HSR&D Citation Abstract
Search | Search by Center | Search by Source | Keywords in Title
Benchmarking Site of Death and Hospice Use: A Case Study of Veterans Cared by Department of Veterans Affairs Home-based Primary Care.
Intrator O, Li J, Gillespie SM, Levy C, Davis D, Edes T, Kinosian B, Karuza J. Benchmarking Site of Death and Hospice Use: A Case Study of Veterans Cared by Department of Veterans Affairs Home-based Primary Care. Medical care. 2020 Sep 1; 58(9):805-814.
The objective of this study was to examine site of death and hospice use, identifying potential disparities among veterans dying in Department of Veterans Affairs (VA) Home Based Primary Care (VA-HBPC).
Administrative data (2008, 2012, and 2016) were compiled using the VA Residential-History-File which tracks health care service location, daily. Outcomes were site of death [home, nursing home (NH), hospital, inpatient hospice]; and hospice use on the day of death. We compared VA-HBPC rates to rates of 2 decedent benchmarks: VA patients and 5% Traditional Medicare non-veteran males. Potential age, race, urban/rural residence and living alone status disparities in rates among veterans dying in VA-HBPC in 2016 were examined by multinomial logistic regression.
In 2016, 7796 veterans died in VA-HBPC of whom 62.1% died at home, 11.8% in NHs, 14.7% in hospitals and 11.4% in inpatient hospice. Hospice was provided to 60.9% of veterans dying at home and 63.9% of veterans dying in NH. Over the 2008-2012-2016 period, rates of VA-HBPC veterans who died at home and rates of home death with hospice increased and were higher than both benchmarks. Among VA-HBPC decedents, younger/older veterans were more/less likely to die at home and less/more likely to die with hospice. Race/ethnicity and urban/rural residence were unrelated to death at home but veterans living alone were less likely to die at home.
Results reflect VA-HBPC's primary goal of supporting its veterans at home, including at the end-of-life, surpassing other population benchmarks with some potential disparities remaining.