Medicare: Modest Eligibility Expansion for Critical Access	 
Hospital Program Should Be Considered (19-SEP-03, GAO-03-948).	 
                                                                 
Critical Access Hospitals (CAHs) are small rural hospitals that  
receive payment for their reasonable costs of providing inpatient
and outpatient services to Medicare beneficiaries, rather than	 
being paid fixed amounts under Medicare's prospective payment	 
systems. Between fiscal years 1997 and 2002, 681 hospitals have  
become CAHs. In the Medicare, Medicaid and SCHIP Benefits	 
Improvement and Protection Act of 2000, GAO was directed to	 
examine requirements for CAH eligibility, including the ban on	 
inpatient psychiatric or rehabilitation distinct part units	 
(DPUs) and limit on patient census, and to make recommendations  
on related program changes.					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-03-948 					        
    ACCNO:   A08515						        
  TITLE:     Medicare: Modest Eligibility Expansion for Critical      
Access Hospital Program Should Be Considered			 
     DATE:   09/19/2003 
  SUBJECT:   Eligibility criteria				 
	     Health care costs					 
	     Health care facilities				 
	     Health care programs				 
	     Health care services				 
	     Hospital administration				 
	     Hospital care services				 
	     Hospitals						 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-03-948

Report to Congressional Committees

United States General Accounting Office

GAO

September 2003 MEDICARE Modest Eligibility Expansion for Critical Access
Hospital Program Should Be Considered

GAO- 03- 948

Using fiscal year 1999 hospital cost report data, GAO identified 683 rural
hospitals as *potential CAHs* based on their having an annual average of
no more than 15 acute care patients per day. About 14 percent (93) of
these potential CAHs operated an inpatient psychiatric or rehabilitation
DPU, which they would have to close to convert to CAH status. Among
existing CAHs, 25 previously operated a DPU but had to close it as part of
becoming a CAH. Among the potential CAHs that operated a DPU, about half
had a net loss on Medicare services, indicating they might benefit from
CAH conversion. Officials in some hospitals expressed a reluctance to
close their DPU, even if conversion would benefit the hospital
financially, as they

believe the DPU maintains the availability of services in their community.
Because inpatient rehabilitation and psychiatric services are
disproportionately located in urban areas, even a small number of rural
DPU closures may exacerbate any disparities in the availability of these
services. Using 1999 Medicare claims data, GAO found 129 potential CAHs
that likely would have been able to meet the CAH census limit of no more
than 15 acute care patients at any given time if not for a seasonal
increase in their patient

census. Seasonal increases in patient census were common among the
hospitals GAO studied, generally occurring during the winter flu and
pneumonia season. For most potential CAHs, their patient census was
typically low enough that a small seasonal increase did not cause them to
exceed CAH limits. For the 129 potential CAHs that would have had
difficulty staying under the CAH limit due to seasonal variation, they
could have accommodated their patient volume and had greater flexibility
in the management of their patient census if the CAH census limit were
changed from an absolute limit of 15 patients per day to an annual average
of 15 patients.

Potential CAHs That May Otherwise Be Eligible to Conversion If Not for
Seasonal Variation in Patient Stays or Because They Operate a DPU

Critical Access Hospitals (CAHs) are small rural hospitals that receive
payment for their

reasonable costs of providing inpatient and outpatient services to
Medicare beneficiaries, rather than being paid fixed amounts under

Medicare*s prospective payment systems. Between fiscal years 1997 and
2002, 681 hospitals have become CAHs.

In the Medicare, Medicaid and SCHIP Benefits Improvement and Protection
Act of 2000, GAO was directed to examine requirements for CAH eligibility,
including the ban on inpatient psychiatric or rehabilitation distinct part
units

(DPUs) and limit on patient census, and to make recommendations on related
program changes. GAO suggests that the Congress

may wish to consider allowing hospitals with a DPU to convert to CAH
status. GAO also suggests

that the Congress may wish to consider changing the CAH limit on acute
care patient census from an absolute limit of 15 patients to an annual
average of 15 patients. The Department of Health and Human Services said
that these modifications to CAH eligibility criteria would provide the
needed

flexibility for some additional facilities to consider conversion to CAH
status, and emphasized the importance of maintaining financial

incentives for efficiency as well as health and safety standards.

www. gao. gov/ cgi- bin/ getrpt? GAO- 03- 948. To view the full product,
including the scope and methodology, click on the link above. For more
information, contact A. Bruce Steinwald at (202) 512- 7119. Highlights of
GAO- 03- 948, a report to the

Senate Committee on Finance, the House Committee on Ways and Means, and
the House Committee on Energy and Commerce

September 2003

MEDICARE

Modest Eligibility Expansion for Critical Access Hospital Program Should
Be

Considered

Page i GAO- 03- 948 Critical Access Hospital Program Letter 1 Results in
Brief 4 Background 6 Existing CAHs Had Fewer Beds and Patients and Lower
Medicare

Margins Than Potential CAHs 12 Ban on CAHs Operating DPUs May Have
Contributed to Diminished Availability of Services in Rural Areas 14
Seasonal Variation in Patient Census Is Common and May Impede CAH
Eligibility for Hospitals Near the CAH Limit 19 Conclusions 22 Matters for
Congressional Consideration 24 Agency Comments and Our Evaluation 24
Appendix I Scope and Methodology 28

Appendix II Comments from the Department of Health and Human Services 31

Appendix III GAO Contact and Staff Acknowledgments 35 GAO Contact 35
Acknowledgments 35 Tables

Table 1: Selected Characteristics of Existing CAHs Prior to Their
Conversion and Potential CAHs, Fiscal Year 1999 13 Table 2: Financial
Performance of Existing CAHs Prior to Their

Conversion and Potential CAHs, Fiscal Year 1999 14 Table 3: Financial
Performance of Potential CAHs with DPUs, Fiscal Year 1999 16 Table 4:
Medicare Margins for DPUs of Potential CAHs, Fiscal Year 1999 17 Table 5:
Seasonal Increase in Average Acute Care Patient Census among Potential
CAHs, by Bedsize, 1999 20 Table 6: Potential CAHs with Estimated Seasonal
Increases in Patient Census That Pushed Them over CAH Limit, 1999 21
Contents

Page ii GAO- 03- 948 Critical Access Hospital Program

Table 7: Financial Performance of Potential CAHs with a Seasonal Increase
in Patient Census That Pushed Them over CAH Limit, Fiscal Year 1999 21
Table 8: Potential CAHs with Seasonal Increase in Medicare

Patients* Length of Stay That Pushed Them over the 4- day CAH Limit, 1999
22 Table 9: Summary of Site Visits and Interviews 30 Figures

Figure 1: Major Eligibility Criteria for Critical Access Hospitals 7
Figure 2: Number of Critical Access Hospitals through Fiscal Year 2002 8
Figure 3: Location of the 681 Critical Access Hospitals, September 2002 9

Page iii GAO- 03- 948 Critical Access Hospital Program Abbreviations

BBA Balanced Budget Act of 1997 BIPA Medicare, Medicaid, and SCHIP
Benefits Improvement and

Protection Act of 2000 CAH Critical Access Hospital CMS Centers for
Medicare & Medicaid Services

DPU distinct part unit EACH essential access community hospital EMS
emergency medical services FORHP Federal Office of Rural Health Policy HHS
Department of Health and Human Services HRSA Health Resources and Services
Administration MSA metropolitan statistical area OMB Office of Management
and Budget PPS prospective payment system RHFTP rural hospital flexibility
tracking project

RPCH rural primary care hospital SCHIP State Children*s Health Insurance
Program TEFRA Tax Equity and Fiscal Responsibility Act of 1982

This is a work of the U. S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

Page 1 GAO- 03- 948 Critical Access Hospital Program September 19, 2003
Congressional Committees

Medicare beneficiary access to hospital services in rural areas has been a
source of concern for policymakers for many years. To bolster the
financial stability of rural hospitals, the Congress approved several
special payment provisions both before and after the implementation of the

Medicare acute care inpatient prospective payment system (PPS) 1 in 1983.
These provisions enhanced Medicare payments to certain groups of rural
hospitals, such as those that are the only source of care in their
community; larger hospitals that serve as referral sites for rural
physicians and community hospitals; and hospitals highly dependent on
Medicare payments. Many rural hospitals have, however, continued to
experience financial difficulties.

In the Balanced Budget Act of 1997 (BBA), the Congress established
additional special payment provisions for Critical Access Hospitals (CAH).
2 When designated as a CAH, a hospital generally receives payment for its
reasonable costs of providing inpatient and outpatient services to
Medicare beneficiaries, rather than being paid the PPS fixed amount for
those services. Thus, the CAH designation provides higher payments to
hospitals whose reasonable costs are higher than their PPS payment. The
CAH program has grown steadily to 681 CAHs at the end of fiscal year 2002.
3 The CAH designation is targeted to small rural hospitals with a low
patient

census and short patient stays. Statutory provisions specifying criteria
for CAHs do not specifically exclude facilities with distinct part units
(DPUs) *separate sections certified to provide inpatient rehabilitation or
psychiatric care. However, statutory and regulatory provisions concerning
1 Under the PPS, hospitals are paid a fixed amount for each hospital
discharge, based on national average costs, adjusted for such factors as
local wage costs and type of illness

treated. 2 Pub. L. No. 105- 33, S: 4201( c), 111 Stat. 251, 373- 374
(1997).

3 CAH enrollment figures were provided by the Rural Hospital Flexibility
Tracking Project (RHFTP), a federally funded national evaluation by a
consortium of five rural health research centers and the Rural Policy
Research Institute.

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 03- 948 Critical Access Hospital Program payment for such DPUs
effectively require them to be operated by hospitals paid PPS rates. Thus,
because CAHs are paid their reasonable

costs, they are effectively banned from having DPUs. Some hospital
officials have raised concerns that because CAHs cannot operate DPUs, it
may be more difficult to ensure that rural beneficiaries have access to
the kind of psychiatric and rehabilitation services these units provide,
if hospitals choose to close their DPU as part of becoming a CAH. In
addition, to be a CAH, a hospital must remain under CAH limits on the
number of hospital beds (* bedsize*) and average patient length of stay,
and can have no more than 15 acute care patients on any given day. Some
hospitals may have difficulty remaining under CAH limits during the entire
year because they may experience fluctuations in patient demand due to
seasonal tourism or illnesses, like influenza or pneumonia, that are more

prevalent at certain times of the year. In the Medicare, Medicaid, and
SCHIP Benefits Improvement and Protection Act of 2000 (BIPA), 4 the
Congress directed us to study CAH eligibility requirements including with
respect to limitations on average length of stay, bedsize, and DPU
operations, and to make related recommendations on program changes. As
agreed with the committees of jurisdiction, we have examined (1) the
characteristics of a group of hospitals prior to their designation as CAHs
compared to a group of small rural hospitals that have not become CAHs,
but were in a position to consider doing so based on their low patient
census, (2) the impact that the effective ban on CAHs operating DPUs has
had on the availability of psychiatric and rehabilitation services in
rural areas and on rural hospitals* decisions to seek CAH conversion,
possible options for Medicare payment to DPUs and CAH eligibility
requirements if CAHs were allowed to operate DPUs, and (3) the extent to
which seasonal variation in patient census or length of stay prevents
hospitals from being eligible for CAH status.

To address these objectives, we analyzed Medicare hospital cost reports 5
from fiscal year 1999, the most recently available audited cost report
data, and Medicare inpatient claims data for 1999. We defined 683 rural
hospitals that had not converted to CAH status as of January 1, 2003, as 4
Pub. L. No. 106- 554, App. F, S: 206, 114 Stat. 2763A- 463, 2763A- 483
(2000). 5 The Medicare cost report is the financial document that
hospitals are required to submit annually to the Centers for Medicare &
Medicaid Services (CMS). The reports include

information about Medicare inpatient and outpatient costs and payments, as
well as information about payments from other revenue sources.

Page 3 GAO- 03- 948 Critical Access Hospital Program *potential CAHs,*
based on their having an annual average patient census of no more than 15
acute care patients. 6 We estimated how many of the 683

potential CAHs might be prevented from converting to CAH status because
they operate a DPU or experience seasonal variation in their patient
census or average length of stay. We also examined the characteristics of
620 hospitals that were not yet CAHs in fiscal year 1999 but have since
converted to CAH status (* existing CAHs*) and compared their
preconversion characteristics to those of potential CAHs in fiscal year
1999. We evaluated how many potential CAHs and existing CAHs experienced
financial losses under the Medicare PPS and likely could benefit from
cost- based reimbursement. Since DPUs are paid under different payment
methodologies from acute care hospitals, we evaluated how many of the DPUs
operated by potential CAHs experienced financial gains or losses under the
payment methodology that applied to them in fiscal year 1999 as well as
the possible impact if cost- based reimbursement were extended to DPUs
operated by CAHs. We also evaluated how many of the potential CAHs with
DPUs could have met CAH bedsize and length of stay criteria in fiscal year
1999 if their DPU beds and lengths of stay

were counted towards the limits. We interviewed officials with the Centers
for Medicare & Medicaid Services (CMS) and the Federal Office of Rural
Health Policy (FORHP), which administers a grant program supporting

CAHs. We interviewed administrators of 24 CAHs and potential CAHs across
10 states, and made site visits to 7 of these hospitals in 3 states. We
also interviewed state staff administering FORHP grants, and conducted an
e- mail survey of state CAH coordinators. 7 We did our work in

accordance with generally accepted government auditing standards from
April 2001 through August 2003. A detailed discussion of our scope and
methodology is in appendix I.

6 Most of the 683 potential CAHs (79 percent) exceeded the CAH bedsize
limit. We did not exclude these hospitals from our definition of potential
CAHs because hospitals have the option of reducing their bedsize in order
to become eligible for CAH conversion. Our inclusion of hospitals with an
average census up to 15 is likely a high estimate of the number of
potential CAHs because hospitals with an annual average of 15 acute care
patients per day may need more than 15 acute care beds to accommodate
variation in their patient census that periodically causes them to exceed
15. 7 New Jersey, Rhode Island, Delaware, and Washington D. C. do not
participate in the CAH program. All but 5 state CAH coordinators
participated in the e- mail survey or were interviewed.

Page 4 GAO- 03- 948 Critical Access Hospital Program Existing CAHs
averaged six fewer beds and about three fewer patients per day prior to
their conversion than did potential CAHs. Existing CAHs had

to make smaller operational changes to qualify for CAH status, such as
reducing bedsize or length of stay, than potential CAHs would have had to
make if they had chosen to convert. While both groups had a median loss on
Medicare inpatient and outpatient services, existing CAHs tended to
experience bigger losses prior to their conversion (8.9 percent) than did
potential CAHs (0.8 percent). Existing CAHs also had a median loss on all
sources of revenue of 0.3 percent before conversion, while potential CAHs
had a median gain of 1.8 percent.

The effective ban on CAHs operating DPUs may have contributed to the
disparity between urban and rural areas in the availability of inpatient
psychiatric and rehabilitation services in fiscal year 1999. While
onequarter of Medicare beneficiaries reside in rural areas, only 8 percent
of rehabilitation hospital and DPU beds and 17 percent of psychiatric
hospital and DPU beds were in rural areas in fiscal year 1999. The
subsequent closure of 25 DPUs by hospitals converting to CAH status may
have exacerbated this difference in availability. Of the 93 potential CAHs
that operated a DPU, about half lost money on Medicare inpatient and
outpatient services, giving them a financial incentive to convert. If,
however, the other financial benefits associated with the DPU exceeded
their losses under the PPS, these potential CAHs would have a
countervailing incentive to stay under the PPS rather than close their DPU
and convert. Some rural hospital administrators told us that, even when it
was financially advantageous to seek CAH status, they were reluctant to

close their DPU because it is needed to maintain access to psychiatric or
rehabilitation services in the community they serve. While allowing
hospitals to convert to CAH status and retain their DPU would alleviate
this concern, extending cost- based reimbursement to DPUs operated by CAHs
diminishes the incentives for efficiency that are inherent in PPS
payments. If DPU patient stays and beds were counted against current CAH
limits without any adjustment, nearly all potential CAHs with DPUs would
have exceeded the limits in fiscal year 1999.

Among hospitals we studied, seasonal fluctuations in patient volume or
length of stay were common, particularly during the winter. Such increases
can be an obstacle for some hospitals considering CAH conversion if it
causes them to exceed the CAH patient census limit of no more than 15
patients at any time or length of stay limit of an annual average of 4
days. We found 129 potential CAHs that likely would have been able to meet
the CAH patient census limit in fiscal year 1999 if not for the seasonal
increase in their patient census. While these 129 hospitals, as Results in
Brief

Page 5 GAO- 03- 948 Critical Access Hospital Program a group, averaged
13.2 patients per day over the entire year, their daily census increased
to an estimated average of 16. 9 during their high season. If the CAH
patient census limit were changed from an absolute limit of 15

acute care patients per day to an annual average of 15, these potential
CAHs would have been able to remain under such a limit because they all
had an annual average below 15. It would not be necessary to increase the
number of acute care beds CAHs are allowed to maintain in order to

implement this relaxation of the patient census limit, since more than
three- quarters of existing CAHs and potential CAHs have swing beds 8
which they could use to accommodate additional acute care patients

beyond 15. About 40 percent of these 129 potential CAHs, however, had
positive Medicare margins, meaning they would have had little financial
incentive to switch from the PPS to the cost- based payment CAHs receive.
In contrast to the CAH patient census limit, the patient length- of- stay
limit

gives CAHs the flexibility to keep some acute care patients beyond the
limit because it is an average.

We suggest that the Congress may wish to consider allowing hospitals with
DPUs to convert to CAH status while making allowances for DPU beds,
patients, and lengths- of- stay when determining CAH eligibility, and that
CAH- affiliated DPUs be paid under the same formulas as other inpatient
psychiatric or rehabilitation providers. We also suggest that the Congress
may wish to consider changing the CAH limit on acute care patient census
from an absolute limit of 15 acute care patients to an annual average of
15 in order to give CAHs greater flexibility in the management of their
patient census. In commenting on a draft of this report, the Department of
Health and

Human Services said that these modifications to CAH eligibility criteria
would provide the needed flexibility for some additional facilities to
consider conversion to CAH status. The department also emphasized several
considerations, including maintaining financial incentives for efficiency
as well as health and safety standards for DPUs, if they are allowed to be
operated by a CAH.

8 A hospital with swing beds can *swing* its beds between hospital and
skilled nursing levels of care, on an as needed basis.

Page 6 GAO- 03- 948 Critical Access Hospital Program CAHs are an outgrowth
of the seven- state Essential Access Community Hospital/ Rural Primary
Care Hospital (EACH/ RPCH) program established

in 1989. The BBA replaced the EACH/ RPCH program with the
stateadministered Rural Hospital Flexibility Program (the *Flex* Program),
which includes the CAH designation. The reimbursement component of the
Flex Program is the responsibility of CMS. The Flex Program also includes
a grant program that supports hospital participation in the program as
well as state emergency medical services systems (EMS), and is the
responsibility of the FORHP within the Health Resources and Services
Administration (HRSA).

The CAH program allows eligible rural hospitals to receive Medicare
payments based on their reasonable costs rather than under a PPS. Under
the Medicare inpatient PPS, hospitals are generally paid a fixed amount
per patient discharge, providing an incentive for hospitals to control
their costs to stay under this fixed amount because they can retain the
difference between the PPS payment and their costs. Under cost- based
reimbursement, hospitals are reimbursed for their reasonable costs, which
does not provide the same incentive to control costs, but benefits
hospitals whose Medicare costs exceed their PPS payments.

In addition to receiving cost- based payment for inpatient services to
Medicare beneficiaries, CAHs receive cost- based payment from Medicare for
skilled nursing care provided in their swing beds and for outpatient care.
9 To become a CAH, a hospital must meet certain criteria with respect to
its location, size, patient census, and patient length of stay (see figure
1). CAHs are also subject to different health and safety regulations,
known as *conditions of participation,* from other acute care hospitals.
10 9 Among 42 states responding to a RHFTP survey, 17 states provide
enhanced Medicaid

payments to CAHs, and 13 states provide enhanced reimbursement for
outpatient services. 10 42 C. F. R. S:S: 485.601 et seq. (2002).
Background

Page 7 GAO- 03- 948 Critical Access Hospital Program Figure 1: Major
Eligibility Criteria for Critical Access Hospitals

Note: The Office of Management and Budget (OMB) defines a metropolitan
statistical area as a core area of at least 50, 000 people together with
adjacent areas having a high degree of economic and social integration
with that core. Nonmetropolitan areas include all counties outside of a
metropolitan area. a The statutory provision outlining the certification
exception does not specify the criteria for a hospital

to be a necessary provider of services. Source: GAO. Day

Day Day Day

1 2 3 4 Length of Stay

A CAH must maintain an annual average length of stay of no more than 4
days.

Rurality

A CAH must be located in a nonmetropolitan area, as defined by the U. S.
Office of Management and Budget, or in a rural census tract of a
metropolitan statistical area (MSA).

Proximity

A CAH must be more than 35 miles from the next nearest acute care hospital
or be certified by the state as a necessary

provider of services a .

Bedsize - Patient Census

A CAH can maintain up to 15 beds, or up to 25 beds if swing beds are
included as long as no more than 15 beds are used for acute care patients
at any given time. A CAH's patient census is limited to no more than 15
acute care patients at any given time.

Hospital

Page 8 GAO- 03- 948 Critical Access Hospital Program Growth in the number
of CAHs has been steady (see figure 2). There is a large concentration of
CAHs in the central states, although 45 states had at

least one CAH as of September, 2002 (see figure 3). 11 Figure 2: Number of
Critical Access Hospitals through Fiscal Year 2002

11 Connecticut, Delaware, Maryland, New Jersey, and Rhode Island did not
have CAHs as of September 2002.

24 36 74

269 478

681

0 200

400 600

800 pre- 1998 1998 1999 2000 2001 2002 Number of CAHS

Source: Rural Hospital Flexibility Tracking Project.

Page 9 GAO- 03- 948 Critical Access Hospital Program Figure 3: Location of
the 681 Critical Access Hospitals, September 2002

Note: Some Critical Access Hospitals may not be visible because they are
obscured by state boundary lines. Source: Rural Hospital Flexibility
Tracking Project.

CAHs (681)

Page 10 GAO- 03- 948 Critical Access Hospital Program Since the inception
of the CAH program, two factors have been important in increasing the
number of hospitals qualifying for the designation. First,

the length- of stay criterion was changed. Until 1999, patient stays at
CAHs were limited to 4 days, after which patients would have to be
transferred to another health care facility or discharged. In 1999, the
Congress relaxed the criterion to require that CAHs keep their annual
average length of stay to no more than 4 days. 12 Second, states have
widely utilized their

authority to designate hospitals as *necessary providers,* thereby
exempting such hospitals from the otherwise applicable CAH criterion that
they be more than 35 miles from the nearest hospital. According to the
Rural Hospital Flexibility Tracking Project (RHFTP), a little more than
half of all CAHs had qualified for the CAH program through state
designation

rather than by meeting the mileage and location requirements, as of
September 2002. 13 Hospitals considering CAH conversion weigh numerous
factors in their

decision, including the impact on hospital finances and community
reaction. Financial impact studies are commonly used to estimate how a
hospital*s reimbursement for services would change under CAH status. The
financial impact may change as Medicare reimbursements to hospitals
changes. For example, Medicare payment for hospital outpatient services
shifted in 2000 from cost- based payment to a new PPS for outpatient
services. Because CAHs are exempt from this PPS and continue to receive
cost- based payment for outpatient services, potential CAHs may factor
into their decision the impact of being paid reasonable costs, rather than
a fixed PPS payment, for outpatient services. They may also consider the
possible reaction from the community and from other health care providers
to CAH conversion. Some communities have been reluctant to support a
hospital*s conversion because they perceive it as the last step before
closure. In other cases, hospital officials reported that their physicians
expressed concern that if a hospital became a CAH, they would occasionally
be unable to admit patients to it because this would bring the CAH over
the patient limit.

12 Medicare, Medicaid and SCHIP Balanced Budget Refinement Act of 1999,
Pub. L. No. 106- 113, App. F, S: 403( a), 113 Stat. 1501A- 321, 1501A-
370- 372. 13 List of CAH conversions by state downloaded from www. rupri.
org/ rhfp- track on September 27, 2002.

Page 11 GAO- 03- 948 Critical Access Hospital Program Clinical research
has indicated better outcomes for patients who are appropriately treated
in inpatient psychiatric or rehabilitation facilities,

such as DPUs, rather than in general acute or post acute care settings.
For example, one study concluded that elderly depressed patients who were
treated in specialty psychiatric DPUs may have received better treatment
for their depression than similar patients who were treated in general
medical wards. 14 Another study found better outcomes among stroke
patients treated in rehabilitation facilities, such as DPUs, than those
treated in nursing homes. 15 As separate sections of hospitals,
psychiatric and rehabilitation DPUs are

subject to specific Medicare regulations regarding the types of patients
they admit and the qualifications of their staff. 16 Psychiatric DPUs may
admit only patients whose condition requires inpatient hospital care and
are described by a psychiatric principal diagnosis. 17 Rehabilitation DPUs
may treat only patients likely to benefit significantly from intensive
therapy services, such as physical therapy, occupational therapy, or
speech therapy. Both types of DPUs must provide a specified range of
services and employ clinical staff with specialized training.

14 G. Norquist et al. *Quality of Care for Depressed Elderly Patients
Hospitalized in the Specialty Psychiatric Units or General Medical Wards,*
Archives of General Psychiatry,

vol. 52, no. 8 (1995). 15 R. L. Kane et al. ** Functional Outcomes of
Posthospital Care for Stroke and Hip Fracture Patients under Medicare,*
Journal of the American Geriatric Society, vol. 46, no. 12 (1998).

16 For a hospital to establish a psychiatric DPU, Medicare regulations
require that a hospital must furnish, through the use of qualified
personnel, psychological services, social work, psychiatric nursing,
occupational therapy and recreational therapy. Inpatient psychiatric
services must be under the supervision of a clinical director, service
chief, or equivalent who is qualified to provide the leadership required
for an intensive treatment program, and who is board certified in
psychiatry. The DPU must have a director of nursing who is a registered
nurse with a master*s degree in psychiatric or mental health nursing or
who is qualified by education and experience, and a director of social
services. There also must be an adequate number of registered nurses to
provide 24- hour- a- day coverage as well as licensed practical nurses and
mental health workers. 42 C. F. R. S: 412.27 (2002). For a hospital to
establish a rehabilitation DPU, Medicare regulations require that a
hospital must

provide rehabilitation nursing, physical and occupational therapy, speech
therapy, plus as needed, social services or psychological services and
orthotics and prosthetics. The unit must have a director of rehabilitation
who is experienced in rehabilitation and is a doctor of medicine or a
doctor of osteopathy. 42 C. F. R. S: 412. 29 (2002).

17 42 C. F. R. S: 412.27( a) (2002). Psychiatric principal diagnoses are
listed in the Third Edition of the American Psychiatric Association
Diagnostic and Statistical Manual and in chapter 5 of the International
Classification of Diseases, 9th Edition Clinical Modification (ICD- 9-

CM). Distinct Part Units

Page 12 GAO- 03- 948 Critical Access Hospital Program The Congress has
required that CMS develop PPSs for both inpatient rehabilitation and
inpatient psychiatric providers, including DPUs, to

replace the payment methodology established by the Tax Equity and Fiscal
Responsibility Act of 1982 (TEFRA). Under TEFRA, providers that had been
exempted from the inpatient PPS, including inpatient rehabilitation and
psychiatric hospitals and DPUs, receive the lesser of either their average
cost per discharge or a provider- specific target amount. 18 In 2002, a
PPS was implemented for inpatient rehabilitation. Because a PPS for
inpatient psychiatric providers has yet to be implemented, psychiatric
DPUs continue to be paid under TEFRA.

The financial incentives associated with TEFRA payments differ from those
associated with cost- based payment. Under TEFRA, Medicare payments are
capped by a provider*s target amount, giving hospitals an incentive to
restrain costs. By contrast, hospitals such as CAHs, which are paid their
reasonable costs, have less incentive to restrain costs because their
payments can increase as their costs increase.

Most existing CAHs prior to their conversion had more beds in fiscal year
1999 than CAHs are allowed. Most were likely able to reduce their bedsize
to 15 (or 25 with swing beds) to become CAHs without adjusting their
patient volume because their average patient census of 4.8 was generally
well below the CAH limit of 15 (see table 1). Likewise, potential CAHs, on

average, exceeded CAH bedsize limits in fiscal year 1999 and had a patient
census well below 15. To meet the CAH limit, existing CAHs, on average,
had to reduce their bedsize by less than potential CAHs would have had to
if they had sought CAH status. Most existing CAHs prior to their
conversion and potential CAHs were below the CAH length- of- stay limit.

18 TEFRA (Pub. L. No. 97- 248, S: 101( a)( 1), 96 Stat. 324, 331- 333)
established this payment methodology for classes of hospitals deemed
exempt from the PPS. The target amount is the PPS- exempt provider*s
Medicare- allowable costs per patient stay in a designated base year,
inflated to the current year by an annual update factor. Existing CAHs Had

Fewer Beds and Patients and Lower Medicare Margins Than Potential CAHs

Page 13 GAO- 03- 948 Critical Access Hospital Program Table 1: Selected
Characteristics of Existing CAHs Prior to Their Conversion and Potential
CAHs, Fiscal Year 1999 Total Average

daily census Average

length of stay (days) Average

bedsize Percentage

with swing beds Percentage

exceeding bedsize limit

Percentage exceeding length- of- stay

limit

Existing CAHs a (pre- conversion) 620 4.8 3. 5 30 85 61 14 Potential CAHs
683 8.1 3. 8 36 78 79 21 Source: Fiscal year 1999 Medicare hospital cost
reports.

a Statistics on existing CAHs include CAH conversions reported through
January 1, 2003, but do not include CAHs that had already converted to CAH
status in fiscal year 1999 or for which cost report data were not
available for fiscal year 1999.

In fiscal year 1999, existing CAHs prior to their conversion generally
experienced greater losses on their inpatient and outpatient Medicare
services than did potential CAHs (see table 2), and therefore had greater
financial incentive to seek conversion. A small majority, 55 percent, of
existing CAHs experienced losses on inpatient Medicare services, while
more than 60 percent of potential CAHs experienced gains. Nearly all

hospitals in both groups experienced losses on their Medicare outpatient
services. Across all revenue sources, existing CAHs prior to their
conversion experienced a 0.3 percent median loss, while potential CAHs
experienced a 1.8 percent median gain.

Page 14 GAO- 03- 948 Critical Access Hospital Program Table 2: Financial
Performance of Existing CAHs Prior to Their Conversion and Potential CAHs,
Fiscal Year 1999 Median margin a (percent) Hospitals with negative margins
Hospitals with positive margins

Existing CAHs (preconversion)

(n= 542) Potential

CAHs (n= 683)

Number (percent) of existing CAHs (preconversion)

Number (percent) of

potential CAHs

Number (percent) of existing CAHs

(preconversion) Number

(percent) of potential

CAHs

Medicare inpatient -2.4 6.0 296 (55) 254 (37) 236 (44) 419 (62) Medicare
outpatient -21.0 -19.6 523 (96) 649 (96) 11 (2) 14 (2) Medicare inpatient
and outpatient -8.9 -0.8 398 (74) 343 (51) 136 (25) 322 (48)

Total facility (all payers) -0.3 1.8 277 (51) 260 (38) 255 (47) 406 (60)

Source: Fiscal year 1999 Medicare hospital cost reports. Notes: For each
of the four calculations of hospital margins, a small number of hospitals
were excluded because of incomplete data or because their margins were
extreme outliers. Three to 17 potential CAHs were excluded among the four
calculations, and 2 to 10 existing CAHs were

excluded. In addition, 78 existing CAHs do not have pre- conversion PPS
margins statistics for fiscal year 1999 because they did not meet criteria
used for the margins calculation. Results do not reflect the effects of
the outpatient PPS, which was implemented in 2000. a A margin is the
difference between a hospital*s revenue and costs, divided by its
revenues.

The effective ban on CAHs operating DPUs may have contributed to the
disparity between urban and rural areas in the availability of inpatient
psychiatric and rehabilitation services in fiscal year 1999. Twenty- five
existing CAHs had to close their DPU as part of becoming CAHs. Of the 93
potential CAHs that operated a DPU (one- seventh of all potential CAHs),
about half lost money on their Medicare inpatient and outpatient services,
giving them a financial incentive to convert. If, however, the other

financial benefits associated with the DPU exceeded their combined losses
on inpatient and outpatient services, these potential CAHs would have had
a countervailing incentive to stay under the PPS, rather than close their
DPU and convert. Some rural hospital administrators told us that, even
when it was financially advantageous to seek CAH status, they were
reluctant to close their DPU because it was needed to maintain access to
psychiatric or rehabilitation services in the community they serve. While
allowing hospitals to convert to CAH status and retain their DPU would
alleviate this concern, extending cost- based reimbursement to DPUs
operated by CAHs diminishes the incentives for efficiency that are
inherent in PPS payments. If DPU patient stays and beds were counted
against current CAH limits without any adjustment, nearly all potential
Ban on CAHs

Operating DPUs May Have Contributed to Diminished Availability of Services
in Rural Areas

Page 15 GAO- 03- 948 Critical Access Hospital Program CAHs with DPUs would
have exceeded either the bedsize or length of stay limit in fiscal year
1999. The closure of 25 DPUs by hospitals that needed to relinquish their
DPU as

part of becoming a CAH may have contributed to the lower availability of
inpatient psychiatric and rehabilitation services in rural areas.
Inpatient psychiatric and rehabilitation providers are concentrated in
urban areas, and DPUs are least common among smaller rural hospitals. Only
8 percent of rehabilitation beds and 17 percent of psychiatric beds were
located in rural areas in fiscal year 1999, while about 25 percent of
Medicare beneficiaries live in rural areas. In fiscal year 1999, 14
percent (93) of potential CAHs operated a DPU. 19 By comparison, 37
percent of larger rural hospitals operated a DPU, and 53 percent of urban
hospitals operated a DPU.

DPUs may be less common in rural areas due to the challenge of finding the
resources needed to open a DPU. Hospital representatives and officials
from rural health organizations said the difficulty in finding the
specialized staff required to operate a DPU likely prevents many small
rural hospitals from opening a DPU.

In fiscal year 1999, nearly half the potential CAHs with a DPU experienced
net gains on their combined inpatient and outpatient payments for Medicare
services (see table 3). These potential CAHs had a financial incentive to
continue under the PPS because this allowed them to continue receiving
Medicare payments that were higher than their costs, rather than being
paid only their reasonable costs as a CAH. The 47 potential CAHs with DPUs
that experienced losses on their combined inpatient and outpatient
Medicare payments would more likely have a financial incentive to seek CAH
status.

19 Eighty- one of the 93 operated only a psychiatric DPU, 7 operated only
a rehabilitation DPU, and 5 operated both types of DPUs. CAH Eligibility
Requirements Led to DPU

Closures in Rural Communities

Many Potential CAHs Had No Financial Incentive to Close DPU

Page 16 GAO- 03- 948 Critical Access Hospital Program Table 3: Financial
Performance of Potential CAHs with DPUs, Fiscal Year 1999 Median margin a
in percentages

(n = 93) Number (percent)

of potential CAHs with negative

margins Number

(percent) of potential CAHs

with positive margins

Medicare inpatient 3.9 35 (38) 56 (62) Medicare outpatient -17.5 88 (97) 0
(0) Medicare inpatient and outpatient -1.1 47 (53) 41 (47)

Total facility (all payers) 0.6 42 (46) 46 (51)

Source: Fiscal year 1999 Medicare hospital cost reports. Notes: For each
of the four calculations of hospital margins, three or fewer hospitals
were excluded because of incomplete data or because their margins were
extreme outliers. Results do not reflect the effects of the outpatient
PPS, which was implemented in 2000. a A margin is the difference between a
hospital*s revenue and costs, divided by its revenues.

Potential CAHs with DPUs can compare the financial benefits of CAH
conversion to the benefits of keeping their DPUs. Some that suffered
losses on their inpatient and outpatient Medicare payments may lack a
financial incentive to become a CAH because DPU revenues help offset those
losses. If the projected increase in revenue under cost- based payment
that a hospital would receive as a CAH is lower than the loss of revenue
from having to close its DPU, the hospital may chose not to convert to CAH
status. Just over half of the DPUs operated by potential CAHs had net
gains on their Medicare payments (see table 4). A DPU may also provide a
financial benefit to the hospital because it enables the hospital to
spread its fixed costs over more services. Several administrators of
potential CAHs with a DPU whom we interviewed stated that their DPU had
contributed positively to the hospital*s financial situation, providing a
revenue source they would be reluctant to relinquish to gain CAH status.

Page 17 GAO- 03- 948 Critical Access Hospital Program Table 4: Medicare
Margins for DPUs of Potential CAHs, Fiscal Year 1999 DPUs of

potential CAHs Number Median

Medicare margin a (percent)

Number (percent) of

DPUs with negative

margins Number

(percent) of DPUs with

positive margins

Psychiatric 86 0.9 28 (33) 47 (55) Rehabilitation 12 0.0 5 (42) 5 (42) All
98 0.9 33 (34) 52 (53)

Source: Fiscal year 1999 Medicare hospital cost reports. Notes: Because 5
of the potential CAHs had both a psychiatric and rehabilitation DPU, there
are a total of 98 DPUs among the 93 potential CAHs. Margin information is
not included for 11 psychiatric DPUs and 2 rehabilitation DPUs due to
incomplete data or the exclusion of units whose margins were at extreme
outliers. Results do not reflect the effects of the inpatient
rehabilitation PPS, which was implemented in January 2002.

a A margin is the difference between a hospital*s revenue and costs,
divided by its revenues.

While hospitals report that the projected financial impact is generally a
key factor in the decision about whether to become a CAH, 20 some
potential CAHs with DPUs also consider how local access to services

would be affected if the DPU were closed. Some rural hospital
administrators told us that, even when it was financially advantageous to
seek CAH status, they were reluctant to close their DPU because they
believed it was needed to maintain access to psychiatric or rehabilitation
services in their community. Several hospital administrators and state
health officials emphasized the need for patients to be near their family
during treatment and the difficulty that some families would have if they
had to travel outside their community to visit family members receiving
treatment. Other administrators said that if their DPU closed, alternative
sources for these services could be as much as 165 miles away. We were
also told of difficulties in several states with referring psychiatric
patients to hospitals because of a lack of available beds or because
referral

hospitals prefer not to take patients with significant behavioral issues
or believe that psychiatric services should be provided in smaller
communitybased facilities.

20 Rural Policy Research Institute, Rural Hospital Flexibility Program
Tracking Project Year Two Report (Columbia, Mo. 1999). Hospitals with DPUs
Expressed Reluctance to

Seek CAH Conversion If Access to Care Could Be Jeopardized

Page 18 GAO- 03- 948 Critical Access Hospital Program If potential CAHs
were allowed to convert to CAH status while retaining their DPU, the
payment methodology applied to the DPUs could remain

unchanged or could be shifted to cost- based payment along with the acute
care hospital services. Hospitals that have been able to keep their DPU
costs below their Medicare payments under the current methodologies
(rehabilitation PPS for rehabilitation DPUs or TEFRA payment for
psychiatric DPUs) would likely prefer no change because they can continue
to keep their net gains; hospitals that have DPU costs exceeding their
current Medicare payments would likely prefer cost- based payment. If CAHs
were allowed to have DPUs and the DPUs were shifted to costbased

payment, diminished incentives for efficiency could result in higher costs
per case. Under cost- based reimbursement, a hospital can receive higher
payments if its costs increase. Under the rehabilitation PPS or TEFRA
methodologies currently applied to DPUs, their payments cannot exceed a
predetermined amount, creating pressure on them to operate efficiently. If
CAHs were allowed to operate DPUs and the DPU beds and patients*

length of stay were counted against the CAH limits, only one of the 93
potential CAHs with DPUs would have met both limits in fiscal year 1999.
Among these 93 potential CAHs, the median bedsize of psychiatric DPUs

was 11 and the median bedsize of rehabilitation DPUs was 13. If their DPU
beds, acute care beds and swing beds were added together, 88 would have
exceeded the CAH bedsize limit. Similarly, psychiatric inpatient stays at
these potential CAHs averaged 11.8 days, and rehabilitation DPU inpatient
stays averaged 13.7 days, both significantly longer than the CAH limit of
an

annual average of 4 days. About eighty percent of the potential CAHs with
DPUs exceeded the CAH length- of- stay limit when the DPU length of stay
and acute care length of stay were counted together. Paying DPUs
Associated

with CAHs Reasonable Costs Would Reduce Incentives to Operate Efficiently

Most Potential CAHs with DPUs Exceeded CAH Bedsize and Length- of- Stay
Limits When DPUs* Patients Were Counted

Page 19 GAO- 03- 948 Critical Access Hospital Program Hospitals we studied
commonly experienced at least a small seasonal increase in their patient
census, most often during winter. Such increases can be an obstacle for
some hospitals considering CAH conversion if it

causes them to exceed the CAH patient census limit of no more than 15
patients at any time, or length of stay limit of an average of 4 days. We
found 129 potential CAHs that likely would have been able to meet the
patient census limit of 15 in 1999 if not for the seasonal increase in
their patient census. About 40 percent of these 129 potential CAHs,
however, had positive Medicare margins, meaning they would have little
financial incentive to switch from the PPS to CAH cost- based payment. In
contrast to the CAH patient census limit, the patient length of stay limit
is an annual average, and gives CAHs the flexibility to occasionally keep
some acute care patients longer than 4 days as long as the average remains
below 4.

Among hospitals we studied, seasonal fluctuations in patient volume were
common. In 1999, over 80 percent of potential CAHs had an increase in
their patient census averaging at least one additional patient per day
during a 3- month period. To assess whether this finding is consistent
with small and medium- size hospitals in general, we analyzed Medicare
patient claims for 2,139 hospitals with an average census of no more than
50 patients and found that about 90 percent had an increase in their
patient

census averaging at least one additional patient per day during a 3- month
period of 1999.

For nearly three- quarters of potential CAHs, the patient volume increase
in 1999 occurred during the winter. This pattern was consistent with
reports from hospital officials that their patient census often increased
during the winter due to a higher incidence of flu and pneumonia. The
seasonal increase in patient census was greater for larger potential CAHs.
For example, potential CAHs with 41 to 60 beds averaged 2.8 patients more
per day during their peak 3- month period, while potential CAHs with no
more than 15 beds averaged 1. 3 patients more per day during this period
(see table 5). Seasonal Variation in

Patient Census Is Common and May Impede CAH Eligibility for Hospitals Near
the CAH Limit

Most Hospitals Experience Higher Patient Census during Winter

Page 20 GAO- 03- 948 Critical Access Hospital Program Table 5: Seasonal
Increase in Average Acute Care Patient Census among Potential CAHs, by
Bedsize, 1999

Patient census Potential CAHs with a high season average

census exceeding thresholds

Bedsize Number of potential CAHs

Estimated 3- month high season average Annual

average Exceeded 15 acute care

patients Exceeded 20

acute care patients

1- 15 45 3.5 2.2 0 0 16- 25 124 7.2 5.5 3 0 26- 40 284 10.5 8.3 40 2 41-
60 195 13.2 10.4 72 3 >60 35 13.3 10.6 14 0

Total 683 10.4 8.1 129 5

Source: GAO analysis of Medicare inpatient claims. Note: Because this
analysis was based on hospitalizations of Medicare patients, rather than
all patients, we used the hospital*s annual ratio of all patients to
Medicare patients to estimate each hospital*s total patient census by
season. (See app. I for a description of our methodology.)

There were 129 potential CAHs that had at least a slight seasonal increase
in 1999 that pushed them over the CAH limit of 15 acute care patients per
day for some portion of the year. These 129 potential CAHs had an average
daily patient census of about 13.2, with none having an annual average

above 15. But these potential CAHs had an estimated average acute care
patient census of 16.9 during their peak season (see table 6), nearly two
patients per day higher than the CAH limit. Because CAH Patient

Census Limit Is Absolute, Potential CAHs Near the Limit May Have
Difficulty Staying under It

Page 21 GAO- 03- 948 Critical Access Hospital Program Table 6: Potential
CAHs with Estimated Seasonal Increases in Patient Census That Pushed Them
over CAH Limit, 1999

Potential CAHs with a seasonal increase in patient census 129

Estimated average increase in patients per day during seasonal increase
3.7 Total annual average daily census 13.2 Estimated total average daily
census during seasonal increase 16.9

Source: GAO analysis of Medicare inpatient claims. Note: Because this
analysis was based on hospitalizations of Medicare patients, rather than
all patients, we used the hospital*s annual ratio of all patients to
Medicare patients to approximate each hospital*s total patient census by
season.

About 40 percent of the 129 potential CAHs with seasonal increases that
pushed them over the CAH patient census limit had net gains on combined
inpatient and outpatient payments for Medicare services (see table 7).
These potential CAHs would have a financial incentive to remain under the
PPS, where they can keep the difference between payments and their costs,
rather than convert to CAH status, where they would be paid only their
reasonable costs.

Table 7: Financial Performance of Potential CAHs with a Seasonal Increase
in Patient Census That Pushed Them over CAH Limit, Fiscal Year 1999 Median

margins a in percent (n= 129)

Number (percent) of hospitals with negative

margins Number

(percent) of hospitals with positive

margins

Medicare inpatient 2.4 57 (44) 72 (56) Medicare outpatient -19.3 122 (95)
5 (4) Medicare inpatient and outpatient -2.7 75 (59) 52 (41)

Total facility (all payers) 2.5 47 (36) 82 (64)

Source: Fiscal year 1999 Medicare hospital cost reports. Note: For each of
the four calculations of hospital margins, two or fewer hospitals were
excluded due to incomplete data or because their margins were extreme
outliers. Results do not reflect the effects of the outpatient PPS, which
was implemented in 2000. a A margin is the difference between a hospital*s
revenue and costs, divided by its revenues.

Significant Number of Potential CAHs with Seasonal Increase in

Patient Census Have No Financial Incentive to Become a CAH

Page 22 GAO- 03- 948 Critical Access Hospital Program Seasonal
fluctuations in patient length of stay were also common among hospitals we
studied. Among the 2,139 hospitals with a patient census of

no more than 50, about three- fourths had a seasonal increase in their
Medicare length of stay of at least one- third of a day. Sixty- five
potential CAHs had an average Medicare patient length of stay below 4 days
(3.8 days) for 9 months of fiscal year 1999, but their average length of
stay

during the other 3 months was high enough (4.8 days) to push their
Medicare annual average over the 4- day CAH limit, to 4.2 (see table 8).
Among the 620 existing CAHs, 60 had an annual average length of stay
greater than 4.2 days before they converted. These existing CAHs have been
subject to the 4- day limit since they became CAHs, suggesting that
potential CAHs with an annual average of 4.2 days would be able to remain
under the limit if they converted.

Table 8: Potential CAHs with Seasonal Increase in Medicare Patients*
Length of Stay That Pushed Them over the 4- day CAH Limit, 1999

Potential CAHs with increase pushing them over the limit 65

Average Medicare length of stay during 9- month period (days) 3.8 Average
Medicare length of stay during 3- month seasonal increase (days) 4.8
Annual average Medicare patient length of stay (days) 4.2

Source: GAO analysis of Medicare inpatient claims.

The relaxation of the CAH length- of- stay limit in 1999 from an absolute
limit of 4 days to an annual average of 4 days has made it easier to meet
because hospitals are able to keep some patients for a longer period, as
long as the hospital*s annual average remains below the limit. Examples of
how a hospital can manage its length of stay during the course of a year
include discharging longer- stay patients to skilled nursing care in the
hospital*s swing beds or transferring them to referral facilities.
Administrative staff of one rural hospital considering CAH conversion
reported that its average length of stay dropped over 3 years from 5.3 to
3.7 days. The decline, in their opinion, was due to factors such as
utilization review, emphasis on community- based services, increased use
of post- acute care, and education of staff.

The ineligibility of hospitals with DPUs or with seasonal increases in
patient stays that push them over a CAH limit impedes CAH conversion for
some hospitals that might otherwise be able to become CAHs. The
ineligibility of hospitals with DPUs may result in the loss of some rural
DPU services if potential CAHs close their DPU as part of becoming a
Remaining under Lengthof-

Stay Limit Is Manageable Because It Is an Average

Conclusions

Page 23 GAO- 03- 948 Critical Access Hospital Program CAH. Hospitals
seeking CAH status may occasionally need to transfer patients to stay
under the CAH limit of 15 acute care patients if they

otherwise periodically exceed 15 due to seasonal increases. Since
inpatient rehabilitation and psychiatric services are less prevalent in
rural areas, enabling rural DPUs to continue operating can help preserve
the availability of services. In fiscal year 1999, 25 hospitals ceased
operation of their DPU as part of becoming a CAH, and beneficiaries in the
affected communities have lost a local provider of these services. Any of
the 93 potential CAHs with a DPU may also relinquish it to convert to CAH
status if hospital officials conclude that shifting to CAHs* cost- based
payment is the best way to maximize revenue and preserve the other
services they offer. Among these 93 potential CAHs, 47 had net losses on
Medicare services in fiscal year 1999, indicating they might benefit from
CAH conversion.

Because it is generally difficult for rural hospitals to staff and
maintain a DPU, it is unlikely that allowing CAHs to operate DPUs would
result in many existing CAHs opening new DPUs, as long as the DPUs
continue to be paid under PPS and TEFRA. If DPUs operated by CAHs were
paid their reasonable costs, however, DPUs would have less financial
incentive to operate efficiently. The experience of rural DPUs under the
new rehabilitation PPS or the forthcoming psychiatric PPS may provide
information about whether Medicare payments under these PPSs will be
appropriate for rural DPUs.

If CAHs were allowed to operate DPUs, they would generally not be able to
stay under the limits on bedsize, length of stay, and patient census if
the DPU beds and patient stays were counted against current limits.
Relaxing the limits for CAHs with DPUs or not counting the DPU beds or
patient stays for purposes of determining whether the CAH meets the limits
would enable some or all potential CAHs with DPUs to convert to CAH
status.

Relaxing the CAH census limit to an annual average of 15 acute care
patients rather than an absolute limit of 15 would accommodate the 129
potential CAHs that exceeded the current limit due to a seasonal increase
as they all had an annual average census below 15. Such a change would

provide CAHs greater flexibility in their management of patient census,
just as the relaxation of the length of stay limit in 1999 to an annual
average of 4 days provided CAHs greater flexibility in their management of
patients* length of stay. CAHs would then not be required to transfer
patients whenever they would otherwise exceed the limit, as long as they
manage their census so that their annual average is below the limit. It

Page 24 GAO- 03- 948 Critical Access Hospital Program would not be
necessary to increase the number of acute care beds CAHs are allowed to
maintain in order to implement this relaxation of the

patient census limit. More than three- quarters of existing CAHs and
potential CAHs have swing beds which they could use to accommodate
additional acute care patients beyond 15, since the limit is 25 beds for
CAHs with acute and swing beds. Among the 129 potential CAHs, about 60
percent had net losses on Medicare services in fiscal year 1999,
indicating they might benefit from CAH conversion, while the 40 percent
with net gains would less likely have the financial incentive to convert.

Many potential CAHs that decide to seek CAH status would need to adjust
their bedsize or length of stay to become CAHs, just as about 60 percent
of existing CAHs needed to reduce their bedsize and 14 percent needed to

reduce their length of stay in fiscal year 1999. CAH status and the
costbased reimbursement that goes with it have proven to be attractive
enough that hospitals have been willing to make the necessary adjustments.

We suggest that the Congress may wish to consider allowing hospitals with
DPUs to convert to CAH status while making allowances for DPU beds,
patients, and lengths- of- stay when determining CAH eligibility, and that
CAH- affiliated DPUs be paid under the same formulas as other inpatient
psychiatric or rehabilitation providers. We also suggest that the Congress
may wish to consider changing the CAH limit on acute care patient census
from an absolute limit of 15 acute care patients to an annual average of
15 to give CAHs greater flexibility in the management of their patient
census.

In commenting on a draft of this report, the Department of Health and
Human Services said that these modifications to CAH eligibility criteria
would provide the needed flexibility for some additional facilities to
consider conversion to CAH status. It stated that the key is to provide
the proper incentives for facilities to convert when they meet the
statutory requirements and when it is the right thing to do for a
particular community.

HHS suggested that we further emphasize several issues regarding CAH
eligibility and payment. (See app. II for the full text of HHS*s written
comments.) HHS pointed out that it is important to consider that the
financial incentives for efficiency under TEFRA payments to psychiatric
DPUs or rehabilitation PPS payments to rehabilitation DPUs would not be
preserved if CAHs were able to claim cost- based reimbursement for their
Matters for

Congressional Consideration

Agency Comments and Our Evaluation

Page 25 GAO- 03- 948 Critical Access Hospital Program DPUs, and therefore
HHS said such DPUs should continue to be paid separately from the CAH. The
department also emphasized that CAHs are required to meet more limited
health and safety standards compared to

other acute care hospitals and raised concerns that any DPUs operated by
CAHs would likewise be subject to more limited health and safety standards
unless the Congress acted to maintain standards currently in place for
DPUs. Furthermore, HHS suggested that we analyze the extent to which
inpatient rehabilitation and psychiatric services are available to

rural residents beyond their local hospitals in order to determine whether
such services are more or less accessible to rural residents than other
specialty services. The department expressed concern that non- CAH
hospitals that are within close proximity to CAHs may perceive unfair
treatment if such CAHs are allowed to operate DPUs. Finally, in commenting
on the relaxation of the CAH acute care patient census limit to an annual
average of 15, HHS proposed that we consider suggesting corresponding
changes to the CAH bedsize limit.

As we noted in the draft report, incentives for efficiency that exist
under the current payment systems for inpatient psychiatric and
rehabilitation services would not be preserved under cost- based
reimbursement. We revised the matters for congressional consideration to
specifically suggest that CAH- affiliated DPUs be paid under the same
formulas as other inpatient psychiatric or rehabilitation providers. We
also agree with HHS that there are differences in conditions of
participation between hospitals and CAHs and that appropriate health and
safety standards should be

maintained for CAH- affiliated DPUs, and we modified the report
accordingly. However, determining what health and safety standards should
be applied to the DPUs of CAHs was beyond the scope of this report. While
we noted differences in the availability of inpatient rehabilitation and
psychiatric services between rural and urban areas in the draft report,
measuring in detail the level of access rural residents have to various
specialty services was beyond the scope of this report. We believe that
the close proximity of non- CAH hospitals to CAHs with DPUs would only
present a fairness issue if such CAH- affiliated DPUs are paid cost- based
reimbursement or if they are subject to less stringent regulations. If
such DPUs operate under the same payment methodologies and regulations as
other DPUs, this would not be an issue. A detailed examination of the
levels of competition between CAH and non- CAH

hospitals was beyond the scope of this report. We clarified in the report
that we are not suggesting any changes to the CAH limits of 15 acute care
beds or 25 total beds when swing beds are included, since most CAHs have
swing beds that could be used when the acute care patient census

Page 26 GAO- 03- 948 Critical Access Hospital Program exceeds 15. HHS also
provided technical comments, which we have incorporated as appropriate.

We are sending copies of this report to the Secretary of Health and Human
Services and interested congressional committees. We will also make copies
available to others upon request. In addition this report is available at
no charge on the GAO Web site at http:// www. gao. gov.

If you have any questions about this report, please call me at (202) 512-
7119. Other major contributors are listed in appendix III.

A. Bruce Steinwald Director, Health Care * Economic

and Payment Issues

Page 27 GAO- 03- 948 Critical Access Hospital Program List of Committees

The Honorable Charles E. Grassley, Jr. Chairman The Honorable Max Baucus
Ranking Minority Member Committee on Finance United States Senate

The Honorable Bill Thomas Chairman The Honorable Charles B. Rangel Ranking
Minority Member Committee on Ways and Means House of Representatives

The Honorable W. J. *Billy* Tauzin Chairman The Honorable John D. Dingell
Ranking Minority Member Committee on Energy and Commerce House of
Representatives

Appendix I: Scope and Methodology Page 28 GAO- 03- 948 Critical Access
Hospital Program To identify potential Critical Access Hospitals (CAHs),
we selected rural, non- CAH hospitals with an annual average patient
census of 15 or fewer

acute care patients, based on patient census figures reported in fiscal
year 1999 Medicare cost reports. 1 Any hospital that had converted to CAH
status as of January 1, 2003 was excluded from the list of potential CAHs.
We defined potential CAHs based on their annual average census, rather
than by bedsize, because average census better represents the bed capacity
a hospital would need to support its current demand for services. If
potential CAHs have more beds than necessary to meet their patient

demand, they can decertify beds in order to meet CAH eligibility criteria.
Our inclusion of hospitals with an average census up to 15 is likely a
high estimate of the number of potential CAHs. Hospitals with an annual
average of 15 acute care patients per day may need more than 15 acute care
beds to accommodate variations in their patient census that periodically
cause them to exceed 15. From the resulting list of 683 potential CAHs, we
identified hospitals

operating rehabilitation or psychiatric distinct part units (DPUs), as
well as those with seasonal variation in patient census or length of stay
that caused them to exceed CAH limits. For our analysis of seasonal
variation in patient census, we used the volume of Medicare patients as a
proxy for

total patient volume because national data on day- to- day variation
inpatient admissions were only available for Medicare patients. We
calculated from hospital cost reports the Medicare share of each
hospital*s total acute care patient volume, and for each hospital
multiplied the CAH

limit of 15 acute care patients by its Medicare share in order to define a
comparable limit based on Medicare patient stays. For example, if a
hospital*s Medicare share of patients was 67 percent in fiscal year 1999,
then a Medicare census of about 10 acute care patients was considered to
be equivalent to a total census of 15 acute care patients. Using Medicare

inpatient claims data for 1999, we defined seasonal variation in daily
census as having a period of 3 consecutive months with an average census
greater than the estimated limit, with the remaining nine months* census
averaging below the estimated limit. We identified 129 potential CAHs as
having a seasonal increase that caused them to exceed the limit for a 3-
month period, while staying under for the remaining 9 months. To estimate
total patient census for these hospitals for each season, we multiplied
their

1 Medicare cost report data for fiscal year 1999 were used because they
were the most current complete data available. There is typically a
several year delay between the start of a fiscal year and the point at
which a complete set of audited hospital cost report data are available
for that year. Appendix I: Scope and Methodology

Appendix I: Scope and Methodology Page 29 GAO- 03- 948 Critical Access
Hospital Program Medicare census by their ratio of total patients to
Medicare patients. We defined seasonal variation in length of stay as
having a period of 3

consecutive months with an average Medicare length of stay greater than 4
days with an average for the remaining 9 months of less than 4 days. In
addition, we identified only those hospitals for which their seasonal
increase in length of stay caused them to exceed the CAH limit of an
average of 4 days.

Because we used Medicare utilization to estimate hospitals* total patient
utilization for each season, the hospitals we identified as having
seasonal variation that causes them to exceed CAH limits may not be
precisely the same set of hospitals that would have been identified if
claims data for all patients had been available. Rather, our analysis
provides an estimate of the proportion of potential CAHs so affected. By
broadly defining seasonal variation, we captured all the hospitals that
have census or length of stay fluctuations around the CAH limits,
regardless of the magnitude of the fluctuation. We calculated Medicare
margins and total facility margins using fiscal year

1999 Medicare hospital cost report data, using methods developed jointly
by the Centers for Medicare & Medicare Services (CMS) Office of the
Actuary and the Medicare Payment Advisory Commission. The reported median
margins are hospital- weighted, meaning that each hospital counts equally
in the calculation of the median, regardless of differences in hospital
size or total revenues.

We interviewed officials at CMS, at the Federal Office of Rural Health
Policy, and state staff administering Flex Program grants in 11 states
(table 9). To get a comprehensive perspective of how current and potential
CAHs are affected by CAH eligibility criteria, we also conducted an e-
mail survey of all state CAH coordinators, and received e- mail responses
or

directly interviewed 42 out of 47. In addition, we interviewed researchers
with the Rural Hospital Flexibility Tracking Project, an evaluation of the
Flex Program funded by the FORHP. We interviewed administrators of 24 CAHs
and potential CAHs across 10 states, and made site visits to 7 of these
hospitals in 3 states. These 10 states were selected based on having
significant CAH enrollment or potential enrollment, and representing

different regions of the country.

Appendix I: Scope and Methodology Page 30 GAO- 03- 948 Critical Access
Hospital Program Table 9: Summary of Site Visits and Interviews State

Interviewed state staff administering Flex Program

grants Hospital site visit

Interviewed hospital administrators

Number of administrators of existing and potential CAHs

interviewed

Alabama X 1 Indiana X X X 2 Iowa X X 2 Kansas X X 2 Mississippi X X X 5
Montana X X 1 Nebraska X North Carolina X X X 2

South Dakota X Texas X X 2 Vermont X X 1 Washington X X 6

Total 11 3 10 24 Source: GAO.

Appendix II: Comments from the Department of Health and Human Services
Page 31 GAO- 03- 948 Critical Access Hospital Program Appendix II:
Comments from the Department of Health and Human Services

Appendix III: GAO Contact and Staff Acknowledgments

Page 35 GAO- 03- 948 Critical Access Hospital Program A. Bruce Steinwald,
(202) 512- 7119 Jean Chung, Chris DeMars, Michael Rose, Margaret Smith,
and Kara Sokol

made key contributions to this report. Appendix III: GAO Contact and Staff

Acknowledgments GAO Contact Acknowledgments

(290053)

The General Accounting Office, the audit, evaluation and investigative arm
of Congress, exists to support Congress in meeting its constitutional
responsibilities and to help improve the performance and accountability of
the federal government for the American people. GAO examines the use of
public funds; evaluates federal programs and policies; and provides
analyses, recommendations, and other assistance to help Congress make
informed oversight, policy, and funding decisions. GAO*s commitment to
good government is reflected in its core values of accountability,
integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through the Internet. GAO*s Web site (www. gao. gov) contains abstracts
and fulltext files of current reports and testimony and an expanding
archive of older products. The Web site features a search engine to help
you locate documents using key words and phrases. You can print these
documents in their entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as *Today*s Reports,* on its
Web site daily. The list contains links to the full- text document files.
To have GAO e- mail

this list to you every afternoon, go to www. gao. gov and select
*Subscribe to e- mail alerts* under the *Order GAO Products* heading.

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to: U. S. General Accounting Office 441 G Street NW, Room LM
Washington, D. C. 20548 To order by Phone: Voice: (202) 512- 6000

TDD: (202) 512- 2537 Fax: (202) 512- 6061

Contact: Web site: www. gao. gov/ fraudnet/ fraudnet. htm E- mail:
fraudnet@ gao. gov Automated answering system: (800) 424- 5454 or (202)
512- 7470 Jeff Nelligan, Managing Director, NelliganJ@ gao. gov (202) 512-
4800

U. S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.
C. 20548 GAO*s Mission Obtaining Copies of

GAO Reports and Testimony

Order by Mail or Phone To Report Fraud, Waste, and Abuse in Federal
Programs Public Affairs
*** End of document. ***