Medicare Hospice Cap


The Hospice Cap is an outdated and too-blunt tool for the post-1998 hospice world, and poses a significant and unnecessary barrier to timely patient access.  The Cap undercuts Congress’ intent that non-cancer patients have equal access to hospice, and breaks Congress’ explicit promise that all eligible patients are entitled to unlimited days of hospice care, until they die in their own good time.

The Cap harms non-cancer patients’ access to hospice; the argument that the Cap is the only tool available to Medicare to guard against hospices admitting ineligible patients is ridiculous. Today, CMS has the knowledge and resources to introduce evidence-based hospice eligibility criteria – National Coverage Determinations (NCDs).  Well-designed NCDs, supplemented by local physician judgment, would:  (a) reduce the enormous current gaps in hospice access among states, diagnoses and races; (why is access 80% for white cancer patients in Arizona, and 10% for black non-cancer patients in New York?);  (b) reduce extraordinarily long individual stays; (c) improve timely admission for the 75% of Medicare hospice users whose stays are currently far too short, and (d) hopefully end the ugly and ill-informed speculation that long individual stays in hospice are the result of hospice misbehavior.

The Cap also poses the largest and most volatile economic threat to most hospices today.   Hospice leaders should make absolutely sure they know what they are doing with respect to their Cap risks.  If they have any doubt at all, they should get the very best advice available as soon as possible, before they need it. A hospice that admits all eligible patients it encounters, that provides those patients with superb (and cost-effective) care as long as they remain eligible until they die in their own good time, that manages its resources wisely, and that creates a great working environment for its employees has still not done enough to insure its own survival.  The Aggregate Hospice Cap is an outdated, too-blunt weapon just waiting to inflict a fatal wound on even the best-intentioned hospice.


In 1982, Congress initiated hospice as a Medicare benefit to offer terminally ill patients and their families an informed choice between interdisciplinary palliative care and conventional hospital-based acute care. Hospice’s objective is to improve the quality of life of each patient and family, when curing the illness is no longer a realistic goal.

In contrast to recent MedPAC claims, there is actually little evidence that Congress’ primary motive was cost savings;  hospice was proposed, and subsequently confirmed, as gold-standard healthcare for the terminally ill.  Claims that hospice would reduce Medicare’s total end-of-life care costs provided needed bi-partisan political support in the difficult economic environment in the early 1980’s, and had the added benefit of being true.

Substitution Benefit
When a Medicare beneficiary elects hospice care, that beneficiary agrees to forego all other Medicare coverage related to his or her terminal diagnosis. Medicare pays the hospice a flat fee per patient per day regardless of the patient’s diagnosis, acuity or the actual cost of care, and Medicare shifts all actual costs of care related to the patient’s terminal illness to the hospice.  In return the hospice agrees to plan and deliver an individualized plan of care to each patient and family.  The hospice is responsible for all costs “related” to the terminal diagnosis.

Hospices’ Revenues and Individual Patients’ Days of Care Capped in 1982
The 1982 law provided that total payments to a hospice provider in any year may not exceed an aggregate Cap, calculated as the product of an individual flat dollar “Cap Amount” (indexed annually for inflation) and the “number of Medicare beneficiaries” served by the hospice in that year.  The intent was to protect Medicare against the aggregate average cost of the new hospice benefit, compared to the aggregate average cost of caring for these patients in a conventional acute care setting.  The Cap was intended as a limitation on total annual payments to a hospice, not a limit on payment for any one individual.  The Cap Amount was (in theory only) based on the cost of caring for a cancer patient for six months.  Because hospices are reimbursed at a flat rate per patient per day, the Aggregate Hospice Cap in effect penalizes hospices as the average length of stay of their patients increases.

Virtually no hospice exceeded the Cap until after 1998, for three reasons:  (i) through the mid-1990’s, hospice served primarily cancer patients, whose hospice stays are predictably short, (ii) physicians were reluctant to certify non-cancer patients as having a life expectancy of six months, because of the inherent uncertainty involved and fear of regulatory 20/20 hindsight should they certify patients who subsequently “live too long,” and (iii) in 1982 Congress also placed a lifetime 210 day Cap on the hospice care any individual could receive. Congress finally removed any Cap on individuals in 1998.

Congress and Medicare Dramatically Expanded the Hospice Benefit in 1998
Between 1982 and 2005, Congress and Medicare repeatedly acted to increase hospice coverage, access and utilization, including a dramatic expansion in 1998:

  • Initially, in 1982, patient eligibility was defined as “terminally ill with a likely prognosis of six months or less,” and each individual beneficiary’s hospice benefit was capped at 210 days, consisting of two 90 day benefit periods followed by a final 30 day benefit period.
  • In 1989 hospice eligibility was liberalized to “…six months or less if the illness runs its normal course” to demonstrate regulatory understanding that terminal prognoses were uncertain and that many terminally ill hospice patients would routinely live longer than 180 days after admission.
  • In 1998, Congress removed any limit on an individual’s hospice days by providing unlimited 60 day eligibility periods, as long as a physician continues to “certify” the patient is terminally ill.  At the same time, Medicare developed objective eligibility criteria for non-cancer diagnoses to encourage physicians to certify eligible non-cancer patients for hospice.

The 1998 expansion reflected Congress’ intent to establish equal access to hospice for the 78% of Medicare decedents who die of non-cancer causes, which in effect quadrupled the number of potential hospice users, and shifted the patient mix toward those who were more likely to remain on the hospice benefit for a longer period before dying.

In September 2000 CMS Administrator Nancy Ann DeParle sent a letter to the nation’s hospices promoting more timely access for non-cancer patients, and explicitly said that Medicare was unconcerned about length of stay as long as patients met CMS’ objective eligibility criteria:

“… There is a disturbing misperception that hospices and beneficiaries will be penalized if a patient lives longer than six months. Nothing could be further from the truth … Let me be clear. In no way are hospice beneficiaries restricted to six months of coverage. There is no limit on how long an individual beneficiary can receive hospice services as long as they meet the eligibility criteria.”

Hospice Growth
In the three year period leading up to 1998, hospice grew less than 5% per year and access to hospice was below 20%.  Not surprisingly, immediately following the 1998 expansion, hospice jumped to a 20% per year growth rate, and continued that growth rate through 2007.  This growth was clearly due to the 1998 expansion of the benefit, the rapid increase in non-cancer admissions and the related growth in length of stay that followed almost immediately.

Interestingly, and contrary to MedPAC’s theories of hospice growth, the overwhelming majority of the growth in hospice days of care during this period was provided not by new hospices, but by hospices well established before 2000, mainly not-for-profit providers. MedPAC’s oft-repeated suggestion that “new-for-profit-hospices-admitting-ineligible patients” were responsible for a material portion of hospice growth between 1998 and 2007 simply isn’t supported by any numbers we’ve ever seen.

From 1998 until 2007, hospice access more than doubled, from about 20% to over 43%.  Hospice access has been largely stagnant from 2007 through 2013 at 43% to 45%.  Today most of America’s terminally ill seniors, more than one million people each year, still receive their end-of-life care in and out of hospitals and other acute care facilities, or they receive no care at all.

Hospice Length of Stay
In 2009 NAHA commissioned an independent healthcare advisory firm to compile the first rigorous analysis of hospice length of stay, including methodologies that measured true patient length of stay across years.  The research compiled Medicare admissions, deaths and length of stay data for 100% of the 3.5 million Medicare beneficiaries who were first admitted to hospice from 1/1/2003 through 12/31/2007.  The data was segmented by year of admission, by state, by diagnoses, by race and by certain other measures.  In addition, NAHA was given access to detailed patient and referral level data from some of its 500 members.

Observations from NAHA’s proprietary national length of stay data base contradict some of MedPAC’s claims, and include:

  • Average length of stay (ALOS) jumped almost immediately following the 1998 benefit changes, from under 60 days in 1998 to 82 days for patients admitted in 2003.
  • ALOS of patients admitted from 2003 through 2007 did not grow, and ALOS of patients admitted since 2008 has probably been declining.  ALOS by quartile also did not increase much after the “class of 2003.”
  • Not surprisingly, cancer patient ALOS is half that of non-cancer patients.
  • ALOS of patients referred to hospice while in a hospital bed is 50% below that of all other patients, regardless of diagnosis.
  • 1st Quartile ALOS is 3 to 4 days, regardless of diagnosis.
  • 4th Quartile ALOS for all patients is about 270 days; 340 days for all non-cancer patients, and 450 days for Alzheimer’s patients.
  • 5% of Alzheimer’s patients stay over two years.

Congress Didn’t Update the Cap in 1998, and “Cap Demands” Exploded
Congress did not modify the Aggregate Hospice Cap in 1998, even as they extended equal access to longer-stay non-cancer beneficiaries, and gave beneficiaries unlimited hospice days.  An unfunded mandate was born.

Cap Demands grew geometrically from 1999 through  2007.  CMS doesn’t disclose total Hospice Cap Demands, but some CMS contractors do release partial data.  Estimates suggest that Cap Demands grew from nearly zero in 1998, to over $50 million in 2003, doubled to $100 million in 2004, redoubled to $200 million in 2005, grew 50% more to $300 million in 2006, and remained in the $200 million range from 2007 through 2010, each year affecting at least 15% of all hospices nationally, and 30% of all hospices in the South where non-cancer access is generally higher.  Since 2012, it appears that Cap Demands have resumed their growth.
CMS Cap Demands require a hospice to repay to Medicare monies the hospice received for providing covered hospice services to Medicare beneficiaries who were statutorily entitled to receive those services.  Hospice may be unique in Medicare as the only significant benefit in which the provider is capped, but the beneficiary is promised unlimited service as long as CMS’ objective eligibility criteria are met.  Demands often amount to 25% or more of a hospice’s annual revenues.  Most hospices don’t have this cash; they spent it caring for eligible terminally-ill beneficiaries.

Regulatory Reform / Litigation
In 1983 HHS wrote regulations that implemented the 1982 Hospice Statute.  The regulations relating to the Cap were inconsistent with the underlying statute but went unchallenged for almost 25 years.  In 2007, NAHA-member hospices began to file suits against Medicare in Federal Court claiming that the 1983 regulation that defined the Cap calculation was inconsistent with the 1982 Statute, and harmed hospices by overstating Cap Demands.  By 2011 over 100 hospices had filed suit in multiple Federal District Courts, including the Washington D.C. District.  No Federal District Court or Federal Appellate Court ruled against the hospices.

The 1982 Statute provided that total payments to a hospice provider in any fiscal year may not exceed an aggregate cap, calculated as the product of a flat “Cap Amount” and the “number of Medicare beneficiaries” in a hospice program in any given accounting year.   The Statute required that the “number of beneficiaries” calculation be prorated so that Cap Allowances for patients who lived across accounting years would be spread across those years:

“ …For purposes of this calculation, the “number of Medicare beneficiaries” in a hospice program in an accounting year is equal to the number of individuals who have made an election under subsection (d) of this section with respect to the hospice program and have been provided hospice care … by the hospice program under this part in the accounting year, such number reduced to reflect the proportion of hospice care that each such individual was provided in a previous or subsequent accounting year or under a plan of care established by another hospice program…”

But, HHS’ 1983 regulation chose to ignore the statute and instead give providers credit for each patient’s Cap Allowance only in the year admitted, regardless of whether the patient lived into the following year.  In 1983 HHS wrote:

“ … The statute specifies that the number of Medicare patients used in the calculation is to be adjusted to reflect the portion of care provided in a previous or subsequent reporting year or in another hospice … With respect to the adjustment necessary to account for situations in which a beneficiary’s election overlaps two accounting periods, we are proposing to count each beneficiary only in the reporting year in which the preponderance of the hospice care would be expected to be furnished rather than attempt to perform a proportional adjustment” 

    (emphasis added)

By not prorating the beneficiary calculation across years for patients who live across years, allowances are not matched with revenues generated and Cap Allowances are routinely “trapped” in early years and understated in later years, thus overstating Cap Demands.

Following years of losses in many Federal District Courts and two Federal Courts of Appeal, HHS announced in April 2011 that it would issue a new regulation with a Proportional Cap calculation.  HHS issued the final regulation in October 2011 and at the same time informally suspended issuing new Cap Demands, while they settled existing Cap litigation.  In May 2012, CMS contractors announced that they would resume issuing Cap Demand letters.

Legislative Reform
NAHA was founded in 2007 as an independent voice for fiscally responsible reform, and initially focused on advocating for regulatory and legislative reform of the hospice Cap.  NAHA-member hospices secured regulatory reform in 2011 through the Federal Courts; legislative reform was more elusive.

There was considerable consensus that the 1982 Cap creates unintended barriers to access for eligible non-cancer patients.  (National hospice organizations are on record as saying the Cap harms patient access.)  But there was no consensus on how to reform the Cap.  NAHA introduced one attempt at fiscally responsible legislative reform with bipartisan support:  The Medicare Hospice Reform and Savings Act of 2009 (HR3454).  HR3454 would have reformed the Cap, improved patient access to hospice, and at the same time reduced Medicare’s total costs by $1 billion to $2.5 billion annually.   But, HR3454 was not incorporated into the 2009 Affordable Care Act and was never brought to the floor for a stand-alone vote in Congress.

We believe legislative Cap reform is unlikely in the foreseeable future.

The “Proportional” Cap methodology is likely to benefit virtually all hospices. The new Proportional Cap calculation more accurately matches allowances to revenue by ensuring that, for each patient served, the hospice provider receives allowances in proportion to the days of service provided to that patient in each year, rather than entirely in that patient’s initial year of service.  Over time, fewer allowances will be “trapped” and lost in prior years.

We have yet to see any credible evidence that, over time, any provider would be harmed by the new Proportional Cap methodology.  Spreadsheets filed in Federal Court by litigating hospices, and even by CMS, show material reductions of Cap liability under the proportional method.

The Old Cap (now euphemistically called “streamlined”) method allocates allowances entirely to the first year in which each patient receives hospice care.  This method tends to overstate allowances in earlier years, and understate allowances and overstate Demands in later years. Surplus allowances are “trapped” and not allowed to flow forward with patients as they live into subsequent years.

The relative complexity of the new Proportional Cap methodology is more than outweighed by the benefits.

“Managing the Cap” is too-often code for restricting access for clearly eligible, non-cancer patients, especially those with Alzheimers.  
When the Cap was legislated in 1982, no thought was given to the wisdom of requiring a hospice to actively manage the average length of stay of its terminally ill patients.  That is because in 1982 each patient was limited by a 210 day lifetime cap and hospices served primarily cancer patients.  There was no “patient mix” to manage, and there were no “long-stay” patients.

The 1998 expansion of hospice coverage changed everything.  It: (i) quadrupled the potential hospice population; (ii) dramatically shifted the patient mix toward those who were more likely to remain on the hospice benefit for a longer period before dying; and (iii) in practice, enabled very long individual patient stays.  Today, 78% of all eligible patients are dying from non-cancer causes.  These “non-cancer patients” are legally (and ethically) entitled to receive hospice care for as long as they live.  On average they’ll live twice as long as cancer patients, and a significant percentage will live far longer than 180 days after admission to hospice.

Today, hospices that want to grow, improve hospice choice in their communities, and avoid potentially fatal Cap Demands must operate within outdated length of stay constraints imposed by a thirty year old Cap Statute that has never been updated, and that was designed for hospices that served only cancer patients.   Most believe the Cap forces hospices to restrict access for some eligible non-cancer patients.  Of course, most hospices don’t know this when they first begin.  But the stakes are high; the penalty for not operating within the 1982 Cap’s length of stay constraints is a financially devastating and often fatal Cap Demand.  On the other hand, if a hospice is seen to be unwisely restricting access for eligible, non-cancer patients, it risks harming the hospice’s “brand,” as well as its revenues and its ability to sustain itself financially.  Catch 22 for many hospices.

The Hospice Cap disproportionately affects small hospices and hospices serving rural and ethnic / minority populations, because those hospices are least likely to have referral relationships with hospital systems and oncologists, and most likely to be serving relatively more non-cancer patients.  Patients referred from hospitals and patients with cancer diagnoses are predictably “short-stay” hospice patients, and generate the excess Cap Allowances necessary to cover longer staying patients.

As we noted above, hospice leaders who are not absolutely certain they know what they are doing with respect to the Cap should seek the very best advice they can find, as soon as possible.   A hospice that admits all eligible patients it encounters, and that provides those patients with superb (and cost-effective) care as long as they remain eligible until they die in their own good time, and that manages its resources wisely, and that creates a great working environment for its employees has still not done enough to insure its own survival.  The Aggregate Hospice Cap is an outdated, too-blunt regulatory weapon  just waiting to inflict a fatal wound on even the best-intentioned hospice.  As two nurses who had started their own hospice to serve their inner city ethnic community once remarked at a NAHA meeting:  “who ever thought we were supposed to ‘manage’ the length of stay of our eligible patients?”




Be Sociable, Share!

Leave a Comment

NAHA Releases Length of Stay Research Data to Members