Note from the Editors: This is the unabridged version of a guest article that appeared in our July 15, 2003 issue. A separate pictures page contains larger versions of the pictures with a full description of each.
Eric Dishman, Manager, Proactive Health Research at Intel, has been leading innovative field research and driving collaboration with other stakeholders to focus on information technologies that can help everyday people be more proactive about their health and that of their loved ones.
Eric is the Chair of the Center for Aging Services Technologies, a cross-industry R&D partnership sponsored by the American Association of Homes and Services for the Aging. He was previously at Interval Research. Eric has a Master of Science from Southern Illinois University, and is a Ph.D. candidate at the University of Utah.
Revolutionary Needs for Broadband
“I hate to tell you this, but we really don’t need another technology to help us watch more television. What we really need is something to help us look in on my mother-in-law who lives alone in upstate New York. She has early stage Alzheimer’s, and the closest person who can help her is Tom’s sister who lives five hours away. Surely we’re not the only ones needing help helping our parents!”
Sheila was part of a digital entertainment study of early broadband adopters done in 1999 by Intel’s social science team, People and Practices Research, which I had just joined. We were experimenting with what I call an “informance” or “informative performance” in which actors play out plausible interactive skits to potential consumers by using cardboard-rendered concept prototypes to help them imagine future technologies. I knew from earlier in-home interviews that Sheila had given up her much-loved life as a schoolteacher to care for her own ailing mother. Just as the actors completed their scene showcasing a wireless digital music jukebox, a portable movie player, and a PVR with remote audio chat, Sheila held forth. And almost every participant erupted with a health-related story: a diabetic who needed to track her food and exercise, a college professor who was forgetting the names of his colleagues, and countless people who were exhausted by caring for an aging parent.
In the sleepless nights following my wake-up call from Sheila, I began reviewing my ten years of interviews with more than a thousand American and European households. Stories of health, wellness, and caregiving needs emerged from almost every discussion, even though we had never set out to study such topics. Women were the most vocal, which is not surprising given that they tend to be the primary audience and actors for caregiving technologies. They yearned for systems to help with security & health monitoring, maintaining independence, and fostering social & intellectual engagement for their aging parents. They saw a different purpose—and future—for the broadband home.
A Worldwide Caregiving Crisis
Sheila is not the only person who needs help helping her parents. We are at the beginning of an unprecedented demographic shift that has the potential to disrupt every business, industry, and economy. Older people already comprise one fifth of the total population of much of Western Europe and Japan, and many countries will at least double their elderly populations by the middle of the century (see the United Nations Population Ageing 2002 Wall Chart for excellent maps comparing 2002 and 2050). Modern medical and agricultural technologies have extended human life-spans into the 80s and 90s, but most countries will soon face an enormous caregiving crisis, if they have not already.
The Administration on Aging report, A Profile of Older Americans: 2002, reveals that there are about 35 million people—one out of every eight—over the age of 65 in the United States. By 2030, those numbers will double to 70 million with older Americans accounting for 20% of the U.S. population. There are currently 5 workers for every 1 retiree in the U.S., a nation already plagued by shortages of professional nurses, but that ratio drops to 3:1 over the next thirty years (Hooyman & Kiyak, p. 27). The 76 million American “baby boomers” who are waiting in the retirement wings already face the double challenge of caring for their parents and their own emerging health problems. Almost a third of the U.S. adult population serves as informal caregivers, most of whom are unpaid women helping an elderly parent, in an often invisible but growing industry which costs hundreds of billions of dollars a year in the U.S. alone (see Informal Caregiving: Compassion in Action).
Home is Where the Health Is
The worldwide age wave presents technology companies with both huge business challenges and new market opportunities. Companies face decreases in worker productivity and enormous increases in the cost of health insurance for their employees. Older people are by far the most frequent and expensive users of healthcare systems worldwide. The $1.4 trillion annual healthcare budget in the U.S. (based on widely reported 2001 figures) is likely to skyrocket as the boomers move into their retirement years. The good news, especially for broadband and home networking companies, is that home care offers the best hope for the healthcare crisis. The key to simultaneously saving costs while maintaining quality care is helping people to change their behaviors at home, at work, and at play—not just when they are at a medical clinic or are prompted by some illness or emergency.
The next generation of elders will demand next generation technologies to help keep them fit, functioning, and having fun from wherever they choose to live. At the same time, healthcare payers and providers have no choice but to look for ways to prevent disease, encourage healthy behaviors, detect illnesses early, drive consumer adherence to commonly accepted care plans and therapies, and support informal caregiving to help offset demands on the formal healthcare system. An always-on, high-speed, safe-and-secure network is a prerequisite for almost any home health intervention/invention. From automatic collection of health and behavioral data to new paradigms of online social support to proactive systems that coach and even cajole people into healthier behaviors, the promise of tomorrow’s broadband home is that it can transform how we care for ourselves and our families.
Proactive Health at Intel
Given the needs of Sheila and the millions like her, Intel Research funded a strategic research program called Proactive Health. Our mission is to catalyze a research ecosystem around information technologies that can help people be more proactive about their health. Our focus is on future elders—on that disruptive demography of “baby boomers”—who are dealing particularly with cognitive decline, cancer, and cardiovascular disease. These three conditions account for more than half a trillion dollars of the annual U.S. healthcare budget. Cardiovascular disease—which afflicts over 60% of seniors—costs the U.S. $351.8 billion a year in health expenditures and lost productivity (see Heart Disease and Stroke Statistics—2003 Update). People who are 55 and older account for 77% of all cancers, a disease which cost the U.S. about $171.6 billion in 2002 (Cancer Facts and Figures 2003). A recent report showed that the 4 million Americans with Alzheimer’s disease cost U.S. businesses more than $61 billion in 2002. Alzheimer’s cases in the U.S. will increase to more than 14 million by the middle of this century, resulting in massive healthcare costs that alone could bankrupt the American Medicare system (Alzheimer’s Disease: The Costs to U.S. Businesses in 2002).
Given the significance of this Alzheimer’s threat, we began our studies by examining cognitive decline. Utilizing methods borrowed from anthropological and other social sciences, we recently completed observations and interviews of fifty U.S. households with conditions ranging from the “normal” memory decline of healthy elders to the extreme cases of stroke-based dementia and advanced Alzheimer’s. We sought to understand what needs, problems, and aspirations our home health inventions should address for the entire household.
Using those findings, we have prototyped numerous “smart home” systems to help address the needs we saw (see a case study of Carl for a detailed example). Now, we are slowly refining these systems for trials in early 2004, moving from the lab back to the lives of real elders with mild cognitive impairment (MCI) and their informal caregivers.
Prototyping a Smarter “Smart Home”
The lives of “Betty” and “Jim,” participants in our field studies, show the need for a home health assistance network that can intelligently adapt to the day-to-day variability of Betty’s decline as well as to Jim’s increasing needs as her primary caregiver. Betty has been forced to retire early from an engineering career since, like most with moderate stage dementia, she now not only forgets names and faces but also the sequences of how to do everyday tasks, such as getting dressed or making a cup of tea.
Jim still works full time but it is all he can do to help her remember to eat, drink, and take her medications; he helps her practice these activities of daily living so she can maintain her independence as long as possible. He is quick to point out “that a good day for Betty is when she is able to make tea for herself—this disease has completely changed our priorities.” It is important to remember that many of the people dealing with these devastating illnesses do have very different priorities and very different opinions about being monitored than those of us who may be healthy. Many of the households we have studied see such technologies as the difference between staying in their own homes versus being institutionalized, and they want the choice to make privacy decisions for themselves.
Using “mote” technology—a small plug-and-play processor and wireless transmitter from our Intel Research Berkeley lab—we have plugged in five kinds of sensors: 1) off-the-shelf motion sensors for activity detection; 2) simple pressure sensors placed in chairs to know whether or not someone is sitting; 3) contact and magnetic switches to know when drawers, cabinets, or objects in the kitchen have been moved; 4) RFID antennas situated between the family room and the kitchen to identify small tags placed in peoples’ shoes; and 5) an IR-tracking camera that detects whether or not a badge-wearing “patient” has fallen at home. All of this raw, real-time data travels through the motes’ wireless network back into a host PC for processing, prioritization, and communication.
To help address the problem of dehydration—at this stage of the illness people often forget to drink enough—our system infers that no one has been in the kitchen or opened the cabinets where the mugs are kept. It waits as long as possible for the user to remember to get something to drink on her own, but once it reaches a certain threshold of concern, the assistant software locates and prompts her first with commercial for tea and finally with an explicit textual prompt on a nearby television.
Many patients may completely forget this prompt as they move towards the kitchen and get distracted by something like the mail on the coffee table, so we use classic “smart home” technologies like X-10 control of the lights and other sound sources such as the stereo to try to keep them on task.
Once in the kitchen, our system again waits to see if the person needs help making tea. With cognitive conditions, it is critical that the machine not prematurely replace the human’s own capacity to act. If someone is slow to start opening cabinets or moving the teapot, the system finally utilizes the kitchen television to ask if she or he needs help. If she says “yes,” it proceeds to monitor her progress, offering her video instructions of only the steps she misses: finding a mug in the cabinet, finding a tea bag, pouring the hot water, or adding the sugar if desired.
Our current prototype is primitive in its inference and assistance abilities. Nonetheless, our ultimate goals for the system are not only to successfully help with tea making and other kitchen activities, but also to longitudinally track how much help was needed, how often, and which steps were most difficult for the user so that we might detect further cognitive decline.
Returning to the case of Betty and Jim, there is some hope that on her more lucid days, Betty can still use the television remote and simple voice commands to interact with such a system, though even these “simple” technologies can prove daunting for her.
Unfortunately, her condition is likely to worsen to the point that she may lose both her verbal and physical capacities. At this stage, the home health network becomes more for someone like Jim, the caregiver, than for Betty. We observed many advanced Alzheimer’s patients sitting most of their daylight hours in the same chair, but the caregivers’ fears of their loved one falling demanded constant vigil and co-presence. These caregivers needed systems such as the chair sensor and fall detector to help monitor the safety of their loved one, thus freeing them to work or rest in other parts of the house. Our current prototype system could alert Jim that “Betty has gotten up” on whatever home device is closest to him, followed by a more urgent alert of “Betty may have fallen” if the system senses from the infrared cameras that she is at floor level.
I should warn that these laboratory prototypes I have described are very preliminary, and the systems engineering challenges to make these technologies work for real elders are enormous. Nonetheless, as digital convergence and the wireless revolution continue to make new home functionalities possible, these systems become much more plausible. As we aim for real home trials, the key technical capabilities we are working towards include:
Conclusion: Evidenced-Based Collaborative Research
If the promise of health and wellness technologies in the home is to be realized, it is critical that we adopt an “evidenced-based” approach to technology R&D modeled after the way pharmaceutical companies develop new drug therapies. Ultimately, home health and wellness technologies must prove their worth in cost savings, disease reduction, or improved quality of care through carefully designed technology trials. We cannot achieve these goals on a large enough scale or quickly enough to meet the needs of the worldwide age wave without the cooperation and collaboration of corporate technology labs, healthcare experts, academic researchers, and government agencies from many sectors. We need large-scale “collaboratories” to:
Over the past year, Intel has worked with the American Association of Homes and Services for the Aging to launch one such initiative called CAST, which stands for the “Center for Aging Services Technology”. CAST is a volunteer-driven collaboration amongst people in technology companies, university labs, government agencies, and elder care providers across the United States to spark more research and development of technologies that will support the health and wellness needs of the imminent age wave.
Technology companies across the board need to understand that there is much more to life than “efficiency”—arguably the super-motive that governed most of the office-centric development of personal computers over the past 30 years—and “entertainment,” which monopolizes most of our industry’s imagination and investment dollars today. In a recent interview in Fortune, Andy Grove warned: “We’re going to have a schism. Keep in mind, revolutions have been waged over taxation and over dividing the economic pie. But this is life and death—some people will get access to this ‘health-care mainframe,’ and everybody else dies” (see Fortune.com article).
The caregiving crisis presents us with both enormous opportunities and obstacles as our planet ages. If companies as well as people are to remain healthy in the midst of these demographic disruptions, then we would all do well to have the healthcare epiphany, to answer Sheila’s call for “help helping our aging parents,” and to notice the caregiving needs that are inexorably and inevitably becoming part of our own everyday lives.
References (in order of appearance)