Prediction of prognosis in hospice patients: Integrating symptom burden from multiple dimensions
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background: Understanding the prognosis of palliative care patients is crucial for appropriate care and hospice admission. Current prediction models are not always suitable in hospice settings. Exploring the relationship between multidimensional symptoms and concerns and prognosis may offer insights. Aim: To unravel which multidimensional symptoms and concerns are predictive of survival at hospice admission. Design: Cohort study (2017–2025). Primary outcomes were 2-week, 4-week and 3-month survival. Multidimensional symptoms and concerns were measured using the Utrecht Symptom Diary-4 Dimensional (USD-4D) on an 11-point scale. Analyses used multivariable Cox regression with backward selection and evaluated models using performance indicators. Setting/Participants: Adults admitted to Dutch hospices who completed at least one USD-4D within 7 days of admission. Results: 1204 patients were included (median age 77 years); 56% were female. Median survival among deceased patients was 23 days. Three models were developed. The 2-week model included anorexia, dyspnoea, depressed mood, well-being, ‘perceived value of life’, ‘perceived balance in life’ and ‘take time for yourself’. The 4-week model included anorexia, dyspnoea, depressed mood, ‘take time for yourself’, ‘thinking about the end of life gives peace of mind’, sleeping problems, dry mouth and pain. The 3-month model included anorexia, dyspnoea, depressed mood, ‘thinking about the end of life gives peace of mind’, sleeping problems, dry mouth, pain and fatigue. Conclusion: This study underscores the dynamic nature of symptomatology in predicting survival. By integrating routinely collected multidimensional symptoms into existing prediction models, their effectiveness in hospice care can be improved, enabling timely admissions and better care planning.
Source: Palliative Medicine
Institutions: Utrecht University, University Medical Center Utrecht
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $4,400
OpenAlex ID: W7236835776
