OBJECTIVE: Public attention has focused on quality of care for the dying. Where one dies is an important individual and public concern.
METHODS: Using the National Mortality Followback Survey (NMFS), we examine site of death, defined as home, hospital or nursing home. Using multi-nomial logistic regression, we tested the rate of functional decline in the last year as a predictor of site of death, controlling for socio-demographics, disease, geographic region, access to care, having a living will, hospice use, social support, insurance and baseline functioning.
RESULTS: Nearly 60% of deaths occurred in acute care hospitals with 21% dying at home and 20% dying in a nursing home. Younger people and African Americans were more likely to die in the hospital. Functional status a year prior to death did not predict dying in the hospital (p>0.05). Persons who lost one or more ADLs in the last 4 months of life were more likely to die at home (i.e. 35% with functional decline compared to 18% without). After adjustment, functional decline in the last 4 months of life remained associated with dying at home (for each ADL decline, the odds of dying at home increased by 8%, OR=1.08, CI 1.001-1.172). Both baseline function at 12 months, and rate of decline were associated with dying in the nursing homes compared to home.
CONCLUSIONS: Rapid physical decline during the last 4 months was associated with dying at home, while earlier functional loss was associated with dying in nursing homes.
Learning Objectives: Participants will recognize the importance of functional decline in predicting site of death
Keywords: End-of-Life Care, Quality
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.