Medicare's Hospice Benefit

This report discusses Medicare's hospice benefit, which was provides care that specializes in the relief of the pain and symptoms associated with a terminal illness and the provision of supportive and counseling services to patients and their families during the final stages of a patient's illness and death.

ȱ
Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ž•’Žȱ˜—Žȱ
™ŽŒ’Š•’œȱ’—ȱ ŽŠ•‘ȱŠ›Žȱ’—Š—Œ’—ȱ
Š›Œ‘ȱşǰȱŘŖŖşȱ
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŝȬśŝŖŖȱ
   ǯŒ›œǯ˜Ÿȱ
ŚŖŚŘşȱ
ȱŽ™˜›ȱ˜›ȱ˜—›Žœœ
Pr
epared for Members and Committees of Congress

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
ž––Š›¢ȱ
Hospice care provides an interdisciplinary approach to services for Medicare beneficiaries with a
terminal illness. This care specializes in the relief of the pain and symptoms associated with a
terminal illness and in the provision of supportive and counseling services to patients and their
families during the final stages of a patient’s illness and death. The benefit covers a broad range
of services, including prescription drugs for pain control and symptom management, skilled
nursing care, physician services, home health aide services, homemaker services, patient
counseling, and family bereavement counseling. Services are provided primarily in the patient’s
home, but may also be provided in institutional settings, such as nursing homes. Hospice care is
provided in lieu of most other Medicare services related to the curative treatment of the terminal
illness.
For a person to be considered terminally ill and eligible for Medicare’s hospice benefit, the
beneficiary’s attending physician and the medical director of the hospice (or physician member of
the hospice team) must certify that the individual has a life expectancy of six months or less.
Beneficiaries electing hospice are covered for two 90-day periods, followed by an unlimited
number of 60-day periods.
Medicare payments to hospices in 2007 totaled $10.1 billion, having more than tripled since
2000. Medicare spending for hospice is expected to continue growing and to more than double by
2018, reaching a projected $21 billion and outpacing the projected growth rates for Medicare
payments in hospitals, skilled nursing facilities, physician services, and home health care. Growth
in spending to date has been driven, in part, by increased utilization of hospice as well as
spending per hospice user. For example, spending per user grew between 2004 and 2005 by 8%.
Growth in spending per user may be in part a result of increasing lengths of stay among certain
hospice providers.
The number of hospices participating in Medicare also grew by 33.4% during the four-year
period from 2003 to 2007. As of 2007, for-profit hospices constituted the majority of these
hospices, and since 2000, made up over 90% of hospices participating in Medicare.
Medicare pays hospices using a prospective payment system containing four categories of daily
rates, which are predetermined, fixed amounts intended to pay for the costs of care for a hospice
beneficiary, on average. These amounts are adjusted annually by the hospice market basket.
Hospice payments are also adjusted for geographical differences. Total payments to hospices may
not exceed an aggregate per beneficiary cap amount.
Some analysts have expressed concerns about Medicare margins earned by certain types of
hospice providers, the growing number of hospices exceeding the aggregate per beneficiary cap,
increasing lengths of stay, and the three-year phase out of the budget neutrality factor authorized
under regulation in August of 2008. All of these topics are discussed in this report, which will be
updated as necessary.

˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
˜—Ž—œȱ
Background ..................................................................................................................................... 1
Description of Hospice Coverage.................................................................................................... 1
Spending and Utilization ................................................................................................................. 2
Participating Hospices ..................................................................................................................... 5
Prospective Payment System........................................................................................................... 6
Unit of Payment ........................................................................................................................ 7
Wage Index Update ................................................................................................................... 8
Aggregate Per Beneficiary Cap................................................................................................. 9
Current Issues ................................................................................................................................ 10
Medicare Margins ................................................................................................................... 10
Increase in Hospices Exceeding the Aggregate Per Beneficiary Cap ......................................11
Increasing Length of Stay ....................................................................................................... 13
Budget Neutrality Adjustment for FY2009............................................................................. 13

Š‹•Žœȱ
Table 1. Hospice Spending and Beneficiary Counts ....................................................................... 3
Table 2. Measures of Medicare Hospice Care, Selected Fiscal Years ............................................. 4
Table 3. Number of Hospices Participating in Medicare, by Type of Hospice ............................... 6
Table 4. Hospice Aggregate Medicare Margins ............................................................................ 10
Table 5. Average Days Per Hospice Patient, by Disease Category, Below-Cap and Above-
Cap Hospices, All Diagnoses ..................................................................................................... 12

˜—ŠŒœȱ
Author Contact Information .......................................................................................................... 14

˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
ŠŒ”›˜ž—ȱ
Medicare’s hospice benefit was established as a trial program under the Tax Equity and Fiscal
Responsibility Act of 1982 (TEFRA, P.L. 97-248) for years 1983 through October 1, 1986, and
made permanent under the Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA,
P.L. 99-272). Using an interdisciplinary team, the hospice benefit provides care that specializes in
the relief of the pain and symptoms associated with a terminal illness and the provision of
supportive and counseling services to patients and their families during the final stages of a
patient’s illness and death. In general, the objective of the benefit is to make a patient’s last days
as comfortable as possible.1
Medicare covers hospice care for terminally ill beneficiaries instead of most other Medicare
services related to the curative treatment of their illness. Beneficiaries who elect hospice may still
receive curative treatments for illnesses or injuries unrelated to their terminal illness, and they
may disenroll from hospice at any time.
ŽœŒ›’™’˜—ȱ˜ȱ ˜œ™’ŒŽȱ˜ŸŽ›ŠŽȱ
The hospice benefit covers a broad range of services, including prescription drugs for pain control
and symptom management, skilled nursing care, physician services, home health aide services,
homemaker services, patient counseling, and family bereavement counseling. Services are
provided primarily in the patient’s home but may also be provided in institutional settings, such as
nursing homes.
For a person to be considered terminally ill and eligible for Medicare’s hospice benefit, the
beneficiary’s attending physician and the medical director of the hospice (or physician member of
the hospice team) must certify that the individual has a life expectancy of six months or less.
Beneficiaries electing hospice are covered for two 90-day periods, followed by an unlimited
number of 60-day periods. The medical director or physician member of the hospice team must
recertify at the beginning of each period that the beneficiary is terminally ill. Services must be
provided under a written plan of care established and periodically reviewed by the individual’s
attending physician and the medical director of the hospice.
Covered hospice services include (1) nursing care provided by or under the supervision of a
registered nurse; (2) physical or occupational therapy or speech-language pathology services; (3)
medical social services; (4) services of a home health aide who has successfully completed a
training program approved by the Secretary of Health and Human Services (HHS); (5)
homemaker services; (6) medical supplies (including drugs and biologicals) and the use of
medical appliances; (7) physician services; (8) short-term inpatient care (including both respite
care and procedures necessary for pain control and acute and chronic symptom management); (9)
counseling, including dietary counseling, for care of the terminally ill beneficiary and for family
adjustment to the patient’s death (bereavement counseling is not a reimbursable service); and (10)
any other item or service that is specified in a patient’s plan of care for which Medicare can pay.

1 For a broader view of end-of-life issues, see CRS Report R40235, End-of-Life Care: Services, Costs, Ethics, and
Quality of Care
, coordinated by Kirsten J. Colello.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
Medicare’s hospice benefit is intended to be principally an in-home benefit. For this reason,
Medicare law prescribes that respite care, or relief for the primary care giver of the terminally ill
patient, may be provided by hospice, yet only on an intermittent, nonroutine, and occasional
basis, and may not be provided consecutively over longer than five days. In addition, a payment
cap has been established in which the aggregate number of inpatient care days provided in any
12-month period to Medicare beneficiaries, for medical reasons other than the reason for claiming
hospice, cannot exceed 20% of the total number of days of hospice coverage provided to these
persons.
Only two covered hospice services—outpatient drugs and biologicals and respite care—are
subject to coinsurance. Outpatient drugs and biologicals are subject to a coinsurance amount that
approximates 5% of the cost of the drug to the hospice program, except that the amount may not
exceed $5 per prescription. For respite care, coinsurance equals 5% of program payments for
respite, but may not exceed Medicare’s inpatient hospital deductible during a hospice coinsurance
period (defined as the period when hospice election is not broken by more than 14 days).
Covered services must be provided by a Medicare-certified hospice. Certified hospices must be
either public agencies or private organizations primarily engaged in providing covered hospice
services and must make services available on a 24-hour basis, in individuals’ homes, on an
outpatient basis, and on a short-term inpatient basis. Hospices must routinely and directly provide
substantially all of the following “core” services: nursing care, medical social services, and
counseling services. The remaining hospice services may be provided either directly by the
hospice or under arrangements with others. If services are provided through arrangements with
other providers, the hospice must maintain professional management responsibility for all such
services, regardless of the facility in which the services are furnished.
The hospice program must also have an interdisciplinary group of personnel that includes at least
one registered professional nurse and one social worker employed by the hospice, one physician
employed by or under contract with the hospice, and at least one pastoral or other counselor.
Medicare also requires each hospice to maintain a volunteer staff sufficient to provide
administrative or direct patient care in an amount equal to at least 5% of the total paid patient care
hours of all paid hospice employees and contract staff.2
™Ž—’—ȱŠ—ȱ’•’£Š’˜—ȱ
Medicare payments to hospices in 2007 totaled $10.1 billion, having more than tripled since 2000
and increased by an average annual rate of 20%. Medicare spending for hospice is expected to
continue growing and to more than double by 2018, reaching a projected $21 billion and
outpacing the projected growth rates for Medicare payments in hospitals, skilled nursing
facilities, physician services, and home health care.3

2 42 C.F.R. § 418.70(e)
3 MedPAC, entitled “A Data Book: Healthcare Spending and the Medicare Program,” June 2008 and Centers for
Medicare and Medicaid Services (CMS), HHS, “Medicare Hospices to See Increase in 2009 Payments, CMS
Announces Changes to Wage Index Calculation also to be Made,” press release, July 31, 2008,
http://www.cms.hhs.gov/apps/media/press/factsheet.asp?Counter=3218&intNumPerPage=10&checkDate=&checkKey
=&srchType=1&numDays=3500&srchOpt=0&srchData=&keywordType=All&chkNewsType=6&intPage=&showAll
=&pYear=&year=&desc=&cboOrder=date.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
Řȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
Growth in spending to date has been driven, in part, by increased utilization of hospice. Between
2000 and 2007, for example, hospice utilization grew by nearly 95%, from 513,800 persons to 1
million persons. This increase represented an average annual growth rate of 10%. Table 1
provides spending and beneficiary count trend data from 2000 to 2007.
Table 1. Hospice Spending and Beneficiary Counts
2000-2007

Average
Annual
Percentage
2000 2001 2002 2003 2004 2005 2006 2007 Change
Spending
$2.9 $3.6 $4.5 $5.7 $6.9 $8.2 $9.2 $10.1 —
(billions)
Percentage
— 24.1% 25% 26.7% 21.1% 18.8% 12.1% 9.8%
20%
change
Beneficiaries
513.8 579.8 643.3 713.4 791.6 869.2 912.6a 1,000

(thousands)
Percentage
— 12.8% 11% 10.9% 11% 9.8% N/A 9.6% 10%
change
Source: Based on Centers for Medicare and Medicaid Services, Standard Analytical Files, 100 percent Final
Action Claims. Aggregate summary tables available at http://www.cms.hhs.gov/ProspMedicareFeeSvcPmtGen/
downloads/FY05update_hospice_expenditures_and_units_of_care.pdf and transcript and presentation materials
from a MedPAC public meeting on November 6, 2008. Numbers from 2007 are reported in MedPAC’s
publication, “Medicare Payment Policy,” March 2009.
Note: Numbers have been rounded. N/A = not available.
a. The 2006 beneficiary number is calendar year and is not directly comparable to all other numbers, which
are fiscal year.
Additional details provided in Table 2 allow for a more complex analysis of spending growth.
Although Table 1 (years 2000-2007) and Table 2 (years 2002-2005) display different year
parameters, both report the same totals for hospice spending and beneficiary counts.
As shown in Table 2, payments to freestanding hospices (i.e., independent hospices that are not a
part of a hospital, skilled nursing facility, or home health agency) experienced the largest payment
increase between 2002 and 2005, almost doubling from a total of $2.9 billion to a total of almost
$5.8 billion. The overall growth in spending between 2002 and 2005 was fueled, in large part, by
the increase in spending on routine home care. Between 2002 and 2005, spending on routine
home care grew by about $3.3 billion. Spending on general inpatient care grew by $267 million,
spending on continuous home care grew by $78 million, spending on physician services grew by
$25 million, and spending on inpatient respite care grew by $5 million. Definitions of these
spending categories can be found under the “Prospective Payment System” section of this report.
Spending growth was also, in part, due to a 10% growth in beneficiaries served overall between
2002 and 2005. A significant difference in spending growth can be seen among provider types.
Whereas the number of beneficiaries served by freestanding providers grew by almost 50%
between 2002 and 2005, the number of beneficiaries served by hospital-based providers grew by
15.4%, and by home health agency-based (HHA-based) providers by 13.8%. No growth in the
number of beneficiaries served by skilled nursing facility-based (SNF-based) providers was
experienced during this period.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
řȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
Furthermore, growth can also be seen in spending per hospice user between 2002 and 2005.
Between 2002 and 2005, the average dollar spent per beneficiary also grew by 33.6%, from
$7,021 per beneficiary in 2002 to $9,382 per beneficiary in 2005. Between 2004 and 2005 alone,
spending per user grew by 8%.4
Since Medicare pays for hospice on a per diem basis, spending per enrollee is largely a function
of the length of time a patient is enrolled (often referred to as length of stay or the average
number of days a beneficiary elects hospice care, as shown in Table 2). The average length of
hospice stays grew by 20.4% between 2002 and 2005. Freestanding hospice providers are among
the providers with the longest length of stays (69 days), compared with SNF-based providers
(54.9 days), HHA-based providers (53.6 days), and hospital-based providers (52.6 days).
Table 2. Measures of Medicare Hospice Care, Selected Fiscal Years
2002-2005

Category
2002 2003 2004 2005
Cash outlays by provider type (in millions)

Freestanding
$2,912 $3,819 $4,725 $5,771
Hospital Based
$649
$756
$876
$967
SNF Based
$22
$28
$32
$29
HHA Based
$934
$1,081
$1,264
$1,389

Total
$4,517 $5,682 $6,897 $8,155
Cash outlays by care type (in millions)
Routine Home Care
$3,930
4,954
$6,056
$7,182
Continuous Home Care
$70
$93
$122
$148
Inpatient Respite Care
$9
$10
$12
$14
General Inpatient Care
$470
$576
$649
$737
Physician Services
$38
$49
$58
$73

Total
$4,517 $5,682 $6,898 $8,154
Number of beneficiaries by provider type

Freestanding
393,061 453,712 510,821 583,254
Hospital Based
101,826
107,206
115,116
117,597
SNF Based
3,367
3,581
3,854
3,335
HHA Based
145,049
148,901
161,777
165,015

Total
643,303 713,400 791,568 869,201
Average dollar amount per beneficiary

Freestanding
$7,408 $8,416 $9,249 $9,894
Hospital Based
$6,369
$7,048
$7,610
$8,220
SNF Based
$6,567
$7,736
$8,392
$8,652

4 MedPAC, “Report to Congress: Reforming the Delivery System,” June 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
Śȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ

Category
2002 2003 2004 2005
HHA Based
$6,440
$7,256
$7,815
$8,417

Total
$7,021 $7,965 $8,713 $9,382
Average number of days a beneficiary elects hospice care

Freestanding
57.2 62.5 66.5 69.0
Hospital Based
47.6
49.7
51.3
52.6
SNF Based
43.9
50.7
53.9
54.9
HHA Based
45.6
48.5
50.8
53.6

Total
53.0 57.6 61.0 63.8
Source: Centers for Medicare and Medicaid Services, Standard Analytical Files, 100 percent Final Action Claims.
Note: Totals may not add due to rounding.
Among other reasons, increasing length of stay overall may be attributed to changes in the patient
mix of diagnoses over time. Specifically, the majority of patients electing hospice in the earlier
years of the benefit had cancer and other relatively acute conditions. These conditions resulted in
more certain diagnoses about life expectancy and relatively shorter lengths of stay. Over time, the
patient mix has changed, and increasingly more persons with other conditions, such as dementia,
nonspecific debility, and congestive heart failure, are electing hospice. These patients have less
certain life expectancy prognoses and tend to have longer lengths of stays.5
Not all hospices experienced increases in their length of stay, and not all lengths of stay have
increased. In fact, MedPAC conducted an analysis of length of stay among providers and found
that only those providers that experienced lengths of stays above the median of 15 days in 2005
experienced significant increases in their lengths of stay. Specifically, MedPAC found that
hospices with lengths of stay that were shorter than the median of two weeks found their length of
stay unchanged between 2000 and 2005. Hospices with lengths of stay above the median,
however, grew longer. For example, length of stay among those hospices with an average length
of stay of 144 days in 2000 grew by 47% to 212 days in 2005.6
Analyses by MedPAC found that hospice providers with the longest stays on average have the
highest Medicare margins (see “Medicare Margins” section).7 This is discussed further in the
“Current Issues” section of this report and is currently being investigated by MedPAC.
Š›’Œ’™Š’—ȱ ˜œ™’ŒŽœȱ
A Medicare Payment Advisory Commission (MedPAC) report showed that in December 2007,
3,240 hospices across the country participated in Medicare. Table 3 shows that this number grew
by 33.4% during the four-year period from 2003 to 2007 (from 2,428 providers in December
2003 to 3,240 providers in December 2007). Of all types of providers, including freestanding,
hospital-based, skilled nursing facility-based (SNF-based) and home health agency-based (HHA-

5 MedPAC, “Report to Congress: Reforming the Delivery System,” June 2008.
6 A Data Book: Healthcare spending and the Medicare program, June 2008.
7 MedPAC transcript of Public Meeting on January 9, 2009.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
śȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
based), the largest growth rate occurred among freestanding hospices. Between 2003 and 2007,
the number of freestanding hospices grew by 68.9%, from 1,200 in 2003 to 2,027 in 2007.
Hospices that were part of skilled nursing facilities had the second highest growth rate, at 12.5%
between this four-year period. Hospices that were part of hospitals increased in number slightly,
and hospices that were part of home health agencies decreased in number (see Table 3).
Table 3. Number of Hospices Participating in Medicare, by Type of Hospice
2003-2007

Number of Hospices
Date Freestanding Hospital-Based SNF-Based HHA-Based Total
12/03
1,200 559 16
653
2,428
12/04
1,393 560 14
653
2,620
12/05
1,615 553 13
671
2,852
11/06
1,824 564 14
652
3,054
12/07
2,027 564 18
631
3,240
Source: Provided to CRS by the Centers for Medicare & Medicaid Services in spring of 2008.
As of 2007, for-profit hospices constituted 50% of the total number of hospices participating in
Medicare, having grown by 119% since 2000. Of all for-profit hospices in 2007, 43% were
freestanding and 8% were provider-based. Although for-profit hospices comprised half of all
Medicare-participating hospices, they served 41% of hospice beneficiaries in that year. In
contrast, non-profit hospices (16% of which were freestanding and 21% of which were provider-
based) comprised 37% of total Medicare-participating hospices and served 54% of hospice
beneficiaries.8
›˜œ™ŽŒ’ŸŽȱŠ¢–Ž—ȱ¢œŽ–ȱ
Medicare payments to hospices are prospectively determined for each case. The prospective
payment methodology for hospice was established in 1983. This prospective payment system
(PPS) pays hospices according to the general type of care provided to a beneficiary on a daily
basis. Unlike other Medicare PPSs, there is no additional adjustment for case mix. Like other
PPSs that pay health care providers for care to Medicare beneficiaries on the basis of
predetermined, fixed amounts, Medicare payments to hospices are intended to pay for the costs of
care for a hospice beneficiary, on average. That is, although the payment is a predetermined daily
rate, a hospice’s actual costs may be above or below that amount for an individual patient.
Facilities have an incentive to manage costs so that, on average, costs do not exceed the PPS
average amounts. Hospices that provide the services at lower costs than the Medicare payment
are able to keep the difference. Recent analyses by MedPAC have raised questions about the daily
rate methodology.9

8 MedPAC, “Medicare Payment Policy,” March 2009.
9 MedPAC, The Medicare Hospice Payment System: A Preliminary Consideration of Potential Refinements, conducted
by Rand for MedPAC, 06-4, June 2006; and MedPAC, Medicare’s hospice benefit: Recent trends and consideration of
payment system refinements, June 2006.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
Ŝȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
Payments for hospice care contain three separate components that are adjusted annually. These
components are daily payment rates, the hospice wage index, and a cap amount (maximum
profit). Payment rates are based on one of four prospectively determined units of payment, which
correspond to four different levels of care (i.e., routine home care, continuous home care,
inpatient respite care, and general inpatient care) for each day a beneficiary is under the care of
the hospice. The units of payments are updated annually by the hospice market basket. The
hospice wage index is used to adjust the unit payment rates to reflect local differences in area
wage levels. Total payments to a hospice are subject to an aggregate cap. These components are
described below.
—’ȱ˜ȱŠ¢–Ž—ȱ
Under the hospice prospective payment, hospices are paid one of four prospectively determined
rates, which correspond to four different levels of care for each day a Medicare beneficiary is
under the care of the hospice. Payment will thus vary by the length of the patient’s period in the
hospice program as well as by the characteristics of the services (intensity and site) furnished to
the beneficiary. Each rate is adjusted for the geographic location in which the service is delivered
to account for variations in area wages, as described below.
The four rate categories are as follows:
Routine Home Care. Routine home care payment is made for a day on which an
individual is at home and is not receiving continuous home care. The routine
home care rate is paid for every day a patient is at home and under the care of the
hospice, regardless of whether the hospice actually visits the home and regardless
of the volume or intensity of the services provided on any given day as long as
fewer than eight hours of care is provided. The FY2009 base routine home care
payment is $139.97. In 2005, 88% of spending on hospice paid for routine home
care (see Table 2).
Continuous Home Care. Continuous home care payment is made for a day on
which an individual receives hospice care consisting predominantly of nursing
care on a continuous basis at home. Home health aide or homemaker services or
both may also be provided on a continuous basis. Continuous home care is
furnished only during brief periods of crisis and only as necessary to maintain the
terminally ill patient at home. Home care must be provided for at least eight
hours before it would be considered to fall within the category of continuous
home care. Payment for continuous home care will vary depending on the
number of hours of continuous services provided. The FY2009 base continuous
home care rate is $816.94 for 24 hours or $34.04 per hour. In 2005, 1.8% of
spending on hospice paid for continuous home care (see Table 2).
Inpatient Respite Care. Inpatient respite care payment is made for a day on
which the individual who has elected hospice care receives care in an approved
facility on a short-term (not more than five days at a time) basis for the respite of
his or her caretakers. For FY2009, the base inpatient respite care rate is $144.79
per day. In 2005, 0.2% of spending on hospice paid for inpatient respite care (see
Table 2).
General Inpatient Care. General inpatient care payment is made for a day on
which an individual receives general inpatient care in an inpatient facility for
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŝȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
pain control or acute or chronic symptom management that cannot be managed in
other settings. Care may be provided in a hospital, skilled nursing facility (SNF),
or inpatient unit of a freestanding hospice. The FY2009 base general inpatient
care rate is $622.66 per day. In 2005, 9% of spending on hospice paid for general
inpatient care (see Table 2).
Hospice daily payment rates for routine home care, continuous home care, inpatient respite care,
and general inpatient care are updated annually by the hospice market basket (MB), a measure of
inflation of goods and services used by hospices. Since FY2003, updates have been at the full
hospital MB percentage increase. The FY2009 payment rates will be updated from FY2008 by
the full hospice MB of 3.6%.10
Hospices bill separately for additional physician services not covered under the payment
categories described above. Medicare pays hospices for amounts equivalent to 100% of the
physician fee schedule for those physician services furnished by hospice employees, or under
arrangements with the hospice, that are not covered under the payment categories above.
ŠŽȱ —Ž¡ȱ™ŠŽȱ
Each of the four payment rates has a labor share and a nonlabor share, reflecting the estimated
proportion of each rate that is attributable to wage and nonwage costs. The labor shares are
adjusted for geographic differences in wages by the hospice wage index to reflect differences in
wages in the area where the beneficiary resides.
The current hospice wage index methodology was implemented in 1997 through the rule making
process. The hospice wage index is updated annually and based on the most current hospital wage
data and any changes to the Office of Management and Budget’s (OMB’s) Metropolitan
Statistical Areas (MSA) definitions. Prior to this date, the wage adjustment used a hospice wage
index based on hospital data from the 1981 Bureau of Labor Statistics (BLS). According to the
Secretary, this BLS data had not been updated since 1983.11 The change in 1997 was intended to
improve the data used to account for disparities in geographic location and improve accuracy,
reliability, and equity of Medicare payments to hospices across the country.12 Until FY2009, the
wage index adjustment was subject to an additional budget neutrality adjustment. In its FY2009
final rule for hospice, CMS proposed a phase-out of this budget neutrality adjustment.13 This rule
is controversial and is mentioned again in the current issues section of this report.

10 Centers for Medicare and Medicaid Services (CMS), HHS, CMS Manual, Transmittal 1570, “Update to Hospice
Payment Rates, Hospice Cap, Hospice Wage Index, and the Hospice Pricer for FY 2009,” press release, August 1,
20008, http://www.cms.hhs.gov/Transmittals/Downloads/R1570CP.pdf.
11 Centers for Medicare and Medicaid Services (CMS), HHS, “Medicare Program: Proposed Hospice Wage Index for
Fiscal Year 2009, Proposed Rule,” 73 Federal Register 2400, May 1, 2008.
12 Health Care Financing Administration HCFA), HHS, “Medicare Program; Hospice Wage: Final Rule,” 62 Federal
Register
4280, August 8, 1997.
13 42 CFR Part 418, Medicare Program; Hospice Wage Index for Fiscal Year 2009; Final Rule, August 8, 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
Şȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
›ŽŠŽȱŽ›ȱŽ—Ž’Œ’Š›¢ȱŠ™ŗŚȱ
Medicare law requires that payments to a hospice for care furnished over the period of a year be
limited to a “cap amount.” The cap amount is a per beneficiary amount applied on an aggregate
basis during the period from November 1 through October 31 of the following “cap year” and is
not adjusted for geographic costs differences. It is calculated by multiplying the yearly per
beneficiary cap amount by the total unduplicated number of persons served by an individual
hospice between September 28 of the previous year and September 27 of the current year. The
result is then compared to the actual amount paid to that hospice to determine whether the hospice
was paid an amount in excess of the cap. For example, the average annual payment per hospice
patient cap, ending October 31, 2008, is $22,386. Say that a hospice received $6 million in
Medicare payments in FY2007 and served 250 patients. To determine whether the hospice
received excess payments, an intermediary would multiply 250 by the cap amount of $22,386
(250 x $22,386) and compare that result ($5.6 million) to the actual amount the hospice was paid
by Medicare ($6 million). If the amount paid by Medicare is greater than the cap amount ($6
million - $5.6 million = $403,462), then the hospice would be required to reimburse Medicare for
this difference ($403,462). If the amount paid by Medicare is less than or equal to the cap
amount, no action would be taken and the hospice could retain all of its payments.
The payment methodology and rates for the aggregate per beneficiary cap were developed using
cost data from 26 hospices that were part of a demonstration project run by the Centers for
Medicare and Medicaid Services (CMS, then the Health Care Financing Administration, or
HCFA), from 1980 to 1982. In that demonstration project, patients receiving hospice care
primarily had a diagnosis of terminal cancer and received the hospice benefit for an average
length of stay of 70 days. The cap amount was codified in law in section 1814(i)(2)(A) of the
Social Security Act. For the accounting year that ended after 1984, the statute specifies that the
cap was $6,500. Since then, this amount has been adjusted annually by the medical expenditure
category of the consumer price index for all urban consumers. The hospice aggregate cap amount
for the 2008 cap year is $22,386.15.15
Fiscal intermediaries bill the hospices for any amounts paid above the cap amount. There is about
a two-year lag time between the time the hospice is paid and the time the hospice is billed for a
refund. The lag is likely a result of the time to collect and process the spending and beneficiary
data needed to make these calculations. In 2005, the amount of payments that were returned to
Medicare by hospice providers totaled $166 million, representing 2% of total Medicare spending
on hospice (8.2 billion) in FY2005).16

14 Hospice payments are also subject to another cap limiting the number of inpatient care days that may be provided
during a 12-month period. Specifically, the aggregate number of inpatient care days provided to Medicare beneficiaries
by a hospice in any 12-month period, for medical reasons other than the reason for claiming hospice, cannot exceed
20% of the total number of days of hospice coverage provided to these persons. See the “Description of Hospice
Coverage” section of this report.
15 Centers for Medicare and Medicaid Services (CMS), HHS, CMS Manual, Transmittal 1570, “Update to Hospice
Payment Rates, Hospice Cap, Hospice Wage Index, and the Hospice Pricer for FY 2009,” press release, August 1,
20008, http://www.cms.hhs.gov/Transmittals/Downloads/R1570CP.pdf.
16 MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
şȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
ž››Ž—ȱ œœžŽœȱ
The following describes a few of the most recent issues of concern to some policymakers. Among
them are disparate Medicare margins across hospices with distinct characteristics, an increasing
number of hospices that are exceeding the aggregate per beneficiary cap, increasing length of
stay, and the three-year phase out of the budget neutrality factor authorized under regulation on
August 8, 2008.
Ž’ŒŠ›ŽȱŠ›’—œȱ
Overall, MedPAC reported that the aggregate margins (measures of the differences between
estimated costs and payments to providers) for hospices for years 2001 through 2005, were
positive—starting at 1.0% in 2001 and reaching 3.4% by 2005. Margins, however, were not the
same for all hospices. Principally, MedPAC found that in 2005, for-profit hospices faired
substantially better than their non-profit counterparts, with for-profit margins reaching 11.8% in
2005, compared with -2.8% for non-profit hospices (see Table 4). New hospices, counted as
those that began participating in Medicare on or after January 1, 2000, had margins five or six
times higher than those hospices that began participating before January 1, 2000. Nearly all of
these hospices have been for-profit.17 Analyses by MedPAC also found that hospice providers
with the longest stays on average have among the highest Medicare margins.18
In addition, significant variation in margins was found across facilities, based on whether they
were freestanding or provider-based. For 2005, MedPAC calculated a 6.3% margin for
freestanding hospices and a -5.6% margin for provider-based hospices.
MedPAC also reported a significant difference in the margins of hospices that exceeded the
aggregate per beneficiary cap before and after the overpayments were returned to the Medicare
program. Prior to the return of excess payments to Medicare, cap hospices showed an aggregate
margin of 18.9%. After the return of excess payments, this margin was reduced to a negative
2.9%. See Table 4 below for data published by MedPAC on Medicare margins.
Table 4. Hospice Aggregate Medicare Margins
2001-2005
Percentage
of Hospices
Category
(2005) 2001 2002 2003 2004 2005
All
100 1.0% 3.1% 4.5%
3.2%
3.4%
Freestanding
59 5.6 6.8 9.0
6.7
6.3
Provider-
41 -10.5 -7.6 -8.9
-7.6
-5.6
based
For-profita 43 12.0 14.6 15.9
12.4
11.8
Nonprofitb 48 -4.4 -3.7 -2.9
-3.6
-2.8

17 MedPAC transcript of Public Meeting on Friday January 9, 2009.
18 MedPAC transcript of Public Meeting on January 9, 2009.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗŖȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
Percentage
of Hospices
Category
(2005) 2001 2002 2003 2004 2005
Urban 64 1.4 3.6 4.9 3.6 3.4
Rural 36
-1.8
0.1
2.5
0
3.3
Below-cap 91 N/A 2.1 3.3 1.8 1.5
Above-cap
9 N/A 30.1 23.0 17.4 18.9
(including
overpayments)
Above-cap
9 N/A 13.3 2.1 -4.6 -2.9
(net of
overpayments)
Source: MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008, Chart 12-14.
Notes: N/A (not available).
a. Totals by ownership do not sum to 100% due to exclusion of government facilities.
b. Ibid.
—Œ›ŽŠœŽȱ’—ȱ ˜œ™’ŒŽœȱ¡ŒŽŽ’—ȱ‘Žȱ›ŽŠŽȱŽ›ȱŽ—Ž’Œ’Š›¢ȱŠ™ȱ
In recent years, a small but growing number of hospices have exceeded Medicare’s aggregate per
beneficiary payment cap. Between 2002 and 2005, the number exceeding the cap grew from 2.6%
(60 hospices) to 7.8% (220 hospices), respectively. By 2006, the number increased to 293, up by
a third from 2005.19 The amount of payments over the cap that were subject to recovery also
increased from $28.2 million in 2002 to $213 million in 2006 ($166 million in 2005), the latter
representing 2.3% of total Medicare payments to hospices in that year.20
MedPAC’s analysis of hospice data provided by four regional home health intermediaries found
that the vast majority of hospices exceeding the cap in 2005 tended to be for-profit (89.1%) and
freestanding (92.3%) hospice providers.21 Hospices that exceeded the cap in 2005 also tended to
have longer lengths of stay than those hospices that did not exceed the cap. The average length of
stay for all disease categories was 54.4 days for non-cap hospices and 104.8 days for cap hospices
in 2005 (see Table 5).
Differences in the patient mix may account for some, but not all, of this variation. For example,
cap hospices tended to have a smaller share of cancer patients (i.e., 20.3% of the patient mix for
cap hospices versus 38.4% of the patient mix for non-cap hospices) who, on average, have
relatively short lengths of stays. Cap hospices also tended to have a larger share of patients with
Alzheimer’s dementia or dementia (i.e., 16% of the patient mix for cap hospices versus 9.5% of
the patient mix for non-cap hospices) who, on average, have relatively long lengths of stays. At
the same time, however, average lengths of stay still tended to be longer even for those with the
same diagnoses. For example, patients in non-cap hospices with circulatory diseases had an

19 MedPAC, “Medicare Payment Policy,” March 2009.
20 MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008.
21 MedPAC, “Report to Congress: Reforming the Delivery System,” Chapter 8, June 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗŗȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
average length of stay of 51.4 days in 2005, compared with patients in cap hospices who had an
average length of stay of 114.2 days (see Table 5).22
Table 5. Average Days Per Hospice Patient, by Disease Category, Below-Cap and
Above-Cap Hospices, All Diagnoses
2005
Below-Cap
Hospices
Above-Cap Hospices

Average
% of
Length of
% of
Difference, in
Disease
Number Total
Stay
Number Total
ALOS, Cap vs.
Category
of Cases Cases
(ALOS)
of Cases Cases ALOS Non-Cap
Cancer (except
194,089 27.2 45.9 4,831 14.5 68.3 48.9%
lung cancer)
Lung cancer
79,560
11.2
43.6
1,914
5.8
53.6
22.9
Circulatory,
77,653 10.9 51.4 5,200 15.7 114.2 122.1
except heart
failure
Heart failure
57,010
8
58.3
4,184
12.6
120.5
106.8
Debility, NOS
51,616
7.2
65.1
2,485
7.5
115.5
77.3
Chronic airway
39,796 5.6 67.4 2,495 7.5 118.9 76.4
obstruction, NOS
Alzheimer’s and
39,572 5.5 81.9 3,184 9.6 129.7 58.4
similar disease
Unspecific
36,770 5.2 66.1 2,567 7.7 107.2 62.1
symptoms/signs
Dementia 28,830
4
71.3
2,136
6.4
119.2
67.3
Genitourinary
23,118 3.2 21.3 579 1.7 37.3 75.3
diseases
Organic psychoses
22,907
3.2
71.6
1,282
3.9
116.1
62.1
Respiratory
18,300 2.6 41.7 444 1.3 89.9 115.9
diseases
Nervous system,
18,179 2.5 77.9 996 3 134.4 72.7
except
Alzheimer’s
Other 14,168
2
43.8
572
1.7
104.3
138.1
Digestive diseases
11,576
1.6
36.5
356
1.1
63.9
75.1
Total
713,144 100 54.4 33,225 100 104.8 92.6
Source: MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008, Chart 12-13.
As mentioned earlier, cap hospices experienced a Medicare margin of 18.9% in 2005, prior to the
return of overpayments. However, profitability for cap hospices is limited by the existence of the
aggregate cap, which requires that overpayments be returned to Medicare. The two-year lag
between the time Medicare makes payments to hospices and the time hospices are required to
return excess payments, however, effectively allows cap hospices to borrow funds from the
Medicare program to use for an unspecified purpose.23 Margins of this size, even temporarily,
may be advantageous to a hospice. At the same time, negative margins after overpayments are

22 MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008.
23 MedPAC, “A Data Book: Healthcare Spending and the Medicare Program,” June 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗŘȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
returned may result in hardship for these cap hospices. It is unclear whether cap hospices will be
able to sustain long-term participation in the program or whether they will develop new business
practices that would not lead them to exceed the cap.
—Œ›ŽŠœ’—ȱŽ—‘ȱ˜ȱŠ¢ȱ
The design of the prospective payment system creates incentives for longer lengths of stay, which
may partly explain the increasing number of hospices that are exceeding the cap. As described
earlier, Medicare makes routine home care payments to hospices every day that an individual is at
home, regardless of the volume or intensity of the services provided under eight hours. Spending
on routine care significantly increased between 2000 and 2005 (Table 2).
Analyses by MedPAC also found that profitability increases almost linearly with increase in the
average length of stay and that patient mix does not entirely account for the significant increase in
length of stay among certain hospices.24 Profitability, however, is limited by the aggregate per
beneficiary cap.
In its March 2009 report, MedPAC recommends reforms to Medicare’s hospice PPS in an attempt
to better realign hospice payments to the need for services, reduce financial incentives associated
with longer stays, and better promote lengths of stays that are consistent with an end-of-life
Medicare benefit rather than a long-term care benefit.25 Among other recommendations, MedPAC
proposes that the constant daily rate be adjusted to include higher payments at the beginning and
end of an episode of care—when the intensity of services delivered is expected to be higher—and
lower payments for days as the length of the episode increases and the intensity of the care
delivered is expected to be lower. MedPAC asserts that this proposal would reduce incentives for
longer lengths of stay, contain overall hospice spending, and be consistent with the spirit of the
hospice benefit design as an end-of-life benefit and not a long-term-care benefit.26
žŽȱŽž›Š•’¢ȱ“žœ–Ž—ȱ˜›ȱŘŖŖşȱ
When the data source used to adjust hospice payments for differences in the cost of labor across
geographic area was changed in 1997 from the 1983 Bureau of Labor Statistics data to the
hospital wage data, a budget neutrality adjustment factor (BNAF) was instituted by the Secretary
to prevent participating hospices from experiencing reductions in total payments as a result of the
change. This BNAF increases payments to those hospices that would otherwise experience a
payment reduction by boosting hospice payments to these providers by amounts that would make
overall payments budget neutral to the levels that they would have received had the Secretary
used the 1983 Bureau of Labor Statistics wage adjustment. According to the proposed rule
published by the Department of Health and Human Services (HHS) in the Federal Register on
May 1, 2008, the BNAF boosts total payments to hospice providers by about 4%.27

24 MedPAC, “Report to Congress: Medicare Payment Policy,” March 2009.
25 Ibid.
26 MedPAC, “Report to Congress: Medicare Payment Policy,” March 2009.
27 Centers for Medicare and Medicaid Services (CMS), HHS, “Medicare Program: Proposed Hospice Wage Index for
Fiscal Year 2009, Proposed Rule,” 73 Federal Register 2400, May 1, 2008.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗřȱ

Ž’ŒŠ›ŽȂœȱ ˜œ™’ŒŽȱŽ—Ž’ȱ
ȱ
According to the final rule, published by HHS in the Federal Register on August 8, 2008, the
BNAF would be phased out over three years, beginning with a 25% reduction in FY2009, an
additional 50% reduction (totaling 75%) in FY2010, and a final 100%, or elimination, in
FY2011.28 The Secretary projects that eliminating the BNAF would save Medicare $2.18 billion
over five years.29
The BNAF phase-out has been controversial. Many policymakers rejected the prospect of
payment reductions for hospice. As a result, the American Recovery and Reinvestment Act of
2009 (P.L. 111-5) included a provision that delayed the implementation of the phase-out of the
budget neutrality adjustment factor during FY2009. Consequently, Medicare payments to hospice
during FY2009 will contain budget neutrality adjustments similar to those in previous years.
Without changes to current law, the phase-out will begin in FY2010, starting on October 2, 2009.
Industry groups have also filed a lawsuit to block implementation of the final hospice payment
rule.30

ž‘˜›ȱ˜—ŠŒȱ —˜›–Š’˜—ȱ

Julie Stone

Specialist in Health Care Financing
jstone@crs.loc.gov, 7-1386





28 Centers for Medicare and Medicaid Services (CMS), HHS, “Medicare Program; Hospice Wage Index for Fiscal Year
2009; Final Rule,” 46464 Federal Register 73, August 8, 2008.
29 Centers for Medicare and Medicaid Services (CMS), HHS, “Medicare Hospices to See Increase in 2009 Payments,
CMS Announces Changes to Wage Index Calculation also to be Made,” press release, July 31, 2008,
http://www.cms.hhs.gov/apps/media/press/factsheet.asp?Counter=3218&intNumPerPage=10&checkDate=&checkKey
=&srchType=1&numDays=3500&srchOpt=0&srchData=&keywordType=All&chkNewsType=6&intPage=&showAll
=&pYear=&year=&desc=&cboOrder=date.
30 One of the litigants in the suit has posted materials about the lawsuit on its website:
http://www.nhpco.org/i4a/pages/index.cfm?pageid=5713.
˜—›Žœœ’˜—Š•ȱŽœŽŠ›Œ‘ȱŽ›Ÿ’ŒŽȱ
ŗŚȱ