U.S. Department of Health and Human Services

Post-Acute and Long-Term Care: A Primer on Services, Expenditures and Payment Methods

David C. Grabowski, PhD

Harvard Medical School

June 2010

PDF Version: http://aspe.hhs.gov/daltcp/reports/2010/paltc.pdf (20 PDF pages)

This report was prepared under contract #HHSP233201000192P between U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and the Harvard Medical School. For additional information about this subject, you can visit the DALTCP home page at http://aspe.hhs.gov/_/office_specific/daltcp.cfm or contact the ASPE Project Officer, Jennie Harvell, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201. Her e-mail address is: Jennie.Harvell@hhs.gov.

The opinions and views expressed in this report are those of the authors. They do not necessarily reflect the views of the Department of Health and Human Services, the contractor or any other funding organization.

TABLE OF CONTENTS

BACKGROUND

NURSING FACILITIES

SKILLED NURSING FACILITIES

HOME HEALTH AGENCIES

INPATIENT REHABILITATION FACILITIES

LONG-TERM CARE HOSPITALS

REFERENCES

BACKGROUND

Medicare and Medicaid are important payers of post-acute and long-term care. In 2008, the Medicare program spent $49.9 billion on post-acute services among fee-for-service (FFS) beneficiaries (see Table 1). Similarly, Medicaid spent $56.3 billion in 2008 on nursing home care. This section of the report reviews the services, expenditures and payment methods for care at nursing facilities (NFs) and across the various post-acute care sites including skilled nursing facilities (SNFs), home health agencies (HHAs), inpatient rehabilitation facilities (IRFs), and long-term care hospitals (LTCHs).

NURSING FACILITIES -- SERVICES, EXPENDITURES, AND PAYMENT METHODS

NFs provide a range of compensatory, rehabilitative, psychosocial and social services to residents. In 2008, roughly 16,000 Medicaid-certified NFs were in operation in the United States. Roughly two-thirds were for-profit, about 60% were chain-owned and approximately 5% were hospital-based. Importantly, nearly 85% of Medicaid-certified NFs are also Medicare-certified SNFs.

Medicaid is the dominant purchaser of chronic-care (long-stay) nursing home services in the United States. Almost half of NF expenditures are paid for by Medicaid, while Medicaid recipients constitute 70% of all nursing home bed days (Rhoades and Sommers, 2000). Medicaid NF expenditures are shared between federal and state governments. In 2007, the Federal Government paid 56.7% of Medicaid NF expenditures, and states paid 43.3%. Other public payers of long-stay nursing home care include the Veteran’s Administration (VA) and other state and local sources outside of Medicaid. Importantly, Medicare pays only for post-acute (short-stay) SNF services (see SNF section below).

In 2008, Medicaid nursing home expenditures totaled $56.3 billion. In 2004, the average Medicaid per diem was $131.66 (Grabowski et al., 2008), which inflates to $148.62 in 2008, using the SNF market basket update. The average length-of-stay for Medicaid-financed nursing home care is about two years, with significant variation around this mean.

Medicaid nursing home reimbursement policies vary dramatically across states and most states have made changes to their reimbursement system multiple times since the onset of the Medicaid program. Over that period of time, most states have transitioned from a cost-based, retrospective reimbursement model to some form of prospective payment. As of fiscal year (FY) 2002, 39 states used purely prospective reimbursement methods to set NF Medicaid payment rates. Among states using a prospective system, rates were set using a class or flat-rate method in four states, facility-specific in 18 states, resident specific in two states, and both facility and resident specific in 14 states (Grabowski et al., 2004). Only two states (Maryland and Wyoming) used purely retrospective systems in FY 2002, while seven states (North Carolina, Virginia, Tennessee, Alabama, Kentucky, Michigan, New York, and Texas) used combination systems, which combine elements of prospective and retrospective payment systems.

In the majority of states, nursing homes are paid a prospective per diem rate by the state Medicaid program. The rate is typically calculated based on a series of cost centers encompassing direct care services, indirect care services, administration and capital.

States have broad discretion in the specific methodology used to formulate Medicaid nursing home payment rates. Historically, comprehensive change has involved movement from retrospective systems to prospective systems. More recent trends, however, include adoption of case-mix systems, which adjust payment for patient acuity, as well as the adoption of fair rental approaches to reimbursing capital expenses, which permit greater control of rate changes and allow less inflation in the valuation of capital than more widely used historical approaches. Recent trends also include adoption of wage-pass through policies through which reimbursement increases are directed toward wages and benefits for direct care workers. In addition, some states have begun to experiment with “pay-for-performance” (P4P) incentives, which provide nursing homes with higher levels of reimbursement based on achievement of desired outcomes. A recent review found seven active state P4P programs in 2007 using a range of performance measures such as staffing, certification, efficiency, quality indicators, resident/family satisfaction, quality-of-life and other outcomes. The payment incentives were relatively modest accounting for less than 5% of the Medicaid per diem. Less radical adjustments include rate freezes, inflation adjustments, ceiling limitations, cost center developments, ancillary service inclusions, efficiency incentives, and cost rebasing (to a more recent cost report), just to name a few.

The majority of states with prospective payments systems have a case-mix adjustment to compensate nursing homes for the admission and care of more disabled (i.e., costlier) patients. Although several case-mix methods are currently in practice, the majority of systems classify residents into homogeneous categories based on their resource utilization. Associated with each of these categories is a case-mix index (CMI), which represents, at least relatively, the time or cost of the average resident in the group (Fries, 1990). A higher CMI indicates a greater degree of complexity and consequently a greater need for input resources. Perhaps the best-known methodology of this type is the Resource Utilization Groups (RUG) system, which is currently in its third version (Fries et al., 1994). The RUG-III system, with 44 distinct resource groups, has been shown to achieve 55.5% variance explanation of total per diem costs for nursing home residents (Fries et al., 1994). Patient characteristics and service use are determined using the Minimum Data Set, a patient assessment instrument. As of 2004, Medicaid nursing home case-mix reimbursement was available in 35 states, up from 19 states in 1991 and just four states in 1981. These states include Arizona, Arkansas, Colorado, Delaware, Georgia, Illinois, Idaho, Iowa, Indiana, Kansas, Kentucky, Louisiana, Massachusetts, Maryland, Maine, Minnesota, Mississippi, Montana, New Hampshire, North Dakota, Nebraska, New Jersey, Nevada, New York, Ohio, Pennsylvania, South Carolina, South Dakota, Texas, Utah, Virginia, Vermont, Washington, Wisconsin, and West Virginia.

Capital represents a much smaller portion of total NF costs (10%-15%) than direct patient care. Most states reimburse for capital on the basis of historical costs and usually include actual interest expenses, lease payments, and sometimes the payment of a return on equity. By freezing the value of a home at its initial construction or renovation costs, however, historical methods fail to recognize appreciation in market value, thereby penalizing owners of functional, though fully depreciated, assets. This discourages long-term ownership and creates incentives for frequent turnover through leases or sales at inflated prices attractive to short-term real estate speculators.

Cognizant of frequent ownership turnover Congress included measures in the Deficit Reduction Act (DEFRA) of 1984 and Consolidated Omnibus Reconciliation Act of 1985 in order to limit asset revaluation for Medicaid reimbursement purposes. These provisions combined with growing awareness of existing problems led states to adopt alternative approaches to reimbursing capital. The result has been movement away from historical and market valuation methods, which permit less control of rate changes by allowing greater inflation in the valuation of capital, to fair rental approaches that pay a simulated rent (or return on the appraised value of a facility’s assets) that is permitted to change with market conditions, and combination systems that combine less inflationary approaches with historical methods. An advantage of the fair rental approach is that it accounts for increases in the value of nursing home assets without requiring turnover or refinancing to occur before owners realize gains on their investments. In the context of health information technologies acquisition, a fair rental approach to capital reimbursement would likely encourage greater investment in the technologies.

Recent data on state methods for reimbursing capital expenses do not exist. However, Harrington and colleagues reported both a marked increase in the use of fair rental methodologies from one state in 1984 to 18 states in 1998, along with a concomitant decrease in the use of other approaches to valuing capital during this time period (Harrington et al., 1999; DuNah et al., 1994). This includes the use of historical methods, which declined from 35 states in 1984 to 30 in 1989 and 22 in 1992 before reversing somewhat after 1993 (to 25 states). By 1998, 26 states used historical approaches, nine fair rental methods, and nine fair rental methods combined with other capital valuation strategies. Less frequently used were the market valuation approach, which bases reimbursement on the price a home would bring on the private market, and the imputed value approach, which uses mathematical formulas to base reimbursement on a composite of different costing methods or calculations.

Under prospective payment, states have two primary means of updating payment rates over time. First, they can rebase rates using Medicaid cost report data, and second, they can use an inflation factor to adjust payment rates in the absence of rebasing. Reimbursement systems that use older cost reports are generally considered more cost containing than reimbursement systems that use newer cost reports because they are tied less closely to nursing homes’ current costs. States can limit rate increases in a variety of ways, including, for example: (1) freezing rates, (2) adjusting rates at less than actual inflation, (3) capping the growth in rates so that rate increases cannot exceed a certain percentage under a cap, and (4) providing monetary efficiency incentives, which typically allow nursing homes with costs below a predetermined amount in one or more cost centers to have a portion of the difference incorporated into their per diem rates. States that have employed efficiency incentives include Alabama, Colorado, Connecticut, Illinois, Iowa, North Dakota and Pennsylvania (U.S. General Accounting Office, 2003).

SKILLED NURISNG FACILITIES -- SERVICES, EXPENDITURES, AND PAYMENT METHODS

SNFs offer skilled nursing care and rehabilitation services, such as physical and occupational therapy and speech-language pathology services, to Medicare beneficiaries following an acute care hospital stay. In 2007, rehabilitation services accounted for 88% of all Medicare days in freestanding SNFs (Medicare Payment Advisory Commission (MedPAC), 2009). The supply of SNFs in the United States has remained relatively constant over the past few years, composed of roughly 15,000 facilities that are two-thirds for-profit and mostly freestanding (93%) (MedPAC, 2009). Importantly, the majority (~93%) of SNFs are also Medicaid-certified NFs. With the Centers for Medicare and Medicaid Services (CMS) approval, certain Medicare-certified “swing bed” hospitals (typically small, rural hospitals and critical access hospitals) may also provide SNF services.

The dominant public payer of SNF services is the Medicare program. A small amount of skilled services may also be covered by state Medicaid programs, but these services are generally paid for in a manner similar to the NF services discussed above. In terms of other public payers, the VA accounts for a very small percentage of overall skilled nursing spending.

In 2008, Medicare spent $22.9 billion on 2.56 million covered SNF admissions. Thus, the approximate Medicare payment per SNF stay was just under $9,000, with an average length-of-stay of 27 days. The cost per day averaged approximately $330 and the median was $259. The three most prevalent diagnoses among Medicare SNF patients in 2007 were major joint and limb reattachment of lower extremity (5.5%), heart failure and shock (3.6%), and simple pneumonia and pleurisy with complication or comorbidity (3.4%) (MedPAC, 2009).

Medicare’s prospective payment system (PPS) for SNF services was implemented based on the start of the facility FY on or after July 1, 1998. Prior to this date, SNFs were paid on a cost-based basis. Under the Medicare SNF PPS, facilities are paid a predetermined daily rate, up to 100 days, but only after a qualifying hospital stay of at least three days. The per diem prospective payment rate for SNFs covers routine, ancillary, and capital costs related to the services provided under Part A of the Medicare program (CMS, 2010). Certain high-cost, low-probability ancillary services (e.g., radiation therapy or cardiac catheterization) are paid separately.

Adjustments to the SNF Medicare payment rates are made according to a resident’s case-mix and geographic factors associated with wage variation (MedPAC, 2009). The Medicare SNF payment rate is adjusted for geographic differences in labor costs using the hospital wage index. Specifically, 70% of the base rate is adjusted for area wages and then added to a non-case-mix adjusted component reflecting the costs of room and board, linens and administrative services. The SNF payment rate is adjusted for patient case-mix using the 53 payment categories in the RUG-III system. Each RUG category has an associated nursing and therapy weight that is applied to the base payment rate. Patients are assigned to a RUG based on their health status and services delivered. Specifically, assignment of a beneficiary to particular RUG is based on the number of minutes of therapy (physical, occupational, or speech) minutes that a patient used, the need for certain services (e.g., respiratory therapy or specialized feeding), the presence of certain conditions (e.g., pneumonia or dehydration), an index based on the patient’s ability to perform independently four activities of daily living (eating, toileting, bed mobility and transferring), and in certain instances, signs of depression. Patient characteristics and service use are determined using the Minimum Data Set, a patient assessment instrument.

Medicare updates SNF payment rates each year based on the projected increase in the SNF market basket index, a measure of the national average price level of goods and services. Over the last five years, the market basket updates have been around 3%. The MedPAC advises Congress each year on the adequacy of the SNF market update, and the Congress can adjust the payment update upwards or downwards based on this recommendation and other factors. Under the recently passed health reform legislation, the SNF market basket update will be reduced by a productivity adjustment beginning in FY 2012. The productivity adjustment will be set at the 10-year moving average of non-farm business productivity. The application of this productivity adjustment may result in a negative market basket adjustment (i.e., lower Medicare payment rates).

HOME HEALTH AGENCIES -- SERVICES, EXPENDITURES, AND PAYMENT METHODS

HHAs provide limited part-time or intermittent skilled nursing care (from a nurse, physical, or speech therapist) to Medicare beneficiaries in their homes. The Medicare HHA population uses a range of services including home health aide services (18% in 2008), skilled nursing care (55%), physical, occupational and speech therapy (26%), and medical social services (1%) (NAHC, 2010; MedPAC, 2009). In 2008, Medicare HHA users numbered 3.2 million, constituting roughly 9% of Medicare beneficiaries (MedPAC, 2010). The total number of HHAs in the United States was 9,283 in 2009, with over 99% of Medicare beneficiaries having access to at least one HHA in their service area (MedPAC, 2009; CMS, 2009).

The major public payer of home health care services is the Medicare program. However, every state Medicaid program is mandated to offer home health services to individuals who qualify for federal income maintenance payments (e.g., Social Security Income and Aid to Families with Dependent Children) and are “categorically needy.” Services must include visits by registered nurses, credentialed home health aide services and medical supplies and equipment. In addition, states may choose to cover physical, occupational, and speech therapies and audiology services. States reimburse agencies using various methodologies including FFS, prospective and cost-based methodologies (Kaiser Family Foundation, 2004).

Medicare expenditures on HHA services totaled $16.9 billion in 2008 over 3.2 million beneficiaries and 6.1 million episodes (MedPAC, 2009; MedPAC 2010). The average Medicare payment per HHA episode in 2008 was roughly $2,800. In 2006, Medicaid home health expenditures totaled $4.6 billion (Kaiser Commission on Medicaid and the Uninsured, 2009). The average Medicaid payment per visit in 2008 was $73.30 (Kaiser Commission on Medicaid and the Uninsured, 2009).

Prior to the Balanced Budget Act (BBA) of 1997, HHAs were paid by Medicare on the basis of their costs, up to pre-established per-visit limits. The 1997 BBA changed Medicare home health eligibility and coverage rules and reformed the payment methodology by instituting a PPS for home health care reimbursement. Implemented on October 1, 2000, Medicare pays HHAs a set payment rate for each 60-day episode of care.

Medicare payments to HHAs are packaged in 60-day blocks referred to as “episodes,” which begin as soon as the beneficiary is enrolled at an HHA. The majority of patients are discharged after a single 60-day payment episode. However, if the patient’s care is not complete in 60 days, then the patient may start another episode without a break in care, assuming the patient remains eligible for services (MedPAC, 2009).

Medicare HHA payments are adjusted according to the patient’s clinical and functional severity, the episode’s timing in a sequence of episodes and the use of therapy during the 60-day episode (MedPAC, 2009). Patient characteristics are determined using the Outcome and Assessment Information Set, a patient assessment instrument. Payments are also adjusted for differences in local prices. Other adjustments that can occur under special circumstances include an outlier payment (when costs exceed payment by a certain amount), low utilization payment adjustment (5 visits in the episode), a change-in-condition adjustment and a partial episode adjustment to account for transfers across agencies (MedPAC, 2009).

For the purposes of case-mix adjustment, Medicare beneficiaries are grouped into 153 home health resource utilization groups based on their clinical and functional status, and according to the types of services they will require (MedPAC, 2009). Payment rates are adjusted for geographic factors using the hospital wage index. Specifically, 77% of the base payment rate is adjusted for area wages and then added to the remaining non-labor portion (23%) of the rate. Capital is part of the non-labor portion of the rate, but it only accounts for a small portion (3%) of the overall HHA payment rate. Depreciation, interest, and other measures of capital related costs are used to estimate prices for capital inputs. The local area adjustment for wages is determined by the beneficiary’s residence. When a patient receives fewer than five visits during the 60-day episode, the HHA receives payments on a per-visit basis. When a patients’ care becomes too costly or requires a large number of visits, an outlier payment can sometimes be made to an HHA. In order to be eligible for this outlier payment, imputed episode costs must be 65% greater than the payment rate. When this occurs, the HHA receives a payment equal to 80% of the difference in addition to the standard episode payment. Medicare outlier payments comprise less than 10% of HHA total PPS payments (CMS, 2009).

Medicare updates HHA payments each year based on the projected increase in the HHA market basket index, a measure of the national average price level of goods and services. Over the period 2006-2010, the market basket updates have ranged between 2% and 3.6%. The MedPAC advises Congress each year on the adequacy of the HHA market update, and the Congress can adjust the payment update upwards or downwards based on this recommendation and other factors. For example, the DEFRA of 2005 eliminated the home health payment update for 2006, effectively “freezing” home health payment rates at the 2005 level. Under the recently passed health reform legislation, the HHA market basket update will be reduced by a productivity adjustment of one percentage point in FY 2011 and FY 2012. Beginning in FY 2015, the productivity adjustment will be set at the 10-year moving average of non-farm business productivity. The application of the productivity adjustment may result in a negative market basket adjustment (i.e., lower Medicare payment rates).

INPATIENT REHABILITATION FACILITIES -- SERVICES, EXPENDITURES AND PAYMENT METHODS

An IRF is a hospital that provides rehabilitation services such as physical, occupational, and speech therapies (CMS, 2010). In 2008, more than 332,000 FFS beneficiaries were enrolled in an IRF, which constituted nearly 1% of all FFS enrollees in the Medicare program during that year (MedPAC, 2010). To qualify for Medicare payment, an IRF must have a preadmission screening process; use a coordinated multidisciplinary team approach to therapy; have an experienced medical director of rehabilitation; and have no fewer than 60% of all patients admitted with at least 1 of 13 conditions (60% rule) (MedPAC, 2009). As of January 1, 2010, CMS issued new admission requirements for Medicare-eligible IRF patients: (1) they should require interdisciplinary care; (2) they should be medically stable enough to participate in intensive therapy; (3) they should require supervision by a physician; and, (4) they should require and expected to benefit from three hours of therapy at the least five times a week (MedPAC, 2010).

In 2008, just over 1,200 IRFs were in operation nationwide. Although an IRF was located in every state and the District of Columbia, some geographic variation exists in the supply of IRFs. An IRF may be a specialized unit with an acute care hospital (roughly 80% of facilities) or a specialized, larger freestanding facility.

The Medicare program is the dominant payer of IRF services, accounting for roughly 60% of all IRF cases nationwide. In 2008, Medicare’s aggregate expenditures for IRFs were $5.84 billion and the average payment per case was $16,649 (MedPAC, 2010). Some of the most common types of IRF cases were stroke (20.4%), hip fracture (16%), major joint replacement (13.1%), debility (9.1%), and neurological disorders (8%). The average length-of-stay for a Medicare IRF patient was 13.3 days in 2008.

Prior to 2002, IRFs were paid by Medicare on their average costs per discharge subject to a facility-specific limit that was adjusted annually. In January 2002, the IRF PPS was implemented under which IRFs are paid a predetermined per-discharge rate based primarily on the patient’s condition. The PPS payment rate covers all operating and capital costs that IRFs are expected to incur in the provision of intensive rehabilitation services.

The unit of payment under the IRF PPS is the discharge episode. Patients are assigned to one of 92 intensive rehabilitation categories called case-mix groups (CMGs) according to the primary condition for which the patient was admitted, afterwards the patients fall into tiers based upon secondary diagnosis and comorbidities (CMS, 2009). Patient characteristics are determined using the Inpatient Rehabilitation Facility Patient Assessment Instrument. Payments are also adjusted for differences in local prices.

For 87 of the CMGs, patients are assigned to a group based on their primary reason for intensive rehabilitation care (e.g., a stroke or burns), their age, their levels of functional and cognitive impairments and their comorbidities. The remaining five CMGs are for individuals discharged before the fourth day--termed short-stay outliers--and for those who die in the facility. Additionally, IRFs may also receive lower Medicare payments when the beneficiary transfers to another facility and when the length-of-stay is less than the standard for patients with the same condition. The base rate under the IRF PPS--$12,958 for FY 2009--is adjusted using a version of the hospital wage index. Specifically, 75% of the base payment rate is adjusted for area wages and the result is then added to the non-labor portion (25%) of the base rate.

Medicare payment rates are increased for IRFs located in rural markets, those that treat low-income patients, and those that are teaching institutions. Because rural IRFs tend to have fewer cases, longer lengths of stay, and higher average costs per case, payments to rural facilities are increased by 21.3%. Payments for IRFs treating low-income patients are adjusted based on the sum of the proportion of total Medicare days furnished to beneficiaries eligible for Supplemental Security Income benefits and the proportion of total patient days furnished to Medicaid patients not covered by Medicare. An IRF’s payments are adjusted if they are teaching institution according to the ratio of their residents to their average daily census.

Medicare has two outlier payment policies for IRFs. Once again, patients with short-stays (4 days) are paid lower rates. The other policy is for high-cost patients when costs exceed a fixed-loss threshold. In these cases, IRFs are paid their regular rate plus 80% of their costs above the fixed-loss threshold. Outlier payments are estimated to account for roughly 3% of overall IRF spending by Medicare (MedPAC, 2009).

Medicare updates IRF payments each year based on the projected increase in the IRF market basket index, a measure of the national average price level of goods and services. Over the period 2006-2010, the market basket updates have ranged between 2.5% and 3.6%. The MedPAC advises Congress each year on the adequacy of the IRF market update, and the Congress can adjust the payment update upwards or downwards based on this recommendation and other factors. Under the recently passed health care reform legislation, the IRF market basket update will be reduced by 0.25 percentage points in FY 2010 and FY 2011 and 0.2 percentage points for FY 2012 through FY 2019. Beginning in FY 2012, a productivity adjustment will be set at the 10-year moving average of non-farm business productivity. The application of the productivity adjustment may result in a negative overall market basket adjustment (i.e., lower Medicare payment rates).

LONG-TERM CARE HOSPITALS -- SERVICES, EXPENDITURES, AND PAYMENT METHODS

LTCHs provide treatment for patients who stay more than 25 days on average and suffer from clinically complex problems, such as multiple acute or chronic conditions (CMS, 2010; MedPAC, 2009). To qualify for Medicare payment, a LTCH must meet Medicare’s conditions of participation for acute care hospitals; have an average length-of-stay greater than 25 days for its Medicare patients; and have an adequate screening process. The largest groups of beneficiaries to receive treatment in a LTCH include those with respiratory ailments (28%), skin ulcers (5.2%), and degenerative nervous system disorders (2.6%) (MedPAC, 2009). No prior acute care hospitalization is required for Medicare admission to an LTCH, although roughly 80% of Medicare LTCH patients are admitted from an acute care hospital.

In 2008, there were 386 LTCHs nationwide with 26,578 Medicare-certified beds. The majority of LTCHs were freestanding facilities (62%), and for-profit owned (73%) and located in urban areas (92%). LTCHS are not distributed evenly across the country, with a strong concentration in northeastern (e.g., Massachusetts, Rhode Island) and southeastern (e.g., Louisiana) states.

Medicare is the predominant payer of LTCH services, accounting for roughly two-thirds of all discharges. In 2008, Medicare spent $4.6 billion on 130,869 LTCH cases. The Medicare payment per case was $35,200, with an average length-of-stay of 26.7 days. LTCHs were historically paid by Medicare on the basis of their average costs per discharge, as long as they did not exceed a facility-specific limit that was adjusted annually. In October 2002, Medicare adopted the LTCH PPS that pays a predetermined per-discharge rate based primarily on the patient’s diagnosis and market area wages.

The Medicare unit of payment under the LTCH PPS is the discharge episode. Medicare sets per-discharge payments for LTCHs according to different CMGs based on the Medicare severity long-term care diagnosis related group (MS-LTC-DRG). Patients are assigned to a MS-LTC-DRG based on their principal diagnosis, up to eight secondary diagnoses, up to six procedures performed, age, sex, and discharge status. The MS-LTC-DRGs correspond to the DRGs used in the acute inpatient PPS, but the relative weights are specific to LTCH patients. The PPS payment rate covers all operating and capital costs. The initial payment level (base rate) for a typical discharge is $39,897 in 2010. The labor portion of the base rate (76%) is adjusted for differences in market area wages using a version of the hospital wage index and then the result is added to the non-labor portion (including capital).

The LTCH PPS system accounts for both short-stay and high-cost outliers. Short-stay outliers are episodes with a length-of-stay up to and including five-sixths of the geometric average length-of-stay for the specific MS-LTC-DRG. For these cases, LTCHs are paid the least of: (1) 100% of the cost of the case; (2) 120% of the MS-LTC-DRG specific per diem amount multiplied by the length-of-stay for that episode; (3) the full MS-LTC-DRG payment; or (4) an amount that is a blend of the inpatient PPS amount for the MS-DRG and the 120% of the LTCH per diem payment amount. High-cost outliers are identified by comparing episode costs to a threshold that is the MS-LTC-DRG payment for the episode plus a fixed-loss amount. In 2010, the fixed-loss amount is $18,425, and Medicare pays 80% of the LTCH’s costs about the threshold.

LTCH patients that are discharged to an inpatient acute care hospital (for 9 days), an IRF (for 27 days), or a SNF (for 45 days) and then return to the same LTCH--termed “interrupted stay” patients--generate a single Medicare LTCH payment. Any LTCH patient readmitted within 3 days is also considered an interrupted stay.

LTCHs are regulated by the “25%” rule, which reduces payments for LTCHs that exceed percentage thresholds for patients admitted from certain referring hospitals during a cost reporting system (MedPAC, 2008). After the threshold is reached, the LTCH is paid the lesser of the LTCH PPS rate or an amount equivalent to the inpatient hospital PPS rate for patients discharged from the host acute care hospital. Basically, this rule is in place to help ensure that LTCHs do not function as units of acute care hospitals. When the rule was first implemented, it applied only to LTCH hospitals within hospitals (HWHs) and satellites. The Medicare, Medicaid and SCHIP Extension Act of 2007 rolled back the implementation of the 25% rule for HWHs and satellites and prevented application of the rule to freestanding LTCHs for three years.

Prior to the recently passed health care reform legislation, the statute did not specify a mechanism for updating payments to LTCHs. In recent years, CMS has used the Rehabilitation, Psychiatric, and Long-Term Care (RPL) market basket index. However, CMS has adjusted the RPL market basket increase downward to account for improved coding practices that have resulted in higher CMIs without corresponding increases in patient severity of illness. For example, the market basket increase of 2.5% in 2010 was adjusted downward to 2% to account for improved coding practices. Under the recently passed health care reform legislation, the LTCH market basket update will be reduced by 0.25 percentage points in FY 2010 and FY 2011 and 0.2 percentage points for FY 2012 through FY 2019. Beginning in FY 2012, a productivity adjustment will be set at the 10-year moving average of non-farm business productivity. The application of the productivity adjustment may result in a negative overall market basket adjustment (i.e., lower Medicare payment rates).

REFERENCES

Centers for Medicare and Medicaid Services (CMS). 2009. Home Health Providers by Type of Ownership and by State: Centers for Medicare and Medicaid Services, Nursing Home Compare.

CMS. 2009. Limitation on Home Health Prospective Payment System (HH PPS) Outlier Payments. Washington, DC: CMS.

CMS. 2009. Medicare Payment Updates for Inpatient Rehabilitation Facilities in Fiscal YR 2010. Washington, DC: CMS.

CMS. Glossary. CMS 2010 [cited February 2nd]. Available from http://www.cms.hhs.gov/apps/glossary.

CMS. Skilled Nursing Facilities PPS. CMS 2010 [cited February 2nd]. Available from http://www.cms.hhs.gov/SNFPPS.

DuNah, R., S. Keo de Wit, C. Harrington, J.H. Swan, and B. Bedney. 1994. State Data Book on Long-Term Care Program and Market Characteristics. Baltimore, MD: Health Care Financing Administration, U.S. Department of Health and Human Services. HCFA Pub No. 03354.

Fries, B.E. 1990. Comparing case-mix systems for nursing home payment. Health Care Financ Rev 11(4):103-119.

Fries, B.E., D.P. Schneider, W.J. Foley, M. Gavazzi, R. Burke, and E. Cornelius. 1994. Refining a case-mix measure for nursing homes: Resource utilization groups (RUG-III). Med Care 32(7):668-685.

Grabowski, D.C., Z. Feng, O. Intrator, and V. Mor. 2004. Recent trends in state nursing home payment policies. Health Affairs W4:363-373.

Grabowski, D.C., Z. Feng, O. Intrator, and V. Mor. 2008. Medicaid nursing home payment and the role of provider taxes. Medical Care Research and Review 65(4):514-527.

Harrington, C., J.H. Swan, V. Wellin, W. Clemena, B. Bedney, and H. Carillo. 1999. 1998 State Data Book on Long Term Care Program and Market Characteristics. San Francisco, CA: Department of Social and Behavioral Sciences, University of California.

Kaiser Commission on Medicaid and the Uninsured. 2009. Medicaid Home and Community-Based Service Programs: Data Update. Washington, DC: Kaiser Family Foundation.

Kaiser Family Foundation. 2005. Medicaid Benefits: Online Database 2004 [cited September 21st]. Available from http://www.dff.org/medicaid/benefits/service.jsp?nt=on&so=0&tg=0&yr=2&ca....

Medicare Payment Advisory Commission (MedPAC). 2008. Long-Term Care Hospitals Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Healthcare Spending and the Medicare Program. Washington, DC: MedPAC.

MedPAC. 2009. Home Health Care Services Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Payment Basics: Home Health Care Services Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Payment Basics: Long Term Care Hospitals Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Payment Basics: Rehabilitation Facilities (Inpatient) Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Payment Basics: Skilled Nursing Facility Services Payment System. Washington, DC: MedPAC.

MedPAC. 2009. Report to the Congress: Medicare Payment Policy. Washington, DC: MedPAC.

MedPAC. 2010. Comment Letter: CMS’s proposed rule entitled: Medicare program; Home health prospective payment system rate update for calendar year 2010. Washington, DC: MedPAC.

MedPAC. 2010. Report to the Congress: Medicare Payment Policy. Washington, DC: MedPAC.

National Association of Home Care and Hospice (NAHC). What Types of Services Do Homecare Providers Deliver? NAHC 2010 [cited]. Available from http://www.nahc.org/consumer/FAQs/services.html.

Rhoades, J., and J. Sommers. 2000. Expenses and Sources of Payment for Nursing Home Residents, 1996. Rockville, MD: Agency for Healthcare Research and Quality. MEPS Research Findings No. 13, AHRQ Pub No. 01-0010.

U.S. General Accounting Office (GAO). 2003. Medicaid Nursing Home Payments: States’ Payment Rates Largely Unaffected by Recent Fiscal Pressures. Washington, DC: GAO. GAO Pub No. GAO-04-143.

To obtain a printed copy of this report, send the full report title and your mailing information to:

U.S. Department of Health and Human ServicesOffice of Disability, Aging and Long-Term Care PolicyRoom 424E, H.H. Humphrey Building200 Independence Avenue, S.W.Washington, D.C. 20201FAX:  202-401-7733Email:  webmaster.DALTCP@hhs.gov

RETURN TO:

Office of Disability, Aging and Long-Term Care Policy (DALTCP) Home [http://aspe.hhs.gov/_/office_specific/daltcp.cfm]Assistant Secretary for Planning and Evaluation (ASPE) Home [http://aspe.hhs.gov]U.S. Department of Health and Human Services Home [http://www.hhs.gov]