Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
Appellate Division
SUBJECT: California
Department DATE: August 5, 1993
of
Health Services Audit Control
No.
A-09-89-00165 Docket No. A-92-70 Decision
No.
1430
DECISION
The California Department of Health Services (California) appealed
a
disallowance by the Health Care Financing Administration (HCFA)
of
$109,880 in federal financial participation claimed under Title
XIX
(Medicaid) of the Social Security Act (Act). HCFA asserted that
the
disallowed amount represented the federal share of a
$223,180
overpayment to the Harold D. Chope Community Hospital (Chope) for
the
hospital's fiscal period ending June 30, 1984 (FPE 1984).
The payment to Chope was one of the hospital payments disallowed
as
overpayments and previously appealed to the Board in California Dept.
of
Health Services, DAB No. 1254 (1991). While upholding HCFA's ability
to
disallow the federal share of firmly established overpayments, even
if
California had not yet recovered these amounts from the providers,
the
Board also afforded California the opportunity to submit
documentation
to HCFA showing that the overpayment amounts had been reduced
by
settlements, recalculations, and administrative appeal decisions, or
had
been refunded to HCFA subsequent to the audit on which the
disallowance
was based. HCFA and California subsequently settled the
disallowance
for all the providers except Chope. Stipulation of
Parties, California
Exhibit (Ex.) C.
As discussed below, we find that there was no basis for
HCFA's
determination that California overpaid Chope for FPE 1984.
California
did not determine that Chope was overpaid, and HCFA did not
establish
that Chope received reimbursement in excess of that to which it
was
entitled under California's State Medicaid plan. Accordingly,
we
reverse the disallowance.
Applicable law
Title XIX of the Act authorizes federal grants to states to aid
in
financing state programs which provide medical assistance and
related
services to needy individuals, including inpatient hospital
services.
Any state that wishes to participate in the Medicaid program
must
develop and submit a plan that meets certain requirements set forth
in
the Act and regulations.
Realizing that many states might have difficulty financing a
Medicaid
program even if subsequently reimbursed by the federal
government,
Congress also established a funding mechanism by which the
Department of
Health and Human Services advances funds to a state, on a
quarterly
basis, equal to the federal share of the estimated cost of the
program.
A state submits a quarterly expenditure report to HCFA which serves
as
the basis for adjustments to future payments to reflect any
overpayment
or underpayment which was made to the state for that or any
prior
quarter. Section 1903(d) of the Act. Specifically, section
1903(d)(2)
of the Act provides that amounts paid to a state shall be reduced
to the
extent of any overpayment which HCFA determines was made to the
state
for any prior quarter and with respect to which adjustment has
not
already been made.
Under Section 1902(a)(13) of the Act, state plans must provide for
payment
of hospital services through the use of reimbursement rates
which the state
finds, and makes assurances satisfactory to HCFA, are
reasonable and adequate
to meet the costs which must be incurred by
efficiently and economically
operated facilities. This provision, known
as the "Boren Amendment,"
was intended to provide the states greater
flexibility in developing methods
of provider reimbursement. Missouri
Dept. of Social Services, DAB No.
1189 (1990); South Dakota Dept. of
Social Services, DAB No. 934 (1988).
Generally, states pay hospitals on
the basis of a reimbursement rate based on
the hospital's costs, such as
the costs per diem or per discharge.
States also have their own
procedures for determining whether a Medicaid
overpayment has been made
to a health provider. These procedures
generally include audits of the
providers to determine whether they have
properly reported costs used to
set such rates, whether patient days or
discharges have been correctly
claimed, and whether the reimbursement
otherwise meets the state plan
and federal requirements. See California
Dept. of Health Services, DAB
No. 1152 at 1-2 (1990).
In numerous cases involving excess or improper payments by states
to
Medicaid providers, this Board has held that, under section
1903(d)(2),
HCFA may require adjustment of the grant award for the federal
share of
firmly established provider overpayments, even if a state has not
yet
recovered these amounts from the providers. The Board reasoned
that
excess or improper payments are not "medical assistance" within
the
meaning of sections 1903(a)(1) and 1905(a) of the Act. See,
e.g.,
California Dept. of Health Services, DAB No. 1015 (1989);
California
Dept. of Health Services, DAB No. 977 (1988); California Dept. of
Health
Services, DAB No. 619 (1985); Massachusetts Dept. of Public Welfare,
DAB
No. 262 (1982). 1/ The Board has upheld disallowances based on
state
overpayment determinations where properly performed state audits
showed,
with a high degree of reliability, that providers had
received
reimbursement in excess of what was permitted under the state plan
and
federal requirements. California Dept. of Health Services, DAB No.
977
(1988); California Dept. of Health Services -- Accounts Receivable,
DAB
No. 334 (1982). A state is provided the opportunity to show that
HCFA
should not rely on its overpayment determination. If there is no
state
overpayment determination upon which HCFA may rely, then HCFA must
make
its own factually and legally supportable overpayment determination.
Background
This appeal concerns application of a 1989 change in California's
method
for calculating inpatient hospital reimbursement. California's
State
Medicaid plan provides several limits on reimbursement for
inpatient
hospital services, including a limit based on the hospital's
all
inclusive "rate per discharge" (RPD). The allowable RPD for
a
particular settlement year is determined by adjusting the RPD for
the
prior, or base, year through application of a "hospital cost
index"
(HCI) reflecting inflation and other cost factors. The HCI
is
calculated using cost information from the midpoints of the
settlement
year and the base year. The allowable RPD for the settlement
year is
then multiplied by the number of Medicaid discharges to compute
the
maximum inpatient reimbursement limit.
The State Medicaid plan does not specify how to determine what RPD the
HCI
is applied to, and in implementing its plan California employed two
different
methods. Under the method which was used to determine the
reimbursement
limit for Chope for FPE 1984, the HCI was applied to the
RPD at the beginning
of the settlement year (which is the same as the
RPD at the end of the base
year), to yield the RPD for the end of the
settlement year. These two
RPDs, one for the beginning and one for the
end of the settlement year, were
then averaged to determine the
allowable RPD for the settlement year.
The parties referred to this
process as the "averaging" method. Using
the averaging method then in
effect, California provided Chope with a notice
of settlement dated
April 20, 1989, which informed Chope that there had been
no overpayment
for FPE 1984. California Ex. D-2.
In a memorandum to the staff of California's Hospital
Reimbursement
Section dated August 7, 1989, California's Rate Development
Branch
announced a prospective change in the method for calculating
the
allowable RPD for the settlement year. The allowable RPD for
the
settlement year would be determined by applying the HCI to the RPD
from
the midpoint of the base year (also described as the average RPD for
the
base year). California Ex. D-1.
According to the August 7, 1989 memorandum, the method was changed so
that
the HCI and the RPD to which it was applied would reflect cost
information
from the same time period. Since the HCI was calculated
based on
information from the midpoint of the base year, California
determined that it
should be applied to the RPD from the midpoint of the
base year as well,
instead of to the RPD from the beginning of the
settlement year, as was the
case under the averaging method. The
memorandum explained that the time
period based on which the HCI was
computed could be skewed by as much as six
months from the time period
affected by the HCI, with the result that only
one-half of adjustments
to certain cost items reflected in the HCI were
actually realized in the
settlement year, the remainder being realized in the
subsequent year.
The memorandum noted that several providers had objected to
the
averaging method, and had advocated applying the HCI to the average
base
year RPD to get the average current year RPD. The memorandum
further
explained that providers would be affected differently by the
change,
with some benefitting from higher rates in all years, some with
lower
rates, and some with a combination of higher and lower rates than
under
the averaging method. Id.
To implement the change, the memorandum explained that while
all
settlement years would be recomputed using the midpoint method,
new
settlements would be issued only for active years, including
settlement
years that were the subject of pending administrative appeals
and
lawsuits. No new appeal rights would be granted for closed
years.
California then sent notices of the change in the calculation method
to
its providers which showed what each provider's reimbursement for
recent
closed years would have been under the new midpoint method. The
notices
further stated that liabilities assessed in previously closed
final
settlement years would not be amended, that no new appeal rights
would
be given, and that there would be neither recoupment nor refunds
of
additional funds for the closed years. The notices stated that
closed
years' settlements had been recalculated for "flow through" effects
and
to assure the providers that the settlements were comparable.
The
notice sent to Chope, dated August 24, 1989, showed that under
the
midpoint method, reimbursement for FPE 1984 would have been reduced
by
$223,180. California Ex. D-4.
Contentions
HCFA asserted that California's notice to Chope showing that it would
have
received less reimbursement for FPE 1984 under the midpoint method
was a
reliable determination that Chope had been overpaid for that year,
and that
California should be required to return the federal share of
the overpayment
to HCFA. HCFA asserted that California changed its
calculation method
because the averaging method was defective, and that
California's original
finding that Chope had not been overpaid was
therefore unreliable. HCFA
noted that the Board has repeatedly held
that HCFA may base a disallowance on
a state's own determination that a
provider was overpaid, so long as the
state's overpayment determination
had a high degree of reliability.
In support of its position, HCFA presented the declaration of an
official
in California's Rate Development Branch who stated that
California had
adopted the midpoint method because the averaging method
was technically
invalid, mathematically incorrect, and inconsistent with
all component parts
of the RPD calculations. HCFA Ex. 1. The official
also indicated
that California had first used the midpoint method but
had switched to the
averaging method in 1981, after which provider
hospitals began complaining
that the rate calculations failed to
accurately reflect their costs. He
stated that California had settled
provider appeals challenging the use of
the averaging method because
California would have lost such appeals.
HCFA argued that it was thus
entitled to rely on the midpoint calculations
that California had
concluded were the more accurate ones. 2/
California argued that its FPE 1984 settlement with Chope was closed
and
final prior to the time that it changed its method for
calculating
reimbursement limits. There was no determination that Chope
had been
overpaid, California asserted, and the notice to Chope showing what
it
would have received under the midpoint method was only an
informational
document issued to provide comparability for future
reimbursement
calculations. California argued that HCFA's determination
that Chope
was overpaid was a "misunderstanding" concerning the significance
of the
recalculation using the midpoint method for FPE 1984. California
Brief
at 2.
Analysis
This case raises the issues of whether California's calculation
showing
what Chope would have received for FPE 1984 under the midpoint
method
was correctly construed by HCFA as a determination that Chope had
been
overpaid, and, if not, whether HCFA has independently shown that
Chope
received an overpayment.
California's final cost settlement for FPE 1984 expressly stated
that
there was no "reimbursement reduction" (overpayment) due from Chope
for
that year. California Ex. D-2. Although California changed
its method
for calculating inpatient reimbursement limits and found that
Chope
would have received less reimbursement for FPE 1984 under the
midpoint
method, we find that this did not amount to a determination that
Chope
had been paid more for that year than what Chope was entitled to
under
California's State Medicaid plan. Rather, California's
determination of
Chope's FPE 1984 reimbursement under the midpoint method was
made in the
context of notifying providers of a prospective change in the
method for
calculating inpatient reimbursement limits and the effect of that
change
on their reimbursement rates. The notice letter showed,
for
"flow-through" purposes only, what hospitals would have received
for
closed cost years under the new method. California Ex. D-4.
As
California noted, the change in the calculation method was
applied
prospectively only to settlement years which were still open or
for
which there were pending administrative appeals and
lawsuits.
California also clearly stated that settlements for closed years
would
not be reopened, and there has been no evidence introduced showing
that
any such settlements were reopened. HCFA also provided no reason
why
this one provider should be singled out for retroactive application
of
the midpoint method to a closed settlement year.
Moreover, we find that this case is distinguishable from cases where
the
Board held that HCFA could use state records of
overpayment
determinations to disallow federal funding. The Board
concluded in
those cases that HCFA could recover the federal share of
firmly
established overpayments where states had determined, with a high
degree
of reliability, that the providers had been overpaid. Generally,
those
cases involved situations where state audits, performed according
to
accepted audit standards, had made findings based on review
of
providers' cost records, the states had attempted to recoup
the
overpayments based on those audits, and the overpayment findings
were
subject to appeal by providers. In this case, by contrast, there
was no
audit by California establishing that Chope received
reimbursement
beyond what it was entitled to under California's State
Medicaid plan
and policies. In addition, since the notice to Chope was
informational
only, the calculations were not subject to challenge (and,
consequently,
closer scrutiny) through an administrative appeals
process. HCFA did
not point to any other California determinations
which established that
Chope received reimbursement for costs unallowable as
charges to federal
funds.
We further conclude that HCFA did not independently establish that
Chope
had received an overpayment. HCFA did not demonstrate that
the
averaging method was inconsistent with the requirements of
California's
State Medicaid plan, or so defective as to be an
unreasonable
implementation of the plan and therefore render the closed
settlements
invalid.
In support of its position that Chope was overpaid, HCFA relied
primarily
upon the declaration of the California official addressing the
reasons behind
the change to the midpoint methodology. HCFA Ex. 1.
The
declaration is too vague about what was required by
California's
rate-setting methodology, and too conclusory in evaluating the
averaging
method, to support a determination that that method was
inconsistent
with the State plan. The declaration, which was signed on
August 8,
1991, prior to this appeal, explains the basis for
California's
settlement of disputed overpayments with six hospital
providers. For
some of these hospitals California had reduced disputed
overpayment
amounts based on recalculations using the midpoint method.
However, the
declaration does not address Chope and provides no basis for
concluding
that California's final cost settlement with Chope should be
reopened.
Additionally, the declaration of an analyst in California's
Hospital
Reimbursement Unit indicates only that the midpoint method
was
considered to give a "more accurate" comparison of a provider's
actual
versus allowable costs. California Ex. D. The fact that
the midpoint
method may have been more accurate than the averaging method
does not
establish that the averaging method was unreliable or inconsistent
with
California's State Medicaid plan.
We note that the record indicates that providers had complained that
the
averaging method resulted in rate limits that failed to
adequately
reflect their cost adjustments and were thus presumably too
low. Here,
by contrast, Chope's reimbursement under the averaging
method was higher
than what would have been available under the midpoint
method for the
one year in question. While the August 7, 1989
memorandum to
California's hospital reimbursement staff indicates that only
one-half
of certain cost adjustments were being realized in any one year
under
the averaging method, it also states that the remainder were realized
in
the following year. This gives rise to the inference that the
averaging
method could have resulted in reasonably accurate settlements over
the
course of several years. California Ex. D-1. Therefore, there
is no
reason to conclude that the averaging method was an
inherently
unreasonable implementation of California's State Medicaid
plan.
As noted above, the Boren Amendment gave states flexibility
in
establishing reimbursement methods for inpatient hospital
services.
California's official interpretation of its reimbursement method
was
that while the midpoint method was preferable, California's State
plan
did not require it to apply the method retrospectively to reopen
closed
settlements. HCFA has presented no persuasive reason why we
should not
defer to that interpretation. To find an overpayment on the
basis of
California's change to the midpoint method would have the effect
of
discouraging states from revising their calculation methods to
more
accurately implement their state plans.
Accordingly, we find that the mere fact that California changed
its
calculation method prospectively to one that may have been
more
mathematically precise did not amount to a determination that Chope
had
been overpaid for FPE 1984. To make a valid determination of
an
overpayment, HCFA must show that the payment was in excess of the
amount
determined payable under the State plan. Here, California did
not
determine that Chope had been overpaid for FPE 1984, and HCFA did
not
determine based on its own review that the rate calculation method
in
effect at that time violated the applicable State plan provision.
Conclusion
For the reasons discussed above, we reverse the disallowance.
Judith A. Ballard
M. Terry Johnson
Cecilia Sparks Ford Presiding Board Member
1. The Board's prior holdings on overpayments issues have
been upheld
in three decisions by United States Courts of Appeals:
Massachusetts v.
Secretary, 749 F.2d 89 (1st Cir. 1984), cert. denied, 472
U.S. 1017
(1985); Perales v. Heckler, 762 F.2d 226 (2d Cir. 1985); and
Missouri
Department of Social Services v. Bowen, 804 F.2d 1035 (8th Cir.
1986).
2. HCFA also argued that California's assertion that the FPE
1984
settlement was closed and that California was unable to recover
any
funds from Chope for that year did not affect California's obligation
to
refund the federal share of the overpayment to HCFA. HCFA noted
that
the Board has held that HCFA may recover overpayments from
states
regardless of whether the overpayments have been recovered from
the
providers. However, since we find that there was no
overpayment
determination upon which HCFA could properly rely, we need not
address
this