認知症高齢者向け赤ちゃん型対話ロボット―介護施設での長期導入の実現―

認知症高齢者の暴言や暴行，徘徊といった問題行動，妄想や意欲低減といった心理的な症状（BPSD：Behavioral and Psychological Symptoms of Dementia）は介護者の負担にとどまらず，社会全体の経済的な負担増として大きな社会課題となっている．BPSDへの対処では，非薬理学的手法が推奨される．我々は，非薬理学療法の1つである人形療法を参考に，人形にインタラクティブな機能を持たせた赤ちゃん型対話ロボットを開発してきた．このロボットは，ミニマルデザイン発想で，構造・機能ともに非常にシンプルに設計されており，利用者・運用者にとって取り扱いが容易で，低廉さの実現を目指した．これまでに短期間の実証実験を行ってきたが，今回，介護施設での長期導入実験を実施し，認知症高齢者の生活の質の向上やBPSD対策だけでなく，介護職員，介護施設への影響調査も行った．その結果，我々が開発したミニマルデザインの赤ちゃん型対話ロボットでも対話仕様と運用の工夫により長期運用の可能性が確認できた．また，ロボットとは直接には関わっていなかった認知症高齢者や介護職員にも受動的に影響を与えるパッシブソーシャルな状況も観測された．導入実験の過程でコロナ禍となり，通常以上に多忙で負担の多い介護現場での実験となったにも関わらず，介護職員のみで継続的に運用されたということの意義は大きい．最後に，この実験を通して分かったこと，今後の課題，なかでも高齢者の特性や周囲の状況に適したインタラクションの設計，赤ちゃん型対話ロボットの運用指針の確立などについて述べる．

BPSD (Behavioral and Psychological Symptoms of Dementia), such as verbal abuse, assault, wandering, delusions, and decreased motivation, are not only a burden on caregivers but also a major social issue that increases the economic burden on society. Non-pharmacological methods are recommended to deal with BPSD. We have developed a baby-like robot with interactive functions based on doll therapy, one of the non-pharmacological therapies. Based on a minimal design concept, this robot is designed to be very simple in both structure and function, aiming for easy handling and low cost for older people and caregivers. In this paper, we conducted a long-term experiment at a nursing home to improve the quality of life and BPSD symptoms of older people with dementia and to investigate the impact on nursing staff and the nursing home. As a result, we confirmed that even with the minimally designed robot we developed, long-term use was possible by designing how to interact with older people with dementia and operate the robot. We also observed passive social situations in which older people with dementia and caregivers, who were not directly involved with the robot, were passively influenced by the robot. It is significant that the robot was operated continuously only by caregivers, even though the COVID-19 pandemic occurred during the process of conducting the experiment and the caregivers were busier and more burdened than usual. Finally, we discuss what we learned through this experiment and future challenges, especially in designing interactions suitable for the characteristics of older people with dementia and their surroundings, and in establishing operational guidelines for the baby-like interactive robot.