Vermonters don’t utilize hospice until too late

The vast majority of Vermonters say they want to die at home, and people who choose hospice typically end their lives where they hoped. However, Medicare spending on hospice in Vermont is low not just because it costs less to die at home.

From 2001-05, Medicare spent an average of $41,514 caring for each chronically ill Vermonter in the last two years of life, according to The Dartmouth Atlas of Healthcare. More than half the amount — $26,886 — went for patient care in hospitals and other facilities. Only $1,382 was spent on hospice care.

Rather, the Medicare outlay reflects some troubling facts: that hospice is comparatively underused in Vermont and that Vermonters enter hospice care very late in the dying process. Hospice proponents argue that outdated Medicare restrictions lead to underutilization, which threatens the financial viability of Vermont’s mission-driven, nonprofit hospices.

The Dartmouth Atlas of Health Care, which tracks regional differences in health care spending and outcomes, found that barely a fifth of Vermonters (20.2 percent) used hospice during their final six months of life during 2001-05. That rate is 10 percent below the national average. In fact, Vermont ranked only a sliver above the lowest 10 percent of states in hospice use.

To understand why hospice is not the default choice for those who want to die at home and free of pain, it helps to know how the practice developed, how it’s related to palliative care and how providers are reimbursed.

Before the mid-1970s, the United States had no established hospice practice. Then, in 1978, some two dozen demonstration sites were chosen to help develop a hospice program for the country. Among them were five Vermont home health agencies.

“The five northern Vermont hospices are the oldest hospices in the United States,” says Church Hindes, president and CEO of the Visiting Nurse Association of Chittenden and Grand Isle Counties, which provides home health and hospice services. “We were all founded as part of the original federal hospice program. This is a very unusual part of the country because our hospice programs are so old.”

Based on the demonstration projects, Congress approved a Medicare hospice benefit in 1982 with the twin goals of improving end-of-life care and containing costs. The Medicare benefit spurred hospice growth; now there are nearly 3,500 providers nationwide.

Despite the deep roots of hospice in Vermont, there’s growing concern that the program is underutilized. In March, the Vermont House passed a hospice and palliative care bill, H.201, with the goal of increasing access to hospice by modifying restrictions that limit its use. The bill is now in the Senate Health and Welfare Committee.

Hospice and palliative care share similar values. Both are multidisciplinary, patient-centered approaches that encompass patients’ psychosocial and spiritual needs. Both focus on keeping patients free from pain and providing the best quality of life possible, not only for patients but for their families. Both types of care can be provided in the patient’s home, a hospital or in other facilities, although hospice services are typically delivered at home. In a recent article in the “New England Journal of Medicine” cited on the Vermont Ethics Network’s website, Drs. Amy S. Kelley and Diane E. Meier explain the difference between nonhospice palliative care and hospice palliative care.

Nonhospice palliative care is offered simultaneously with life-prolonging and curative therapies for persons living with serious, complex, and life-threatening illness. Hospice palliative care becomes appropriate when curative treatments are no longer beneficial, when the burdens of these treatments exceed their benefits, or when patients are entering the last weeks to months of life.

Medicare’s 30-year-old hospice insurance benefit had two restrictions that providers today regard as barriers to access.

First, to qualify for Medicare’s hospice benefit, patients must have a terminal diagnosis and be expected to live no longer than six months. Second, they must agree to forego further curative treatment. (Medicare places no age limit on hospice eligibility — young people who are diagnosed with a terminal disease and have a life expectancy of six months or less are eligible for hospice benefits.)

Indeed, Medicare’s hospice benefit has failed to keep pace with evolving needs. It reflects the way Americans died 30 years ago, not the way they die today, and the mismatch creates a particular hardship for small, rural and nonprofit hospice providers.

End of life care in Vermont

In Vermont, hospice services are delivered by home health agencies, which also provide palliative home health care. While hospice utilization is low, Vermont ranks second in the country in the number of home health agency visits (39) received by the dying in their last two years of life, the Dartmouth Atlas shows.

Medicare’s restriction of hospice benefits to patients with a life expectancy of six months or less has become a major barrier because the leading causes of death have changed dramatically since the program was created.”

Hindes confirms what the statistics suggest: “Medicare-funded palliative care has been the longstanding ‘work-around’ for the Medicare hospice restrictions. Much end-of-life care in Vermont is through Medicare-funded palliative care … delivered under the parameters of the usual Medicare home health benefit.”

Medicare covers 100 percent of the cost of both services, with the exception of prescription drugs, and virtually everyone who needs end-of-life care can get some version of it: The last time the state did a survey to determine how many Vermonters were uninsured, it found only 60 people age 65 or older who lacked insurance coverage.

Hospice offers benefits that home health palliative care lacks, however, so it’s important that the dying be able to access hospice services and not have to rely on a work-around. For example, hospice programs have volunteers who provide respite for families and companionship for patients. Not all palliative care programs have volunteers. And Medicare’s hospice benefit pays for prescription drugs, while the home health benefit doesn’t cover them. A patient who lacks a Medicare drug plan could incur very high costs for medications at the end of life, and even patients with a drug plan would have the expense of copayments.

Patients and their families indicate a high level of satisfaction with hospice — 99 percent of respondents to a statewide survey done for the Vermont Assembly of Home Health Agencies said they would recommend hospice to others. So why aren’t more Vermonters using it?

How we die and hospice eligibility

Medicare’s restriction of hospice benefits to patients with a life expectancy of six months or less has become a major barrier because the leading causes of death have changed dramatically since the program was created. Alzheimer’s disease, for example, first appeared in the National Vital Statistics Reports of mortality factors around 1980.

A study of Medicare hospice data trends from 1998-2008 documents the changes in terminal diagnoses, “with relatively fewer cancer patients and relatively more non-cancer patients as a percentage of total hospice patients.” The summary notes that the percentage of patients with a diagnosis of terminal cancer “dropped from 52.8 percent in 1998 to 31.1 percent in 2008.” At the same time, it says, there has been “a notable increase in the number of neurologically-based diagnoses” and “a marked increase in non-specific diagnoses such as “Debility, Not Otherwise Specified,” and “Adult Failure to Thrive.”

The largest changes were “failure to thrive” (up 64 percent); “debility, not otherwise specified” (up 62.7 percent); “Alzheimer’s” (up 56.7 percent); “non-Alzheimer’s dementia” (up 56.1 percent); and “Parkinson’s” (up 55.2 percent).

Hindes notes the irony that the original Vermont demonstration projects helped develop the Medicare restrictions, including the six-month prognosis for hospice admission, which they now want changed.

“(A six-month prognosis) is fairly easy to determine if you have a well-known brain cancer — there are ample statistics,” he said. “On the other hand, if you have a very slow-moving degenerative disease such as Parkinson’s or ALS (amyotrophic lateral sclerosis, or Lou Gehrig’s disease) – ALS is inevitably a fatal diagnosis, but it is a disease that moves very differently person to person. So it would be almost impossible, unless someone was very near the end of their life, to say any particular client has a diagnosis of six months or less.”

That means the client would be ineligible for hospice, “even though the patient, the family, everybody knows they are dying,” Hindes said. He added that such a patient would be offered palliative care through the home health program.

Dr. Allan Ramsay, the director of palliative care at Fletcher Allen Health Care, added that making an accurate assessment that a patient’s life will end within six months is also “very difficult for benign conditions like dementia, heart failure and chronic obstructive pulmonary disorder.

“The medical director has to fill out a form that documents why you’re saying that. You could say progressive cancer, weight loss, shortness of breath, but for each condition there’s a set of criteria that has to be met.”

Ramsay noted that there’s an additional obstacle to getting a physician to certify a six-month prognosis: “Most doctors, when you say ‘six months,’ they get cold feet about having that discussion with a patient or a family. Doctors overestimate prognosis — that’s proven — and the longer relationship you have with a patient, the worse you are about establishing a realistic prognosis.” While six months is daunting, a year, he said, “sounds like an acceptable time.”

H.201 would remove the obstacle by seeking a Medicare waiver to extend eligibility to patients with a prognosis of 12 months. It would also change Medicare’s requirement that hospice patients give up curative care, which acts as a barrier for other patients.

“This is more likely to occur with young adults with cancer, where no one — no one — wants to stop trying,” Hindes observed. Perhaps the patient has been through various chemotherapy regimens, but she, her doctor and the family want to try a new clinical trial. “In the meantime the patient is getting weaker and weaker and clearly the end is in sight,” he said, “but no one wants to give up.”

Other patients are reluctant to renounce palliative chemotherapy or radiation because they fear pain. H.201 would allow “concurrent care,” that is, it would allow patients to receive hospice care while retaining access to therapies that Medicare classifies as curative, such as chemotherapy, radiation and dialysis.

Hospice economics 101

For several years MedPAC — the Medicare Payment Advisory Commission — has been studying the economics of hospice. In its March report to Congress, the agency observed that the number of hospices and of hospice patients, the average length of stay and total Medicare spending on hospice is growing nationwide, and that the growth has been driven by the entry into the field of for-profit providers.

Vermont has 10 nonprofit home health and hospice agencies. (According to Hindes, while the number is “very stable,” the size of the VNA’s program has tripled in the past 15 years.)

In 2009, Bayada Nurses, a privately held company, received a Certificate of Need to acquire a 100 percent interest in a Vermont home health agency that was financially stressed. Bayada is now developing home health and hospice services statewide. Bayada Hospice has administrative offices in Norwich and Rutland and will be opening a third office in Burlington in a few months.

If you look at entire heath care system, there are only two systems of care that have accomplished everything that we want to accomplish with overall healthcare reform. One of them is palliative care, because we have very high quality ratings, we’re patient- and family-centered, and the goals of care really dictate the plan, and we are very cost-effective. The other one is hospice.”
Allan Ramsay

MedPAC data show that from 2000-09, the number of for-profit hospices increased 142 percent nationwide, while the number of nonprofits decreased by 1 percent. During the same period, the average length of stay grew from 54 days to 84 days. Medicare spending on hospice quadrupled, due to “greater numbers of beneficiaries electing hospice and by longer stays among hospice patients with the longest stays.”

In analyzing the hospice benefit, MedPAC discovered that Medicare’s payment structure makes long hospice stays more profitable than short ones, creating an incentive for extremely long stays. In 2009, beneficiaries in the top 10 percent of the “length of stay” category had stays of around 240 days. The authors of the MedPAC report observed that “the incentives in the current hospice payment system for long stays may have led to inappropriate use of the benefit among some providers” – that is, the providers with the longest stays may have been pushing the envelope in certifying patients’ life expectancies.

“Hospice visits tend to be more frequent at the beginning and end of a hospice episode and less frequent in the intervening period,” the report explains. But Medicare pays a flat daily rate that doesn’t reflect the more intense — and more costly — services required at each end of a stay. The payment structure has a predictable impact on hospice providers’ margins. In 2008, hospices that ranked in the highest fifth for average length of stay had margins of 7 percent; those in the lowest fifth had margins of -10.1 percent. In addition, hospices that ranked in the highest fifth for patient volume had average margins of 6 percent; those with the lowest, margins of -9.8 percent. Urban hospices had margins of 5.6 percent; rural ones had margins of 1.3 percent. For-profits had margins of 10 percent; nonprofits had average margins of 0.2 percent.

MedPAC urges changing the payment structure to provide lower daily payments as the length of stays increase and higher daily payments at the beginning and end of stays, making payments correspond to costs. It also recommends that Congress increase the hospice payment rate by one percent.

Hospice economics in Vermont

The average length of stay at Vermont’s small, rural nonprofit hospices was 7.9 days in 2001-05. The MedPAC analysis suggests that they are on an unsustainable trajectory.

“Hospice costs are U-shaped,” explained Ramsay, who has been a hospice medical director off and on for 20 years. “You spend a lot of money when you admit someone. You hope they stay in for a long enough period of time so that you save some money, (and) you spend all the money at the end when you’re providing the most intense services. When the length of stay decreases, which it has, hospice programs struggle to survive.”

Vermont’s nonprofit hospices are mission-driven rather than profit-driven, and they aren’t in a position to game the system: They don’t have big margins, and they adhere closely to the six-month-prognosis rule — they can’t risk having to pay money back to Medicare for incorrectly certifying patients.

The restrictions are “hurting states like us who primarily provide not-for-profit services, as opposed to huge hospice programs in Arizona and Florida and California where they can distribute the costs and they can document things through very sophisticated ways to justify longer hospice stays for people with benign diseases,” Ramsay said. “We are struggling with admission criteria that … don’t fit the current way we provide health care to those with life-limiting conditions.”

When Ramsay testified at the hearings on H.201, he stressed the importance of getting a waiver to extend hospice eligibility to patients with a life expectancy of 12 months, “because then you can get heart failure patients, and COPD patients and dementia patients into hospice care with longer stays which will help our home health agencies survive. When the length of stay is declining, we will lose money on our hospice patients.”

The arguments for ensuring the health of Vermont’s hospice services extend beyond providing compassionate care for the dying, as central as that is. Ramsay maintains that hospice and palliative care serve as a model for health care reform.

“If you look at entire heath care system, there are only two systems of care that have accomplished everything that we want to accomplish with overall healthcare reform,” he said. “One of them is palliative care, because we have very high quality ratings, we’re patient- and family-centered, and the goals of care really dictate the plan, and we are very cost-effective. The other one is hospice.”

The model of the patient-centered medical home, as good as it is, doesn’t deal with the highest-cost patients, he maintains. “Palliative care gets involved in that 10 percent of Medicare recipients who consume 80 percent of the costs,” he said. “There’s no other part of this fee-for-service system that’s accomplished these goals. It’s just us.”