Jun-E Zhang

Objective: To evaluate patient safety culture (PSC) and its associated factors and staff perceptions of adverse event reporting in Chinese nursing homes. Design: A mixed-methods study based on a secondary analysis of a multisite research program, integrating a 2021–2022 cross-sectional survey with a secondary thematic analysis of focus group interviews. Setting and participants: Eleven nursing homes in South China, including 674 frontline staff (quantitative phase) and 35 managers, nurses, and caregivers (qualitative phase). Methods: PSC was assessed using the Nursing Home Survey on Patient Safety Culture. Quantitative data were analyzed using descriptive statistics, univariate analysis, and multiple linear regression. Interview data were analyzed using Braun and Clarke’s thematic analysis. Results: The overall positive response rate (PRR) for PSC was 86.5%, with lower PRRs for Adherence to Workflow (72.7%) and Staffing (67.0%). PSC was statistically associated with income satisfaction (B = 0.174, β = 0.189, p = 0.020), job satisfaction (B = 0.176, β = 0.208, p = 0.046), and years of work experience in nursing home care (B = −0.098, β = −0.113, p = 0.014). The model explained 35.0% of the variance ( R ² = 0.350, adjusted R ² = 0.345). Thematic analysis identified four themes: workforce shortages and safety anxiety, non-standardized workflows and interprofessional role conflicts, cultural barriers to adverse event reporting, and staff-driven improvement strategies. Conclusions and implications: Despite positive overall PSC, structural deficiencies were observed in staffing and adherence to workflow. Cultural factors and organizational pressures may be associated with the underreporting of adverse events. The findings suggest priorities for future intervention development, including resource optimization, workflow standardization, and non-punitive reporting systems. However, these findings require testing in longitudinal or interventional studies.

To evaluate the effects of a dyadic symptom management intervention on health outcomes of postoperative esophageal cancer patients and their caregivers. A quasi-experimental study was conducted involving 84 esophageal cancer patients and their caregivers at a cancer center in Southern China. Participants in the control group received standard perioperative care, whereas those in the intervention group underwent four structured educational sessions and received symptom management consultations during the first postoperative month. Primary outcomes included patients' symptom burden and difference in symptom assessment between patients and their caregivers. Secondary outcomes included dyadic communication, depression, anxiety, and quality of life. Generalized estimating equation models were applied for data analysis. Semi-structured interviews were conducted for the process evaluation. Statistically significant differences were observed between groups in dyadic symptom severity assessments, patient symptom severity and interference, the 31-day unplanned readmission rate, as well as dyadic anxiety, depression, and the mental component summary. No significant differences were found in dyadic symptom interference assessments, family communication, or the physical component summary of the dyads. Qualitative process evaluation identified three main themes: (1) security and gratitude, (2) adapting together, healing together, and (3) sustainable support and resources. Our findings suggest that the nurse-led dyadic symptom management intervention may improve symptom management and psychological outcomes (e.g., anxiety and depression) among postoperative esophageal cancer patients and their family caregivers. This study was retrospectively registered with the Chinese Clinical Trial Registry on 10 June 2025 (ChiCTR2500104063).

Background: Falls are a growing long-term care challenge in ageing societies. Intrinsic capacity provides a multidomain framework for conceptualising fall risk in later life. However, traditional analytic approaches are less well suited to capturing interdependence among intrinsic capacity indicators and identifying those most directly connected to short-term fall occurrence.Methods: We conducted a prospective observational study of 871 residents from four nursing homes in southern China. Ten baseline intrinsic capacity indicators were assessed, and falls were recorded over 6 months. Mixed graphical models with LASSO regularisation were used for network outcome analysis, with fall occurrence modelled as the prospective outcome node. Robustness was examined using non-parametric bootstrap resampling, conservative network estimation, a covariate network, conventional logistic regression, and exploratory age- and sex-stratified networks.Findings: During follow-up, 181 residents (20·8%) experienced at least one fall. Fall occurrence was directly connected with walking speed (edge weight=−0·215), balance (edge weight=−0·200), depressive symptoms (edge weight=0·136), handgrip strength (edge weight=−0·099), and chair standing (edge weight=−0·011). Bootstrap assessment of edge-weight accuracy and sensitivity analyses using conservative network estimation and conventional logistic regression generally supported the main findings.Interpretation: Intrinsic capacity indicators differed in their proximity to 6-month fall occurrence. Locomotor and psychological indicators may offer practical entry points for prioritising residents who require closer fall surveillance or more comprehensive geriatric assessment. These findings support the integration of brief intrinsic capacity-based assessment into long-term care quality improvement as a feasible approach to targeted fall surveillance within multidomain fall-risk evaluation and prevention.