應用創新慢性腎臟病健康照護機器人進行數位學習之使用者經驗探討

Abstract

背景：人工智慧（Artificial Intelligence, AI）科技的發展之下，機器人的發展日漸蓬勃，且已經被應用在許多智慧健康服務的業務上。隨者慢性腎臟病罹病人數越來越多，隨之而來的護理人力短缺、衛教時間不足，以及沉重的健保、財政負擔，都是臺灣正面臨的重大議題。為了提升病人的照護品質、自我健康管理的能力，應用創新科技在健康促進與衛生教育的領域已是刻不容緩。目的：本研究旨在探討慢性腎臟病病患使用慢性腎臟病健康照護機器人進行數位學習的使用者經驗。方法：共計招募慢性腎臟病病患92人。參與者接受20分鐘的創新衛教後會進行問卷填寫。了解其在系統可用性得分、個人涉入程度、學習動機跟學習成效的表現。問卷蒐集完畢後，以結構方程模型－偏最小平方法（Partial Least Squares Structural Equation Modeling, PLS-SEM）進行統計及分析。結果：使用者在系統可用性得分74.701分，介於「good」到「excellent」之間，顯示使用者能接受慢性腎臟病健康照護機器人的使用。並在個人涉入程度、學習動機及學習成效的變項都有高度的回饋。使用者經驗中的系統可用性與個人涉入程度可以與學習動機及學習成效建立PLS-SEM結構模型。本研究假設14條直接路徑，共10條顯著，分別是「系統可用性」與「個人涉入程度」到「學習動機」的四個構面（T=4.416, P<.01; T=5.422, P<.01; T=6.621, P<.01; T=6.003, P<.01; T=3.500, P<.01; T=4.173, P<.01; T=2.427, P<.05; T=3.306, P<.01）；以及學習動機的「引起注意」及「感到滿意」構面到「學習成效」（T=2.729, P<.05; T=2.092, P<.05）。間接路徑的8條假設路徑上，共2條顯著，分別為「系統可用性」透過「引起注意」及「感到滿意」到學習成效（T=2.402, P<.05; T=2.101, P<.05）。 結論：研究對象使用本研究工具「慢性腎臟病健康照護機器人」後，在系統可用性的得分上呈現高於平均的可接受程度，在系統可用性高的情境下，使用者可能因為工具能引起使用者興趣並維持注意力，或是使用者操作後能獲得內在或外在的成就感，進而提高使用後的學習成效。結果顯示本研究工具可應用在實際情境上，未來可以將研究擴及不僅是慢性腎臟病患者的使用，並比較不同情境下，使用次數、時間、使用者在資訊吸收上的情形。Background：In the era of AI, robotic technology has emerged and be used in smart healthcare services. With the increasing of patients with chronic kidney disease (CKD), patient education for CKD carriers the substantial responsibility upon economic burden, because the lack of time and nursing shortage on health education are both big issues in Taiwan. It is a critical strategy to use technology improving the patients' self-management and the health education situation on site. Objectives: The study explored the user experience of an application of CKD Education Robot on Patients with CKD. Methods: The study recruited a total of 92 participants. The participants received a 20-minute digital education. All the participants completed the measurements consisted of System Usability Scale (SUS), Revised Personal Involvement Inventory (RPII), ARCS (Attention-Relevance-Confidence-Satisfaction) model of Motivation (ARCS) and learning effectiveness of health education programs after education. Partial Least Squares Structural Equation Modeling (PLS-SEM) was used to analyze the collected data. The data was analyzed with Smart PLS 3.3.3.Results: Based on the responses concerning System Usability, Revised Personal Inventory Involvement, ARCS and Learning Effect in the context of the CKD Education Robot, most participants agree or strong agree that the Education Robot were acceptable and motivational, they can involvement in the using process. We well built a structural model. SUS and RPII significantly influenced each construct of ARCS model (T=4.416, P<.01; T=5.422, P<.01; T=6.621, P<.01; T=6.003, P<.01; T=3.500, P<.01; T=4.173, P<.01; T=2.427, P<.05; T=3.306, P<.01), while Attention and Satisfaction significantly influenced Learning Effect (T=2.729, P<.05; T=2.092, P<.05). In addition, indirect affects were observed from System Usability to Learning Effect via Attention and Satisfaction respectively (T=2.402, P<.05; T=2.101, P<.05). Conclusion: The study reveals that the CKD Education Robot were positively accepted by the participating patients. The users improve learning effect in context of high system usability, with sense of achievement, or in arousing their interest and maintaining their attention. This indicates that the robotic education material is applicable for patients’ education with CKD. In the future, we can expand the study into not only patients with CKD and compare the differences in deferent using situation.