Conectar
To develop and validate a machine learning-based risk prediction model for delirium in older inpatients.
A prospective cohort study.
A prospective cohort study was conducted. Eighteen clinical features were prospectively collected from electronic medical records during hospitalisation to inform the model. Four machine learning algorithms were employed to develop and validate risk prediction models. The performance of all models in the training and test sets was evaluated using a combination of the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, Brier score, and other metrics before selecting the best model for SHAP interpretation.
A total of 973 older inpatient data were utilised for model construction and validation. The AUC of four machine learning models in the training and test sets ranged from 0.869 to 0.992; the accuracy ranged from 0.931 to 0.962; and the sensitivity ranged from 0.564 to 0.997. Compared to other models, the Random Forest model exhibited the best overall performance with an AUC of 0.908 (95% CI, 0.848, 0.968), an accuracy of 0.935, a sensitivity of 0.992, and a Brier score of 0.053.
The machine learning model we developed and validated for predicting delirium in older inpatients demonstrated excellent predictive performance. This model has the potential to assist healthcare professionals in early diagnosis and support informed clinical decision-making.
By identifying patients at risk of delirium early, healthcare professionals can implement preventive measures and timely interventions, potentially reducing the incidence and severity of delirium. The model's ability to support informed clinical decision-making can lead to more personalised and effective care strategies, ultimately benefiting both patients and healthcare providers.
This study was reported in accordance with the TRIPOD statement.
No patient or public contribution.
Chronic heart failure (CHF) is a progressive life-limiting condition that necessitates early implementation of advance care planning (ACP). However, patients and caregivers encounter emotional, informational, and cultural barriers to effective ACP engagement. This meta-synthesis consolidates qualitative evidence to deepen our understanding of ACP practices in CHF care.
This study aimed to explore experiences of CHF patients and their caregivers in ACP, which is defined as a proactive decision-making process to establish future treatment plans based on patients' values. The study also aimed to identify barriers and facilitators influencing ACP decisions and assess the impact of flexible, personalized ACP approaches on care quality.
Using qualitative meta-synthesis, we analyzed 10 qualitative studies on CHF patients' and caregivers' ACP experiences. Data were thematically synthesized to identify emotional, relational, and practical factors that influence engagement in ACP.
Three themes emerged: (1) heart failure patients and caregivers face difficulties in ACP (difficulties from patients, difficulties from the family, and difficulties from the society), (2) multidimensional drivers and impacts of ACP (advance care planning drivers, acceptance and implementation of ACP, emotions and effects of ACP), (3) flexible, personalized ACP delivers tangible benefits (timing and effectiveness of ACP discussions, patients and caregivers have personalized needs for ACP, and patients and caregivers affirm ACP benefits).
ACP plays a critical role in improving end-of-life care quality and reducing emotional and decision-making burdens on caregivers. Flexible and personalized ACP strategies supported by trained healthcare professionals more effectively meet the unique needs of patients and families. To overcome persistent barriers and promote broader ACP adoption, healthcare systems should prioritize provider communication training, ACP education, and support systems tailored to diverse cultural contexts.
To construct and evaluate a novel nomogram for predicting the risk of dual dimensional frailty (comorbidity between physical frailty and social frailty) in older maintenance haemodialysis.
A cross-sectional investigation was conducted. A total of 386 older MHD patients were recruited between September and December 2024 from four haemodialysis centres in four tertiary hospitals in Sichuan Province, China. LASSO regression and binary logistic regression were employed to determine the predictors of dual dimensional frailty. The prediction performance of the model was evaluated by discrimination and calibration. The decision curve was utilised to estimate the clinical utility. Internal validation with 1000 bootstrap samples was conducted to minimise overfitting.
In the overall sample (386 cases), a total of 92 (23.8%) of patients exhibited dual dimensional frailty. Five relevant predictors, including physical activity, self-perceived health status, ADL impairment, malnutrition, and self-perceptions of aging, were identified for constructing the nomogram. Internal validation indicated excellent discriminatory power and calibration of the model, while the clinical decision curve demonstrated its remarkable clinical utility.
The novel nomogram constructed in this study holds promise for aiding healthcare professionals in identifying physical and social frailty risks among older patients on maintenance haemodialysis, potentially informing early and targeted interventions.
This nomogram enables nurses to efficiently stratify dual-dimensional frailty risk during routine assessments, facilitating early identification of high-risk patients. Its visual output can guide tailored interventions, such as exercise programmes, nutritional support, and counselling, while optimising resource allocation.
Data were collected from self-reported conditions and patients' clinical information.
STROBE checklist was employed.
To construct a symptom network of maintenance hemodialysis patients and identify the core symptoms and core symptom clusters. Finally, this study provides a reference for accurate symptom management.
A correlational cross-sectional design. A total of 368 patients who underwent maintenance hemodialysis were enrolled from two hemodialysis centers in Chengdu, Sichuan Province, China. A symptom network was constructed with the R coding language to analyze the centrality index. Symptom clusters were extracted by exploratory factor analysis, and core symptom clusters were preliminarily determined according to the centrality index of the symptom network.
The most common symptoms in maintenance hemodialysis patients were fatigue, dry skin, and pruritus. In the symptom network, headache had the highest mediation centrality (rB = 2.789) and closeness centrality (rC = 2.218) and the greatest intensity of numbness or tingling in the feet (rS = 1.952). A total of six symptom clusters were extracted, including pain and discomfort symptom clusters, emotional symptom clusters, gastrointestinal symptom clusters, sleep disorder symptom clusters, dry symptom clusters, and sexual dysfunction symptom clusters. The cumulative variance contribution rate was 69.269%.
Fatigue, dry skin, and pruritus are the sentinel symptoms of maintenance hemodialysis patients, headache is the core symptom and bridge symptom, and pain symptom clusters are the core symptom clusters of MHD patients. Nurses can develop interventions based on core symptoms and symptom clusters to improve the effectiveness of symptom management in maintenance hemodialysis patients.
Understanding the core symptoms and symptom groups that plague maintenance hemodialysis patients is critical to providing accurate symptom management. To ensure that maintenance hemodialysis patients receive effective support during treatment, reduce the adverse effects of symptoms, and improve the quality of life of patients.
Non-pharmacological interventions have been used in the rehabilitation of stroke survivors, but their effects on stroke survivors' quality of life (QoL) are unknown.
This review aimed to summarize the existing evidence regarding non-pharmacological interventions for QoL in stroke survivors and to evaluate the effectiveness of different types of interventions.
We systematically searched databases including PubMed, Embase, Web of Science, Cochrane Library, China National Knowledge Infrastructure, Chinese BioMedical Literature Database, China Science and Technology Journal Database, and Wanfang data from the earliest available records to March 2023. Randomized controlled trials which explored the effects of non-pharmacological interventions on QoL in stroke patients were included. The meta-analysis was conducted to evaluate the effectiveness of different interventions on QoL. The Review Manager 5.3 was used to conduct the meta-analysis and the revised Cochrane risk-of-bias tool was used to assess the methodological quality of trials.
A total of 93,245 records were identified, and 34 articles were reviewed and summarized, of which 20 articles were included in the meta-analysis. The summary of the findings of the included studies revealed fitness training, constraint-induced movement therapy (CIMT), physical exercise, music therapy (MT), and art-based interventions may have positive effects on QoL. The fitness training improved total QoL, especially in physical domains including physical functioning (mean difference [MD] = 10.90; 95% CI [7.20, 14.59]), role physical (MD = 10.63; 95% CI [6.71, 14.55]), and global health (MD = 8.76; 95% CI [5.14, 12.38]). The CIMT had a slight effect on general QoL (standardized mean difference [SMD] = 0.48, 95% CI [0.16, 0.80]), whereas significantly improved strength (MD = 8.84; 95% CI [1.31, 16.38]), activities of daily living/instrumental activities of daily living (ADL/IADL; MD = 10.42; 95% CI [2.98, 17.87]), and mobility (MD = 8.02; 95% CI [1.21, 14.83]). MT had a positive effect on the mental health domain (SMD = 0.54; 95% CI [0.14, 0.94]).
Our findings suggest that fitness training and CIMT have a significant effect on improving physical QoL, while MT has a positive effect on improving psychological QoL. Future studies may use comprehensive and multicomponent interventions to simultaneously improve the patients' physical, psychological, and social QoL.
FreshRSS 