Purpose: This study aimed to develop a machine learning-based predictive model for hospital length of stay incorporating clinical, nursing, and healthcare system factors to optimize hospital resource allocation, improve patient-centered care, and enhance nursing workflow efficiency.
Methods: This retrospective study analyzed a large dataset of inpatient electronic medical records from a private tertiary hospital. The dataset was used to develop predictive models for long-term versus short-term hospitalization. The modeling process involved several machine learning algorithms, and their performance was evaluated using standard statistical metrics. The most significant predictive variables were identified through an analysis of their feature importance.
Results: Among the tested models, the Random Forest algorithm exhibited the highest predictive accuracy, demonstrating strong performance in predicting hospital length of stay. Key influencing factors included the number of consultations, postoperative recovery time, duration of stay in the intensive care unit, the use of third-generation antibiotics, and the need for infection isolation. Patients requiring ventilator care, intensive care unit admission, and specific powerful antibiotics were more likely to experience prolonged hospitalization. Additionally, nursing-related factors such as fall risk and pressure ulcer risk were significantly correlated with an extended hospital stay.
Conclusion: This study demonstrates that machine learning models can effectively predict hospital length of stay, aiding in hospital resource management, nursing workforce allocation, and patient safety interventions. The integration of predictive analytics into healthcare systems can support early risk assessment, personalized discharge planning, and overall hospital efficiency.
Purpose: To examine the effectiveness of a practice-based reflection training programme in enhancing reflective practice competency and humanistic caring ability among undergraduate nursing interns.
Methods: A convenience sample of 95 undergraduate nursing interns from a tertiary hospital were recruited to join a quasi-experimental study. Participants in the control group (n = 48) underwent a traditional training programme between September 2021 to April 2022. Those in the intervention group (n = 47) participated in a practice-based reflection training programme which contained 8 sessions between September 2022 to April 2023. The two groups were compared in terms of their level of reflective practice competency measured by the Nursing Reflective Practice Questionnaire (NRPQ) and humanistic caring ability measured by the Nursing Humanistic Caring Assessment Tool (NHCAT). Additionally, each intern in the intervention group submitted a reflective essay on humanistic caring which was analyzed by theme analysis.
Results: After the intervention, the intervention group showed statistically significantly higher mean scores compared to the control group for reflective practice competency (160.94±11.02 vs 147.60±13.95, p < .001) and humanistic caring ability (124.02±7.35 vs 115.52±8.31, p < .001). By theme analysis of 45 reflective essays, the most commonly reported theme related to humanistic care was "responsibility" (present in 38 essays, mentioned 153 times), followed by "empathy" (present in 36 essays, mentioned 149 times) and "communication skills/competence" (present in 32 essays, mentioned 125 times).
Conclusions: The practice-based reflection training programme was shown to be effective to enhance reflective practice competency and humanistic caring ability among undergraduate nursing interns in China. Further studies are needed to examine its long-term impacts on nursing interns.
Purpose: To explore how insight into illness mediates and moderates the relationship between psychotic symptoms and self-care ability in patients with chronic schizophrenia.
Methods: A cross-sectional convenience sampling method was employed, and 203 participants were recruited from seven rehabilitation units in northern Taiwan between April and December 2020. Psychotic symptoms were measured using the Positive and Negative Syndrome Scale (PANSS), insight into illness was assessed with the Schedule for Assessment of Insight in Psychosis (SIP), and self-care ability was evaluated using the Self-Care Ability Assessment Scale. Mediation and moderation analyses were conducted using Models 4 and 1 of the PROCESS Macro in SPSS, controlling for demographic variables.
Results: Insight into illness emerged as both a mediator and a moderator in the relationship between psychotic symptoms and self-care ability. Specifically, insight into illness mediated the relationship by partially alleviating the adverse effects of psychotic symptoms on self-care ability. Furthermore, insight into illness acted as a moderator, with greater insight significantly reducing the impact of psychotic symptoms on an individual's capacity for self-care.
Conclusion: Improving insight into illness is vital to help patients with chronic schizophrenia manage psychotic symptoms and improve self-care. Mental health nurses should prioritize interventions aimed at improving insight, which may significantly improve self-care and overall quality of life.
Purpose: To identify the predictors of dietary behaviors and health outcomes in middle-aged men using social cognitive theory as a framework.
Methods: This cross-sectional study analyzed data on 244 middle-aged men in South Korea between November 2022 and January 2023. The data were collected using structured online questionnaires that assessed self-efficacy, outcome expectations, self-regulation, social support, dietary behaviors, body mass index (BMI), and health-related quality of life (HRQoL). Analyses included descriptive statistics, Pearson's correlation analysis, independent t-tests, one-way analysis of variance with Scheffe's post-hoc test, and path analysis.
Results: The findings indicate that self-efficacy and negative outcome expectations significantly influence dietary behaviors, which in turn affect BMI and HRQoL. Moreover, autonomous motivation, controlled motivation, and social support were identified as direct determinants of HRQoL. The model accounted for 24% of the variance in dietary behaviors and 38% in HRQoL.
Conclusion: To improve the dietary behaviors and health outcomes in middle-aged men, it is imperative to build self-efficacy and address perceived barriers to healthy eating. Additionally, comprehensive and tailored interventions that foster self-regulation and social support can improve the HRQoL of middle-aged men.
Purpose: To comprehensively integrate qualitative findings on the reasons of late diagnosis in HIV/AIDS patients and to encourage timely clinic consultations for advisory purposes.
Methods: A qualitative systematic review utilizing a meta-aggregation approach. Extensive searches were conducted across PubMed, Embase, Cochrane Library, Web of Science, CINAHL, ProQuest, Scopus, Medline, CNKI, VIP, Chinese Biomedical, and Wanfang databases. The search was completed on December 10th, 2024. Studies were screened according to predefined inclusion and exclusion criteria. Quality assessment was performed using the Joanna Briggs Institute Critical Appraisal Checklist for Qualitative Research. The review adhered to PRISMA 2021 guidelines.
Results: A total of 13 articles were included, 113 original findings were extracted and categorized into 13 new categories, which were further synthesized into three integrated findings: (1) lack of proper knowledge about HIV; (2) complex psychological distress and (3) inadequate medical resources and limited services.
Conclusion: This meta-synthesis of qualitative research focused on the reasons for late diagnosis in people living with HIV(PLWH). It is based on an in-depth exploration of patients' personal feelings and experiences, as well as insights into their genuine desires. Patients' lack of disease knowledge may amplify fears, leading to anxiety, depression, and other adverse psychological conditions that reduce willingness to seek medical care and contribute to late HIV diagnosis. Furthermore, this reduced healthcare-seeking behavior contributes to underutilization of healthcare resources and impairs system efficiency. Consequently, the risk of late-stage consultation for serious conditions such as AIDS increases. Given these diagnostic challenges, enhancing early detection among PLWH is critical to reducing viral transmission and improving the quality of life of those infected.
Clinical trial registration: As it was based entirely on previously published studies, this study protocol had registered on the PROSPERO website (CRD42025631287).
Purpose: Home care and digital healthcare are essential strategies for addressing health challenges in an aging society. This study aims to understand the behavioral intention to use digital healthcare and its influencing factors based on the Unified Theory of Acceptance and Use of Technology (UTAUT).
Methods: This cross-sectional study surveyed 180 home care nurses in South Korea using online questionnaires. Participants were from hospital-based home care (HHC), home hospice (HH), and community health center-based home care (CHC). The survey assessed their experience with and intention to use 19 digital health technologies and measured key factors based on the Unified Theory of Acceptance and Use of Technology (UTAUT): performance expectancy, effort expectancy, social influence, attitude toward digital healthcare, and concerns.
Results: Among the 19 digital health technologies, big data/artificial intelligence (AI) solutions (87.2%) and AI-assisted care sensors (85.6%) had the highest behavioral intention to use. Usage experience and intention were negatively correlated across all technologies (κ = -0.103 to -0.703, p < 0.001). Behavioral intentions differed significantly between groups (χ2 = 6.354, p = .042), with HH nurses reporting the highest intention (93.8%), followed by HHC (87.0%) and CHC (75.0%) nurses. Effort expectancy increased adoption likelihood (odds ratio [OR] = 2.256, 95% confidence interval [CI]: 1.102-4.617, p = .026), while social influence was the strongest predictor (OR = 2.931, 95% CI: 1.152-7.462, p = .024).
Conclusions: Home care nurses generally showed strong behavioral intentions to adopt digital healthcare. However, the negative association between experience and intention suggests the need to enhance user satisfaction when implementing digital health technologies. Since effort expectancy and social influence significantly affect adoption, reducing perceived effort and strengthening social support are essential for successful integration.

