Robots Supporting Care for Elderly People
Sandra Bedaf, Claire Huijnen, Ren4e van den Heuvel, and Luc de Witte
Aging-Related Challenges 310
A Framework for Robotics Supporting Care for Elderly People 312
Critical Review of the Technology Available 313
My Spoon 320
Critical Review of the Available Utilization Protocols 323
Review of User Studies, Outcomes, and Clinical Evidence 324
Discussion and Reflection 327
After completing this chapter, readers will be able to
- 1. Describe problematic activities of elderly people that hinder their independent living.
- 2. Describe robotic solutions to assist elderly people in those activities.
- 3. Explain a robotics framework that is used to classify robots supporting care for elderly people.
- 4. Describe specific robots, illustrating different areas within the proposed robotics framework.
The Western population is aging. In 2014, the percentage of people aged 65+ in Europe was 18.5% (Eurostat 2016). Per 100 persons of working age (15-64 years old), 28.1 persons were aged 65+. Twenty years ago, this number was lower, with 21.6 persons aged 65+ per 100 persons of working age. The elderly population is expected to continue to grow in the upcoming years not only in Europe, but also in many countries all over the world. Most elderly people are healthy and participate actively in society, but a substantial number have chronic diseases or other conditions that hinder them in daily life. For those, the performance of everyday tasks may become difficult, challenging, or even impossible due to the negative consequences of their decreasing abilities. Different kinds of solutions are being created for people who are (becoming) impaired in their capabilities for independent living in the hope of regaining or sustaining their independence. Examples are informal or formal care. However, social structures are changing, which results in informal caregivers being less inclined or able to provide care. In addition, professional caregivers are unable to answer the growing demand for home care as societies are facing an increasing shortage of care staff (Cameron and Moss 2007). For those who are unable to continue to live at home, institutional arrangements for dependent living are offered. It is known that the causes of institutionalization of an elderly person are complex (Miller and Weissert 2000). However, institutionalized care is often a final recourse due to its high costs and because elderly people usually prefer to stay at home as long as possible. “Aging in place” is also the best solution from the societal point of view.
Assistive technology (AT) can play a major role in supporting elderly people in their daily life. AT, such as (powered) wheelchairs, stair lifts, and in general home accessibility adaptations, have come a long way in supporting individuals in their independence (Vlaskamp, Soede, and Gelderblom 2011) and are increasingly becoming accepted. A new field within AT that emerged from the ongoing development of technology is robotics, which has the potential to support care and independence in many ways (Bekey et al. 2006). The technological developments of robotics are promising; however, only a limited number of care robots designed for elderly people have become commercially available (Bedaf, Gelderblom, and De Witte 2015). It seems that the envisioned role of the robots and the type of tasks they perform are often primarily guided by technical feasibility and to a lesser degree by the target users’ needs (Butter et al. 2008).
When developing a robot supporting care for elderly people, it is important to know which activities are most likely to threaten independent living when it becomes difficult. Bedaf et al. (2014) made an inventory of these activities based on a systematic literature search and focus group sessions with older people and formal and informal caregivers. The International Classification of Functioning, Disability, and Health (ICF) (World Health Organization 2001) of the World Health Organization, which provides a structured taxonomy for the description of human functioning, was adopted to group the variety of activities found in this study. The ICF domains mobility, self-care, and social isolation seemed to be the most problematic and threatening for the independence of elderly people (Bedaf et al. 2014). However, no single activity could be selected as the main activity resulting in institutionalization. One needs to take into consideration that this study by Bedaf et al. (2014) focused on elderly people with no cognitive decline even though there is a growing number of elderly people who experience problems in this area. For this group, probably a different set of activities would prove to be a threat to independent living, as they would likely experience more problems related to cognitive decline (e.g., need for reminders to take medication).
In general, a robot is an embodied system that can be programmed to perform different automated tasks involving physical movement or force exertion (see Chapter 1). Looking at the three problematic activity domains mentioned and considering the expected problems with people with cognitive decline, robots can support a number of these tasks, but for other tasks, it is more questionable. For example, several activities within the domains self-care (e.g., washing, toileting, dressing, and eating) and mobility (e.g., climbing stairs, lifting and carrying objects) involve physical movement or force exertion. These activities may therefore be interesting for a robotic system. The third problematic domain, social isolation, or the provision of reminders may be less suitable for solution by a robotic system as there is most likely no need for physical movement or force exertion. There are already several (low-cost) non-robotic Information and Communication Technology (ICT) technologies available to support social interaction (e.g., Skype, tele-home care systems), and it is easier to place a tablet in every room than to create an expensive tablet on wheels that can navigate through the house without problems. Nevertheless, there are exceptions, such as the seal-like social robot PARO (PARO Robots 2014), which also addresses social isolation and has received some attention (Gelderblom et al. 2010).
The domain of robots for the care of elderly people is broad and includes many different robots with varying goals and intentions, as shown in a number of earlier reviews on socially assistive robots (SARs) for care of elderly people (Bedaf, Gelderblom, and De Witte 2015; Bemelmans et al. 2012; Broekens, Heerink, and Rosendal 2009; Kachouie et al. 2014). To categorize these different efforts, a care robotics framework is suggested in this chapter that can help to distinguish between the different types of robots and their focus. The framework proposes two dimensions: the level of social interaction on the one hand and the targeted end user on the other hand.
With respect to the first dimension, distinction is made between physically assistive robots (PARs) and SARs. Assistive robots (ARs) are robots that provide aid or support to a human user (Feil-Seifer and MatariC 2005). An adequate definition of a PAR is one that gives aid or support to a human user through physical interaction. SARs share with PARs the goal of assisting human users, but the assistance is provided through social interaction rather than physical interaction. SARs can have a role in assisting people similar to the role guide dogs have for visually impaired people. In short, Feil-Seifer and Mataric (2005) defined an SAR as the intersection of AR and socially interactive robotics (SIRs), which was first used by Fong et al. (2003) to describe robots with the goal to develop close and effective interactions with a human for the sake of interaction itself (also see Chapter 1 for definition of SARs). Because of the emphasis on social interaction, SARs have a similar focus to SIRs. With SARs, however, the robot’s goal is to create close and effective interaction with a human user for the purpose of giving assistance and achieving measurable progress in convalescence, rehabilitation, learning, and so on (Fong, Nourbakhsh, and Dautenhahn 2003). As such, SARs can also be seen as a subsection of SIRs. A review of SAR systems concluded that different roles could be distinguished, ranging from companionship to therapeutic play partner, coach, or instructor (Rabbitt, Kazdin, and Scassellati 2015).
An example of a PAR is the mealtime assistance robot My Spoon (http:// www.secom.co.jp/english/myspoon/). This robot supports the user physically with the activity of eating without any social interaction. A good example of a SAR (or a SIR) is the interactive seal-like robot PARO. PARO is a therapeutic robot for people with dementia. It can perceive people and its environment, and by interaction with people, it responds as if it is alive.
Both the My Spoon and the PARO focus on the user. However, there are also robots that focus more on (providing support for) the caregivers. This
Figure 9.1 Care robotics framework for care of elderly people. (Robot images reproduced with the permission of the ACCOMPANY Project, Focal Meditech BV, the VictoryaHome Consortium, Vecna Technologies, Inc., and RIKEN-TRI.) is the second dimension in our framework in which a distinction is made between user-oriented robots and caregiver-oriented robots. One example of a caregiver-oriented robot is the RIBA (Robot for Interactive Body Assistance; and its follow-up version, ROBEAR [Robot for Interactive Body Assistance]) (http://rtc.nagoya.riken.jp/RIBA/index-e.html). This robot can crouch down and pick up a patient off the floor, a strenuous action that caregivers must do frequently.
The two dimensions described can be schematically described in a matrix that can function as a framework for positioning different robots for the care of elderly people. In Figure 9.1, some examples of specific robots are positioned in this matrix. These robots are illustrative examples to understand different kinds of support that robots can give. These are discussed in more detail in the remainder of this chapter.