Wood Kerry V, Frings Daniel, Flood Chris, Thomas Nicola
Department of Psychology, London South Bank University, London, UK.
Digit Health. 2025 Jun 17;11:20552076251321016. doi: 10.1177/20552076251321016. eCollection 2025 Jan-Dec.
Managing outpatient appointments is challenging, with missed appointments wasting capacity. Artificial Intelligence (AI) machine learning-driven automated reminders offer a solution, but their success relies on patient and staff engagement, highlighting the need for impact assessment.
To investigate the acceptability of AI machine learning-driven appointment management for patients and staff, identifying barriers and facilitators.
Semi-structured interviews with seven staff and twelve patients. Despite scheduling efforts and incentives, practical constraints limited the sample size and generalizability. Interviews were analysed separately using Thematic Analysis, with one researcher coding and categorizing data, followed by discussions to refine themes and validate quotes.
Five themes emerged. Patients: ethical concerns, AI understanding, reminder efficacy, user satisfaction, and usability. Staff: AI understanding and hesitancy, barriers and drivers, technology experiences, appointment management, and sustainability. Barriers included privacy concerns, limited interactivity, fragmented integration, and operational challenges. Facilitators were perceived prediction accuracy and reminder usefulness. Patients valued usability, convenience, and reminders but sought better interactivity and integration. Staff emphasized ethics, operations, and sustainability, with motivation linked to reduced DNAs. Both valued accuracy and reliability, highlighting the need for tailored strategies.
This study explores patient and staff perceptions of AI in NHS appointment management. Despite high trust in data security, privacy concerns, inefficiencies, and limited interactivity hinder adoption. Accuracy and convenience drive engagement. Findings highlight the need for better integration, clarity, interactivity, and accessibility to enhance user experience and AI adoption in healthcare.
管理门诊预约颇具挑战性，错过的预约会造成资源浪费。人工智能（AI）机器学习驱动的自动提醒提供了一种解决方案，但其成功依赖于患者和工作人员的参与，这凸显了进行影响评估的必要性。
调查患者和工作人员对人工智能机器学习驱动的预约管理的可接受性，识别障碍和促进因素。
对7名工作人员和12名患者进行半结构化访谈。尽管进行了排班努力和激励措施，但实际限制因素限制了样本量和普遍性。使用主题分析法分别对访谈进行分析，由一名研究人员对数据进行编码和分类，随后进行讨论以完善主题并验证引述内容。
出现了五个主题。患者方面：伦理问题、对人工智能的理解、提醒效果、用户满意度和可用性。工作人员方面：对人工智能的理解和犹豫、障碍和驱动因素、技术体验、预约管理和可持续性。障碍包括隐私担忧、互动性有限、整合碎片化和运营挑战。促进因素是感知到的预测准确性和提醒有用性。患者重视可用性、便利性和提醒，但寻求更好的互动性和整合。工作人员强调伦理、运营和可持续性，积极性与减少爽约相关。双方都重视准确性和可靠性，凸显了制定量身定制策略的必要性。
本研究探讨了患者和工作人员对国民保健服务（NHS）预约管理中人工智能的看法。尽管对数据安全高度信任，但隐私担忧、效率低下和互动性有限阻碍了其采用。准确性和便利性推动了参与度。研究结果凸显了需要更好的整合、清晰度、互动性和可及性，以提升医疗保健中的用户体验和人工智能采用率。
