Aim. To elaborate on the relationship between work engagement, perceived organizational support, and the turnover intention of nurses by analysing some potential moderators. Background. Nurses’ turnover intention is negatively impacted by their level of work engagement and perceptions of organizational support. However, it is challenging to reach a consistent conclusion. Methods. Data were acquired from six electronic databases. Each study was evaluated using the quality assessment tool for cross-sectional studies of the Agency for Healthcare Research and Quality (AHRQ). STATA 15.0 was used to analyse the data, and a random effects model was used. The groups that included two or more studies were added to the moderator analysis. Results. A total of 40 study articles involving 23,451 participants were included. The turnover intention of nurses was inversely associated with work engagement (coefficient: −0.42) and perceived organizational support (coefficient: −0.32). A substantial moderating role was played by cultural background, economic status, working years, and investigation time ( ). Conclusion. Work engagement and organizational support significantly reduced turnover intention among nurses. Considering the acute shortage of nurses worldwide, nurses with lower wages, fewer working years, and lower levels of work engagement should be given more attention and support from their organizations. Implications for Nursing Management. The meta-analysis suggested that managers should give their employees a more organizational support and promote their work engagement to motivate nurses’ retention intention and maintain a stable workforce with little employee turnover.
Background: The rapid advancement of AI has introduced new opportunities for palliative care. However, its context-specific applicability and the trade-offs related to it use for older adults with multimorbidity, functional decline, and complex care needs remain unclear. Objective: This review aimed to characterize the applicability of AI in palliative care for older adults and synthesize its potential benefits and limitations. Methods: A scoping review was conducted following the framework of Arksey and O’Malley. Literature searches were performed in PubMed, Web of Science, CINAHL, Embase, and Scopus. Eligible studies were screened, and data were synthesized as a narrative synthesis incorporating thematic analysis. Results: Eleven studies were included, primarily comprising retrospective predictive model development and validation studies, as well as AI-based clinical information extraction studies. Applications were examined in hospital and community settings and drew on diverse routinely collected and population-based data sources, including electronic health records, clinical databases, administrative claims and health insurance databases, and population-based longitudinal survey datasets. Traditional machine learning, deep learning, and natural language processing approaches were applied. AI applications encompassed prediction and identification, monitoring and data integration, clinical decision support, and health care system optimization. Reported potential roles occurred across the data management, application performance, and clinical practice levels and included multisource information integration, more efficient data use, identification of potential palliative care beneficiaries and health risks, prognostic prediction, and quantitative support for clinical decision-making. Potential cost savings were suggested but not directly evaluated. Reported limitations relevant to real-world implementation included insufficient data reliability and availability, weak model generalizability, and restricted applicability. Human-centered limitations were infrequently examined and included difficulty recognizing patients’ emotions and the continuing need for human intervention. Conclusions: AI has the potential to support palliative care for older adults, but its clinical effectiveness and implementation effects remain uncertain. In this complex care context, the potential benefits of AI should be recognized while its inherent limitations are carefully considered. Future research should prioritize external model validation, real-world implementation studies, interoperable data systems, and the integration of patient-centered and contextual information. In clinical palliative care, AI should be positioned as an assistive tool, complementing rather than replacing clinical judgment and humanistic care.