Background: Early unplanned hospital readmissions are associated with increased harm to patients, increased medical costs, and negative hospital reputation. With the identification of at-risk patients, a crucial step toward improving care, appropriate interventions can be adopted to prevent readmission. This study aimed to build machine learning models to predict 14-day unplanned readmissions. Methods: We conducted a retrospective cohort study on 37,091 consecutive hospitalized adult patients with 55,933 discharges between September 1, 2018, and August 31, 2019, in an 1193-bed university hospital. Patients who were aged < 20 years, were admitted for cancer-related treatment, participated in clinical trial, were discharged against medical advice, died during admission, or lived abroad were excluded. Predictors for analysis included 7 categories of variables extracted from hospital’s medical record dataset. In total, four machine learning algorithms, namely logistic regression, random forest, extreme gradient boosting, and categorical boosting, were used to build classifiers for prediction. The performance of prediction models for 14-day unplanned readmission risk was evaluated using precision, recall, F1-score, area under the receiver operating characteristic curve (AUROC), and area under the precision–recall curve (AUPRC). Results: In total, 24,722 patients were included for the analysis. The mean age of the cohort was 57.34 ± 18.13 years. The 14-day unplanned readmission rate was 1.22%. Among the 4 machine learning algorithms selected, Catboost had the best average performance in fivefold cross-validation (precision: 0.9377, recall: 0.5333, F1-score: 0.6780, AUROC: 0.9903, and AUPRC: 0.7515). After incorporating 21 most influential features in the Catboost model, its performance improved (precision: 0.9470, recall: 0.5600, F1-score: 0.7010, AUROC: 0.9909, and AUPRC: 0.7711). Conclusions: Our models reliably predicted 14-day unplanned readmissions and were explainable. They can be used to identify patients with a high risk of unplanned readmission based on influential features, particularly features related to diagnoses. The operation of the models with physiological indicators also corresponded to clinical experience and literature. Identifying patients at high risk with these models can enable early discharge planning and transitional care to prevent readmissions. Further studies should include additional features that may enable further sensitivity in identifying patients at a risk of early unplanned readmissions.

The leading causes of death among the older adults have shifted from cancer to non-cancer conditions such as ischemia heart diseases, stroke, and dementia, affecting end-of-life care needs. This study examined the difference in the proportion of palliative care utilization in relation to specific causes of death and compared the trends in palliative care utilization between older adult decedents with and without cancer in Taiwan from 2010 to 2020. The study utilized data from the Health and Welfare Data Science Center in Taiwan, covering demographic and healthcare variables for 588,010 decedents aged 65+ years who died between 2010 and 2020. We used Poisson regression to investigate the temporal trends in palliative care utilization during the last six months of life. Multivariable logistic regression models were constructed to analyze cancer, and non-cancer causes of death associated with palliative care utilization. The proportion of palliative utilization in cancer deaths started at 21.7% in 2010 and increased to 63.2% by 2020 with the beta coefficient of 0.09 (95% CI: 0.09-0.09). The proportion of palliative utilization in non-cancer deaths started at 0.8% in 2010 and increased to 23.5% by 2020, with the beta coefficient of 0.26 (95% CI: 0.26-0.26). Compared to deceased cancer patients, deceased non-cancer individuals were less likely to have received palliative care (OR = 0.12, 95% CI: 0.12-0.13). Efforts to ensure equitable access to palliative care for non-cancer individuals should focus on expanding services, enhancing provider education, and promoting cultural sensitivity to meet the growing need for palliative care integration. This study was supported in part by grants from the National Cheng Kung University Hospital (NCKUH-11305001, NCKUH-V101-10) and the National Health Research Institutes (NHRI-13A1-CG-CO-04-2225-1). The sponsors had no influence on the study.

Background: Fear of falling (FOF) is a common and major health concern in older adults. The consequences of FOF include reduced physical performance, social activity, and health-related quality of life. Ramsay Hunt syndrome (RHS) is a herpes zoster-related facial nerve dysfunction accompanied by an erythematous vesicular rash on the ear or mouth that may complicate ipsilateral facial paralysis and otalgia, vertigo, tinnitus, hearing loss, and meningoencephalitis. However, repeated falls and subsequent FOF due to RHS have not been reported in older adults. Case presentation: A 65-year-old woman diagnosed with RHS experienced repeated falls during hospital admission and after discharge. Despite recovery of balance and no subsequent falls, the patient presented with persistent FOF at the geriatric outpatient follow-up visit 1 year after the RHS episode. The fear sensation impaired the patient’s instrumental daily activities and was confirmed by documentation of serial comprehensive geriatric assessments, especially the Timed Up and Go test scores. Conclusions: RHS may cause repeated falls and FOF, leading to impairment in daily activities and psychosocial function in older adults. Therefore, clinicians should be mindful of falls and FOF when caring for older patients with RHS and should develop multidimensional strategies for fall prevention and FOF.

Advance care planning (ACP) has the potential to improve the outcomes of end-of-life care for residents in nursing homes. The aim of this study was to determine whether an ACP program was beneficial for nursing home residents by assessing end-of-life indicators. An experimental study with a retrospective chart review was conducted. In total, 37 residents in the intervention group participated in an institutional advance care planning program for 1 year, and their chart data over 1 year were collected following the completion of the program; 33 residents in the control group had died within 1 year before the start date of program, and their chart data were reviewed retrospectively. Chi-square and t tests were used to examine four indicators of the quality of end-of-life care. Compared with the control group, the intervention group had a higher proportion of do-not-resuscitate directives, hospice care before death, and deaths in the nursing home, and fewer hospitalizations and deaths in an emergency department. ACP programs may improve the quality of end-of-life care for nursing home residents in Taiwan. Further research across different long-term care facilities is warranted.