Technology is currently critical to home health care. Future advances in home health care technologies have the potential not only to facilitate the role of home health care within the overall health care system but also to help foster community-based independence for individuals. In one panel at the workshop, three speakers spoke about a wide range of technologies, including the existing base of evidence on the impact of technologies, the challenges facing the development and use of technologies for home health care, and how to facilitate the use of technology to achieve an ideal state of home health care in the future. Chapter 8 touches further on the role of information technology and the use of electronic health records.
University of Washington
Computing is pervasive and ubiquitous, it is wearable, and it is used in every aspect of our lives, said Demiris. Advances in computing and information technology are also affecting home health care. Although information technology has been used in health care for a long time to bridge geographic distances and give people access to expert opinions without having to travel, ultimately, computing and information technology may actually be used to improve the quality of home health care services and enable their redesign, said Demiris. For example, Demiris noted that a MEDLINE search for papers on the use of technology in home health care published
in 2003 found 556 papers; a decade later, the number was 1,390. Finally, not just the field of technology for home health care but also the evidence that goes with it have continued to grow, he said.
Active versus Passive Technologies
One distinction that can be made among types of technologies is whether they are active or passive. Someone must operate an active technology, Demiris said, whether it is hardware or software. Active and passive technology can also be distinguished according to whether the user needs training. Passive technologies are, for example, cameras, sensors, or other devices embedded in the residential infrastructure that allow an individual to be monitored without requiring that individual or another person to operate them. As described in the literature, both types of technology have a variety of functions, Demiris noted, including monitoring and the provision of assistance in the various areas described in the following sections.
Active home telemonitoring devices can capture vital signs, weight, or symptoms and report them to a remote provider or a home health agency. Passive telemonitoring technologies include bed sensors that capture restlessness, sleep interruptions, or pulse and respiration during sleep.
Monitoring of Patient Function and Detection of Emergencies
Active technologies include devices that can detect falls and that people can wear. They also include personal emergency response systems that a person can also wear and that allow the person to press a button to summon help when he or she falls. Passive systems may be embedded in the carpet and can detect falls or near falls.
Alarm systems can actively detect fires or floods; passive systems can use motion and heat sensors, for example, to distinguish between heat that occurs during meal preparation and heat that builds up when a person turns on the stove and forgets about it.
Camera systems allow the remote monitoring of residential spaces and visitors; passive systems use sensors to capture the level of activity and whether unusual patterns of activity are occurring.
Social interactions can be made through the active use of social networks, and software can assess self-perceived social connectedness. Passive sensor-based systems track the number of visitors, the amount of time spent inside and outside the home, and sedentary behavior.
Cognitive and Sensory Activity
Active technologies can generate alerts and reminders, aid with the location of lost objects, or dispense medications. Passive automated features operate in the background and trigger warnings, alerts, and reminders or turn on lights.
The Evidence Base
Although both the amount and kind of evidence on home health care technologies are increasing—from evidence from pilot studies to evidence from some longitudinal research and randomized controlled trials—“it is still not really clear what seems to work and what doesn’t work,” Demiris said, especially because the findings are sometimes contradictory. The pace of technological advance is faster than traditional research grant cycles, so that by the time a study has been planned and funding has been acquired, the technology to be studied may be outdated. Furthermore, research projects rarely enable the tailoring of technologies on the basis of the situation in the home or the needs of the individual patient, because everyone in the study needs to receive the same intervention.
Technology interventions do not always have to be complex and sophisticated, Demiris said. Simple tools can sometimes be effective. For example, when the members of his group looked at ways to support informal hospice caregivers (e.g., family members, friends, neighbors), they used video conferencing technology to solve the caregivers’ problems. This technology is being tested in a three-arm clinical trial with groups that receive (1) friendly visits and usual care (the controls), (2) standard care plus the problem-solving therapy in person, and (3) standard care plus the problem-solving therapy by video. If the results for Groups 2 and 3 are equivalent, the video intervention might be a cost-saving alternative. Although the cost
results are not yet available, Demiris said that families are embracing the technology and “finding it [to be] a convenient and effective way to communicate” with the hospice teams.
Simple, low-cost technologies can allow health care workers to capture information more efficiently, and sometimes new technologies let them obtain new data, Demiris said. For example, traditional telehealth allows the capture of a person’s blood pressure or weight, and some of the new technologies provide information about a person’s lifestyle or behavior, eliminating the need to rely on self-reports.
One example of research on smart home technologies for people living in retirement communities involves the use of stove sensors, motion sensors, and sensors that detect water and electricity consumption, Demiris said. The last two are being used in an ongoing study to assess the overall mobility of people in their own homes. Now researchers also have data on sleep quality, restlessness, sedentary behavior, hygiene patterns, meal preparation, and so on. Again, these data should be much more accurate than those obtained from self-reports. Sophisticated algorithms may detect patterns of an individual’s normal routine and deviations from those patterns, and clinicians can be presented with these summative data.
Usability and interface design are important considerations, especially for older adults, and technology designs need to be tested with these populations, Demiris said. For example, a multiuser kiosk used by people who wanted their vital signs assessed also asked them some questions, and researchers were challenged to create a system that could continuously tailor the interface to accommodate the functional, hearing, visual, or cognitive limitations of the users. Focus groups revealed that many older adults had visualization needs very different from those of their family members and clinicians, not only according to the type of information and level of detail desired but also according to the use of colors or features that one group found distracting and another group found helpful.
Studies that survey users about the acceptability of technologies may produce high levels of satisfaction, but Demiris believes that acceptance is more complex than that. For example, obtrusiveness is a broader concept than whether a system is active or passive. A technology may have undesirable features that are too psychologically or physically prominent. “Privacy is a huge issue, but it’s not the only one,” he said. The challenge of obtrusiveness is not only people’s perceptions about what may happen to information about them but also the issue of function and how the equipment works, that is, whether it makes annoying noises or requires a lot of maintenance.
According to Demiris, other home technology issues need to be resolved, including
- Whether a technology will reduce the number of face-to-face visits;
- The technology’s effects on the client’s self-concept and the potential stigma of having it in the home (“turning the home into an ICU [intensive care unit],” one patient said);
- The technology’s effects on people’s daily routines;
- Whether any positive behavior changes are sustainable; and
- Affordability (in research studies, the technology is usually free).
In sum, although technology has a lot of potential and some evidence suggests that it can actually improve home health care services, “it’s really about finding technology as a tool to meet clinical needs,” Demiris said, rather than having it be “viewed as a solution to any of the problems we are facing.”
Almost Family, Inc.
According to Kaushal, Almost Family, Inc., which was founded in 1976 and which operates in rural areas, is the fourth largest home health care provider in the United States, operating in 14 states and serving between 25,000 and 30,000 patients daily. Because of Almost Family’s size, it can pilot test a variety of ways to improve care in its provider-based environment, and it works strategically with physicians, accountable care organizations, and hospitals.
A common denominator in the workshop so far, said Kaushal, is that in home health care technologies can be value-added services that aid with patient centeredness, sustainability, and reimbursement. Telehealth can have a wide range of complexity, from the simplicity of the telephone all the way to the extreme complexity of smart homes. Providers, however, do not know the best way to provide evidence-based telemedicine, he said. From the patient perspective, telehealth can range from being as simple as a telephone call, joining an online support group, or obtaining health information and self-management tools online to having email and online communication with health care providers. From the provider perspective, telehealth can range from the use of electronic health records and remote monitoring of vital signs and symptoms all the way to doing consultations and patient visits by video.
Technology can be helpful to teams of caregivers—physicians, nurses, therapists, social workers, and others, all of whom are delivering some aspect of care—by creating vital links that facilitate communication, coordination, and improved collaboration. Kaushal’s organization tries to create
those links within the company, as well as outside the company with the families, payers, external clinicians, and referral sources that it deals with every day.
Almost Family’s goal, he said, is to keep patients independent in their homes, while reducing emergency care and hospitalizations and improving quality of life and perceived patient satisfaction, along with reducing the total cost of care. Almost Family does this not only by following an evidence-based medical care plan but also by paying attention to functional issues—ambulation, bathing, transfer, and so on—in what is a highly regulated industry.
Meanwhile, home health care takes care of people with complex and multiple chronic conditions (e.g., heart failure, chronic obstructive pulmonary disease, diabetes). Telehealth applications can enhance the management of this high-risk population, Kaushal said, by providing daily monitoring (between in-person visits), timely intervention as the patient’s condition warrants, triage of clinical needs, and reinforcement of the treatment and discharge plan. The collaboration benefits include, for example, the ability to target visits and interventions to the patients and at times when they are the most needed and to tailor self-management training and health education for both patients and caregivers.
These patient management improvements assist physicians with both the coordination of care and the communication of health care problems and in the long run can save clinicians’ time through improved coordination and collaboration, Kaushal said. For hospitals, they can reinforce the discharge plan, allow the hospital to recognize key indicators for readmission, contribute to the stabilization of patients after hospitalization, and generally support care transitions. The communication component encourages good communication and relationships with discharge planners.
For the system as a whole, telecommunications applications have the potential to gather and compile useful data so that health care systems can learn more about what home health care applications produce the most desired outcomes, Kaushal said. To test the impact of telehealth, his company worked with two of its partners, Medtronic and Cardiocom, on a 12-month study of some 566 patients with congestive heart failure. The intervention involved post-acute care and post-episode calls, used an interdisciplinary team approach, and created some champions in every study market. The study demonstrated the positive effects of telehealth on hospitalization rates and patient satisfaction and reduced 30-day hospital readmission rates to about half the national average (12.6 percent versus 24 percent nationally).
What Almost Family learned from this study, Kaushal said, was that it was not only technology or communication, coordination, and collaboration that made the difference but also the need to align staff, standardize care processes around clinical best practices, and then conduct focused
training for clinicians around those processes. “Technology became a tool instead of a focus,” he said. These best practices involved patient education as well.
According to Kaushal, positive outcomes like these can lead to organizational growth in several ways: by providing a competitive advantage, by enabling expansion of service coverage to other disease entities, and by creating opportunities for new business and new partnerships. Kaushal noted several lessons from the experience, including the following:
- Sites that had strong clinical care champions had the best outcomes.
- An integrated hospital–physician–home health care approach to the delivery of care produced the highest number of enrollees.
- Operational planning (program design; incorporation of new programs into ongoing operations; leadership support; broad stakeholder involvement; and definition of clear goals, timelines, and deliverables) is important.
- Vendor systems must be scalable.
- Monitoring of the communication between the members of the team and clinicians in the field directly improved patient care.
- A focus needs to be placed on communication that takes into account the situation, background, assessment, and recommendation (the SBAR approach).
- Success is different for different people, including different patients.
In the future, Kaushal believes that when standardization begins to occur and reimbursement is aligned with value, innovation will be accelerated if telehealth applications show that they can improve the quality of health care and not just save costs so that systems do not have to rely solely on local execution. “Because we live in the environment of reimbursement, if reimbursement is not there, sometimes good ideas fail,” he said.
Wendy J. Nilsen
National Institutes of Health
Nilsen reported on a workshop1 held on September 9 and 10, 2014, at the National Institutes of Health (NIH) in collaboration with the Computer Community Consortium that explored technologies for aging in place and the gaps in products, policies, and research. The meeting was not about
1 See http://www.cra.org/ccc/visioning/visioning-activities/aging-in-place/411-aging-in-place-workshop (accessed December 8, 2014).
“technology for technology’s sake,” Nilsen said; “technology is a reality.” She pointed to Eric Dishman’s visionary conception of health and the need to be thinking about the research that will point the way to that future. The workshop brought together health care researchers, technology researchers, industry, and government to discuss a series of questions around four topics:
- How should technologies be designed for the aging population that wants to age at home? People who want to age in place are a diverse population, so should technologies be designed for the entire aging population or the 5 percent?
- What kind of sensing innovations are needed? The design of sensors and the development of algorithms and precise timing to make them useful are tough, so what information should sensors be conveying?
- How can people be helped to use technology to identify potential transition periods, and how can they be helped through health care transitions?
- How can non–health care technologies be used to support health?
Nilsen noted that the workshop started from the assumption that technology could enhance health outside of hospitals and nursing homes by improving and sustaining health and increasing the quality of life; by allowing people to live at home longer; by reducing health care costs, especially the cost of unnecessary hospitalizations and rehospitalizations; and by reducing the strain on the health care workforce and on family caregivers. Further, the participants looked for ways to use technology’s strengths to facilitate communication and data collection.
The workshop participants found that it was hard to talk to each other, Nilsen said. The people involved with technology wanted to see more investment in basic science, the people involved with health care wanted better technology now, and participants in general tended to focus on the technologies relevant to their own particular area of expertise. The technologists tend to think that all health care professionals are clinicians, she said, and the clinicians tend to think that all people involved with technology are programmers, but many partitions exist in both fields. “We really need to think about each other as complementary disciplines that work together for a common goal,” she said.
Nilsen said that it will be necessary to transform thinking about home health care from an approach that it is “health care at home” to one in which it is “smart homes where people can be healthy.” Technology in the home needs to be seen as an alternative form of care, she said, and not just an add-on to current systems. Such a rethinking implies a balance between
personalization and universal design, as well as stronger human factors research to support that balance. She gave as an example the glucose monitor, saying that glucose monitors are often not designed for people who have diabetes because they have tiny buttons and even smaller type. “If you have problems with your vision from long-term diabetes, you can’t see eight-point type,” she said. The new technologies developed must be useful to the patient, the caregiver, and the care team by giving them actionable data useful to them, Nilsen said.
Other technologies that can serve everyone include, for example, tablet computers that have very simple operating systems for people who have never used computers. Furthermore, although users can take advantage of the health care applications that operate on tablet computers, they also can use the tablets for email and social networking. In other words, Nilsen said, the technology is designed so that people can use the technology in the way in which they intend to use it, with health care woven in. Ideally, integrated technologies free up time for conversation. For example, if home health aides did not have to take a blood pressure, she said, they might have time to find out what is going on in their patients’ lives that may be having a greater impact on their health and well-being than their blood pressure.
For the long term, the health care system will need both more personalized technology and evidence-based, generalizable solutions that can be adapted to individual needs, Nilsen said. Other long-term considerations are the need to “future-proof” the technology—that is, the need to anticipate the rapid pace of technological change and not become stuck in what is possible today—and avoid the creation of new digital divides, in which only some people—for example, wealthier or better-educated individuals—can benefit, she said.
Another difficult question applies to monitoring technologies and, especially, cameras. Nilsen asked whether those technologies are there to keep an eye on the person so that the family knows what the individuals is doing and whether they are safe, or whether they are there to help the person be independent longer. This is one of the facets of obtrusiveness that George Demiris raised earlier.
Assessment of the usefulness of various technologies may require new research methods, such as “continuously evolving evaluations where technology can be evaluated on the fly,” Nilsen said. Randomized clinical trials may still have a place for the evaluation of specific outcomes and the development of best practices. According to Nilsen, the following developments are needed to evaluate the evidence:
- New robust methods of analysis and sensor-driven decision analysis to create predictive, personalized models of health;
- Methods for rapidly generating evidence that will allow validation of the effectiveness and reliability of technologies, including addition of technology into existing trials or cohort studies;
- Test beds that will enable the efficient, economical, and systematic exploration of the use of technologies and that will involve the community in research;
- Development of a generalized matrix for cost analysis that uses, for example, the amount of the caregiver’s time that is saved as one of a set of standard measures of a technology’s cost-saving performance; and
- A mind set that thinks about technologies broadly, including their integration into other health care services and embedding of health care tools and data collection into the “Internet of things.”
Again, she said, what is needed is a balance between the information needed now and the data that need to be collected to move home health care into the future.
An open discussion followed the panelists’ presentations. Workshop participants were able to make comments and ask questions of the panelists. The following sections summarize the discussion session.
Design and Use of Technologies
Mary Brady, U.S. Food and Drug Administration (FDA), echoed the suggestion that one factor affecting whether technologies are actually used by people is how well they can be incorporated into the patient’s life. Technologies that are too obtrusive are less acceptable, she said. One of the ways to find this out, she suggested, is through better human factors engineering and testing of the usability of a technology throughout the development process. Brady further noted that FDA has a guidance for industry related to premarket concerns.2 She said that many types of technology that were not designed for nonclinical and home use are being used outside the clinical environment (where care is needed 24 hours per day, 7 days per week). The guidance asks industry to consider the design, the users, and the physical environment in which a device will be used.
Cindy Krafft, American Physical Therapy Association, said that some of the technologies in development have great potential to support patient
2 See http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm331675.htm (accessed December 8, 2014).
functionality and activities of daily living. Still, she said, home health care agencies cannot make assumptions about their patients’ ability to use the technology. Krafft gave the example of a case in which telehealth was used to report weight on a daily basis. However, she noted, home health care nurses need to be sure that patients can get on and off the scale by themselves; often, they cannot, which can lead to inaccurate data reporting.
Rehabilitation has been a great beneficiary of technological advances, Nilsen said, in part because technology can provide data between in-person visits. She gave as an example a program funded by the National Science Foundation in which physical therapists remotely gathered data from a Wii Balance Board.
Peter Boling, Virginia Commonwealth University, warned against the creation of technology silos, as has been done “with every other component of health care.” He noted examples of the problems that silos engender, including delays in receiving monitoring data because of the way an information routing system is set up. He described a paraplegic patient who lay in bed for 3 days without food or water because he could not reach the phone and could not afford the $30 per month needed for a wearable alert device. Although Boling was confident that technologies can work, he said that they often fail because they are implemented to serve the developer’s business model and not the needs of the population that could benefit. “The vision is out there, but the application is far behind,” he said.
Krafft noted that a tremendous opportunity exists to expand rehabilitation therapies to prevent patient decline and keep patients out of higher-cost care. Unfortunately, she said, the current payment methods discourage the widespread use of rehabilitation methods that might replace a reimbursable therapy visit. Demiris agreed that telehealth and similar technologies raise a concern that they will be used to replace actual visits. They could be a great convenience for patients and families, or they could result in a diminution of services. In the early days of telehealth, he said, advocates emphasized that these services would be an add-on (they would not replace visits), but later it became clear that, if that were true, they would not be cost-effective. Rather than visits lost or visits added, it is important to think about the effective coordination of the services needed to achieve agreed-upon outcomes. Technology is only one tool among many in service redesign, he said. Nilsen agreed, saying that rehabilitation technologies are often useful in providing data between in-person visits. At Almost Family, Inc., Kaushal said, physicians, nurses, and therapists focus on a plan of care instead of visits or technology. Each patient needs a plan of care that is integrated across clinical disciplines, that includes the appropriate technology, and
that each discipline is following. “That is the biggest challenge we face in everyday life,” he said.
Thomas E. Edes, U.S. Department of Veterans Affairs (VA), noted that in the VA’s home-based primary care program, patients who receive 10 or more visits are recognized as having more complex conditions and are moved into a higher cost category. Of the 10 visits, half can be telehealth visits. To measure productivity within Almost Family, Inc., Kashaul said that the organization has developed an equivalency measure, in which some number of telehealth visits has been equated to one in-person visit. With this system (which is still being tested), staff who are paid per patient visit do not feel penalized for telehealth visits, he said. Kathryn H. Bowles, Visiting Nurse Service of New York Research Center and the University of Pennsylvania School of Nursing, said a further step would be to think about ways to reward clinicians who become more efficient. She described a randomized trial that she led in which the nurses were asked to replace some in-person visits with visits via the use of video technology. The trial found that nurses who were efficient could save time. They were then assigned additional patients, however, and so got more work. She said that one way to overcome that disincentive might be to base rewards on patient outcomes and counting telehealth encounters into the productivity standards rather than just the number of patient visits per day.
Evaluating New Technologies
Bruce Leff, Johns Hopkins University, questioned whether randomized controlled trials are really a robust method for evaluating these technologies. Technologies are not used in isolation, he said, but are used in the context of people, processes, and the planning of care. They are implemented in a very specific setting with a specific clinical team and social milieu. Moreover, trials take a long time and are very expensive. Leff noted that at present, Sarah Szanton is evaluating the CAPABLE program that she described (see Chapter 6) in two different ways: one way is through a classic trial funded by NIH in which she cannot change any of the protocol, and the other way is through the use of a challenge grant from the Center for Medicare & Medicaid Innovation, in which the evaluation is constantly being reinvented as it goes along.
Demiris acknowledged these challenges, adding that the technology itself may become outdated, given how long it takes to plan and carry out a randomized trial. Furthermore, technology does not operate independently of everything else. “So when you introduce, let’s say, access to the Internet for a specific educational intervention,” he said, “you can’t prohibit patients from browsing other websites” that also may affect their behavior and you cannot prevent a wearable device that is being tested from being used
for additional purposes. For these reasons, the use of “pragmatic clinical trials” or greater flexibility in the research protocol may be useful and allow the inevitable evolution of technology over the period of the research (Demiris, 2011).
Nilsen listed other kinds of trials—optimization trials, adaptive trials, and continuously evolving trials—that can be considered for the evaluation of technology. Moreover, ways to shorten the amount of time required for traditional randomized trials may exist. NIH is working on speeding up subject recruitment, for example, as well as shortening the time needed to obtain outcomes.