Division of Psychology and Mental Health, School of Health Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, The University of Manchester, Manchester, UK.
Greater Manchester Mental Health NHS Foundation Trust, Manchester, UK.
Health Expect. 2024 Dec;27(6):e70116. doi: 10.1111/hex.70116.
Globally, many people experience mental health difficulties, and the current workforce capacity is insufficient to meet this demand, with growth not keeping pace with need. Digital devices that passively collect data and utilise machine learning to generate insights could enhance current mental health practices and help service users manage their mental health. However, little is known about mental healthcare professionals' perspectives on these approaches. This study aims to explore mental health professionals' views on using digital devices to passively collect data and apply machine learning in mental healthcare, as well as the potential barriers and facilitators to their implementation in practice.
Qualitative semi-structured interviews were conducted with 15 multidisciplinary staff who work in secondary mental health settings. Interview topics included the use of digital devices for passive sensing, developing machine learning algorithms from this data, the clinician's role, and the barriers and facilitators to their use in practice. Interview data were analysed using reflexive thematic analysis.
Participants noted that digital devices for healthcare can motivate and empower users, but caution is needed to prevent feelings of abandonment and widening inequalities. Passive sensing can enhance assessment objectivity, but it raises concerns about privacy, data storage, consent and data accuracy. Machine learning algorithms may increase awareness of support needs, yet lack context, risking misdiagnosis. Barriers for service users include access, accessibility and the impact of receiving insights from passively collected data. For staff, barriers involve infrastructure and increased workload. Staff support facilitated service users' adoption of digital systems, while for staff, training, ease of use and feeling supported were key enablers.
Several recommendations have arisen from this study, including ensuring devices are user-friendly and equitably applied in clinical practice. Being with a blended approach to prevent service users from feeling abandoned and provide staff with training and access to technology to enhance uptake.
The study design, protocol and topic guide were informed by a lived experience community group that advises on research projects at the authors' affiliation.
在全球范围内，许多人都经历着心理健康问题，而当前的劳动力能力不足以满足这一需求，增长速度跟不上需求。被动收集数据并利用机器学习生成见解的数字设备可以增强当前的心理健康实践，并帮助服务用户管理他们的心理健康。然而，人们对这些方法在心理健康护理专业人员中的看法知之甚少。本研究旨在探讨心理健康专业人员对使用数字设备被动收集数据和在心理健康护理中应用机器学习的看法，以及在实践中实施这些方法的潜在障碍和促进因素。
对在二级精神卫生机构工作的 15 名多学科工作人员进行了定性半结构式访谈。访谈主题包括使用数字设备进行被动感应、从这些数据中开发机器学习算法、临床医生的角色，以及在实践中使用这些方法的障碍和促进因素。使用反思性主题分析对访谈数据进行分析。
参与者指出，医疗保健用数字设备可以激励和赋予用户权力，但需要谨慎行事，以防止产生被抛弃的感觉和扩大不平等。被动感应可以提高评估的客观性，但它引发了对隐私、数据存储、同意和数据准确性的担忧。机器学习算法可以提高对支持需求的认识，但缺乏背景知识，有导致误诊的风险。服务用户面临的障碍包括获得设备的机会、可及性以及对被动收集数据产生的见解的影响。对于工作人员而言，障碍包括基础设施和工作量增加。工作人员的支持促进了服务用户对数字系统的采用，而对于工作人员来说，培训、易用性和得到支持是关键的促进因素。
从这项研究中提出了一些建议，包括确保设备用户友好且在临床实践中公平应用。采用混合方法，以防止服务用户感到被抛弃，并为工作人员提供培训和获取技术的机会，以提高采用率。
该研究的设计、方案和主题指南是在作者所在机构的一个具有丰富经验的社区团体的建议下制定的。
