DEPARTMENTAL GRANT APPEALS BOARD
Department of Health and Human Services
SUBJECT: South Dakota Department of Social Services
Docket No. 87-56
Audit Control No. 08-60218
Decision No. 934
DATE: February 5, 1988
DECISION
The South Dakota Department of Social Services (State) appealed
a
determination by the Health Care Financing Administration
(HCFA)
disallowing $2,461,795 in federal financial participation (FFP)
claimed
under Title XIX of the Social Security Act (Act). The claims
were for
services provided to Medicaid recipients in the State's two
State-owned
and operated intermediate care facilities for the mentally
retarded
(ICFs/MR) during the State's fiscal years 1981 through 1984.
Based on an audit, HCFA found that the State had not calculated
the
Medicaid reimbursement rate for the two ICFs/MR in accordance with
the
approved State plan. Specifically, HCFA found that the State had
not
followed its plan in determining the number of patient days used
to
compute the per diem rate for services, resulting in an overpayment
of
$2,151,211 of FFP. In addition, HCFA found that the State had failed
to
adjust the rates for "anticipated costs" that were included
in
calculating the rates but were not actually incurred by the
facilities,
resulting in an overpayment of $310,584 of FFP. After the
State
presented evidence at a hearing that, in calculating the
disallowance
for patient days, HCFA had erroneously included bed days related
to
infirmary beds, HCFA agreed to reduce the disallowance by $840,439.
As we discuss below, we find that the State has shown that
the
interpretation of its plan it applied in determining the number
of
patient days to be used in calculating the rates was reasonable and
was
supported by consistent State administrative practice. The
audit
finding was based on an accounting principle which does not apply in
the
particular circumstances here. We also conclude that the State
plan
establishes a prospective rate- setting system and cannot reasonably
be
read to require retrospective adjustments reflecting a
comparison
between anticipated costs used in calculating the rate and actual
costs
incurred during the rate year. Consequently, we reverse
the
disallowance.
Background
In order to qualify for FFP, a state's claim for the costs of
medical
services must be in accordance with the approved Medicaid state
plan.
Section 1903(a) of the Act. The plan must fulfill certain
statutory and
regulatory requirements, and be approved by the
Secretary. Prior to
1980, states were required under section
1902(a)(13) of the Act to
reimburse skilled nursing facility (SNF) and
intermediate care facility
(ICF) services (including those in a ICF/MR) at
rates determined on a
"reasonable cost- related" basis, using methods and
standards approved
by the Secretary. Section 962 of Public Law 96-499,
the Omnibus
Reconciliation Act of 1980, amended section 1902(a)(13)(A) of the
Act to
require that state plans provide for payment of such services--
through the use of rates (determined in
accordance with methods
and standards
developed by the State . . . ) which the
State
finds, and makes assurances
satisfactory to the Secretary, are
reasonable and adequate to meet the costs which must be
incurred
by efficiently and economically
operated facilities in order to
provide
care and services in conformity with applicable State
and
Federal laws, regulations, and
quality and safety standards. . .
.
This provision, known as the "Boren Amendment," was intended to
provide
the states greater flexibility in developing methods of
provider
reimbursement. The legislative history indicates that Congress
intended
to keep requirements on states to the minimum level necessary to
assure
accountability, and not to burden states with unnecessary
paperwork
requirements. 48 Fed. Reg. 56047 (Dec. 19, 1983),
citing S.REP. No.
96-471.
Both before and after the Boren Amendment, regulations
implementing
section 1902(a)(13) provided that (1) a state Medicaid plan must
set out
the methods and standards to be used by the state Medicaid agency
in
determining rates, and (2) payments to providers must be at
rates
determined in accordance with those methods and standards. See
42
C.F.R. 447.201; 447.273 (1980); 42 C.F.R. 447.201, 447.252 (1981).
During State fiscal years 1981 through 1984, South Dakota provided in
its
state plan for a prospective system of reimbursement. Under
this
system, the State used the costs of a base year, adjusted for
inflation
and divided by an "occupancy rate" to determine per patient/per
diem
rates to be used during a rate year. In general, these rates were
not
subject to adjustment to reflect actual costs during the rate year.
The plan provision related to the first issue here stated:
"The
occupancy rate used in calculating per diem rates shall be the
greater
of actual or 95%." State's Exhibit (Ex.) 12, p. 3
(unnumbered). An
audit performed by the Office of Audit, Office of
Inspector General,
examined reimbursement rates for two State- owned and
operated ICFs/MR,
Custer State Hospital (Custer) and Redfield State Hospital
and School
(Redfield) during fiscal years 1981 through 1984. The
auditor found
that, in calculating the rates for Custer and Redfield for
certain
years, the State had used actual patient days of service provided
during
the base years. According to the auditor, the State should have
used
95% of the certified bed capacities of the facilities, which was
greater
than the actual patient days. Since the occupancy rate is used
in the
denominator of the calculation, the auditor found that, as a result
of
the State's use of actual days rather than 95% of certified
bed
capacity, the rates were overstated.
The second State plan provision at issue provided for allowances
for
"anticipated costs." The auditor found that, in calculating the
rates
for Custer and Redfield, the State had included an allowance for
costs
of anticipated new positions and salary increases but that many of
these
costs were not incurred by the facilities in the year the funding
was
received.
Based on the audit findings, HCFA disallowed FFP in the difference
between
the reimbursement amounts for services provided to Custer and
Redfield during
the audit period at the rates used by the State and at
the rates the auditor
found should have been used.
Below, we first discuss the factors used in previous Board decisions
in
analyzing state plan provisions. We then analyze issues related to
the
interpretation of the occupancy provision in the South Dakota
State
plan. 1/ Finally, we analyze issues related to the
"anticipated costs"
provision.
General Considerations
The State here argued based on the Board's decision in Arkansas Dept.
of
Human Services, DGAB No. 540 (1984) that the State's interpretation
of
its plan should be sustained if reasonably supported by the language
of
the provision or by factors such as administrative practice.
A
comprehensive examination of Board decisions indicates, however,
that
this is not a rigid standard as the State implied and that the Board
has
considered many factors in similar cases, including the context in
which
they arise. See, e.g., Louisiana Dept. of Health and Human
Resources,
DGAB No. 492 (1983); Georgia Dept. of Medical Assistance, DGAB No.
601
(1984).
In considering whether a state has followed its approved state plan,
the
Board first examines the language itself. If the provision
is
ambiguous, the Board will consider whether the state's
proposed
interpretation gives reasonable effect to the language of the plan
as a
whole. The Board will also consider the intent of the
provision. A
state's interpretation cannot prevail unless it is
reasonable in light
of the purpose of the provision and program
requirements. Lacking any
documentary, contemporaneous evidence of
intent, the Board may consider
consistent administrative practice as evidence
of intent. The
importance of administrative practice is in part
determining whether the
state in fact was applying an official interpretation
of a plan
provision or has advanced an interpretation only as an
after-the-fact
attempt to justify acting inconsistently with or simply
ignoring its
plan.
In the context of rate-setting, it is a relevant factor that under
the
Boren Amendment states have flexibility to develop reimbursement
methods
and that plans will be .approved so long as the state provides
the
requisite assurances. Thus, a post-Boren Amendment case differs
from
that decided in Arkansas Dept. of Human Services, DGAB No. 357
(1982).
On review of that decision, the Claims Court applied the principle
of
contra proferentum (construing against the drafter) based on
the
rationale that HCFA could not fulfill its duty to evaluate and approve
a
rate-setting system unless the provisions were clear. Arkansas v.
United
States, No. 150-85c (Cl.Ct., Feb. 20, 1986). While we agree with
HCFA
that the Boren Amendment did not give the states flexibility to
ignore
the methods set out in a state plan, the shift in emphasis under
the
Boren Amendment requires that greater weight be given to what the
state
intended, since HCFA will approve a state's proposed methods as long
as
the state has made the requisite assurances.
The Board has said, however, that rate-setting for a state-owned
facility
is subject to closer scrutiny. Massachusetts Dept. of Public
Welfare,
DGAB No. 730 (1986); DGAB No. 867 (1987). The purpose of this
closer
scrutiny is to determine whether the state in fact has a
consistently applied
interpretation or is somehow giving preferential
treatment to state-owned
facilities. Medicaid rates do not affect state
costs for such facilities,
only the extent of federal participation in
those costs. In the
Massachusetts cases, what happened was particularly
questionable since the
interpretations the state advanced there
permitted liberal retroactive
adjustment of prospective rates, and since
for most facilities involved, the
reimbursement would have been adequate
to cover the facilities' costs at the
unadjusted rates.
Here, while the State's rate-setting system was developed prior to
the
Boren Amendment, the State subsequently provided assurances that
its
system, as interpreted by the State, met the Boren Amendment
standard.
During Board proceedings, the State presented undisputed evidence
to
show that the amounts the State paid Custer and Redfield did not
exceed
the Boren Amendment standard and, indeed, were low compared to
other
states. See, e.g., State's Ex. 1, Att. C. The State also
presented
undisputed evidence that, for every year but one, the prospective
rates
paid did not exceed the facilities' actual costs. Thus, this case
is
distinguishable from the Massachusetts cases.
We also note that neither provision here was drafted to meet a
specific
program requirement. HCFA has not alleged that the State's
rate-setting
was inconsistent with any federal requirement or policy.
Indeed, HCFA
approved a State plan amendment in 1984 clarifying how the State
applied
its occupancy provision to State-owned ICFs/MR, so there is no
question
that the State's methods were approvable.
The Occupancy Provision
During the entire audit period, the State plan provided that--
The occupancy rate used in calculating
per diem rates shall be
the greater of
actual or 95%.
Attachment 4.19 D, para. 1, State's Exhibit (Ex.) 12.
The State said that its consistent interpretation of the
occupancy
provision was that it called for use of 95% of "the total number of
ICF
or SNF patient days that is possible in accordance with the
State
mandated or approved bed-utilization policy in effect at the facility
in
question." State's brief, p. 28. The State argued that this
was the
proper interpretation because it was consistent with the intent of
the
provision, which was to penalize any facility that maintained
excessive
empty beds, by providing that its rates would be set as if the
costs it
had incurred had been spread over a larger number of patient days
than
had actually been registered. The State argued that its
interpretation
had been consistently applied, not only to State- owned
facilities, but
to private facilities as well.
HCFA said that the provision called for use of 95% of "certified
bed
capacity." HCFA relied primarily on the audit finding that the
State
had generally used a figure corresponding to certified bed capacity
in
calculating rates for Custer and Redfield. HCFA also said that the
cost
report form used by the State supported its position. HCFA did
not
advance any reason why use of this figure was required, nor did
HCFA
present any affirmative evidence regarding the intent of the
provision,
but disputed the State's evidence regarding intent and
administrative
practice.
Below, we first examine the language of the plan and of the provider
cost
report form. We then discuss the State's evidence regarding intent
and
administrative practice. Finally, we examine the State's
allegation
that HCFA's interpretation was inappropriate.
The language of the plan and cost report
The State plan contained no definition of the terms "occupancy"
or
"occupancy rate" as used in the occupancy provision. The State
admitted
that it had not promulgated to its providers any written explanation
of
the provision.
The provision on its face is ambiguous; the reader automatically
questions
"95% of what?" The cost report form which the State required
its
providers to use provided a partial answer. The form (Exhibit
14)
required providers to report the "total bed capacity" at the
beginning
of the year as well as the numbers and dates of any beds added
or
subtracted during the year. This yielded a figure called
"total
possible resident days." The form also calls for "total resident
days,"
indicating parenthetically that this means physical resident days
plus
paid reserve bed days. The report calls for entering as a bottom
line
figure the "total resident days" or 95% of "total possible
resident
days," whichever is larger.
We agree with HCFA that the cost report form is the best evidence of
how
to interpret the occupancy provision. Moreover, the comparison
of
"actual resident days" with "possible resident days" is a common
sense
reading of the phrase "actual or 95%." The difficulty is that the
cost
report form simply does not resolve the ambiguity entirely. The
term
"bed capacity" is not defined. HCFA would have us interpret
"bed
capacity" to mean "certified bed capacity" according to State
Department
of Health certification forms. But neither the plain
language of the
plan nor any federal purpose requires use of certified beds
and, as
discussed below, the State showed that it led to inappropriate
results
as applied here. 2/ Moreover, HCFA could have easily required
use of
certified bed capacity by simply having the State add the
word
"certified" to the cost report form before approving the plan
initially.
.On the other hand, we do not think that the fact that no
particular
source for bed capacity is specified means that the State
could
arbitrarily choose any figure, particularly where this could lead
to
manipulation of rates for State-owned facilities. We sustain
the
State's interpretation only because it relates to a principle,
bed
utilization policy, against which the choice of a figure can be
measured
objectively and which derives from the purpose of the provision,
as
discussed below.
We also reject HCFA's argument that the lack of any written
interpretation
of the State plan leaves "an almost unavoidable inference
[that] the state
simply failed to change its state plan in response to
the changing
circumstances found in those ICF/MR's." Tr. III, p. 483.
No such
inference is warranted here. As discussed below, the State
presented
credible evidence to show that it did not simply ignore its
plan when faced
with the question of how to apply the provision to
State-operated ICFs/MR,
but officially adopted an interpretation it
considered permissible.
The purpose of the provision
At the hearing, the State presented testimony by the State's
Medicaid
Director, who had directed the development and initial
implementation of
the State's prospective methodology for reimbursing ICFs
and SNFs in
1975. At that time, the only ICFs and SNFs participating in
the State's
Medicaid program were privately-owned; the State-operated
facilities
were not certified as ICFs/MR until 1977. The Medicaid
Director
testified that the purpose of the occupancy provision was "to
encourage
the nursing homes to occupy all their beds because there was a
shortage
of beds in the State of South Dakota and, of course, for the purpose
of
cost containment." Transcript, Vol. I (Tr. I), p. 21. The
provision
keeps a facility from shifting the costs of excess beds to
occupied
beds. He further explained that the State's bed utilization
policy for
State-operated ICFs/MR was different from most other facilities;
while
it was State policy to encourage most facilities to keep their
beds
full, the State had a policy to encourage ICFs/MR
to
"deinstitutionalize" residents, i.e., move them out of the
institutions
into the least restrictive environment possible. Tr. I, p.
62; see,
also, Tr. I, pp. 190- 201. 3/
HCFA presented no affirmative evidence concerning the purpose behind
the
occupancy provision, but questioned why, if the State's bed
utilization
policy was the lodestar for interpreting the provision, it was
not
mentioned in the State plan. As the State argued, State plans are
not
as detailed as HCFA's argument suggests they should be; nothing in
the
applicable federal provisions requires that the purpose of a
provision
be explained in the plan. The State's explanation of the
provision
presented to us makes sense and we have no reason to doubt
it. We do
note, however, that we would not consider the State's
interpretation
reasonable if it had the effect of allocating to occupied beds
excess
fixed costs not benefiting those beds, contrary to the cost
containment
purpose of the occupancy provision. To show that the
calculation here
did not have that effect, the State provided uncontroverted
evidence to
show that emptied space did not remain idle and that, in general,
the
facilities were under extreme pressure to reduce costs, and did so.
Tr.
I, pp. 146- 148; 160-177. 4/
The State's practice
To show consistent administrative practice, the State
presented
testimony by the person who was the State's Administrator of
Contract
and Auditing Services from 1976 until 1986. In that capacity,
he had
direct .responsibility for determining the rates for SNFs and
ICFs,
including ICFs/MR. The Administrator testified to the
following:
o While the State usually used certified bed capacity for
the
occupancy figure, it would use licensed capacity
if that was lower
than certified capacity and that,
also, the figure would be reduced
if a facility
participated in the State's "supervised living" or
"adult handicapped" programs. Tr. I, pp. 76-82.
o The "working definition" of the term "occupancy" in the
State plan
was "licensed beds by the Health
Department, less any adjustments
that conform to the
State utilization plan." Tr. I, p. 82.
o The State calculated rates for the State-operated
ICFs/MR, from the
time they were certified, using
figures which treated the actual
number of patient
days as being equal to 100% of the "total
possible
resident days" on the cost reports.
o This interpretation of how the occupancy provision should
apply to
State-operated ICFs/MR resulted from
meetings the Adminis- trator
had with officials from
Custer and Redfield when those facilities
joined the
program in the mid-1970s. The officials had
proposed
use of an occupancy figure based on what
they projected the next
year's occupancy could
be. The Administrator met with the
Secretary
and Deputy Secretary of the State Department of
Social
Services. They determined that use of a
projected occupancy would
not be consistent with the
State plan, but that use of 100% of the
actual
patient days would be permissible. They further
determined
that use of 95% of certified bed capacity
would be inappropriate
because the facilities "were
sending patients to the least
restrictive area, in
accordance with our plan, on a daily -- weekly
basis, and we took the position that as a patient left, then
that
bed [for rate- setting purposes] was
decertified as of that day."
Tr. I, p. 86.
5/
HCFA did not present any testimony or other evidence directly in
conflict
with that presented by the State, but sought to undermine the
credibility of
the State's witness by showing that, during the course of
the audit, he had
not articulated the State's position as he did at the
hearing. The
audit workpapers contain notes which the auditor testified
were his notes
concerning conversations with the Administrator taking
place on three
different dates. The auditor testified from these notes
that the first
time the Administrator had raised the concept of
deinstitutionalization was
at the third meeting, on January 29, 1985.
The audit notes from the first two
meetings with the Administrator
state:
12/4/84: "He did concur that for one year, the wrong
patient
day
figure
was used at Custer, but he had an excuse . . .
He
blamed
it on the State Department of Health requiring
Custer
to
reduce the number of beds from 3 a room to 2 a
room.
Under certain circumstances they could board 3 to a room
if
they
put on extra staff (variance given by
Health
Department). He felt this variance should not be
considered
as part of the capacity when computing the rate."
12/7/84: "[He] was told health dept reduce need on daily
basis as
people move out at Redfield [Therefore] what ever it was
was
100%
at Redfield note: (cost rpts say 100%)"
HCFA Ex. D.
The auditor expressed his view that, in these meetings, the
Administrator
was distinguishing between Custer and Redfield with
respect to the patient
day issue and that the point made about two or
three to a room was
irrelevant. Tr. III, pp. 411-412.
When the Administrator's testimony is considered as a whole, we do
not
think these statements are inconsistent in a way which undermines
his
credibility on the key point at issue here. The Administrator
clarified
that his remark concerning use of the wrong figure at
Custer
acknowledged only that the actual count of patient days was wrong
for
one year, as discovered in a State audit. Tr. III, pp.
474-475. His
other alleged statements do not contradict his testimony
regarding what
the State's interpretation was and how it was applied; rather,
they go
to the underlying rationale for that interpretation. The
State
variously explained that rationale as being that the
deinstitutionalized
beds were considered to have been in effect decertified,
or simply that
certified bed capacity should not play a role when considering
"total
possible resident days" for ICFs/MR in view of the
State's
deinstitutionalization policy. In any event, we cannot say that
the
auditor's notes clearly reflect what the Administrator was
saying.
While normally we consider audit notes reliable, given the
standards
under which an auditor operates, the auditor's testimony
on
cross-examination here reveals he simply did not understand the
points
which the State was trying to make regarding the interpretation of
the
occupancy provision. Tr. III, pp. 417-441; 452-455.
On the whole, we found the Administrator to be credible in testifying
that
the State had consistently applied the occupancy provision in light
of the
State's bed utilization policy. The State presented
documentary
evidence, the cost reports from Custer and Redfield, to show that
it had
consistently applied its interpretation to both facilities, contrary
to
what the auditor had found. 6/ Also, the Administrator's
testimony
concerning the events taking place in 1977 is corroborated by
statements
in an affidavit of Redfield's Business Manager. State's Ex.
3, p. 4.
We also reject HCFA's argument that the State's evidence concerning how
it
used a reduced bed capacity for private facilities participating in
the
"supervised living" and "adult handicapped" programs is meaningless
because
the State admitted that costs associated with beds set aside for
those
programs were deleted from the numerator of the rate calculation.
The major
issue here is whether the denominator must always be certified
bed capacity,
as HCFA contended. The State's evidence shows that the
State did not
apply the occupancy provision to always require this, even
in the case of
some private facilities. The question naturally arises,
in light of how
the private facilities were treated, whether costs
should be eliminated from
the numerators for Custer and Redfield's
rates. As indicated above,
however, the State presented testimony that
when a bed in Custer or Redfield
was emptied due to
deinstitutionalization of a resident, the beds were moved
out and the
resulting space was used for services programs provided to the
remaining
residents or as office space for staff. Tr. I, pp.
146-147. Thus, the
fixed costs associated with the space provided a
benefit to the
remaining residents actually receiving services.
The State also presented testimony to dispute HCFA's contention that
the
State was providing preferential treatment to State- operated
ICFs/MR
since it was using certified bed capacity for private ICFs/MR.
That
testimony showed that the private ICFs/MR were small facilities
into
which the deinstitutionalized residents were placed, so that
these
facilities always remained at near to 100% of certified capacity.
Tr.
I, pp. 98-99. Thus, the deinstitutionalization policy simply did
not
affect these facilities in the same way that it affected
the
State-operated ICFs/MR.
HCFA's use of certified bed capacity
At the hearing, the auditor testified that he had chosen "certified
bed
capacity" because the number of certified beds was a figure which
could
be obtained from the State Department of Health and because it is
a
generally accepted accounting princi- ple that figures obtained
from
independent sources are more reliable. Tr. III, pp. 454-455.
The
auditor appeared to think that this accounting principle
controlled,
irrespective of what the State plan required. While we do
not dispute
that such a principle exists, we fail to see how it applies in
this
context; the issue is not which State department has the most
reliable
figures on certified bed capacity, but what figure is called for by
the
State plan in calculating total possible resident days.
Moreover, the State presented evidence that the auditor's use of
certified
bed capacity was inappropriate. The certified bed count for
Redfield
included infirmary beds, used solely for temporary stays by
residents whose
beds in the main part of the facility were set aside
until their
return. After considering the effect of this, HCFA agreed
to reduce the
disallowance. HCFA noted that it had been advised that
Redfield was the
only facility of its kind in the State which has
infirmary beds included in
its certified bed capacity. HCFA then
stated: "Given these unique
circumstances, the Agency has concluded that
it is reasonable not to include
the number of infirmary beds in
Redfield's total certified capacity, in
computing the maximum amount
payable for services rendered at Redfield. . .
." HCFA letter of Nov.
5, 1987.
While HCFA framed its concession in terms of simply agreeing
that
certified bed capacity had been overstated, we agree with the State
that
this concession undercuts HCFA's objection to the
State's
interpretation. The practical effect of the State's policies
regarding
Custer and Redfield was that deinstitutionalized beds would no
longer be
considered beds available for occupancy, irrespective of their
inclusion
on the Department of Health certification forms. Thus, just
as it makes
sense to reduce capacity for infirmary beds, it makes sense not
to
include deinstitutionalized beds when calculating "total
possible
resident days."
Finally, the State presented uncontroverted evidence to show that use
of
certified beds would result in rates below the Boren Amendment
standard.
Thus, it appears as though the State's assurances that the
Boren
Amendment standard was met were given with its consistently
applied
interpretation of the occupancy provision in mind, and
retroactively
requiring the State to use certified beds would result in
paying
inadequate rates.
Accordingly, we conclude that the State did follow its State
plan
occupancy provision when calculating rates for Custer and Redfield.
The "Anticipated Costs" Provision
The second State plan provision at issue here provided:
Allowances may be made for known future
costs not incurred during
the reporting
period if they will be incurred prior to the end
of
the next fiscal year. An
explanation of costs of this nature
must
be attached to the report if they are to be
given
consideration.
State's Ex. 12, para. 16.
The audit found that additional costs had been included in
Medicaid
reimbursement rates for Custer and Redfield for fiscal years
1981
through 1983 for anticipated new positions and for salary
increases.
The audit found that many of these costs were not actually
incurred in
the rate year and that this resulted in the rates being
overstated. The
audit report also stated:
One ICF/MR business administrator told
us that he knew that some
of the new
positions could not be filled, but the extra
money
would be used to offset declining
revenue due to the ICF/MR's
declining
enrollment. SDDSS discovered the problem
with
anticipated costs and had
eliminated them from the fiscal year
1984 rates because they were not being incurred.
State's Ex. 4, pp. 6-7.
In response to the draft audit report, the State showed that (1)
the
allowances for anticipated costs paid to Custer and Redfield during
the
audit period exceeded the actual increased salary and benefits
costs
incurred by the two facilities during that audit period by an
amount
that corresponds to only $22,000 in FFP; (2) during some fiscal
years,
increased costs relating to salaries and benefits exceeded
the
allowances made for anticipated costs; and (3) the auditor
had
erroneously found that some of the new employees specifically listed
as
the basis for the anticipated cost allowance had not been hired when,
in
fact, they had been hired. The audit report responded by adjusting
the
disallowance "to the extent that anticipated costs were incurred for
the
stated purposes." State's Ex. 4, App. D, p. 1. The report
stated,
however, "To the extent that costs were incurred for other purposes
or
not incurred at all, we are recommending their disallowance."
Id. HCFA
adopted the audit report recommendation, disallowing $310,584
in FFP.
The State argued that HCFA's position was inconsistent with the concept
of
a prospective rate-setting system, which does not provide for
retrospective
adjustment to reflect actual costs incurred during the
rate period. 7/
The State acknowledged that its auditors had made an
adjustment to one year's
rate for Custer to reduce the anticipated costs
allowance for costs not
actually incurred, but argued that this was an
aberration from its normal
practice, which was to not make such
adjustments. The State presented
an affidavit by the Administrator of
Contract and Auditing Services (the
person responsible for rate-setting
from 1976 to 1986), who acknowledged that
his audit staff had
recommended an adjustment for Custer's rate for 1982
based on a finding
that Custer had not incurred costs it had anticipated.
State's Ex. 2, p.
11. He further attested:
That adjustment was not based on any
provision in the State Plan
or on any
State law or regulation. On reflection, it is
evident
to me that the Custer adjustment
relating to fiscal year 1982
rates was
inconsistent with the requirements of South
Dakota's
Medicaid State Plan.
State's Ex. 2, p. 11.
At the hearing, the Administrator testified that it was the State's
intent
that anticipated costs would simply be a component of the
prospective rate,
not subject to adjustment, and that the State had no
procedures for making
annual checks on whether anticipated costs were
actually incurred. He
further testified that, although the prospective
rate might be adjusted if
State auditors found an error in the cost
report (or the calculation of the
rate), failure to actually incur an
anticipated cost was not considered an
error in the cost report. Tr. I,
pp. 111-114.
HCFA said that it did "not quarrel with the basic proposition that, to
the
extent that a rate is indeed prospectively set, it is by definition
not
subject to later adjustment based upon actual experience (unless, as
the
State allows, it was established based upon improper data)."
HCFA
brief, p. 16 (emphasis added). HCFA took the position, however,
that
the anticipated costs provision constituted an exception to the
State's
generally-prospective methodology. HCFA raised three general
points in
support of this position: one based on the language of the
anticipated
costs provision, one based on what HCFA called the State's
"admitted
practice," and one based on general policy considerations,
including a
reference to what HCFA called an "intentional misjudgment" by
the
Redfield facility.
Below, we first discuss HCFA's argument that the anticipated
costs
provision constituted an exception to a system which was
otherwise
prospective in nature. We then address whether there was
an
"intentional misjudgment" by Redfield which requires an adjustment.
HCFA's exception argument
HCFA relied primarily on the language of the provision permitting
an
allowance for anticipated costs "if those costs will be incurred."
HCFA
argued, "This language puts a condition on the allowability of
these
costs, making them subject to retrospective readjustment where
those
costs are not incurred; only if they will be incurred are they
allowed."
HCFA brief, p. 14 (emphasis in original).
Prior Board decisions have recognized that states could establish
a
"hybrid system," which was primarily prospective in nature, but
which
permitted some retrospective adjustment for actual costs. HCFA
itself
has taken the position, however, that the method by which such
a
retrospective adjustment would be made would have to be specified in
the
state plan. See Arkansas, DGAB No. 357, pp. 19-20.
Moreover, HCFA's
own description of prospective rate- setting systems, in a
guide on
"Institutional Reimbursement," states: "Such rates may, if
specified in
the plan, be adjusted (in part) for actual costs experienced
during the
rate year." State's Ex. 15, p. IR-13. There is no evidence
here that
the State intended to adopt a hybrid system; to the contrary, the
State
plan states: "All payment shall be final, subject only to
future
adjustment for erroneous cost or statistical reporting discovered
during
the course .of an audit." State's Ex. 12, para. 14 (emphasis
added).
Adjustments for erroneous cost or statistical reporting are
consistent
with a prospective-only system. Under a prospective system,
the
prospective rate is an estimate; the expectation is that it will
not
correspond precisely to the actual costs incurred during the rate
year
by any specific provider.
The "will be incurred" language in the anticipated costs provision is
a
condition placed on obtaining an allowance for anticipated costs.
The
allowance is made at the time the prospective rate is set, and
the
explanation of the costs is to be made when the facility reports on
its
prior year costs before the rate is set. Moreover, it is
inconsistent
with what HCFA has said elsewhere for HCFA here to base a
retrospective
adjustment on language which is not specific and which does not
set out
a method for a retrospective adjustment. Without such a method
being
specified, the question arises whether the adjustment should
compare
specific cost items, as the auditors did here, or only
general
categories of costs. Obviously, this can make a significant
difference
in the amount of any adjustment. (HCFA did not contest the
State's
assertion here that the State incurred increased salary and
benefit
costs in an amount close to the anticipated cost allowances for
that
category of costs during the audit period.) The question also
arises
whether upward adjustments to rates should be made if, as the
State
established here, in some years increased costs in a category exceed
the
anticipated costs allowance for that category.
HCFA's second argument in support of viewing the anticipated
costs
provision as an exception to prospective rate-setting was based on
what
HCFA called "the State's own admitted practice in making a
downward
adjustment itself." HCFA brief, p. 14. HCFA argued that,
even if this
was an isolated variance as the State said, "it constitutes a
clear
acknowledgment of the reasonable- ness of the Agency's interpretation
of
the State plan provision." HCFA brief, p. 15. We do not think that
the
record here establishes a "practice" of the State making
retrospective
adjustments for anticipated costs. Moreover, while even
the one
instance does undercut the State's position somewhat, it is not
a
sufficient basis for adopting HCFA's interpretation. The State's
view
that no retrospective adjustment was intended is directly supported
by
the wording of the plan; HCFA relied on language which is much
more
ambiguous. Moreover, the State provided uncontradicted evidence
from a
State official involved in developing the State's reimbursement
system,
that no retrospective adjustment was intended. State's Ex. l,
pp.
10-11; see, also, Tr. I, p. 65. There is no evidence that the
one
instance of retrospective adjustment was based on any official
State
interpretation to the contrary.
HCFA's final argument in support of its position that the
anticipated
costs provision provided an exception to the State's prospective
system
was that the State's interpretation had no foundation in any
legitimate
public policy. "Where a facility claims an 'anticipated
cost,' but then
does not incur it," HCFA said, "no public policy is served by
permitting
it to benefit from at least its misjudgment in proposing the
inclusion
of this cost," and, indeed, "by permitting facilities to keep
such
windfalls, misjudgments (including intentional misjudgments such
as
found by the auditors here) are encouraged." HCFA brief, p.
15.
According to HCFA, no harm is suffered by a facility if it does not
in
the end receive any reimbursement for a cost which was not incurred.
Contrary to what HCFA argued, the State did advance a programmatic
reason
why there should be no retrospective adjustment here. The choice
of a
prospective system is within the State's flexibility to choose a
less
burdensome reimbursement system than one which requires
retrospective
adjustment based on actual costs in the rate year; yet,
HCFA's position would
require the State to set up a procedure to compare
anticipated and actual
costs and make retrospective adjustments.
Moreover, part of the rationale for
a prospective system is as an
incentive to a provider to keep costs down
below the prospectively set
rate; provider planning is thus based on knowing
in advance the per diem
rate which will be paid. Here, the record shows
that the anticipated
cost allowance led the facility, where it could not
actually hire
persons for the specified positions, to hire other individuals
to
improve the quality of services provided to residents. Tr. I,
pp.
158-159. In the absence of any specific provision for
retrospective
adjustment, the facility would not have known that it would
have had its
rate reduced so that the funding to do this would have to be
repaid.
Adjustments for erroneous cost reporting
As noted above, the State plan permitted adjustment to the
prospective
rate originally determined if an audit revealed erroneous
cost
reporting. This provision, together with the wording of the
anticipated
costs provision, in our view permits an adjustment if there is a
finding
that a provider claimed an anticipated costs allowance by
improperly
reporting costs which it did not know it would incur, at the time
it
requested the allowance. HCFA implied that we should find here
that
Redfield improperly reported anticipated costs it did not intend
to
incur.
We do not think the record here supports such a finding. HCFA's
view
that the auditor had found such an intentional misjudgment was based
on
the auditor's notes concerning a conversation he had with the
Business
Manager of Redfield. The notes, from a conversation taking
place on
2/7/85, state:
He said he knew that Redfield could not
fill all of the positions
for which
additional funding was requested, but he felt that
this
extra money (profit) would offset
declining revenue due to
Redfield's
declining enrollment.
HCFA Ex. D, p. 1.
In recalling this conversation at the hearing, the auditor testified
that
the Business Manager had said that because of the difficulty of
obtaining
people in Redfield they "most likely wouldn't be able to fill
all of the
positions." Tr. III, p. 408. The Business Manager testified
that,
while he did voice concerns to the auditor about the fact that
Redfield's per
diem rates were below costs, he did not recall indicating
that Redfield was
claiming reimbursement for the anticipated positions
with no intent in
filling those positions. Tr. III, p. 154. He
explained all of the
efforts which Redfield took to try to fill open
positions, and asserted that
Redfield had never requested an allowance
for positions they did not expect
to fill. He further explained that,
for any new position, Redfield
would need new authority from the State
legislature (which was difficult to
obtain) and that Redfield construed
that authority as meaning that if they
were not successful in hiring a
professional staff person as authorized, they
could "take the next best
thing and maybe hire a certified aide. . . ."
Tr. III, pp. 158-159.
We found the Business Manager to be a very credible witness. Moreover,
the
auditor's testimony regarding the conversation does not support a
finding
that the Business Manager "knew that some of the new positions
could not be
filled," nor that such knowledge existed at the time the
anticipated costs
allowance was requested. The auditor also
acknowledged that the word
"profit" in the audit notes was probably his
word. Tr. III, p.
442. His testimony indicates that his view of this
issue was colored by
his view that, because Redfield had not actually
filled the specific
positions for which they obtained an "anticipated
costs" allowance, Redfield
had obtained a profit from the rates paid by
the State. Tr. III, pp.
442-448. The record does not support a finding
that there was any
unauthorized profit or windfall here, however. The
State presented
undisputed evidence that the rates paid by the State met
the Boren Amendment
standard and that in all but one year costs exceeded
reimbursement.
We also note that HCFA did not contest the State's allegation that,
during
some fiscal years, increased salary and benefit costs exceeded
the allowances
made for anticipated costs. In our view, this indicates
that the State
was applying the anticipated costs provision in a way
which, over time,
resulted in a fair reimbursement to the facilities.
Thus, we conclude that the State plan did not require adjustment of
the
rates for Custer and Redfield to reflect a comparison of the amount
of
the anticipated cost allowance to the amount of such costs
actually
incurred.
Conclusion
For the reasons stated above, we reverse the disallowance.
________________________________
Cecilia
Sparks Ford
________________________________ Donald
F.
Garrett
________________________________ Judith
A.
Ballard Presiding Board Member
1. In view of how we resolve these issues, we do not find it
necessary
to reach certain issues, raised by the State, concerning whether
HCFA
had contemporaneous knowledge about how the State was calculating
the
rates for Custer and Redfield and was now unfairly reversing an
earlier
position that the State was following its State plan.
2. HCFA did not argue here that use of certified bed capacity is
the
commonly accepted means of determining occupancy rate, or that the
State
otherwise should have been aware that it should use certified
bed
capacity.
3. The State's evidence showed that its policy
of
deinstitutionalization was based on federal initiatives and that
it
involved not only moving residents out of the large institutions
but,
also, improving services for those who remained.
4. In addition to the testimony which shows that empty beds were
moved
out of the State-operated facilities and the space used for
programming
purposes or staff offices, the record contains attachments to
the
facility certification forms, indicating that living units were
being
"phased out," as resident population was reduced. State's Ex.
46.
5. The State presented evidence that showed not only that it
was
unrealistic to expect the Department of Health, which
certified
facilities, to revise the certification forms each time a resident
was
moved out, but, also, that the State had another reason for not
wishing
to do this. Under the deinstitutionalization policy, the State
was
required to maintain the capability of returning former residents to
the
institution if they did not succeed in the less restrictive
environment.
Tr. I, pp. 198, 203-205.
6. The auditor found that the State had violated the State plan,
as
interpreted by the auditor, for only some of the years in question
for
Custer. The State presented the cost reports which showed that it
had
used actual patient days for the entire period in question;
the
confusion arose because for part of the period the actual patient
days
for Custer were greater than 95% of certified bed days. After
examining
the cost reports at the hearing, the auditor agreed that the State
had
used actual patient days. Tr. III, p. 440.
7. For a discussion of prospective rate-setting, see Illinois
Dept. of
Public Aid, DGAB No. 467 (1983). There, the Board concluded
that HCFA's
application of an actual cost ceiling on reimbursement for
two
State-owned facilities was inconsistent with the
prospective
rate-setting system set out in the State plan and with HCFA's
own
guidance on prospective