JMIR Publications

ABSTRACT

Background:

Technological advances in robotics, artificial intelligence, cognitive algorithms, and virtual coaches have contributed to the development of devices capable of responding to some of the challenges resulting from demographic aging. Numerous studies have explored the use of Robotic Coaching Solutions (RCS) for supporting healthy behaviours in older adults (OAs) and have shown their benefits on the quality of life and functional independence of OAs at home. However, the use of RCS by potentially vulnerable individuals raises many ethical questions. The establishment of an ethical framework to guide the development, use and evaluation practices regarding RCS for OAs seems highly pertinent.

Objective:

The objective of this article is to highlight the ethical issues related to the use of RCS for healthcare and health prevention purposes in OAs and draft recommendations for researchers and healthcare professionals interested in using RCS for OAs.

Methods:

We conducted a narrative review of the literature to identify publications including an analysis of the ethical dimension and recommendations on the use of RCS for OAs. A qualitative analysis methodology inspired by a Health Technology Assessment (HTA) model was carried out. Were included all article types such as theoretical papers, research studies and reviews dealing with ethical issues and/or recommendations for implementation of these RCS in a general population, particularly among OAs, in the healthcare sector and published after 2011 in either English or French. The research was performed between August and December 2021 using Pubmed, Cinahl, Embase, Scopus, Web of science, IEEE explore, Springer link and PsyInfo databases. Selected publications were analysed under the prism of the EUnetHTA Core Model®, version 3.0, around five ethical topics: benefit-harm balance, autonomy, privacy, justice, equity, and legislation.

Results:

In the 25 publications analyzed, the ethical concerns most cited were the risk of accidents, the lack of reliability, the loss of control, the risk of deception, the risk of social isolation, data confidentiality and liability in case of safety problems. Recommendations included consulting the opinion of target users, collecting their consent, and training professionals in the use RCS. Proper data management, anonymization and encryption appeared to be essential to protect RCS users’ personal data.

Conclusions:

Our analysis supports the interest of RCS for OAs because of their potential contribution to individuals’ quality of life and well-being. This analysis highlights many ethical issues linked to the use of RCS for health-related goals. Future studies should take into account the organizational consequences of implementation of RCS as well as the influence of cultural and socioeconomic specificities of the context of experimentation. We suggest implementing a scalable ethical and regulatory framework to accompany the development and the implementation of RCS on various aspects related to the technology, the individual, or the legal aspects.