrodermal activity after the fact, Picard and the student determined that pacing did, in fact, help—and the next time she was preparing for a speech, her friend let her pace.
Sensors don’t even need to be worn. The Cardiocam, also an Affective Computing initiative, can measure heart rate and breathing rate through a webcam, while Affectiva’s Affdex reads emotion through a webcam. Both offer opportunities for telemedicine and remote patient monitoring.
6. Wellness Counseling: Sometimes, People Like Talking to Computers
If patients are being monitored by computers, then having computers talk to them isn’t much of a stretch—and, notes Timothy Bickmore, professor of computer and information science at Northeastern University, patients are receptive to the idea.
Bickmore’s lab, the Relational Agents Group, has run 12 clinical trials involving roughly 2,500 patients. These trials have encompassed exercise promotion (both for older, low-literacy patients and for Parkinson’s patients), a patient portal for promoting follow-up care, collecting family medical history and explaining discharge summaries to hospital patients. In all cases, patients interacting with an animated character, or relational agent, within their program were more active participants than those not working with an agent.
Analysis: Taking Healthcare IT Seriously Demands Culture Changes
Related: Top Challenges Facing Healthcare CIOs
The agents themselves use a dialogue engine that’s based on a branched hierarchical transition network that contains several thousand nodes in all, and that uses XML to denote synchronized nonverbal behavior such as vocalized pauses, head nods and hand gestures. That nonverbal behavior came from watching patient-provider interactions to determine when, for example, a nurse points to a patient’s discharge summary and what she’s pointing to. Patients don’t talk back; instead, they use a touch screen to choose an answer to the agent’s question and await a response.
Feedback from the clinical trials says patients like working with the agents because the interaction format is familiar, even if they don’t normally use computers. In addition, the agents don’t get impatient or condescending if they’re prompted to repeat questions. (Explaining a hospital discharge summary can sometimes take 30 minutes—time few ER nurses have.) Finally, agents have been programmed to engage in conversation—chatting about husbands, for example, or asking if the patient is a Boston Red Sox fans&mash;and this helps patients relate to them.
Brian Eastwood is a senior editor for CIO.com. You can reach him on Twitter @Brian_Eastwood or via email. Follow everything from CIO.com on Twitter @CIOonline, Facebook, Google + and LinkedIn.