The WSJ has a piece focusing on the increasing use of hospice among Medicare beneficiaries with dementia that observes that the long stays of such patients has increased the cost to Medicare as compared to the shorter, more predictable stays by persons with cancer, for example.
Between 2005 and 2013, about 107,000 patients received hospice care for an average of nearly 1,000 days spread out over four or more calendar years, according to a Wall Street Journal analysis of Medicare billing records. They cost Medicare 14% of its overall hospice spending, even though they accounted for just 1.3% of its hospice patients.
Undoubtedly true, but what does that mean?
Here is a paper published 12 years ago noting the problem of the poor quality of life of patients dying with dementia, and suggesting that hospice could care for patients with more than just cancer. We addressed one problem (poor quality of life for those dying with dementia) and created another (longer stays in hospice for patients with dementia). And the cost issue that is the hook of the WSJ piece has always been paramount for hospice because it was sold in the early 1980s at least partly based on its ability to reduce costs as compared to normal care, which has been shown to be the case for more traditional hospice users (but not for very long users). Cost is a fair metric, but why should hospice be the only part of the Medicare program expected to improve quality of life and save costs? That is quite a standard.
The problem that is missed in the WSJ framing is that the United States does not have a coherent long term care financing approach (there are echoes of this in the story–but it is the most important thing to understand about this story in my opinion).
Families self finance LTC with their wealth and time, and when there is no more wealth, Medicaid pays for the elderly to live in a nursing home until they die. There are myriad problems with this approach, and Medicaid pays for about half of all nursing home days in the United States. The lack of a coherent LTC system in this U.S. shows up in all sorts of places–long hospice stays, readmission rates, persons in NHs when they could be cared for at home, etc. and we try to fashion fixes for these ‘canaries in the mine’ that miss the underlying problem.
Long periods of hospice use for home based patients with dementia (and other diseases) are more directly signals of the lack of a coherent LTC financing approach. The WSJ highlights such a patient:
Helen Blincoe, a 100-year-old from Loma Linda, Calif., bounced in and out of hospice care from 2009 until last year. Currently, her main health problem is dementia, and she is in relatively stable condition. On a recent day, she sat upright in an easy chair, her walker nearby.
During the nearly 850 days she spent in hospice care, her services consisted mostly of visits by home-health aides. Nurses and social workers also saw her, but less frequently.
Hospice was not envisioned for an 850 day stay. The country has most certainly not envisioned how 70 year old daughters care for their 100 year old Mother who is a widow. I will go out on a limb and say that the hospice care paid for this person is almost certainly the most valuable thing the Medicare program could pay for on her behalf. The fact that her care doesn’t fit into the Medicare hospice benefit and that there is not a ‘high touch, low tech’ LTC benefit shows that Medicare’s benefit package doesn’t cover what is most important for many of its beneficiaries.
What to do? A preferable approach would be flexibly provided support that enabled patients to stay in their home, likely reimbursed at a lower per diem rate than what hospice pays per day. Covering what I call “high touch, low tech” home based services are what many elderly persons need, and past work we have done at Duke suggests Medicare beneficiaries would be willing to forego some medical treatments to get such care.
Hospice isn’t perfect, but the most obvious warts are at least partly driven by having the health insurer of elderly persons in the U.S. not cover LTC. We need to address this problem head on.