Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

Social isolation is a common phenomenon among the elderly. Retirement, widowhood, and increased prevalence of chronic diseases in this age group lead to a decline in social relationships, which in turn has adverse consequences on health and well-being. The coronavirus COVID-19 crisis worsened this situation, raising interest for mobile telepresence robots (MTR) that would help create, maintain, and strengthen social relationships. MTR are tools equipped with a camera, monitor, microphone, and speaker, with a body on wheels that allows for remote-controlled and sometimes autonomous movement aiming to provide easy access to assistance and networking services. We conducted a narrative review of literature describing experimental studies of MTR involving elderly people over the last 20 years, including during the COVID-19 period. The aim of this review was to examine whether MTR use was beneficial for reducing loneliness and social isolation among older adults at home and in health and care institutions and to examine the current benefits and barriers to their use and implementation. We screened 1754 references and included 24 research papers focusing on the usability, acceptability, and effectiveness of MTR. News reports on MTR use during the COVID-19 period were also examined. A qualitative, multidimensional analysis methodology inspired by a health technology assessment model was used to identify facilitating and limiting factors and investigate if and how MTR could reduce social isolation in elderly people. Reviewed studies provide encouraging evidence that MTR have potential in this regard, as experiments report positive feedback on MTR design and core functionalities. However, our analysis also points to specific technical, ergonomic, and ethical challenges that remain to be solved, highlighting the need for further multidimensional research on the design and impact of MTR interventions for older adults and building on new insights gained during the COVID-19 pandemic.

Handwriting recognition is a mature technology with numerous successful applications. By contrast, handwriting analysis for health assessment is more recent and has attracted far less research efforts. In this work, we propose an original approach for characterizing early Alzheimer, based on the analysis of online handwritten cursive loops. Unlike the literature, we leverage local kinematic parameters, namely horizontal and vertical velocities. In order to keep all the dynamics along each loop, we propose a temporal clustering of the loops considered as time series, by exploiting an elastic distance (Dynamic Time Warping) and a K-medoids algorithm. For classification, we consider a Bayesian formalism that fuses the contributions of the obtained clusters, by taking into account the discriminative power of each. Our approach is assessed on a balanced dataset containing two cognitive profiles, Early-stage Alzheimer Disease and Healthy persons, each consisting of 27 persons, collected at Broca Hospital in Paris. The classification performance with our local approach significantly outperforms the state-of-the-art which is based on global kinematic features, with a relative improvement of 50%.

Web-based programs have been developed for informal caregivers of people with Alzheimer's disease (PWAD). However, these programs can prove difficult to adopt, especially for older people, who are less familiar with the Internet than other populations. Despite the fundamental role of usability testing in promoting caregivers' correct use and adoption of these programs, to our knowledge, this is the first study describing this process before evaluating a program for caregivers of PWAD in a randomized clinical trial. The objective of the study was to describe the development process of a fully automated Web-based program for caregivers of PWAD, aiming to reduce caregivers' stress, and based on the user-centered design approach. There were 49 participants (12 health care professionals, 6 caregivers, and 31 healthy older adults) that were involved in a double iterative design allowing for the adaptation of program content and for the enhancement of website usability. This process included three component parts: (1) project team workshops, (2) a proof of concept, and (3) two usability tests. The usability tests were based on a mixed methodology using behavioral analysis, semistructured interviews, and a usability questionnaire. The user-centered design approach provided valuable guidelines to adapt the content and design of the program, and to improve website usability. The professionals, caregivers (mainly spouses), and older adults considered that our project met the needs of isolated caregivers. Participants underlined that contact between caregivers would be desirable. During usability observations, the mistakes of users were also due to ergonomics issues from Internet browsers and computer interfaces. Moreover, negative self-stereotyping was evidenced, when comparing interviews and results of behavioral analysis. Face-to-face psycho-educational programs may be used as a basis for Web-based programs. Nevertheless, a user-centered design approach involving targeted users (or their representatives) remains crucial for their correct use and adoption. For future user-centered design studies, we recommend to involve end-users from preconception stages, using a mixed research method in usability evaluations, and implementing pilot studies to evaluate acceptability and feasibility of programs.

Introduction: The successful integration of socially assistive robots in geriatric care settings hinges on the attitudes and opinions of healthcare professionals. This study explored their needs, expectations, and perceptions regarding robot use, including facilitating factors and barriers to implementation. Methods: Twenty professionals participated in semi-structured interviews that covered topics such as robot applications, perceived value, acceptance criteria, prerequisites for deployment, ethical considerations, and design attributes. These interviews were analyzed using a deductive qualitative approach guided by the European Network for Health Technology Assessment model (version 3.0). Results: Results indicated that professionals generally viewed robots favorably, noting their usefulness in various non-intimate tasks like meal assistance and toileting. Ease of use was identified as a critical factor for robot adoption in geriatric care. Furthermore, the necessity of educating all stakeholders and providing comprehensive training to professionals emerged as essential for successful implementation. The discussions also included financial and ethical issues related to the introduction of these technologies. Conclusion: These findings will contribute to develop guidelines for designing and deploying socially assistive robots that align with the preferences and requirements of geriatric care professionals.

Precautionary measures (e.g., social distancing, mask wearing, washing hands regularly) to limit the transmission of the Coronavirus 19 (COVID-19) have been put in place worldwide. However, a limited understanding of precautionary measures and low compliance with them has been observed in older adults with neurocognitive disorders, persons with intellectual disability, or mental illness. The objective of this study is to create and evaluate a robot-mediated activity to deliver information on COVID-19 precautionary measures in an accessible and engaging way using the humanoid robot Nao. An interactive scenario explaining and demonstrating COVID-19 precautionary measures is created using the verbal and non-verbal behaviours of the robot. The scenario (≈5 min) is presented to 124 users of a geriatric hospital, including the following: older patients (n = 45), accompanying persons (n = 39), and health professionals (n = 40). The data regarding perceived usefulness, acceptability, and accessibility are collected using a questionnaire. A video analysis of the participants’ behaviour during the interaction with the robot is performed to examine the quality of engagement in the activity. The results show a good acceptance, satisfaction, and perceived usefulness of the robot-mediated activity. These findings suggest that robot-mediated interventions using humanoid robots can be an effective tool for the delivery of health promotion information.