Trends in End-of-Life Care for Medicare Beneficiaries

With all the focus by some policymakers on end-of-life care, we can perhaps be forgiven for thinking that all they really want is for us to die sooner and quicker to save money.  But I am sure there is genuine concern and compassion for the quality of care and quality of life for patients who are nearing the end.  Research in the Journal of the American Medical Association examines trends in end-of-life care for Medicare beneficiaries.   (JAMA Article)  The presumption is that most people would like to die in their homes, surrounded by loved ones and peaceably passing on to the great pasture in the sky, or something like that.  In the 1980s about half of Medicare patients died in the hospital.  By 2000 that had declined to about one-third.  By 2015, only 20% died in the hospital.  A large percent continued to die in nursing homes; from 27% in 2000 to 25% in 2015.  The flip side is an increase in hospice use, hospice being a cheaper setting and one more focused on palliative care–keeping the patient comfortable and comforted while they slip away.  Hospice use was rare in the 80s, and has increased to the point that by 2000, about 22% had used hospice and by 2015 slightly over half of Medicare decedents experienced a hospice stay.  And those stays have increased.  Many more patients are staying for over 30 days at a hospice as well.

On the other hand, there are still many patients who end up hospitalized shortly before the die, including increasing use of ICU stays.  This is a particular problem with cancer patients, who often receive intensive care right up til the time they die.  This study focussed on decedents from 2000 to 2015 and compared Medicare Advantage with the fee-for-service program.  Site of death and patterns of care in the period before death were analyzed.  As noted above, location of death continued to shift to “more pleasant” environments, from hospital and nursing homes to hospice and home.  Medicare Advantage enrollees had “better” outcomes; less deaths in nursing homes, less use of hospital of in the last 90 or 30 days of life and more use of hospice services.  The study does not cover this, but it would be very interesting to see differences in utilization patterns among beneficiaries who do and don’t have a surrogate caregiver.  Patient preferences should generally be honored, but many of these terminal patients are not really capable of directing their care.  Surrogate caregivers and decision-makers may opt for more aggressive treatment.  This is an important issue to address.