[PDF] Home health care services - Free Download PDF (2024)

Download Home health care services...

C ha p t e r

Home health care services

ONS

(The Commission reiterates its previous recommendations on improving the home health payment system. See text box, pp. 207–209.)

C H A PTE R

Home health care services

Chapter summary

In this chapter

Home health agencies provide services to beneficiaries who are homebound and need skilled nursing or therapy. In 2011, about 3.4 million Medicare beneficiaries received home care, and the program spent about $18.4 billion on home health services. The number of agencies participating in Medicare reached 12,199 in 2011.

• Are Medicare payments adequate in 2013? • How should Medicare payments change in 2014?

Assessment of payment adequacy The indicators of payment adequacy for home health care are generally positive. Beneficiaries’ access to care—Access to home health care is generally adequate: Ninety-nine percent of beneficiaries live in a ZIP code where a Medicare home health agency operates and 98 percent live in a ZIP code with two or more agencies. •

Capacity and supply of providers—The number of agencies continues to increase, with over 700 new agencies and 12,199 total agencies in 2011. Most new agencies were concentrated in a few states, and for-profit agencies accounted for the majority of new providers.

•

Volume of services—In 2011, the volume of services was level, and total payments declined by about 5 percent, or $1 billion. The decline in payments was attributable to a reduction in the Medicare base rate.

Report to the Congress: Medicare Payment Policy | March 2013

187

The lower spending comes after several years of increases, as total spending between 2002 and 2011 increased by 92 percent. Between 2002 and 2010, the average number of 60-day episodes per home health user increased from 1.6 to 2.0, indicating that beneficiaries who use home health care stayed on service for longer periods of time. Quality of care—Quality was steady or showed a small improvement in measures of beneficiary function. Providers’ access to capital—Access to capital is a less important indicator of Medicare payment adequacy for home health care because it is less capital intensive than other sectors. According to capital market analysts, the major publicly traded for-profit home health companies had sufficient access to capital markets for their credit needs, although terms were not as favorable as in prior years. For smaller agencies, the significant number of new agencies in 2011 suggests that they had access to the capital necessary for start-up. Medicare payments and providers’ costs—For over a decade, payments have consistently and substantially exceeded costs in the home health prospective payment system. Medicare margins for freestanding agencies equaled 14.8 percent in 2011 and averaged 17.7 percent in 2001 through 2010. Two factors have contributed to payments exceeding costs: Fewer visits are delivered in an episode than is assumed in Medicare’s rates, and cost growth has been lower than the annual payment updates for home health care. Medicare margins are estimated to equal 11.8 percent in 2013. The Commission reiterates recommendation from prior years In 2011, the Commission made a multiyear recommendation for home health payments, and this report reiterates that recommendation (Medicare Payment Advisory Commission 2011). The Patient Protection and Affordable Care Act of 2010 includes reductions in payment for home health care, but these policies will leave home health agencies with margins well in excess of cost. Overpaying for home health services has negative financial consequences for the federal government and raises the Medicare premiums beneficiaries pay. Implementing the Commission’s prior recommendation for rebasing would reduce payments more swiftly and better align Medicare’s payments with the actual costs of home health agencies. ■

188

Home health care services: Assessing payment adequacy and updating payments

Background Medicare home health care consists of skilled nursing, physical therapy, occupational therapy, speech therapy, aide services, and medical social work provided to beneficiaries in their homes. To be eligible for Medicare’s home health benefit, beneficiaries must need part-time (fewer than eight hours per day) or intermittent skilled care to treat their illnesses or injuries and must be unable to leave their homes without considerable effort. Medicare requires that a physician certify a patient’s eligibility for home health care and that a patient receiving service be under the care of a physician. In contrast to coverage for skilled nursing facility services, Medicare does not require a preceding hospital stay to qualify for home health care. Unlike most other services, Medicare does not require copayments or a deductible for home health services. In 2011, about 3.4 million Medicare beneficiaries received home care, and the program spent about $18.4 billion on home health services. Medicare spending for home health care has doubled since 2001 and currently accounts for about 5 percent of fee-for-service (FFS) spending. The number of home health agencies (HHAs) participating in Medicare reached 12,199 in 2011. Medicare pays for home health care in 60-day episodes. Episodes delivered to beneficiaries in rural areas receive a 3 percent payment increase for 2010 through 2015. Payments for an episode are adjusted for patient severity based on patients’ clinical and functional characteristics and some of the services they use. If they need additional covered home health services at the end of the initial 60-day episode, another episode commences and Medicare pays for an additional episode. (An overview of the home health prospective payment system (PPS) is available at http:// medpac.gov/documents/MedPAC_Payment_Basics_12_ HHA.pdf.) Coverage for additional episodes generally has the same requirements (e.g., beneficiary must be homebound and need skilled care) as the initial episode. Medicare also pays for services rendered in the home under Part B via the fee schedule for physicians and other health professionals, though the aggregate amount of services provided under Part B is relatively small compared with the volume of services under the home health benefit. For example, in 2011 Part B paid for 2.6 million physician visits in the home, compared with 59 million skilled nursing visits under the home health benefit. Though utilization of the Part B fee schedule for services in the home is less frequent, several features may make it an appropriate substitute for some home health beneficiaries.

Services provided in the home under the Part B fee schedule do not have to meet the homebound requirement, though the provider does need to document why the service must be provided at home instead of in an office or other professional setting. The fee schedule also covers a broader range of service than home health care, such as mental health, imaging, laboratory testing, and physician management services. Beneficiaries can receive most Part B services during a home health episode, with the exception of outpatient physical therapy. Therapy services are covered under the home health PPS, the agency must bill for them through the PPS, and billing for them under the fee schedule is not permitted in most cases.1 Medicare payments for home visits under the Part B fee schedule are generally lower than payments for similar visits in the home health benefit, although many services covered under the home health benefit do not have an equivalent or similar service in the fee schedule. Physical therapy is one example because both home health care and the fee schedule cover this service. The Medicare allowed charge for a 45-minute home visit to provide therapeutic exercises under the fee schedule in 2010 was $89. The beneficiary would pay 20 percent of this amount, with the remainder paid by the program. Under the home health PPS, the average per visit payment would be about $193.2 Part of the discrepancy between the payment systems reflects the differences in services covered and costs included in each payment system. Home health care covers some services, such as some medical supplies, as part of the PPS. These elements are billed separately under the Part B fee schedule. Other differences may arise due to the costs Medicare includes in its rate calculations. For example, HHAs that reimburse mileage to traveling staff could include these costs in their Medicare allowed costs, while travel costs for physicians are not included in the costs considered in development of the Part B fee-schedule rates for home visits. Also, some of the commonly provided home services are evaluation and management visits, which the Commission has suggested are undervalued. Even with these considerations, the magnitude of the differences is substantial. Medicare has typically overpaid for home health care by 15 percent to 23 percent since 2001, and some of the discrepancy likely reflects the disconnect between payments and costs in the home health PPS. If home health PPS payments were lowered to be closer to actual agency costs, the difference between the fee-schedule rates for home services and the home health PPS would decline.

Report to the Congress: Medicare Payment Policy | March 2013

189

TABLE

9–1

Changes in supply and utilization of home health care, 1997–2011 Percent change

Agencies

1997

2000*

2011

10,917

7,528

12,199

$17.7

$8.5

$18.4

–52

117

3.6

2.5

3.5

–31

258.2

90.6

118.0

–65

Total spending (in billions) Users (in millions) Number of visits (in millions)

1997–2000 –31%

2000–2011 62%

Visit type (percent of total) Skilled nursing

41%

49%

51%

Home health aide

–37

–50

Therapy

101

Medical social services

–2

Number of visits per user

72.6

36.8

36.2

–49

–7

Percent of FFS beneficiaries who used home health services

10.5%

–30

Note:

7.4%

9.5%

FFS (fee-for-service). *Medicare did not pay on a per episode basis before October 2000.

Source: Home health standard analytical file; Health Care Financing Review, Medicare and Medicaid Statistical Supplement, 2002; and Office of the Actuary, CMS.

Use and growth of home health benefit has varied substantially due to changes in coverage and payment policy The home health benefit has changed substantially since the 1980s. Implementation of the inpatient PPS in 1983 led to increased use of home health services as hospital lengths of stay decreased. Medicare tightened coverage of some services, but the courts overturned these curbs in 1988. After this change, the number of agencies, users, and services expanded rapidly in the early 1990s. Between 1990 and 1995, the number of annual users increased by 75 percent and the number of visits more than tripled to about 250 million a year. From 1990 to 1995, spending increased from $3.7 billion to $15.4 billion. As the rates of use and lengths of stay increased, there was concern that the benefit was serving more as a long-term care benefit (Government Accountability Office 1996). Further, many of the services provided were believed to be inappropriate or improper. For example, in one analysis of 1995–1996 data, the Office of Inspector General (OIG) found that about 40 percent of the services in a sample of Medicare claims did not meet Medicare requirements for reimbursem*nt, with most of the errors due to the services not meeting Medicare’s standards for a reasonable and

190

necessary service, the patient not meeting the homebound coverage requirement, or the medical record not documenting that a billed service was provided (Office of Inspector General 1997). The trends of the early 1990s prompted increased program integrity actions, refinements to eligibility standards, temporary spending caps through an interim payment system (IPS), and replacement of the cost-based payment system with a PPS in 2000. Between 1997 and 2000, the number of beneficiaries using home health services fell by about 1 million, and the number of visits fell by 65 percent (Table 9-1). The mix of services changed from predominantly aide services in 1997 to mostly nursing visits in 2000, and therapy visits increased from 10 percent of visits in 1997 to 33 percent in 2011. Total spending for home health services declined by 52 percent between 1997 and 2000. The reduction in payments had a swift effect on the supply of agencies, and by 2000, the number of agencies had fallen by 31 percent. Between 2001 and 2010, the number of home health episodes rose from 3.9 million to 6.8 million. The number of agencies in 2011 was over 12,000, over 1,000 agencies higher than the supply at the earlier peak of spending in 1997. Almost all

Home health care services: Assessing payment adequacy and updating payments

FIGURE

Medicare margins....

8-1 Fig ure 9–1

Medicare margins of freestanding home health agencies since 2001

Medicare margin (in percent)

10 2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

Source: Medicare cost reports.

the new agencies since implementation of the PPS have been for-profit providers. Note: Note and Source are in InDesign.

and 2010 (Figure 9-1). The high overpayments have led the Commission to recommend that home health rates be lowered to a level consistent with costs (Medicare Payment Advisory Commission 2011). These high margins likely have encouraged the entry of new HHAs, as the total number of agencies participating in Medicare has increased by an average of about 532 agencies a year since 2002 (Figure 9-2, p. 192).

Source: The steep declines in services under the IPS do not appear to have adversely affected the quality of care beneficiaries received; one analysis found that patient satisfaction with home health services was mostly unchanged in this period (McCall et al. 2003, McCall et al. 2004). An analysis NotesBudget about graph: of the Balanced Actthis of 1997 changes related to Protection and Affordable Care Act of 2010 post-acute care (PAC), including the home health IPS andin the Patient • Data is in the datasheet. Make updates datasheet. changes to payment for home health services changes for other PAC services, concluded that the rate • WATCH FOR GLITCHY RESETS WHEN YOU UPDATE DATA!!!! of adverse events generally improved or did not worsen In 2010, the Commission recommended that Medicare • IPS Thewas column totals werebyadded manually. when the in effect. A study the Commission lower home health payments to make them more also concluded quality of care tick had marks not declined • I hadthat to the manually draw and axis lines because kept resetting I changed consistent withthey costs, referred to as when payment rebasing. any data. between the IPS and the PPS (Medicare Payment Advisory TheitPatient • I can’t delete the legend, so I’ll just have to crop out in Protection InDesign.and Affordable Care Act of 2010 Commission 2004). The similarity in quality of care under (PPACA) includes several intended address tool, they selection to selectreductions items for modiﬁcation. Otherwise if reductions you use the blacktoselection the IPS • andUse the direct PPS suggests that tool the payment home health care’s high Medicare payments, but these default Budget when you change theagencies data. to in the Balanced Act of 1997 led policies may not achieve the Commission’s goal of making reduce costs and excess utilization without a measurable • Use paragraph styles (and object styles) to format. payments more consistent with actual costs. PPACA calls difference in the quality of patient care. base rate reductions to be phased • Data was from: R:\Groups\MGA\data bookfor 2007\data book 2007 chp1 in over four years beginning in 2014. The law sets a maximum reduction Home health margins for freestanding HHAs have of 3.5 percent a year, for a cumulative reduction of 14 been very high since the PPS was implemented, as Medicare margins averaged 17.7 percent between 2001

Report to the Congress: Medicare Payment Policy | March 2013

191

FIGURE Figure

8-2 9– 2

New agencies Annual number of new home health agencies in Medicare, 2002–2011

1,200

Number of new agencies

1,100 1,000 900 800 700 600 500 400 300 2002

2004

2006 Year

Source: CMS Providing Data Quickly database.

2008

2010

discharge. Community-based organizations must agree to have formal partnerships with area hospitals and consumer groups. In both of these demonstrations, HHAs may serve as providers for participating beneficiaries when they return home. Models that focus on chronic care needs and care coordination may also have a role for home health services. For example, the Independence at Home demonstration will test the effectiveness of delivering comprehensive primary care services at home (Center for Medicare & Medicaid Innovation 2012). The demonstration makes payments to home care physicians for delivering services at home for frail elderly populations with multiple chronic conditions. Practices that achieve quality and cost-saving goals can receive bonus payments under the demonstration. Home care physicians frequently serve community-dwelling homebound Medicare beneficiaries and use Medicare HHAs as a supplement to physician home visits. Other delivery system reforms may seek to use home health services as a substitute for hospitalizations. For example, some providers have tested a “hospital at home” approach in which patients are diverted from the emergency department and sent home with intensive home health services to address their urgent care needs (Cryer et al. 2012). This approach can be appropriate for patients who need intensive assistance to stabilize a condition but do not require the full scope of emergency department or inpatient care.

percent but offsets this reduction with the payment update Note: Note and Source in InDesign. for each year. With this offset, the maximum reduction is roughly halved to 7 percent over the four-year period. With margins that typically exceed 15 percent, these lower reductions could leave HHAs with a significant profit margin. The Commission’s policy would reduce Ensuring appropriate use of home health payments over a two-year period and graph: would not offset Notes about this care is challenging reductions with increases the payment Make update. updates The • Data is infrom the datasheet. inPolicymakers the datasheet. have long struggled to define the role of the Commission’s proposed reductions would likely bring home health benefit • I inhad force return thePPACA items on the x-axis. They will reﬂow inif Medicare I update (Benjamin the data. 1993). From payments more linetowith costs than the rebasing the outset, there was a concern that setting a narrow policy policy. • I had to manually draw tick marks and axis lines because they kept resetting when I changed any data could result in beneficiaries using other, more expensive • Use direct selection tool to select items for modiﬁcation. if you black tool, services, while aOtherwise policy that was too use broadthe could leadselection to Some PPACA initiatives may expand the role of default when you change the data. wasteful or ineffective use of home health care (Feder home health services Lambrew 1996). Medicare relies on the skilled care Use paragraph object styles) toand format. PPACA also •includes several newstyles models(and of care that may and homebound requirements as primary determinants of have a potential role for home health services. Some of home health eligibility, but these broad coverage criteria these models are designed to improve PAC (Center for permit beneficiaries to receive services in the home even Medicare & Medicaid Innovation 2012). For example, the when they are capable of leaving home for medical care, Bundled Payment for Care Improvement demonstration which most home health beneficiaries do (Wolff et al. tests models that include PAC as a part of an acute care 2008). Medicare does not provide any incentives for bundle or as a stand-alone bundle. Another initiative, beneficiaries or providers to consider alternatives to home the Community-Based Care Transitions Initiative, uses health care, and beneficiaries, once they meet program community-based organizations, such as area agencies on coverage requirements, can receive an unlimited number aging, to provide and manage care for beneficiaries after

192

Home health care services: Assessing payment adequacy and updating payments

of home health episodes. In addition, the program relies on agencies and physicians to follow program requirements for determining beneficiary needs, but there is some evidence that they do not consistently follow Medicare’s standards (Cheh et al. 2007, Office of Inspector General 2001). Even when enforced, the standards permit a broad range of services. For example, the skilled care requirement mandates that a beneficiary need therapy or nursing care to be eligible for the home health benefit. The intent of the skilled services requirement is that the home health benefit serves a clear medical purpose and is not an unskilled personal care benefit. However, Medicare’s coverage standards do not require that skilled visits be the majority of the home health services a patient receives. For about 9 percent of episodes in 2010, most services provided are visits from an unskilled home health aide. These episodes raise questions about whether Medicare’s broad standards for coverage are adequate to ensure that skilled care remains the focus of the home health benefit. A recent review by the Department of Health and Human Services OIG suggests that a significant number of HHAs had questionable patterns of payment (Office of Inspector General 2012). The review found that about 25 percent of HHAs in Medicare had unusual utilization or payment trends in 2010. For example, over 400 agencies had an unusually high rate of beneficiaries who received five or more episodes in a consecutive set of home health episodes. OIG cited 257 agencies for providing an unusually high number of therapy visits, which increases the episode payment under the home health PPS. About 80 percent of the agencies considered to have questionable billing practices were in four states: California, Florida, Michigan, and Texas. Some of these states have experienced rapid growth in the number of HHAs participating in Medicare. In 2010, the Commission made a recommendation to curb wasteful or fraudulent home health services (Medicare Payment Advisory Commission 2010). This recommendation calls on CMS to use its authorities under current law to examine providers with aberrant patterns of utilization for possible fraud and abuse. Medicare has implemented increased screening requirements for new agencies but has not implemented all the tools available under current law. For example, many areas with fraud concerns have a supply of agencies that many believe far exceeds the legitimate need for services. PPACA permits Medicare to implement temporary moratoriums on the

enrollment of new agencies in areas believed to have a high incidence of fraud, but it has yet to use this authority. A moratorium on the enrollment of new providers in these areas would prevent new agencies from entering markets that may already be saturated. Medicare also has the authority to require HHAs to hold surety bonds, but it has not exercised this authority and made surety bonds a requirement.3 A recent court case between the Department of Health and Human Services and the Center for Medicare Advocacy will require the program to clarify the language in its benefit manual regarding the coverage of services needed to maintain or prevent deterioration of a patient’s current condition. Coverage will hinge on existing requirements: that the beneficiary needs skilled care and meets the homebound requirement. Until CMS revises the benefit manuals, specifies instructions, and trains claims contractors and providers, it is hard to estimate the impact this change will have on utilization. However, given the rapid growth the benefit has experienced in the past, it remains possible that utilization could increase. The home health benefit provides a valuable service to beneficiaries and the Medicare program, particularly when it substitutes for a higher level of PAC or helps community-dwelling beneficiaries avoid hospitalization. However, the broad program standards and fragmented nature of the FFS program do not encourage effective targeting of the benefit to meet these goals and provide opportunities for fraud, waste, and abuse that have proven difficult to eliminate. Many of these issues might be more easily addressed if the payment and delivery of home health care were more closely integrated with the other sources of care typically provided during an episode. For example, accountable care organizations at risk for the cost of a beneficiary’s Medicare spending would have an incentive to use home health care when it could reduce overall costs but to avoid the excessive utilization observed in many parts of the country.

Are Medicare payments adequate in 2013? The Commission reviews several indicators to determine the level at which payments will be adequate to cover the costs of an efficient provider in 2013. We assess beneficiary access to care by examining the supply of home health providers and annual changes in the volume

Report to the Congress: Medicare Payment Policy | March 2013

193

TABLE

9–2

Number of home health agencies continues to rise Average annual percent change

Number of agencies

2002

2004

2006

2008

2010

2011

2002–2010

7,057

7,804

8,955

10,040

11,654

12,199

Agencies that opened

399

656

828

780

831

730

9.8

–14.0

Agencies that closed

277

183

176

167

181

218

–4.6

15.0

2.0

2.1

2.5

2.8

3.3

3.4

6.1

3.7

Number of agencies per 10,000 beneficiaries Note:

6.5%

2010–2011 4.6%

Agencies’ census includes all agencies operating during a year, including agencies that closed or opened.

Source: CMS’s Providing Data Quickly database and 2012 annual report of the Boards of Trustees of the Medicare trust funds.

of services. The review also examines quality of care, access to capital, and the relationship between Medicare’s payments and providers’ costs. Overall, the Medicare payment adequacy indicators for HHAs are positive. Beneficiaries’ access to care: Almost all beneficiaries live in an area served by home health care Supply and volume indicators show that almost all beneficiaries have access to home health services. In 2011, almost all beneficiaries (99.5 percent) lived in a ZIP code served by at least one HHA, 98 percent lived in a ZIP code served by two or more HHAs, and over 80 percent lived in a ZIP code served by five or more agencies. These findings are consistent with our review of access from prior years.4 Capacity and supply of providers: Agency supply surpasses previous peak In 2011, there were 12,199 HHAs participating in Medicare, a net increase of about 512 agencies compared with the previous year. Most new agencies in 2011 were for-profit agencies. The number of agencies exceeded the previous record of the 1990s when supply exceeded 10,900 agencies. The high rate of growth is a particular concern, as the new agencies appear to be concentrated in areas with fraud issues, including California, Florida, and Texas. These states, like most, do not have state certificateof-need laws for home health care, which can limit the entry of new providers.5 Since 2004, when 99 percent of beneficiaries lived in a ZIP code served by an HHA, the number of agencies per 10,000 FFS beneficiaries has risen 60 percent, from 2.0 to

194

3.4 (Table 9-2). Some of this growth is due to a decrease in the number of FFS enrollees as more beneficiaries enroll in Medicare Advantage, but even when managed care beneficiaries are included with FFS, the number of agencies per beneficiary has increased by about 35 percent since 2004. Supply can vary significantly among states. In 2010, Texas averaged 9.6 agencies per 10,000 beneficiaries, whereas New Jersey averaged 0.4 agency per 10,000 beneficiaries. Some of this variation in supply is likely due to certificate-of-need laws, as New Jersey has certificate-of-need laws and Texas does not. The extreme variation demonstrates that the number of providers is a limited measure of capacity, as agencies can vary in size and capability. Also, because home health care is not provided in a medical facility, agencies can adjust their service areas as local conditions change. Even the number of employees may not be an effective metric, because agencies can use contract staff to meet their patients’ needs. Growth in episode volume slows after many years of rapid growth In 2011, total spending for home health care dropped by about 5 percent, and most of this reduction was due to a decline in the home health episode base rate.6 The average payment per episode declined by about 5 percent, while the number of episodes and beneficiaries using home health care held steady between 2010 and 2011 (Table 9-3). This steady level of utilization is in sharp contrast to the utilization trends in prior years. Between 2002 and 2010, the number of episodes increased by 66 percent, from 4.1 million to 6.8 million episodes. Between 2002 and 2010, the share of beneficiaries using home health

Home health care services: Assessing payment adequacy and updating payments

TABLE

9– 3

Share of beneficiaries using home health services continues to rise Average annual percent change 2002

2006

2008

2010

2011

35.0

36.1

35.5

36.0

36.3

Home health users (in millions)

2.5

3.0

3.2

3.4

Share of beneficiaries using home health care

7.2%

8.4%

8.9%

Episodes (in millions):

4.1

5.5

1.6

1.8

0.12

0.14

FFS beneficiaries (in millions)

Per home health user Per FFS beneficiary Payments (in millions) Per home health user Per FFS beneficiary Note:

2002–2010

2010–2011

0.4%

0.9%

3.4

3.9

0.7

9.5%

3.5

–0.1

6.1

6.8

6.9

6.6

0.1

1.9

2.0

2.6

–0.7

0.16

0.19

6.2

–0.8

$9.6

14.0

16.9

19.4

18.4

9.2

5.2

$3,803

$4,606

$5,359

$5,679

$5,367

5.1

–5.9

$274

$387

$479

$543

$507

8.9

–6.0

FFS (fee-for-service).

Source: MedPAC analysis of home health standard analytical file.

care increased from 7.2 percent to 9.5 percent but was steady in 2011 relative to the prior year. The cause of the lack of growth may be, at least in part, related to the new requirement, effective 2011, that a certifying physician or an allowed nonphysician practitioner had a face-to-face encounter with the patient when authorizing home care. Office visits or telehealth encounters with a physician or nurse practitioner up to 90 days before or 30 days after the beginning of home health care qualify for the requirement. The change was intended to ensure that beneficiaries receive a complete evaluation when home health care is ordered and that physicians do not rely solely on information provided by HHAs when making decisions about patient care. It is possible that the additional scrutiny required by this examination led to fewer referrals for home health care. Home health care stays have grown longer and less focused on post-acute care since 2002

The average number of episodes per user has increased by 22 percent since 2002, rising from 1.6 to 2.0 episodes per user by 2010. Though the trend is flat for 2011, the increase in episodes per user in 2002 through 2010 indicates that beneficiaries are receiving home health care for longer periods of time and suggests that home health care is serving more as a long-term care benefit for

some beneficiaries. This concern is similar to those in the mid-1990s that led to major program integrity activities and payment reductions. The increase in these episodes coincides with Medicare’s PPS incentives that encourage additional volume: The per episode unit of payment and the payment system has an adjustment that raises payments for the third and later episodes in a consecutive spell of home health episodes.7 The rise in the average number of episodes per beneficiary also coincides with a shift away from using home health care as a PAC service. Over the 2001 to 2010 period, the number of episodes that were not preceded by a hospitalization or PAC stay increased by 117 percent (Table 9-4, p. 196). In 2001, about 52 percent of all episodes were not preceded by a hospitalization or PAC stay, but by 2010 the share had increased to 66 percent of all episodes. A corresponding decrease occurred between 2001 and 2010 in episodes preceded by a hospitalization or PAC stay, decreasing from 48 percent to 34 percent. These episodes increased at a lower pace of 26 percent in 2001 through 2010. A review of utilization, demographic, and clinical characteristics suggests that beneficiaries who use home health care primarily for PAC differ on several metrics compared with community-admitted users.8

Report to the Congress: Medicare Payment Policy | March 2013

195

TABLE

9–4

Increase in home health episodes more rapid when episode not preceded by hospitalization or PAC stay Number of episodes (in millions)

Percent of episodes Cumulative growth

2001

2010

2001

2010

First

1.6

1.9

15%

40%

27%

Subsequent

0.3

0.5

Subtotal

1.9

2.4

First

0.8

1.3

Subsequent

1.3

3.2

148

Subtotal

2.1

4.5

117

3.9

6.8

100

Episodes preceded by a hospitalization or PAC stay:

Episodes not preceded by a hospitalization or PAC stay (community-admitted episodes):

Total Note:

PAC (post-acute care). “First” and “subsequent” refer to the timing of an episode relative to other home health episodes. “First” indicates no home health episode in the 60 days preceding the episode. “Subsequent” indicates the episode started within 60 days of the end of a preceding episode. “Episodes preceded by a hospitalization or PAC stay” indicates the episode occurred fewer than 15 days after a hospital (including long-term care hospitals), skilled nursing facility, or inpatient rehabilitation facility stay. “Episodes not preceded by a hospitalization or PAC stay” (community-admitted episodes) indicates that there was no hospitalization or PAC stay in the 15 days before episode start. Numbers may not add due to rounding.

Source: CMS Datalink file, 2010.

Community-admitted home health care users accounted for about 50 percent of all home health users, but they accounted for 4.5 million episodes in 2010 (64 percent of all episodes). PAC users averaged 1.4 episodes in 2010, while community-admitted users averaged 2.6. episodes for the year. About 94 percent of the episodes provided to community-admitted patients were not preceded by a hospitalization or prior PAC use. About 42 percent of the episodes provided to community-admitted users were for dual-eligible Medicare and Medicaid beneficiaries; in contrast, the comparable share for PAC users was 24 percent. Community-admitted users also had a larger share of episodes with high numbers of visits from home health aides—for example, aide services were the majority of services provided in 11 percent of the episodes for community-admitted users compared with 4 percent for PAC users. Community-admitted users had fewer hospitalizations and physician visits compared with PAC users, but this result was likely due, at least in part, to a hospitalization being a criterion for being categorized as a PAC user. About 74 percent of community-admitted users were White, compared with 86 percent of PAC users.

196

Community-admitted users also tended to be older, with relatively more users in the 85 or older age group and relatively fewer in the 65–74 age group. Communityadmitted users had 3.8 chronic conditions on average, compared with 4.2 for PAC users. The mix of conditions also varied, with 29 percent of community-admitted users having Alzheimer’s disease and dementia compared with 21 percent for PAC users. The rate of chronic conditions was lower for community-admitted users for most other conditions, such as heart disease, chronic obstructive pulmonary disease, and heart failure. The differences between the two populations suggest that Medicare is serving different populations within the home health benefit. PAC users appear to be a more clinically severe population, as measured by the number of chronic conditions; the smaller number of episodes per user for this population indicates they remain in home health care for relatively shorter periods of time. By contrast, community-admitted users had some characteristics that were more suggestive of long-term care needs. Community-admitted users consume almost twice as many episodes per user as PAC users, indicating

Home health care services: Assessing payment adequacy and updating payments

Volume of therapy services is influenced by incentives in Medicare’s payment system

The number of therapy visits a beneficiary receives during a home health care episode is one of the factors that determines Medicare’s payment for a home health episode. Generally, providing more therapy visits raises the episode payment. The Commission has long had a concern that allowing actual utilization to drive payment creates an incentive for agencies to provide more services to increase payment, and changes in episode volume have generally reflected the incentives for therapy payment in the payment system. The Commission recommended that Medicare redesign the payment system to rely solely on patient characteristics, and not the number of services provided, for setting payment, but CMS has yet to implement this recommendation (Medicare Payment Advisory Commission 2011). A review of historical trends in the volume of therapy services indicates that payment incentives significantly influenced provider behavior. From 2001 to 2007, CMS had a single payment adjustment for therapy that increased payment for episodes with 10 or more therapy visits. In this period, the growth rate for episodes that just met the threshold was almost double the growth for all other home health episodes. This trend led to concerns that providers were deliberately targeting the 10-visit threshold. Responding to these concerns, CMS implemented changes in 2008 that lowered payments for episodes with 10 to 13 therapy visits and increased payments for episodes in the 6 to 9 and 14 or more therapy visit ranges. The subsequent changes in therapy utilization reflected the new incentives: Episodes with 10 to 13 therapy visits decreased 27 percent, while those with 6 to 9 therapy visits and 14 or more visits increased by 43 and 27 percent, respectively (Figure 9-3). This was the largest one-year shift in therapy volume since the PPS was implemented. Since 2008, the growth in episodes has followed this pattern, with episodes with 14 or more visits growing significantly.

Figure FIGURE

Changes the distribution... Annual in episode volume for episodes with select numbers of home health visits

9–3 9-3

20 2002 2007 2008 15 Percent of episodes

that they remain in home health care for a longer period of time. The higher rates of home health aide services for community-admitted users suggest that they need more assistance with activities of daily living. The high share of community-admitted users who were also Medicaid eligible suggests that some of this utilization may be due to state Medicaid programs leveraging the Medicare home health benefit to provide long-term care. Such an approach can permit states to shift the costs of at least some of their long-term care expenses to the Medicare program.

2011

0 6 to 9

10 to 13

14 or more

Number of therapy visits Note:

From 2002 to 2007, CMS had a single payment adjustment that increased payment for episodes with 10 or more therapy visits. In 2008, CMS added payment adjustments that lowered payments for episodes with 10 to 13 visits and raised them for episodes with 6 to 9 visits and 14 or more visits. These revised thresholds remain in effect.

Source: MedPAC analysis of 2011 home health standard analytical file.

Notes about this graph:

• Data is in the datasheet. Make updates in the d Volume in • 2011 decreased foryears therapy I reformatted the fromepisodes the x-axis. affected by new review requirement

• I had to manually draw tick marks and axis line

In 2011, CMS implemented a requirement for agencies to • Use selection items for modi review the need for direct additional therapytool careto at select two points default whenbefore you change the data. in a home health episode: the 14th therapy visit and again before•the 20th therapy visit. That year, CMS Use paragraph styles (and objectalso styles) to form implemented a new requirement for tighter supervision of therapy services provided under the home health care benefit. In these assessments, the therapist must review the patient’s progress and determine whether the patient will benefit from additional therapy visits. Medicare targeted these visit intervals because under the current PPS, the payments increase substantially for episodes at the 14th and 20th therapy visits. The additional review is intended to serve as a safeguard against manipulation of therapy visits to garner increased payment.

Report to the Congress: Medicare Payment Policy | March 2013

197

TABLE

9–5

Utilization by type of county, 2011 Number of home health episodes per 100 FFS beneficiaries

Type of county Urban

All states

Top 5 states

All other states

17.7

33.5

14.3

Micropolitan

16.0

37.7

11.7

Rural, adjacent to urban

18.1

40.2

12.7

Rural, nonadjacent to urban

16.3

43.6

11.9

All rural

16.5

39.2

11.9

17.5

34.7

13.7

Rural, by subcategory

National (all counties) Note:

FFS (fee-for-service). “Top 5 states” category includes the states with the highest rates of episodes per beneficiary in 2011: Florida, Louisiana, Mississippi, Oklahoma, and Texas. An urban county includes a city that has a population of more than 50,000. A micropolitan county has a population of 10,000 to 50,000.

Source: MedPAC analysis of home health standard analytic file and 2011 beneficiary annual summary file.

198

Claims data for 2011 suggest that these requirements had some impact, as the number of episodes with visits at and beyond these thresholds decreased relative to 2010 (data not shown in Figure 9-3). For example, the number of episodes with 14–17 therapy visits decreased by 9.5 percent and the number of episodes with more than 20 therapy visits decreased by 9.2 percent. The decline in 2011 is a reversal of the trend in 2008 through 2010, when episodes with 14 or more therapy visits were growing rapidly.

In 2012, CMS also raised the payment-relative weights for episodes with fewer than six therapy visits and lowered them for episodes with six or more therapy visits but retained the number of visits furnished as a payment factor. This adjustment at least partially addresses the Commission’s past concerns that therapy services may be overvalued, but agencies can still garner higher payments by providing additional therapy visits.

Episodes with more than 6 and fewer than 14 therapy visits, accounting for 45 percent of episodes that include any therapy visits, have no requirement for additional review.9 The volume of these episodes continued to rise in 2011. This lack of scrutiny is problematic because agencies can significantly raise their Medicare payment by increasing visits within this range. For example, the payment for low-severity episodes increases by 20 percent when the number of therapy visits increases from five visits to six visits. While administrative actions such as additional review may reduce these incentives, these efforts require more resources by agencies and Medicare. Eliminating the use of therapy visits as a payment factor, as the Commission recommended, would eliminate the need for administrative resources to scrutinize therapy use and would ensure that financial incentives did not trump patient needs when determining the amount of therapy to provide in a home health care episode.

Home health care utilization tends to vary more in different regions of the nation than between urban and rural areas within regions or states. In 2011, the national average for home health care episodes per 100 Medicare FFS beneficiaries was about 17.5 (Table 9-5). The average utilization for rural-nonadjacent counties—counties not adjacent to an urban area—was slightly higher than this average and slightly lower for micropolitan counties and rural counties adjacent to an urban area. While rural areas generally had utilization similar to that in urban areas, frontier counties—those that average six or fewer individuals per square mile—had significantly lower utilization. In 2008, the most recent year for which the Commission has data, utilization in frontier counties averaged 8 episodes per 100 beneficiaries, about half the average rate of utilization in other rural areas (data not shown in Table 9-5).

Adjacent urban and rural areas have comparable total utilization

Home health care services: Assessing payment adequacy and updating payments

Regions or states with utilization that is high relative to the national average typically have above-average utilization in both rural and urban counties, and states or regions with utilization below the national average generally have below-average utilization in urban and rural areas. For example, utilization in both urban and rural areas of Wisconsin is well below the national average. In 2011, rural Wisconsin areas averaged 5.6 episodes per 100 beneficiaries, compared with 7.6 episodes per 100 beneficiaries in urban Wisconsin areas. In contrast, utilization in both urban and rural areas of Texas is above average. Rural areas of Texas average 43.5 episodes per 100 beneficiaries, and urban areas average 41 episodes per 100 beneficiaries. Home health care utilization is concentrated in select states

The highest utilization of home health services is concentrated in a few areas of the country. The top five states (Florida, Louisiana, Mississippi, Oklahoma, and Texas) account for about 35 percent of all home health care episodes despite accounting for only 17 percent of beneficiaries. The utilization in these five states is 34.7 episodes per 100 FFS beneficiaries, compared with 13.7 episodes per 100 FFS beneficiaries for all other states (Table 9-5). Large differences in utilization occur in both rural and urban areas. Urban areas in the top five states have a rate of utilization more than double that in the other states, and rural areas in the top five states have rates of utilization double or triple the rates in rural areas in the other states. The concentration of high utilization in a few areas of the country has raised concerns that some of this utilization may be due to fraud and abuse. It is hard to distinguish between fraudulent and legitimate home health care services in Medicare claims data. However, a comparison of areas with remarkably high spending compared with national benchmarks provides some indication of the potential impacts if utilization in these areas could be curbed. As an example, the 25 counties with the highest utilization (Table 9-6, p. 200) had an average utilization of 88 episodes per 100 beneficiaries. If policies to reduce fraud could lower utilization to 18.5 episodes per 100 beneficiaries (the 75th percentile), the total number of episodes in these counties would have declined by about 290,000 episodes, or about 80 percent of these counties’ total utilization in 2011. Medicare spending would have been lower by about 4.3 percent or $783 million in 2011.

Rural areas with high utilization benefit most from Medicare’s rural add-on payment

The high level of utilization in many rural areas results in a maldistribution of the add-on payments Medicare makes for rural home health services. In 2010, PPACA implemented an add-on payment of 3 percent for each home health care episode provided to beneficiaries in rural areas, presumably to bolster access to home health services. The use of such a broadly targeted add-on, providing the same payment for all rural areas regardless of access, results in rural areas with the highest utilization drawing a disproportionate share of the add-on payments. For example, 70 percent of the episodes that received the add-on payments in 2011 were in rural counties with utilization significantly higher than the national average (equal to or greater than the 60th percentile of episodes per FFS beneficiary among all counties). The rural counties in the bottom 40 percent of utilization, those clearly below the national average, accounted for 13 percent of the episodes eligible for the add-on payment. The Commission noted in our June 2012 report that Medicare should target payment adjustments for rural areas to those areas that have access challenges (Medicare Payment Advisory Commission 2012). The large share of payments made to rural areas with above-average utilization does nothing to improve access to care in those areas and raises payments in markets that appear to be more than adequately served by HHAs. Some of the counties with aberrant patterns of utilization suggestive of fraud and abuse are rural—for example, 22 of the 25 top spending counties in 2011 are rural areas (Table 9-6, p. 200). Agencies in these 25 counties received approximately $28 million from the rural add-on that was in effect in 2011. Higher payments in areas without access problems may encourage the entry or expanded operations of agencies that seek to exploit the financial incentives of the Medicare program. More targeted approaches that eliminate rural add-on payments to areas without access problems could be pursued. Quality of care: Quality measures generally held steady Medicare reports several quality measures on its Home Health Compare website, from which we obtained recent trends for measures associated with function and care management (Table 9-7, p. 201). In general, the share of beneficiaries showing improvement in these measures has increased since 2004, and a similar trend is observed for most measures in 2011 and 2012. However, these data

Report to the Congress: Medicare Payment Policy | March 2013

199

TABLE

9–6

State

Counties with the highest rates of home health care use in 2011

County

Share of FFS beneficiaries using home health services

Episodes per user

Episodes per 100 FFS beneficiaries

Duval

34.4%

4.5

154.8

Brooks

33.6

4.2

142.6

Jim Hogg

32.3

4.0

128.3

Starr

31.7

4.1

129.5

Willacy

29.6

3.5

103.8

Miami–Dade

28.9

2.8

79.8

Jim Wells

28.6

4.1

117.3

Claiborne

27.1

3.1

83.2

Zapata

26.9

3.9

104.5

Madison

26.2

4.4

114.3

Hidalgo

26.2

3.7

97.7

Choctaw

25.7

4.2

107.2

McCurtain

24.3

4.5

108.5

Webb

24.1

3.8

92.8 106.9

East Carroll

23.6

4.5

Cameron

22.9

3.3

75.5

Avoyelles

22.6

4.1

91.4

Red River

22.2

4.0

90.0

Hanco*ck

22.2

3.1

68.6

Pushmataha

22.0

4.1

89.3

Latimer

21.8

4.5

98.1

Washington

21.6

3.7

81.1

Falls

21.6

3.3

71.1

Kleberg

21.3

3.8

79.9

Sharkey

21.0

3.7

76.9

9.5

2.0

17.5

National average Note:

FFS (fee-for-service). Counties with fewer than 100 home health users have been excluded.

Source: MedPAC analysis of the 2011 home health standard analytical file and the 2010 Medicare denominator file.

are collected only for beneficiaries who do not have their home health care stays terminated by a hospitalization, which means that the beneficiaries included in the measure are probably healthier and more likely to have positive outcomes. Providers’ access to capital: Adequate access to capital for expansion Few HHAs access capital through publicly traded shares or public debt, like issuing bonds. HHAs are not as capital intensive as other providers because they do not require extensive physical infrastructure, and most are too small to attract interest from capital markets. Information on

200

publicly traded home health care companies provides some insight into access to capital but has limitations. Publicly traded companies may have businesses in addition to Medicare home health care, such as hospice, Medicaid, and private-duty nursing. Also, publicly traded companies are a small portion of the total number of agencies in the industry. For these reasons, access to capital is a smaller consideration than in other sectors the Commission reviews. Analysis of for-profit companies indicates that they have adequate access to capital in 2011, though on terms less favorable than in previous years. The PPACA changes

Home health care services: Assessing payment adequacy and updating payments

TABLE

9– 7

Average agency performance on select quality measures 2004

2005

2006

2007

2008

2009

2011

2012

Share of an agency’s beneficiaries with improvement in: Transferring

47%

49%

50%

51%

52%

Bathing

Walking

Medication management

Pain management

Note:

The measures for walking, medication management, and pain management changed in 2011 and are not comparable to data from prior years. Data are risk adjusted for differences between home health agencies in the mix of patients they serve.

Source: MedPAC analysis of CMS Home Health Compare data.

in home health care policy implemented in the 2011 and 2012 PPS regulations have trimmed revenues for the home health care industry. In addition, several federal investigations have been launched into the therapy billing practices of some of the publicly held home health companies. These factors have weakened investor outlook for these firms and made lenders more cautious in the terms they offer home health firms seeking capital, but for-profit HHAs still appear to have access to capital for their operating needs. For smaller or nonpublic entities, the entry of new providers indicates that access to capital for privately held agencies is adequate. In 2011, over 700 new HHAs entered Medicare; most of these agencies were for profit.

and to reduce Medicare’s historically high payments for this service. The average cost per episode in 2011 was unchanged from 2010. Low cost growth or no cost growth has been typical for home health care, and in some years we have observed a decline in cost per episode. The ability of HHAs to keep costs low has contributed to the high margins under the Medicare PPS.

The low capital requirements for home health care services allow the industry to react rapidly when the supply of agencies changes or contracts. For example, during the interim payment system (1997–2000), when payments dropped by about 50 percent in two years, many agencies exited the program. However, new agencies entered the program (about 200 agencies a year) and existing agencies expanded their service areas to enter markets left by exiting agencies. Because of these adjustments, reviews of access found that access to care remained adequate during this period despite a substantial decline in the number of agencies (Liu et al. 2003).

Margins for hospital-based agencies in 2011 were –10.9 percent. The lower margins of hospital-based agencies are chiefly due to their higher costs, some of which may be due to overhead costs allocated to the HHA from its parent hospital. Potential lower inpatient costs due to shorter hospital stays may more than compensate for any losses from operating an HHA. The Commission includes hospital-based HHAs in the analysis of inpatient hospital margins because these agencies operate in the financial context of hospital operations. Operating an HHA may permit a hospital to discharge its patients earlier, thereby lowering hospital costs for inpatient services.

Medicare payments and providers’ costs: Payments decreased in 2011 but costs remained steady

The negative margins for hospital-based agencies may be an issue for counties that have only these types of agencies. A number of hospital-based agencies experienced small but consistent declines in recent years and may cause access challenges for counties with no other source of home health care.

In 2011, average payments per episode declined by about 5 percent, a result of several policies intended to address changes in coding practices unrelated to patient severity

Medicare margins remained high in 2011

In 2011, HHA Medicare margins in aggregate were 14.8 percent for freestanding agencies (Table 9-8, p. 202). Financial performance varied from –0.3 percent for the agency at the 25th percentile of the margin distribution to 22.8 percent for the agency at the 75th percentile.

Report to the Congress: Medicare Payment Policy | March 2013

201

TABLE

9–8

Medicare margins for freestanding home health agencies, 2010 and 2011 2010

All

2011

19.1%

14.8%

Majority urban

19.1

Majority rural

19.4

Percent of agencies, 2011

Percent of episodes, 2011

100%

14.8

15.3

Geography

Top 5 states in utilization

14.4

11.4

All other states

19.8

15.3

For profit

20.3

15.7

Nonprofit

15.1

12.2

Government*

N/A

First (smallest)

10.2

6.6

Second

11.2

8.3

Third

13.5

10.1

Fourth

17.7

13.4

Fifth (largest)

22.0

17.4

Type of control

Volume quintile

Note:

N/A (not applicable). “Top 5 states in utilization” category includes the states with the highest rates of episodes per beneficiary in 2011: Florida, Louisiana, Mississippi, Oklahoma, and Texas. *Government-owned providers operate in a different context from other providers, so their margins are not necessarily comparable.

Source: MedPAC analysis of Home Health Cost Report files from CMS.

As mentioned earlier, there are several parts of the country with unusual patterns of home health care utilization. A review of Medicare margins for agencies operating in the five states with the highest rates of utilization indicates that they have lower margins than agencies in other areas—11.4 percent compared with 15.3 percent (Table 9-8). The higher margins of agencies in states with more typical patterns of utilization also indicate that, on average, agencies in these states have margins higher than the 2011 national average of 14.8 percent. Conversely, margins for agencies in the five states with high utilization are below the national average. While the margins of agencies in these five states exceed the margins of many other categories of Medicare providers, their belowaverage performance actually reduces the national average Medicare margin for HHAs. Since an individual HHA can serve a mix of urban and rural patients, we determine an agency’s rural or urban designation based on where most of its episodes are provided. In 2011, rural providers had higher margins

202

than urban providers, but this finding is not surprising, as PPACA included a 3 percent add-on for episodes delivered in rural counties beginning in March 2010. Margins did not vary significantly among subcategories of rural agencies (Table 9-9). There did not appear to be a relationship between the share of episodes preceded by a hospitalization or PAC use and Medicare margins for most agencies, but agencies with the lowest share of posthospital or PAC episodes had lower margins than other quintile groups (Table 9-10). Agencies that admitted the fewest posthospital episodes (or, conversely, had the greatest share of communityadmitted episodes) had an average margin of 14 percent. Margins for agencies that admitted higher shares of posthospital or PAC patients—those in the second through fifth quintiles—averaged margins of 19 percent to 21 percent. Agencies in Texas, a state with an aberrant pattern of utilization relative to national benchmarks, accounted for a disproportionate share of agencies in the first quintile of this measure.

Home health care services: Assessing payment adequacy and updating payments

The share of episodes qualifying for therapy payments (episodes with six or more therapy visits) is consistent with the Commission’s past conclusion that these episodes were overvalued under the case-mix system in effect in 2010 (Table 9-10). Agencies with the lowest share of these episodes had margins of 13 percent, while those with the highest share of these episodes averaged margins of 25 percent. Medicare made changes to the case-mix system that lowered payments for therapy episodes in 2012. Under the revised system, the margins for agencies in the lower quintiles would likely be higher and the margins for agencies in the upper quintiles would be lower. There was also a limited relationship between the share of an agency’s episodes provided to beneficiaries dually eligible for Medicare and Medicaid and Medicare margins. Agencies with the highest share of dual eligibles (fifth quintile) had margins of 13 percent, while the margins averaged 20 percent to 21 percent for agencies in the first through fourth quintiles. Similar to results for the share of agencies preceded by a hospitalization or PAC use, agencies in Texas accounted for about one-third of the providers in the highest quintile. Historically, Medicare margins have varied widely among HHAs. To better understand the factors driving this variation, in a prior analysis the Commission examined the characteristics of high-margin and low-margin agencies in 2007. The analysis concluded that the greatest difference between high-margin and low-margin agencies was the average cost per visit and that the quality of care and patient severity did not differ significantly among these two groups. Agencies with lower costs had better profit margins, suggesting that cost efficiency was an important determinant of agency profits (Medicare Payment Advisory Commission 2010).

TABLE

9–1 0

Medicare margins of freestanding home health agencies based on select characteristics, 2010 Average Medicare margin

All agencies

19%

Agencies ranked by share of episodes preceded by a hospitalization or PAC service (quintile) First (low share)

Second

Third

Fourth

Fifth (high share)

Agencies ranked by share of episodes qualifying for therapy payments (quintile) First (low share)

Second

Third

Fourth

Fifth (high share)

Agencies ranked by share of episodes provided to dual-eligible beneficiaries (quintile) First (low share)

Second

Third

Fourth

Fifth (high share)

Note:

PAC (post-acute care). Table displays average Medicare margins for groups of agencies in each quintile. Each agency was assigned to a quintile by computing the share of its episodes with a given characteristic as a percentage of all of the agency’s episodes. Weighted Medicare margins were calculated for each agency group.

Source: MedPAC analysis of cost reports and CMS Datalink file.

TABLE

9– 9

Medicare margins for subcategories of rural agencies, 2011

Subcategory

Margin

Micropolitan

15.0%

Rural, adjacent to urban or micropolitan

15.2

Rural, not adjacent to urban or micropolitan

14.7

Note:

A micropolitan county has a population of 10,000 to 50,000. Table excludes some rural agencies that lacked sufficient data for determining rural subcategory.

Source: MedPAC analysis of home health cost report files from CMS.

Efficient HHAs serve patients similar to patients served by all other HHAs

The Medicare Modernization Act of 2003 requires that the Commission consider the financial performance of an efficient provider in its review of payment adequacy. We examined the quality and cost efficiency of freestanding HHAs to identify a cohort that demonstrates better performance on these metrics relative to its peers (Table 9-11, p. 204). The measure of cost is a risk-adjusted cost per episode, and the measure of quality is a risk-adjusted measure of hospitalization. Our approach categorizes an HHA as efficient if the agency is in the best third on at

Report to the Congress: Medicare Payment Policy | March 2013

203

TABLE

9–11

Performance of relatively efficient home health agencies, 2007–2009

Provider characteristics Number of agencies Share of for-profit agencies Medicare margin 2010 2009 Quality Hospitalization rate (2009)

All 2,223 66% 19.4% 18.7% 28%

Costs and payments Average payment per episode (2009) Cost per visit, standardized for wages and CMI (2009)

Relatively efficient provider 320 63% 23.8% 24.8% 23%

All other providers 1,903 67% 18.5% 17.6% 29%

$2,815

$2,803

$2,817

132

115

135

17.5

16.8

17.6

Visits per episode Total visits per episode (2009) Share of visits by type Skilled nursing visits Aide visits MSS visits Therapy visits

49% 15% 1% 35%

Size, 2009 (number of 60-day payment episodes) Mean Median Share of episodes, 2009 Low-use episode Outlier episode Community-admitted episodes Therapy episodes

11% 2% 58% 38%

Share of agencies by region New England Middle Atlantic South Atlantic East North Central East South Central West North Central West South Central Mountain Pacific

8% 9% 22% 19% 5% 9% 8% 9% 10%

1,111 714 12% 2% 51% 38% 10% 11% 17% 13% 2% 9% 1% 9% 27%

31% 77.4 21%

30% 77.6 13%

Beneficiary demographics, 2009 Share of episodes provided to dual eligible Medicare/Medicaid beneficiaries Average age Share of episodes provided to rural beneficiaries Note:

1,003 575

48% 16% 1% 35% 985 552 11% 2% 60% 38% 7% 9% 23% 20% 6% 9% 9% 9% 8%

31% 77.3 22%

CMI (case-mix index), MSS (medical social services). A home health agency is classified as relatively efficient if it is in the best third of performance of quality or cost and is not in the bottom third of either measure for three consecutive years (2007–2009). Quality is measured using a risk-adjusted measure of hospitalization and cost is measured using risk-adjusted cost per episode. Sample includes freestanding agencies with complete data for three consecutive years. Agencies in highutilization areas were excluded. Low-use episodes are those with 4 or fewer visits in a 60-day episode. Outlier episodes are those that received a very high number of visits and qualified for outlier payments. Community-admitted episodes are those episodes that were not preceded by a hospitalization or prior post-acute care stay. Therapy episodes are those with six or more therapy visits.

Source: Medicare cost reports and home health standard analytic file.

204

52% 12% 1% 35%

Home health care services: Assessing payment adequacy and updating payments

least one measure (either low cost per episode or a low hospitalization rate) and is not in the bottom third of the other measures for three consecutive years (2007–2009). About 14 percent of agencies met these criteria in this period. The analysis indicates that relatively efficient HHAs can provide above-average quality while incurring belowaverage costs. Relatively efficient agencies had margins that were 6 percentage points to 7 percentage points higher with a hospitalization rate that was 20 percent lower compared with other HHAs, and the average cost per visit was 15 percent lower compared with other HHAs.10 The median relatively efficient agency was larger than the median in the all-other-agency cohort. Relatively efficient HHAs provided about 0.8 fewer visit per episode but provided a similar mix of nursing, aide, therapist, and social work visits. Relatively efficient providers were also typically larger in size than other agencies. The agencies had about the same share of high-cost outlier episodes and low-use episodes, suggesting they serve about the same share of beneficiaries at the extremes of utilization. Relatively efficient agencies had more episodes that were not preceded by a hospitalization but about the same share of episodes that qualified for additional therapy payments. The Commission’s criteria for identifying efficient providers exclude all providers operating in areas that have unusually high rates of utilization. Therefore, it is not surprising that relatively efficient agencies were found more frequently in areas with lower utilization, such as New England and the Pacific region. Areas of the country with questionable patterns of utilization—such as the South Atlantic, East South Central, and West South Central—accounted for a smaller share of agencies. Relatively efficient agencies appear to serve beneficiaries with characteristics similar to those other agencies serve. The share of episodes provided to dual-eligible beneficiaries was similar. The mean beneficiary age was also similar for the two cohorts of agencies. A smaller share of the episodes provided by relatively efficient providers was for beneficiaries in rural areas. The high margins of relatively efficient agencies reinforce that Medicare overpays for home health care. Relatively efficient agencies in 2009 had a Medicare margin about 40 percent higher than the margin for all other agencies. They were typically larger and had lower costs per visit, indicating some economies of scale. The relatively

efficient agencies achieved these profits even though they served mostly similar patients, provided a similar mix of services, and had about the same average payment per episode as other agencies. Average providers can achieve high margins in Medicare, but relatively efficient providers reap even higher profits. Projecting margins for 2013 In modeling 2013 payments and costs, we incorporate policy changes that will go into effect between the year of our most recent data, 2011, and the year for which we are making margin predictions, 2013. The major changes are: •

payment updates in 2012 and 2013, equal to market basket minus 1 percent (per PPACA) for each year;

•

reductions to account for coding improvements in 2011 (–3.79 percent) and 2012 (–1.32 percent);

•

3 percent add-on in effect for rural areas in 2012; and

•

assumed episode growth of 0.5 percent a year for 2012 and 2013, higher than the trend for 2011.

On the basis of these policies and assumptions, the Commission projects a margin of 11.8 percent in 2013. Medicare has always overpaid for home health services under PPS Payments for home health care have substantially exceeded costs since Medicare established the PPS. In 2001, the first year of PPS, margins equaled 23 percent. The high margins in the first year suggest that the PPS established a base rate well in excess of costs. The base rate assumed that the average number of visits per episode would decline about 15 percent between 1998 and 2001, while the actual decline was about 32 percent (Table 9-12, p. 206). By providing fewer visits than anticipated, HHAs were able to garner extremely high average payments relative to the services provided. Margins have stayed high since 2001 because annual increases in payment have exceeded growth in costs. The Commission’s review of the annual change in cost per episode suggests that cost growth has been minimal, typically less than 1 percent. In some years, a decline has been observed. Average payments per episode have generally increased from year to year, driven by market basket increases and increases in the average case-mix index.

Report to the Congress: Medicare Payment Policy | March 2013

205

TABLE

9–12

Medicare visits per full episode before and after implementation of PPS Visits per episode

Change in:

Type of visit

1998

2001

2011

1998–2001

Skilled nursing

14.1

10.5

9.5

–25%

Therapy (physical, occupational, and speech–language) Home health aide Medical social services Total Note:

2001–2011 –10%

3.8

5.2

6.1

13.4

5.5

2.9

–59

–48

0.3

0.2

0.1

–36

–31

31.6

21.4

18.6

–32

–13

PPS (prospective payment system). The PPS was implemented in October 2000.

Source: Home health standard analytic file.

This structural mismatch between payment levels and cost growth led to the Commission recommending in March 2010 that Medicare rebase payments to be closer to costs (Medicare Payment Advisory Commission 2010). PPACA has mandated some reductions for home health care that begin to reduce payments, but these reductions would leave HHAs with margins well in excess of cost. Overpaying for home health care has negative financial consequences for the federal government and the beneficiary; implementing the Commission’s prior recommendation for rebasing would better align Medicare’s payments with the actual costs of HHAs. The need to reset the base rate in Medicare is particularly acute because high margins exist across the range of agency types. Urban, rural, for-profit, and nonprofit agencies have margins in excess of 12 percent. While some agencies have margins significantly lower than average, the Commission’s review of agencies in 2007 found that these differences are primarily due to their

206

higher costs. These higher costs do not appear to be related to patient severity, as low-margin agencies, for most measures, did not serve more severely ill patients.

How should Medicare payments change in 2014? A review of the Commission’s indicators suggests that access is more than adequate in most areas and that aggregate Medicare payments are well in excess of costs. Our recommendations from 2011 included multiyear payment changes intended to restructure the incentives of the home health benefit as well as address the high Medicare margins. These recommendations call for expanded efforts to fight fraud, improving beneficiary and provider incentives, and rebasing home health payments (see text box for a summary of recommendations from 2011 and 2012). ■

Home health care services: Assessing payment adequacy and updating payments

Strengthening incentives for effective and efficient use of the home health benefit

n 2011, the Commission noted several problems with the home health care benefit and made several recommendations to reduce fraud, improve provider and beneficiary incentives, and eliminate the high overpayments under the home health care prospective payment system. Recommendation 8-1, March 2011 report The Secretary, with the Office of Inspector General, should conduct medical review activities in counties that have aberrant home health utilization. The Secretary should implement the new authorities to suspend payment and the enrollment of new providers if they indicate significant fraud. The Patient Protection and Affordable Care Act of 2010 (PPACA) expanded Medicare’s authority to stop payment for fraudulent or suspect services, and last year the Commission recommended that the Secretary exercise this new authority to curb fraud in home health care. So far, it does not appear that the Secretary has used this authority in any broad capacity. For many years, the Commission has published a list of counties with questionable utilization patterns (Table 9-6, p. 200). As the Commission recommended in our March 2011 report, these counties would be appropriate areas for the Secretary to exercise new PPACA authorities for investigating and interdicting home health fraud. Implications 8-1 Spending

•

The Congressional Budget Office has already scored savings from the PPACA provision, so its baseline already assumes savings for the new authorities. Implementing this authority would lower home health spending if fraud were discovered. CMS and the Office of Inspector General would incur some administrative expenses.

Beneficiary and provider

•

Appropriately targeted reviews would not affect beneficiary access to care or provider willingness to serve beneficiaries.

Recommendation 8-2, March 2011 report The Congress should direct the Secretary to begin a two-year rebasing of home health rates in 2013 and eliminate the market basket update for 2012. Medicare has overpaid for home health since establishment of the prospective payment system (PPS) in 2000. The higher payments create financial incentives that may encourage providers to deliver services even when they are unnecessary or of low value. Payments should be rebased as soon as practicable, with a short period of time that allows for an appropriate transition to the lower level of payments (e.g., no more than three years). Our recommendation would also eliminate the market basket update during rebasing. In addition, the Commission believes that our recommendation to eliminate the use of therapy thresholds in the PPS should be implemented along with rebasing. This change would ensure that providers do not attempt to offset rebasing with higher payments by increasing the number of therapy visits they provide. The need to rebase is particularly acute because Medicare’s coverage guidance for the home health care benefit is under revision. A recent court case between the Department of Health and Human Services and the Center for Medicare Advocacy will require the program to clarify the language in its benefit manual regarding the coverage of services needed to maintain or prevent deterioration of a patient’s current condition. Until CMS revises the benefit manuals, specifies instructions, and trains claims contractors and providers, it is hard to estimate how this change will affect utilization. If these changes broaden access to care, then expenditures could increase. The Commission expects that a rebasing may cause some agencies to leave the Medicare program, but this effect may be offset by the entry of new providers. The barriers to entry in home health care are lower than in other Medicare services. It does not require extensive capital expenditures like facility-based providers, and many states do not require certificate-of-need analysis to establish a new home health agency. (continued next page)

Report to the Congress: Medicare Payment Policy | March 2013

207

Strengthening incentives for effective and efficient use of the home health benefit Implications 8-2 Spending

•

This recommendation would reduce Medicare spending by $750 million to $2 billion in 2014 and by $5 billion to $10 billion over five years.

Beneficiary and provider

•

Some reduction in provider supply is likely, particularly in areas that have experienced rapid growth in the number of providers. Access to appropriate care is likely to remain adequate, even if the supply of agencies declines.

who need therapy over patients who do not and have encouraged providers to deliver services that are of marginal value to a beneficiary. Our recommendation would use patient characteristics to set payment for therapy, the same approach Medicare currently uses for setting payment for all other services covered in the home health PPS. Implications 8-3 Spending

•

Recommendation 8-3, March 2011 report

The approaches are designed to be implemented in a budget-neutral manner and should not have an overall impact on spending.

The Secretary should revise the home health casemix system to rely on patient characteristics to set payment for therapy and nontherapy services and should no longer use the number of therapy visits as a payment factor.

Beneficiary and provider

The Commission is concerned that Medicare’s home health PPS encourages providers to base therapy regimens on financial incentives and not patient characteristics. The PPS uses the number of therapy visits provided in an episode as a payment factor: The more visits a provider delivers, the higher the payment. The higher payments obtained by meeting the visit thresholds have led providers to favor patients

Recommendation 8-4, March 2011 report

•

Patients who need therapy may see some decline in access, but these services would be available on an outpatient basis after the home health episode ended.

The Congress should direct the Secretary to establish a per episode copay for home health episodes that are not preceded by hospitalization or post-acute care use. The health services literature has generally found that beneficiaries consume more services when cost sharing (continued next page)

208

Home health care services: Assessing payment adequacy and updating payments

Strengthening incentives for effective and efficient use of the home health benefit is limited or nonexistent, and some evidence suggests that the additional services do not always contribute to better health. The lack of cost sharing is a particular concern for home health care, because PPS pays for care on a per episode basis that rewards additional volume. The lack of a cost-sharing requirement stands in contrast to most other Medicare services, which generally require the beneficiary to bear some of the costs of Medicare services.

episode copayment for services not preceded by a hospitalization or other post-acute use.11 The high rate of volume growth for these types of episodes, which have more than doubled since 2001, suggests there is significant potential for overuse. The addition of a copayment would allow for beneficiary cost consciousness to counterbalance the permissiveness of the benefit’s use criteria and the volume-rewarding aspects of Medicare’s per episode payment policies.

One concern with cost sharing is that it can lead beneficiaries to reduce their use of effective as well as ineffective care. Although some studies have found evidence of adverse effects of reduced care due to cost sharing (Chandra et al. 2010, Rice and Matsuoka 2004), the RAND Health Insurance Experiment, concluded that, on average, nonelderly patients who consumed less health care because of cost sharing suffered no net adverse effects (Newhouse 1993). The Commission’s review of the impact of medigap insurance generally found that beneficiaries with this insurance had higher total Medicare spending (Medicare Payment Advisory Commission 2009). The results of the RAND Health Insurance Experiment and the Commission’s study suggest that a home health care copay would decrease utilization for home health care and result in lower overall Medicare spending.

Implications 8-4

To encourage appropriate utilization, the Commission recommended that Medicare add an

Spending

•

A copay of $150 per episode (excluding low-use and posthospital episodes) would reduce Medicare spending by $250 million to $750 million in 2014 and by $1 billion to $5 billion over five years. Expenditures for services would decrease because some beneficiaries who would otherwise use home health services might decline them. Since many of these services are funded by Part B, decreases in spending growth would reduce Part B premiums.

Beneficiary and provider

•

Some beneficiaries might seek services through outpatient or ambulatory care, for which Medicare already has cost-sharing requirements. Some beneficiaries who need relatively few services would have lower cost sharing if they substituted ambulatory care for home health care. ■

Report to the Congress: Medicare Payment Policy | March 2013

209

Endnotes 1 The exceptions pertain to therapy services that require equipment that is not available in the home, such as whirlpool therapy and other treatments requiring specialized equipment. 2 Medicare pays for most services under the home health PPS through a bundled 60-day payment that does not have payment amounts for individual services. The per visit payment amounts for home health services indicated here have been estimated using a pro-rata share of the average full episode payment in 2010, $2,877. This amount was divided among the different visit types (nursing, aide, therapy, and social work) based on each discipline’s share of standardized costs in the average home health episode. Costs were standardized with the per visit payment amounts Medicare uses to reimburse episodes with fewer than five visits, referred to as the low utilization payment adjustment (LUPA). The LUPA rate is useful because it allows the weights for allocating the payment to each discipline to reflect the relative costliness of each discipline (i.e., that nursing is more costly than aide services). However, the payment levels included in the 60-day episode payment are set separately from the LUPA rates, so LUPA rates cannot be used as a proxy for the per visit amount assumed in the full 60-day payment. 3 Surety bond firms review the organizational and financial integrity of an HHA and agree to cover the Medicare obligations, up to a set amount, for those agencies that the surety bond firm believes are low risk. 4 Our measure of access is based on data collected and maintained as part of CMS’s home health compare database as of November 2012. The service areas listed are ZIP codes where an agency has provided services in the past 12 months. This definition may overestimate access because agencies need not serve the entire ZIP code to be counted as serving it. At the same time, the definition may understate access if HHAs are willing to serve a ZIP code that did not receive a request in the previous 12 months. The analysis excludes beneficiaries with unknown ZIP codes.

6 In 2011, Medicare implemented a –3.79 adjustment to account for changes in agency coding practice that appeared unrelated to severity. PPACA also reduced the payment update by 1 percent and had a base rate reduction of 2.5 percent. The combined impact of all these adjustments lowered the 60-day episode payment rate by 5.2 percent. 7 The Commission’s review of margins has generally found that larger agencies have higher margins, suggesting some economies of scale for HHAs. These economies, combined with Medicare’s per unit payment system, suggest that agencies with higher episode volume can achieve higher profits. 8 Home health care users were categorized into PAC users and community-admitted users based on the share of their episodes in 2010 that were preceded by a hospitalization or other PAC use. Users with more than 50 percent of their episodes preceded by a hospitalization or PAC use were categorized as primarily PAC users; those with less than 50 percent of their episodes preceded by these events were categorized as primarily community-admitted users. 9 The home health care PPS includes additional payments for therapy at the 6th, 14th, and 20th therapy visits (with incremental increases in the intervals between these numbers). In 2011, CMS implemented a requirement that agencies conduct additional reviews shortly before the 14th and 20th therapy visits but made no similar requirement for the 6th therapy visit. 10 This risk-adjusted measure of hospitalization includes those that occur at the end of a home health stay or within 30 days of the end of a stay. 11 The recommendation applied only to full episodes—those that include five or more visits.

5 Certificate-of-need laws vary from state to state, and not all states have them. In general, the laws require that an area have a demonstrated need for additional health care services before a new provider is permitted to enter the market.

210

Home health care services: Assessing payment adequacy and updating payments

References Benjamin, A. E. 1993. An historical perspective on home care policy. Milbank Quarterly 71, no. 1: 129–166. Center for Medicare & Medicaid Innovation, Centers for Medicare & Medicaid Services, Department of Health and Human Services. 2012. Center for Medicare & Medicaid Innovation website. http://www.innovations.cms.gov/. Chandra, A., J. Gruber, and R. McKnight. 2010. Patient costsharing and hospitalization offsets in the elderly. American Economic Review 100, no. 1 (March 1): 193–213. Cheh, V., N. Duda, B. L. Carlson, et al. 2007. Evaluation of the Home Health Independence Demonstration: Barriers to a successful experiment were multifaceted and difficult policy issues remain. Prepared for the Centers for Medicare & Medicaid Services. Princeton, NJ: Mathematica Policy Research, Inc. Cryer, L., S. B. Shannon, M. Van Amsterdam, et al. 2012. Costs for ‘hospital at home’ patients were 19 percent lower, with equal or better outcomes compared to similar inpatients. Health Affairs 31, no. 6 (June): 1237–1243. Feder, J., and J. Lambrew. 1996. Why Medicare matters to people who need long-term care. Health Care Financing Review 18, no. 2 (Winter): 99–112. Government Accountability Office. 1996. Medicare: Home health utilization expands while program controls deteriorate. GAO/ HEHS–96–16. Washington, DC: GAO. Liu, K., D. Wissoker, F. Porell, et al. 2003. Agency closings and changes in Medicare home health use: 1996–1999. Report prepared for the Office of Disability, Aging, and Long-Term Care Policy; Office of the Assistant Secretary for Planning and Evaluation; Department of Health and Human Services. Contract #HHS–100–97–0010. Wasington, DC: ASPE. McCall, N., J. Korb, A. Petersons, et al. 2004. Decreased home health use: Does it decrease satisfaction? Medical Care Research and Review 61, no. 1 (March): 64–88. McCall, N., J. Korb, A. Petersons, et al. 2003. Reforming Medicare payment: Early effects of the 1997 Balanced Budget Act on postacute care. Milbank Quarterly 81, no. 2: 277–303.

Medicare Payment Advisory Commission. 2012. Report to the Congress: Medicare and the health care delivery system. Washington, DC: MedPAC. Medicare Payment Advisory Commission. 2011. Report to the Congress: Medicare payment policy. Washington, DC: MedPAC. Medicare Payment Advisory Commission. 2010. Report to the Congress: Medicare payment policy. Washington, DC: MedPAC. Medicare Payment Advisory Commission. 2009. Report to the Congress: Improving incentives in the Medicare program. Washington, DC: MedPAC. Medicare Payment Advisory Commission. 2004. Report to the Congress: Medicare payment policy. Washington, DC: MedPAC. Newhouse, J. P. 1993. Free for all? Lessons from the RAND Health Insurance Experiment. Cambridge, MA: Harvard University Press. Office of Inspector General, Department of Health and Human Services. 2012. Inappropriate and questionable billing by Medicare home health agencies. OEI–04–11–00240. Washington, DC: OIG. Office of Inspector General, Department of Health and Human Services. 2001. The physician’s role in Medicare’s home health services. OEI–02–00–00620. Washington, DC: OIG. Office of Inspector General, Department of Health and Human Services. 1997. Review of Medicare home health services in California, Illinois, New York, and Texas. A–04–99–01994. Washington, DC: OIG. Rice, T., and K. Y. Matsuoka. 2004. The impact of cost-sharing on appropriate utilization and health status: A review of the literature on seniors. Medical Care Research and Review 61, no. 4 (December): 415–452. Wolff, J., A. Meadow, C. Weiss, et al. 2008. Medicare home health and the role of physicians. Presentation at Academy Health, Medicare and Post-Acute Care. June 8.

Report to the Congress: Medicare Payment Policy | March 2013

211