As hospice for nursing home patients grows dramatically, a new study from the Regenstrief Institute and the Indiana University Center for Aging Research compares the characteristics of hospice patients in nursing homes with hospice patients living in the community. The study also provides details on how hospice patients move in and out of these two settings.
Longer lengths of hospice care, rising costs and concerns over possible duplication of services have led to increased scrutiny by policymakers of hospice patients living in nursing homes. Nursing home patients have longer hospice stays compared to individuals living in the community.
The researchers, led by Regenstrief Institute investigator and IU Center for Aging Research scientist Kathleen Unroe, M.D., MHA found that nursing home hospice patients were older, more likely to be women, more likely to be on both Medicare and Medicaid, and more likely to have dementia than individuals receiving hospice services outside a nursing home.
Nationally, an estimated 40 percent of Medicare recipients die with hospice. In the study, one in three Medicare recipients died with hospice. Cancer was the leading primary diagnosis for all hospice patients, although nursing home hospice patients were as likely to have a dementia diagnosis as a cancer diagnosis both nationally and in the study.
“Hospice is a poor fit for many people in nursing homes due to the way eligibility criteria currently are configured,” Dr. Unroe said. “Yet nearly a third of hospice patients in the United States live in nursing homes, and the number is growing. Our goal is to understand more about who uses hospice and where they live so that policymakers can make informed decisions as they contemplate redesign of the hospice benefit.”
Using 11 years of data, the researchers initially planned to contrast only the two groups: those who received hospice care in nursing homes and those who received hospice care in the community. But according to Dr. Unroe, it became apparent that that there were four groups that needed to be studied:
- Patients receiving hospice care only in nursing homes.
- Individuals who received hospice services only in non-nursing-home settings.
- Crossover patients who used hospice in both settings.
- A transition group who received hospice care within 30 days of transition in or out of a nursing home.
Medicare spending on crossover and transition hospice patients was higher than Medicare costs for hospice patients in nursing home or in the community, according to the new study.
While nursing home patients have longer hospice stays compared to individuals living in the community, the number of days on hospice was significantly higher for the crossover group than for any other group. Nearly a third of crossover patients had hospice stays greater than six months. Overall, for the four groups, 28 percent of patients had a hospice stay of less than one week.
“Policymakers need to look at all four groups and consider potential consequences of regulatory decisions for these patients,” Dr. Unroe said.
The 3,771 patients in this study were identified through Eskenazi Health (Wishard Health Services at that time), one of the largest safety-net health systems in the country. Data from Medicare, Indiana Medicaid and other sources capturing use of hospice services inside and outside Eskenazi Health was evaluated.
“Hospice Use Among Nursing Home and Non-Nursing Home Patients” has been published online in advance of publication in a future print issue of the Journal of General Internal Medicine. Study authors, in addition to Dr. Unroe, are Regenstrief Institute investigator Greg A. Sachs, M.D.; M.E. Dennis, B.A.; Susan E. Hickman, Ph.D., of the IU School of Nursing; Timothy E. Stump, M.A.; and Regenstrief Institute investigators Wanzhu Tu, Ph.D., and Christopher M. Callahan, M.D. Dr. Callahan is the founding director of the IU Center for Aging Research.
The study was supported by grants from the National Palliative Care Research Center (41836550) and the National Institute on Aging (RO1 AG031222 and K24 AG024078).