End‐of‐life experience of demented elderly patients at home: findings from DEATH project

Abstract

Abstract Background: As the number of demented elderly patients choosing to die at home continues to rise, it is important to ascertain how cognitive impairment is associated with symptom experience and end‐of‐life care received at home. A number of studies have suggested that patients with dementia often receive poor end‐of‐life care, with inadequate pain or other symptom control, so far however, very little research has dealt with community settings. The aim of the present study is to compare symptom experience and end‐of‐life care received by home patients based on cognitive function. Methods: The present data was obtained from the Dying Elderly at Home (DEATH) project, a multicenter observational study conducted in Japan. The following information was collected: decedent characteristics, observed symptoms and end‐of‐life care provided during the last 48 h of life. A total of 98 decedents with dementia and 112 decedents without dementia were included in the analysis. To assess the differences in characteristics and clinical course among decedents, the survey data was divided into two groups: dementia and non‐dementia. Results: Demented decedents were prone to incontinence or cough, but experienced relatively little pain, nausea and vomiting. Demented decedents were given more antibiotics but less opioids than non‐demented decedents. Also, they received a larger volume of intravenous infusion at 24–48 h before death. After controlling for age and other differences in baseline characteristics, dementia was determined to be a significant independent predictor of incontinence or uncontrolled pain. After controlling for age, other differences in baseline characteristics and symptom experience, dementia was determined not to be a significant independent predictor of use of antibiotics or opioids. Conclusions: Our findings suggest that dementia itself was an independent predictor of incontinence or less pain, and that it was not a significant independent predictor of use of end‐of‐life in a home setting.