Presentation
Designing Robotic Support for Health and Wellness of Older Adults
SessionHome Use Environment (DH9)
DescriptionIntroduction |
Robots have the potential to support the health and wellness of older adults by promoting independence, enhancing safety, and lowering healthcare costs associated with mild cognitive impairment (MCI), and mobility impairments for older adults. Over 35% of people age 65+ in the United States, are estimated to live with a cognitive or physical impairment. They need support and often require other people for assistance with cognitive (reminder to do something, cue procedures) or physical (deliver item, pick up object) activities. However, other people may not always be available, and robots have the potential to fill that gap.
Part of the challenges affecting the acceptability of robotic support in carrying out assistive tasks are issues related to the usability of the robot in the intended context of use for the target users. Using a participatory design approach, we assessed the feasibility of a robot to support older adults with mobility impairments (OAwMI) and older adults with cognitive impairments (OAwCI). We focused on the utility and benefits of using a robot to support the health and wellness of older adults performing their everyday activities in the home environment. The goal is to utilize this user-centered approach in designing robotic support that is usable and useful for older adults with a range of cognitive and physical impairments.
Method |
Research Collaboration: This project is a collaborative research effort between Hello Robot, the company that developed the robot (called Stretch), and the University of Illinois Urbana-Champaign, whose research personnel provided the human factors and neurocognition expertise for the research design.
Participants and Recruitment: A total of nine participants - 6 OAwCI and 6 OAwMI were recruited to participate in the study. Their ages ranged from 60 years to 97 years. For OAwCI participants, additional inclusion criteria included scoring in the range of 22-37 for the Telephone Interview for Cognitive Status - Modified (TICS-M) and self-reported difficulty with thinking, memory, or concentration. For OAwMI participants, additional inclusion criteria included having serious difficulty for at least 10 years in walking or climbing stairs (or inability to do so), difficulty raising a 2-liter bottle from waist to eye level, and/or difficulty using hands and fingers for prehension tasks. Exclusion criteria for both populations were having a diagnosis of Alzheimer’s or other dementia and/or a history of significant psychological illness.
Robot and Environment: The robot used was Stretch, a mobile robot manipulator designed to support everyday activities using a lightweight telescoping arm mounted on a mobile base. We conducted the study in the home simulation of the McKechnie Family LIFE Home, at the University of Illinois Urbana-Champaign.
Measures and Materials: We conducted a semi-structured interview to identify areas of daily living where the robot could support the participants. We evaluated through questionnaires, the users’ likelihood to use Stretch for various tasks on a Likert scale of 1-5, (least likely to very likely). A robot trust questionnaire was used to evaluate their willingness to trust the robot before and after interacting with the robot (Likert scale of 1-5, from strongly disagree to strongly agree). We also used the NASA Task Load Index (TLX) questionnaire to evaluate the workload associated with interacting with the robot (scale of 1 – 21, from very low to very high).
Procedure: The study was composed of three main sections: a video introduction, an observation, and an interaction. The video introductory section was prepared to introduce the robot and its capabilities to the participant through watching video recordings of the robot's activities. The participants were then engaged in an interview that explored their first impressions and overall perceptions of the robot. In the observation section of the study, participants observed Stretch interacting with a research team member to perform tasks such as providing reminders, picking, and delivering items in the home simulation area. During this section, the participants shared their thoughts out loud about the robot and what kind of tasks it should do for them. In the interaction section, participants interacted directly with the robot. They received a delivery of a bottle of water from the robot and participated in a video call session with a research team member through the tablet on the robot. The robot was remotely controlled by another research team member throughout the study and the participants were informed about this.
Results and Discussion |
Perceptions of Task Support: Physical support tasks participants mentioned included retrieving and delivering items, cleaning, and other chores in the kitchen. Cognitive support tasks included reminders for medications, healthcare appointments, and wellness checkups.
Perceptions on Trust: None of the participants had a drop in rating for any of the trust items after interacting with Stretch. The participants’ willingness to trust either stayed consistent or increased by about 20% after interaction with Stretch.
Perceptions on Workload: The participants’ workload perception after interacting with the robot reflects a generally low workload demand (M=4.44, SD=2.11). The participants particularly reported low levels of frustration (M=2.50, SD=1.69) and physical demand (M=2.58, SD=1.77).
Overall Perceptions of the Participants: Participants shared their overall impressions of the use of the robot for health-related support. Those who had lower mobility impairments made comments like “it would be useful to pick up things that are dropped, especially with my balance issues - It could be asked to retrieve something”, “it can help me get trip hazards off the floor”. Some who had upper mobility impairments shared, “it would be very useful to help me open jars or bottle covers – tasks I cannot do now due to the condition of my hand”. Those who had cognitive impairments shared comments such as , “medication reminders is a huge support for me”. Most of the participants valued the potential of the robot to help in situations of a health emergency such as a fall. Most of their comments were related, as expected, to their specific health needs. They shared their thoughts not only on the tasks that would support their health but how the robot should carry out these tasks, the degree of robot autonomy that would facilitate an active and healthy lifestyle for them as well as ways they would like to control and communicate with the robot.
Conclusion |
Overall, our user-centered design approach provided a unique opportunity to identify the needs of older adults with mobility impairments and cognitive impairments and the type of tasks the robot could do to support their health needs. The next stage of the research will explore robot support for care providers who are involved in the care of this population. We expect this, along with the outcome of the study with older adults to yield practical details that will inform design guidelines for robots supporting older adults’ health and wellness.
Acknowledgment |
This research is funded by the National Institute on Aging (National Institutes of Health) Phase II Small Business Innovation Research Grant #2R44AG072982-02
Robots have the potential to support the health and wellness of older adults by promoting independence, enhancing safety, and lowering healthcare costs associated with mild cognitive impairment (MCI), and mobility impairments for older adults. Over 35% of people age 65+ in the United States, are estimated to live with a cognitive or physical impairment. They need support and often require other people for assistance with cognitive (reminder to do something, cue procedures) or physical (deliver item, pick up object) activities. However, other people may not always be available, and robots have the potential to fill that gap.
Part of the challenges affecting the acceptability of robotic support in carrying out assistive tasks are issues related to the usability of the robot in the intended context of use for the target users. Using a participatory design approach, we assessed the feasibility of a robot to support older adults with mobility impairments (OAwMI) and older adults with cognitive impairments (OAwCI). We focused on the utility and benefits of using a robot to support the health and wellness of older adults performing their everyday activities in the home environment. The goal is to utilize this user-centered approach in designing robotic support that is usable and useful for older adults with a range of cognitive and physical impairments.
Method |
Research Collaboration: This project is a collaborative research effort between Hello Robot, the company that developed the robot (called Stretch), and the University of Illinois Urbana-Champaign, whose research personnel provided the human factors and neurocognition expertise for the research design.
Participants and Recruitment: A total of nine participants - 6 OAwCI and 6 OAwMI were recruited to participate in the study. Their ages ranged from 60 years to 97 years. For OAwCI participants, additional inclusion criteria included scoring in the range of 22-37 for the Telephone Interview for Cognitive Status - Modified (TICS-M) and self-reported difficulty with thinking, memory, or concentration. For OAwMI participants, additional inclusion criteria included having serious difficulty for at least 10 years in walking or climbing stairs (or inability to do so), difficulty raising a 2-liter bottle from waist to eye level, and/or difficulty using hands and fingers for prehension tasks. Exclusion criteria for both populations were having a diagnosis of Alzheimer’s or other dementia and/or a history of significant psychological illness.
Robot and Environment: The robot used was Stretch, a mobile robot manipulator designed to support everyday activities using a lightweight telescoping arm mounted on a mobile base. We conducted the study in the home simulation of the McKechnie Family LIFE Home, at the University of Illinois Urbana-Champaign.
Measures and Materials: We conducted a semi-structured interview to identify areas of daily living where the robot could support the participants. We evaluated through questionnaires, the users’ likelihood to use Stretch for various tasks on a Likert scale of 1-5, (least likely to very likely). A robot trust questionnaire was used to evaluate their willingness to trust the robot before and after interacting with the robot (Likert scale of 1-5, from strongly disagree to strongly agree). We also used the NASA Task Load Index (TLX) questionnaire to evaluate the workload associated with interacting with the robot (scale of 1 – 21, from very low to very high).
Procedure: The study was composed of three main sections: a video introduction, an observation, and an interaction. The video introductory section was prepared to introduce the robot and its capabilities to the participant through watching video recordings of the robot's activities. The participants were then engaged in an interview that explored their first impressions and overall perceptions of the robot. In the observation section of the study, participants observed Stretch interacting with a research team member to perform tasks such as providing reminders, picking, and delivering items in the home simulation area. During this section, the participants shared their thoughts out loud about the robot and what kind of tasks it should do for them. In the interaction section, participants interacted directly with the robot. They received a delivery of a bottle of water from the robot and participated in a video call session with a research team member through the tablet on the robot. The robot was remotely controlled by another research team member throughout the study and the participants were informed about this.
Results and Discussion |
Perceptions of Task Support: Physical support tasks participants mentioned included retrieving and delivering items, cleaning, and other chores in the kitchen. Cognitive support tasks included reminders for medications, healthcare appointments, and wellness checkups.
Perceptions on Trust: None of the participants had a drop in rating for any of the trust items after interacting with Stretch. The participants’ willingness to trust either stayed consistent or increased by about 20% after interaction with Stretch.
Perceptions on Workload: The participants’ workload perception after interacting with the robot reflects a generally low workload demand (M=4.44, SD=2.11). The participants particularly reported low levels of frustration (M=2.50, SD=1.69) and physical demand (M=2.58, SD=1.77).
Overall Perceptions of the Participants: Participants shared their overall impressions of the use of the robot for health-related support. Those who had lower mobility impairments made comments like “it would be useful to pick up things that are dropped, especially with my balance issues - It could be asked to retrieve something”, “it can help me get trip hazards off the floor”. Some who had upper mobility impairments shared, “it would be very useful to help me open jars or bottle covers – tasks I cannot do now due to the condition of my hand”. Those who had cognitive impairments shared comments such as , “medication reminders is a huge support for me”. Most of the participants valued the potential of the robot to help in situations of a health emergency such as a fall. Most of their comments were related, as expected, to their specific health needs. They shared their thoughts not only on the tasks that would support their health but how the robot should carry out these tasks, the degree of robot autonomy that would facilitate an active and healthy lifestyle for them as well as ways they would like to control and communicate with the robot.
Conclusion |
Overall, our user-centered design approach provided a unique opportunity to identify the needs of older adults with mobility impairments and cognitive impairments and the type of tasks the robot could do to support their health needs. The next stage of the research will explore robot support for care providers who are involved in the care of this population. We expect this, along with the outcome of the study with older adults to yield practical details that will inform design guidelines for robots supporting older adults’ health and wellness.
Acknowledgment |
This research is funded by the National Institute on Aging (National Institutes of Health) Phase II Small Business Innovation Research Grant #2R44AG072982-02
Event Type
Oral Presentations
TimeWednesday, March 2711:30am - 12:00pm CDT
LocationSalon A-2
Digital Health