Toward Responsible Artificial Intelligence in Long-Term Care: A Scoping Review on Practical Approaches. (6th December 2021)

• Record Type: Journal Article
• Authors: Lukkien, Dirk R M; Nap, Henk Herman; Buimer, Hendrik P; Peine, Alexander; Boon, Wouter P C; Ket, Johannes C F; Minkman, Mirella M N; Moors, Ellen H M
• Editors: Heyn, Patricia C
• Is Part Of: Gerontologist. Volume 63:Number 1(2023)
• Page Start: 155
• Page End: 168
• Publication Date: 2021-12-06
• Subjects: Ethics -- Intelligent technology -- Responsible innovation; Older people -- Periodicals; Older people -- United States -- Periodicals; 305.2605
• DOI: 10.1093/geront/gnab180
• Languages: English
• ISSNs: 0016-9013

Abstract

Abstract: Background and Objectives: Artificial intelligence (AI) is widely positioned to become a key element of intelligent technologies used in the long-term care (LTC) for older adults. The increasing relevance and adoption of AI has encouraged debate over the societal and ethical implications of introducing and scaling AI. This scoping review investigates how the design and implementation of AI technologies in LTC is addressed responsibly: so-called responsible innovation (RI). Research Design and Methods: We conducted a systematic literature search in 5 electronic databases using concepts related to LTC, AI, and RI. We then performed a descriptive and thematic analysis to map the key concepts, types of evidence, and gaps in the literature. Results: After reviewing 3, 339 papers, 25 papers were identified that met our inclusion criteria. From this literature, we extracted 3 overarching themes: user-oriented AI innovation; framing AI as a solution to RI issues; and context-sensitivity. Our results provide an overview of measures taken and recommendations provided to address responsible AI innovation in LTC. Discussion and Implications: The review underlines the importance of the context of use when addressing responsible AI innovation in LTC. However, limited empirical evidence actually details how responsible AI innovation is addressed in context. Therefore, we recommend expanding empirical studies on RI at the level of specific AI technologies and their local contexts of use. Also, we call for more specific frameworks for responsible AI innovation in LTC to flexibly guide researchers and innovators. Future frameworks should clearly distinguish between RI processes and outcomes.