DEPARTMENTAL GRANT APPEALS BOARD
Department of Health and Human Services
SUBJECT: Pennsylvania Department of Public Welfare
Docket No. 85-237
Audit Control No. 03-50208
Decision No. 848
DATE: March 13, 1987
DECISION
The Pennsylvania Department of Public Welfare (State) appealed
a
determination by the Health Care Financing Administration (HCFA,
Agency)
disallowing a total of $1,906,614 in federal financial
participation
(FFP) claimed under the State's Medicaid Program.
1/ The Agency
disallowed $512,965 in FFP based on a determination that
the
expenditures were unallowable because the payments were made for
medical
services on behalf of Medicaid recipients who were eligible for
the
Medicare Part B program but were not enrolled in it. The
remaining
$1,393,649 in FFP was disallowed because the expenditures were
for
medical services on behalf of Medicaid recipients who were enrolled
in
the Medicare Part B program, but whose medical services eligible
for
Medicare were instead paid for by the Medicaid program.
We uphold in principle both parts of the disallowance; however,
while
Pennsylvania clearly was overpaid a substantial sum, it also is
clear
from the record that the disallowance amount as it now stands
is
inaccurate. We conclude that the State has the burden of defining
the
amount of an allowable claim, but, in the particular facts of this
case,
we also conclude that HCFA cannot arbitrarily deny the State
an
opportunity to establish the amount of the disallowance by
some
reasonably accurate means short of a full recreation of every detail
of
the claim. Thus, we uphold the disallowance subject to a
time-limited
opportunity for the State to present a reasonably-based
calculation of
the disallowance amount to HCFA. Below, we discuss some
of the
alternate approaches to such an accounting which have been
discussed
during proceedings in this case, among which HCFA surely should be
able
to find something which meets its needs. We therefore remand the
case
for disposition by the parties as outlined in our conclusions below.
This decision is based on the briefs and exhibits of the parties and
an
evidentiary hearing. 2/
Background.
A. The Buy-in Program.
Part B of Medicare, Title XVIII of the Social Security Act (Act),
is
entitled "Supplementary Medical Insurance Benefits for the Aged
and
Disabled," commonly referred to as SMIB. Part B of Medicare
generally
pays for physician and outpatient services for persons enrolled
in
Medicare, who may either pay their own premiums or have their
premiums
paid for them by their state.
To encourage enrollment in Medicare by the states of all persons
eligible,
the Act in section 1903(b)(1) prohibits the payment of FFP for
medical
services to any Medicaid recipient where the services could have
been
reimbursed under Part B, but were not because the Medicaid
recipient was not
enrolled in Medicare. The same provision appears in 42
CFR 431.625(c)(2).
The Act also provides in section 1843 for facilitating the enrollment
in
Medicare of those who are eligible for it as well as Medicaid. A
state
may under this provision enter into a contractual agreement with
the
Secretary of the Department of Health and Human Services (HHS) to
enroll
in Medicare Part B all its Medicaid beneficiaries who are also
eligible
for Medicare. This agreement is known as a "buy-in"
agreement.
Pennsylvania entered into such an agreement.
Once a person eligible for both Medicaid and Medicare is enrolled
in
Medicare, and the premiums are paid, the cost of any Part B
services
(less any coinsurance and deductibles) is paid for entirely by
Medicare,
rather than under the state's Medicaid program in which the state
shares
the cost. FFP is also available to the state under Title XIX in
the
costs of
the Medicare Part B premiums, and in the coinsurance and
deductibles.
3/
B. The Audit and the Issues.
In late 1983 auditors of the Office of Inspector General (OIG), HHS,
began
an on-site audit of Pennsylvania's Medicare Buy-In Program. The
purpose
of the audit was to determine if the State was excluding from
Medicaid FFP
costs of services provided to:
(1) Medicare enrollees; and
(2) Medicaid recipients eligible for but
not enrolled in
the
Buy-In
Program.
(State Appeal File, 6a)
The disallowance was based entirely on the audit report, which in turn
was
based on a computer tape prepared by the auditors identifying the
claims
erroneously paid by Medicaid; the tape was left with the State
and
subsequently destroyed. We consider below the following issues:
o Is the State entitled to FFP in Medicaid payments for
services to
Medicaid recipients eligible for but not
enrolled in Medicare?
o Is the State entitled to FFP in Medicaid payments for
services to
Medicaid recipients who were enrolled in
Medicare?
o Was the amount of the disallowance correctly determined
by the
audit report? If not, what is the
State's burden of proving its
entitlement to
FFP?
o What remedies are available for determining the quantum
of any part
of the disallowance which we cannot
resolve based on the record
before us?
I. Payments by Medicaid for Medicaid recipients eligible for but
not
enrolled in Medicare Part B are
unallowable expenditures for FFP.
The Agency disallowed $512,965 in FFP for payments to Medicaid
recipients
eligible for but not enrolled in Medicare Part B. The Agency
cited 42
CFR 431.625(c)(2) as the authority for this part of the
disallowance.
The State quoted the pertinent part of this regulation in
its opening brief,
stressing the enrollment requirement:
No FFP is available in State Medicaid
expenditures that could
have been paid
under Medicare Part B but were not BECAUSE
THE
PERSON WAS NOT ENROLLED IN PART
B.
(State brief, p. 6; emphasis by State)
The State cited this regulation as it did because at this point in
the
briefing it argued that the regulation was not applicable to this
case.
The State assumed at this time that the entire disallowance was
for
Medicaid payments for persons who were in fact enrolled in
Medicare.
The State failed to realize that the first part of the
disallowance
pertained to Medicaid payments for persons eligible for but not
enrolled
in Medicare Part B.
In subsequent submissions, the State maintained that failure to enroll
the
Medicaid recipients in Medicare Part B was due partly to State error
and
partly to Agency error. The State maintained that 95 percent of
the
non-enrollments were due to Agency error. The State argued that
the
auditors' workpapers indicated that the Agency was at fault for
most
non-enrollment since Pennsylvania is an "accrete" state, 4/ and
the
Agency has the responsibility for enrolling 95 percent of the
Medicare
population. (State's reply brief, p. 12) Further, the
State argued
that the Agency had shifted its argument from applying 42 CFR
431.625(c)
to arguing that any unenrolled individuals were eventually
enrolled, and
the State could have done a retroactive billing to recover
these funds.
The State maintained that it had done such a billing, and
therefore the
State was taking reasonable measures to recover the funds.
The Agency argued that, regardless of the reasons for the
non-enrollment,
the statute prohibits FFP in services provided to
individuals who were
eligible to be enrolled in the Part B program but
were not. (Agency's
supplemental brief, p. 4) The Agency maintained
that it accreted to the
buy-in program only SSI recipients who were
covered by the State Plan as
categorically needy, and that the State has
provided no evidence to show that
the Agency did not discharge its
obligation. Moreover, the Agency
argued that if any lag time occurred
in notifying the State of an accretion
action, this factor would not
prevent Medicare payment for Part B services
rendered to an individual
on or after the Part B entitlement date because
retroactive billings are
permitted. (Agency's supplemental brief, p.
4)
Section 431.625(c)(2) provides a very limited exception to the
general
policy that no FFP is available in State Medicaid expenditures
that
could have been paid for under Medicare Part B but were not because
the
person was not enrolled in Part B. The State has not submitted
any
evidence to show that it fits within the exception. 5/
Further, the
State did not present any evidence to show why, as the Agency
stated, it
could not make retroactive billings for recipients after
their
entitlement date, nor did the State show that the Agency did
not
discharge its obligation to the State under the accretion
program.
Therefore, based on the applicable regulation, we uphold this
portion of
the disallowance in principle. We find, however, as stated
below, that
the amount of the disallowance was overstated.
II. The State's payments on behalf of Medicaid recipients who
were
enrolled in Medicare Part B are
unallowable expenditures for FFP
under
Medicaid.
The second part of the disallowance concerns medical bills for persons
who
were in fact enrolled in Medicare but whose bills were paid
by
Medicaid. As distinguished from the first part of the
disallowance
discussed above, there is no statute or regulation which says in
so many
words that a state can never receive FFP in the Medicaid payments
in
such a situation. We
must examine the doctrine of third party liability (TPL) to
determine
whether any FFP is available for medical services of Medicaid
recipients
enrolled in Medicare. A "third party" is defined in the
Medicaid
regulations as "any individual, entity or program that is or may
be
liable to pay all or part of the medical cost . . ." of a
Medicaid
recipient. (42 CFR 433.136) Medicare is not mentioned by name
in this
definition, but the parties agree that Medicare is recognized as
a
liable "third party," if the recipient is enrolled in it.
Medicaid is the payor of last resort of the medical bills of
Medicaid
recipients; there is no disagreement on this general
principle. So
here, where we have dual eligibility, that is, Medicaid
recipients
enrolled in Medicare, their medical bills are to be paid first
by
Medicare, and only as a last resort by Medicaid.
The statutory provision on third party liability is section 1902(a)(25)
of
the Act, tracked and implemented in 42 CFR Part 433, Subpart D.
Under 42 CFR
433.138 the State must take "reasonable measures to
determine the legal
liability of third parties" to pay for medical
services to Medicaid
recipients. Section 433.139 gives the State certain
options for payment of
claims for medical services where there may be a
liable third party.
Section 433.140 provides that FFP is not available
if the State does not meet
the requirements of 433.138 and 433.139.
Section 433.139 provides:
(a) The agency has the following options for payment of claims:
(1) It may pay the amount remaining,
under the agency's payment
schedule, after the
amount of the third party's liability has been
established . . . .
(2) It may pay
the full amount allowed under the
agency's
payment
schedule for the claim and seek reimbursement from
any
liable third party
to the limit of legal liability. If the
agency chooses this option, it must seek reimbursement from
the
third party within 30 days after the end
of the month in which
payment is made.
(b) If, after a claim is paid, the agency learns of the
existence of
a liable third party, it must seek reimbursement from the
third party
within 30 days after the end of the month it learned of
the existence
of the liable third party.
(c) An agency must suspend or terminate an effort to
seek
reimbursement from a liable third party if it determines that
the
effort would not be cost effective because the amount it
reasonably
expects to recover will be less than the cost of recovery .
. . .
The State maintained from the beginning of this appeal that the
third
party regulations governed; it admitted that once a client was
enrolled
in Part B, then Medicare became a "third party resource."
(State brief,
p. 6, p. 20) The State at one point said that if it paid
for a Part B
service for a Medicare enrollee with Medicaid funds, it was
permitted to
retain FFP until reimbursement was received from Medicare,
citing 42 CFR
433.140. (Id., p. 20) However, it is apparent that
the State never
seriously argued for such an extreme position. Its
position is clearly
set out elsewhere in the same brief, citing the same
regulation, where
it said that Agency regulations permitted the State to
retain FFP in the
payments only "so long as the State pursued its TPL
responsibilities."
(Id., p. 7)
The State argued that it did pursue its TPL responsibilities since
it
complied with the "reasonable measures" requirements of 42 CFR
433.138
and 433.139. The State maintained that the term "reasonable
measures"
was defined by the Agency in the Medical Assistance
Manual,
SRS-AT-76-90, which provides:
Taking "reasonable measures" means
having a specified system for
identifying, investigation, and recovering from liable
third
parties.
The State argued that it took "reasonable measures" because it
diverted
over $167 million in Medicare claims by forcing providers to
claim
Medicare reimbursement before billing the Medicaid program.
Further,
the State maintained that it detected and rejected over $14 million
in
Medicaid claims which should have been billed to Medicare before
billing
Medicaid. The State argued that this should be compared with
the $2.4
million for enrolled clients which the auditors found that the
State
paid due to deficiencies in its system. (State's reply brief, p.
5) Due
to technical problems, the State argued that it did not know
that
Medicare was a resource in all cases; however, the State maintained
that
once it learned that Medicare was a resource, it sought
reimbursement
promptly. The State contended that it had begun taking
measures to bill
Medicare for claims paid by the
State long before the audit. The State noted that the Agency's
1983
report recognized the State's efforts. Moreover, the State argued
that
it had billed Blue Shield (the Part B carrier) in accordance with 42
CFR
433.139; however, it had not been successful in pursuing a billing
of
the intermediaries (primarily for outpatient bills) because the
Agency
would not permit such a procedure. The State argued that the
Agency's
suggestion that it recover from providers, rather than from
Medicare,
would not be cost-effective and would be burdensome on the
providers.
(State's reply brief, p. 7)
Further, the State argued that while the Agency relied on the
State's
difficulty in posting a Medicare resource indicator on its computer
file
as evidence that the State did not take reasonable measures to
identify
Medicare as a resource, the State's difficulties were attributable
to
SSA's practice of overlaying the state welfare identification
number
with other data which it sent the State. This made it impossible
for
the State to post the Medicare resource indicator uniformly to
its
files. Moreover, the State argued that it was not required to have
a
Medicare resource indicator at all. The State maintained that
the
indicator was a "fall back" procedure. Future models of the
Medicare
resource indicator procedure would be designed to edit the
providers'
claims against the Medicaid program records of recipients'
Medicare
coverages; however, the State's failing was only on an
optional
requirement.
In addition, the State argued that the Agency reviewed its Medicare
TPL
system at least twice during the audit period. The State contended
that
even though deficiencies were identified, the Agency found its
TPL
system to be reasonable and adequate. The State maintained that
the
Agency was aware of one of the two major deficiencies noted in
the
audit, the State's failure to edit for certain providers, from
a
previous review which covered two-thirds of the audit period.
The Agency argued that, to the extent that the Third Party
Liability
regulations applied, the State did not meet its TPL
obligations. The
Agency maintained that the State's internal record
keeping methods were
inadequate to identify Part B enrollees in order for
their claims for
Part B services to be billed to Medicare. Further, the
Agency
maintained that the State could have reasonably identified
such
enrollees from the monthly HCFA buy-in tapes, the monthly
Beneficiary
and Earnings Data Exchange System (BENDEX) updates, the monthly
state
data
exchange (SDX) files, 6/ and from the information generated by
the
State's own buy-in and Medicaid actions. (Agency's brief, p.
15)
We do not need to consider each one of the State's arguments in
detail
since the State admitted the difficulties with its TPL system in
its
comments on the draft report (Agency Exhibit 3) and the final
report
(Agency Exhibit 14). As mentioned later in this decision, the
State
admitted in its response to the draft report that Medicare should
have
paid a "high percentage" of the cases questioned by the auditors.
With
one exception, to be discussed, the substance of the comments to
the
draft and final audit reports is the same: the State made mistakes,
but
the State should not have to pay the questioned FFP; a
bookkeeping
transaction between Medicare and Medicaid should take care of all
the
financial problems.
The only possible blame not admitted by the State was the claimed
practice
of SSA of overlaying the state welfare identification number
with other
data. The Agency presented evidence at the hearing to show
why there
should not have been such a problem. (Tr., pp. 72-73) The
"double and
triple indicator" problems mentioned on pp. 113a and 114a of
the State's
appeal file were an aberration that would not affect the
ordinary case.
In any event, when the Secretary of the State Department of Public
Welfare
wrote the Administrator of HCFA on July 15, 1985, seeking to
resolve the
audit findings by a bookkeeping transaction, he questioned
only the amount of
the disallowance, and said flatly: "The findings of
this audit are not
in dispute." (Agency Exhibit 10, p. 2)
At the hearing there was testimony about the small average amount of
the
claims. The Board raised the question whether the State plan had
any
provision for suspending efforts to collect these small claims.
Under 42 CFR 433.139(c), the State plan must:
(1)(i) Specify the threshold
amount or other guide- line that
the
agency uses in determining whether to seek reimbursement
from
a liable third party; or (ii)
Describe the process by which the
agency determines
that seeking reimbursement would not be
cost
effective; and
(2) Specify a dollar amount or period of time
for which it will
accumulate billings
with respect to a particular liable
third
party, in making the decision
whether to seek recovery.
The Pennsylvania State plan provides, under Guidelines for
Seeking
Reimbursement from Liable Third Parties:
(a) Reimbursement from a liable
third party is sought for each
claim
unless:
* * *
(3)
the appropriate third party insurer cannot
be
determined.
(b) Pursuit of reimbursement for
claims of less than one hundred
dollars
($100.00) is terminated if no response is received
after
one client contact is
attempted.
(c) Third party claims are
forwarded to appropriate insurers on
a
continuous basis.
(Agency Exhibit 26)
The State's witness admitted that there is no dollar limit in the
State
plan which applies where no client contact must be made. Where
no
contact is necessary, "it is the state's responsibility to try and
seek
reimbursement for any claim." (Tr., p. 52) Counsel for the State
also
admitted that subsection "c" of the State plan applied, and "Medicare
is
a third party claim and our obligation is to forward it to
the
appropriate insurer." (Id., p. 53)
After reviewing the State plan, given the State's position that the
TPL
regulations apply to this portion of the disallowance, we find that
the
State had no provision in its plan for suspending efforts to
recover
Medicare Part B payments.
There is some doubt whether, even if the State plan had a provision
for
suspending efforts to recover Medicare Part B payments, the State
could
still recover FFP. The State in all these cases knew or should
have
known that Medicare was the third party payor. Under 42
CFR
433.139(a)(2), if the State used the so-called "pay and chase"
option,
to pay first and then seek reimbursement, the State would still have
to
bill Medicare within 30 days after the end of the month in which
payment
was made.
The fact that the State may have in the past diverted millions of
dollars
in Medicare claims by forcing providers to claim Medicare
reimbursement
before billing the Medicaid program does not affect this
disallowance.
The State is in effect making a de minimis defense: the
amount of
Medicaid funds claimed erroneously here is only a small
fraction of the
amount we might have claimed if we had not required
billing to Medicare
originally. This disallowance must be reviewed
based on the facts here,
not what the State did or did not do in other
cases.
Finally, in response to the State's argument, noted above, that the
Agency
was aware of its procedures during the disallowance period, the
reviews
conducted in the disallowance period were not tantamount to
audits.
Program reviews involve examination of program administration,
the emphasis
of which is on examination of case records and quality
control systems
monitoring recipient eligibility. (45 CFR 201.10)
Audits are reviews of
state agency operations conducted to ensure that
funds are being "properly
expended," (45 CFR 201.12(a)(2)) and to
determine "the allow- ability of
costs." (45 CFR 201.12(b)) Oklahoma
Department of Human Services,
Decision No. 809, November 18, 1986. The
Agency's reviews did not
preclude the Agency from doing an audit and
imposing a disallowance.
The Agency's basis for this part of the disallowance shifted during
the
briefing from relying only on the TPL regulations to including
an
overpayment argument that was first stated by the Agency in
its
supplemental brief. The Agency argued that the assertion by the
State
that it took reasonable measures to enforce third party
liability,
including Medicare liability, was inaccurate, but it also
overlooked the
fundamental fact:
that by claiming FFP in both the Part B
premium payments and in
the costs of
Part B covered ser- vices already paid for by
those
premiums, the State has received
duplicate FFP for the same
services . .
. . Irrespective of any efforts it may have made
to
collect from other liable third-
parties, the fact remains that
the State
has received double FFP for the disallowed claims
of
the Part B enrollees.
(Supplemental brief, pp. 1-2)
The Agency referred to two California decisions by the Board
upholding
disallowances for duplicate payments to providers rendered to
Medicaid
patients for whom the State had also made premium payments to
prepaid
health plans. (See California Department of Health Services,
Decision
No. 170, April 30, 1981, and Decision No. 389, February 28,
1983) The
Agency also pointed out that the Board has consistently held
that in
cases of over- payment the Agency does not have to wait until the
State
recovers from others in order to collect from the State. See,
e.g.,
Pennyslvania Department of Public Welfare, Decision No. 765, July
10,
1986, at p. 5. The State in its post hearing brief did not contest
the
Agency's argument that there was an overpayment:
The State agrees that it should not
receive double FFP through
payment of
both the premium and the claim itself . . . .
(p. 4)
Thus, whether under the third-party statute and regulations, or under
the
overpayment doctrine, we uphold in principle the disallowance for
FFP where
the Medicaid recipient was in fact enrolled in Medicare. We
do,
however, as indicated below, find the amount of the disallowance to
be
overstated.
III. The Audit Report overstated the amount of the disallowance.
We cannot uphold the disallowance in full since it is obvious that it
is
overstated. This is apparent both from the results of the Blue
Shield
billing and an examination of the methodology used by the
auditors.
1. The Blue Shield Billing.
In September 1984 the State billed Pennsylvania Blue Shield for all
the
claims involved in the disallowance, which it thought would be
payable
through this Medicare carrier. (Tr., p. 24) These were the
"actual
claims that were part of the audit disallowance," and were "right
from
the auditors' tape." (Tr., p. 36) Approximately 60 percent
of these
claims were paid. 7/ (Tr., p. 25) The State did a
review of 1,000 of
the rejected cases and found that only approximately two
percent of
the claims were even possibly payable by Medicare. (Tr., p. 26)
The
significance of this fact, if correct, is that there was obviously
a
major flaw in the audit. If the auditors were correct, that
these
claims were all for services for Medicaid recipients enrolled in
or
eligible for Medicare, then when the State billed Blue Shield for
them,
a very substantial number of them should have been paid by Blue
Shield
with Medicare funds, 8/ as discussed below in considering the sample
run
by Blue Shield.
We need not rely on the State's review to show that a large number of
the
claims which the auditors said Medicare should have paid were not
so
payable. Blue Shield, which is not a party here, also did a study
of
5,000 claims selected from the claims it received from the State.
9/
The Blue Shield review showed that only 25.1% of the claims were paid
as
billed. More significant were the reasons for rejection of many of
the
claims. Duplicate claims and charges amounted to 13.1%. The
only other
reasons for rejection which amounted to any substantial
percentage
pertained directly to Medicare eligibility. Those claims
which were
rejected because they were for services before the Medicare
entitlement
dates of the recipients amounted to 28.7%. The questioning
of the Blue
Shield witness, as to the difference between "entitlement"
and
"enrollment" was not conclusive. (Tr., pp. 104-110) In the
other major
category, services were received by patients who were not
entitled to
Part B Medicare in 19% of the claims submitted; these patients
had never
been enrolled in Medicare. (Agency Exhibit 27, Tr., pp.
91-92)
Therefore, 47.7% of the sample claims were not paid because
the
recipient was not enrolled in Medicare at the time the services
were
rendered. It appears that the claims submitted to Blue Shield
did
include those eligible for enrollment, as well as those
actually
enrolled. (See n. 13) Blue Shield had no way of knowing
if the
recipient was eligible but not enrolled. (Tr., p. 90) Some
of the
47.7% of the sample claims must have been for those not enrolled
but
eligible. Presumably the much larger proportion of the
claims
disallowed were for those enrolled, since the audit report
identified
345,609 claims paid by Medicaid for those enrolled in Medicare,
and only
111,345 claims paid by Medicaid for those eligible but not
enrolled.
(Audit report, State appeal file, p. 10a)
Let us assume that roughly one-fourth of the claims submitted to
the
carrier had been disallowed because they were for those eligible
for
Medicare but not enrolled at the time of service. Therefore, if
the
auditors were correct, then Blue Shield should have
rejected
approximately 25% of the claims submitted because of non-enrollment
at
the time of service. There is then something definitely wrong with
the
audit report where Blue Shield rejected 47.7% of the sample
claims
submitted because of non-enrollment at the time of service.
We consider next where the auditors apparently went wrong in
the
methodology they employed.
2. The Audit Methodology.
The testimony at the hearing established that the auditors included in
the
second part of the disallowance some payments (by Medicaid) for
Medicaid
recipients who were actually not enrolled in Medicare when the
services were
rendered.
If we did not have an enrollment date .
. . we would have
determined a date in
which the person was eligible which may
have
been earlier than when the person
was actually enrolled.
(Tr., p. 159)
This misclassification could occur for one of several reasons. The
first
reason was the failure of the auditors to allow for the availability
of
FFP for retroactive coverage of medical services for up to three
months
prior to a Medicaid application, under 42 CFR 431.625(c)(2).
(See n. 5
above) In fact, the auditor who testified at the hearing
admitted that
he did not know of this part of the regulation during the
audit. (Tr.,
p. 155)
The second reason was the failure of the auditors to consider a
possible
gap in the eligibility for the disabled. A person must be
continuously
disabled for at least 24 months to be eligible for Medicare
on
disability. 10/ The auditors took the beginning date of
disability,
and added 24 months to that date as the "Medicare begin
date." If there
was any interruption in the person's disability, so
that it was not
continuous, the audit did not take this into account.
(Tr., pp.
126-129, p. 155) The Agency admitted that the disability
cases were the
bulk of the individuals identified as eligible for but not
enrolled in
Medicare, or the first part of the disallowance. (State
appeal file, p.
30a)
A principal reason advanced by the State for the deficiencies in the
audit
pertaining to both parts of the disallowance was that the auditors
used the
State's procedure codes without fully understanding their
purpose. The
auditors used the so-called Y indicator, placed after
certain State procedure
codes, to determine which Medicare Part B
services were paid by Medicaid.
(Tr., pp. 130-131) The State argued
that the Y code was designed to
pend claims which might be Medicare
payable, and needed to be confirmed
manually, while the auditors assumed
that all claims isolated by this code
were in fact Medicare payable.
(State's brief, pp. 11-12) The use of
the code was to determine which
services were eligible for Medicaid payment;
this would apply to both
parts of the disallowance.
3. The Agency's Admissions.
We need not rely only on the above analyses to conclude that there
were
errors in the auditors' methodology, because the Agency admitted it.
On November 12, 1985, less than a month after the disallowance letter,
the
Acting Regional Administrator of HCFA wrote the State, after
reviewing a
sample of 100 cases where the State had
received "a non-enrollment indicator from Blue Shield." The
Agency's
conclusion was as follows:
Based on our sample review, we
determined that the [audit] report
and
the associated computer tapes are somewhat
misleading
regarding the enrollment
status of particular recipients.
(Agency Exhibit 16) 11/
The letter then went on to try to explain why the "somewhat
misleading"
audit report and tapes should make no difference as far as
the
disallowance was concerned.
However, our review also showed that in
the vast majority of
cases the
recipients, though not enrolled in Medicare Part
B,
were eligible for that program.
Thus, these recipients were
misclassified in the report (i.e., these individuals should
have
been reported as eligible but not
enrolled), but the total amount
of the
overpayment is unaffected . . . .
(Id.)
In summary, the Agency admitted that the report and the tape had
errors,
but claimed it made no difference. It was true that recipients
were
"misclassified" by being reported as enrolled in Medicare, when in
fact
they were not. But, contended the Agency, since they were eligible
to
be enrolled in Medicare, it did not matter. In effect, the Agency
was
saying that the Medicaid payments for persons who were not enrolled
in
Medicare but were eligible for it belonged in the first part of
the
disallowance rather than the second.
There are two things wrong with this approach. The first is that
the
Agency does not even make a pretense that all those
erroneously
identified in the audit as enrolled in Medicare were in fact
eligible
for it. The letter above said that this was true "in the vast
majority
of cases." We cannot determine what percentage of the cases
constitutes
a "vast majority," which can theoretically be shifted from one
part of
the disallowance to the other; and, conversely, we cannot determine
what
percentage of
cases erroneously classified as enrolled in Medicare was not in the
"vast
majority," so that the disallowance as to them is simply wrong.
The other fault with the Agency approach is that the Agency cannot
freely
transfer items from one part of a disallowance to another and say
it does not
matter. We of course have frequently permitted a federal
agency to
change the grounds it relied on in a disallowance, or to
change the amount
involved, if the appel- lant has had the opportunity
to rebut the
disallowance as changed. What we have here is something
more than
that. Here the Agency, recognizing that an audit overstates
the
unallowable claims, seeks to move an unidentified portion of one
part of the
disallowance to another. 12/
4. The Burden of Proving Entitlement to FFP.
The State admitted in its opening brief that "ultimately the State has
the
burden of proving the veracity of its claim for federal funding."
(State's
brief, p. 17) The question in this respect is how far the
State must go
to meet its burden of proving its claim. According to the
State there
are more than 400,000 claim items at issue. (State's brief,
p. 13)
The Agency took the position that the State always has the burden
of
documenting its claim for FFP, citing prior Board decisions,
including
New York State Department of Social Services, Decision No. 673 (p.
6),
July 19, 1985, and Decision No. 520 (p. 15), February 29, 1984.
The
Agency stated it was willing to make a downward adjustment in
the
disallowance only where the State could document each individual
claim
as allowable.
The Agency is absolutely right about the principle articulated in
the
Board's decisions. However, this begs the question in this
case: how
must the State go about meeting its general burden of
documentation?
The problem we are faced with is that we are dealing with a
situation
which differs from the normal disallowance of FFP based on lack
of
documentation. In a typical Medicaid disallowance a State has
claimed
FFP for medical services which were either not covered under
Medicaid,
or were rendered to persons not Medicaid eligible. Here the
services
were presumably covered under Medicaid, and were rendered to
persons
who, by definition, were eligible for Medicaid. The problem was
that
the persons receiving the services were also either enrolled in
Medicare
or eligible to be enrolled.
The burden of the State here is not strictly that of
original
documentation of its claim; it rather accompanies a burden of
refuting
the basis for the disallowance of the claims. The Agency does
not
dispute that, in the absence of Medicare, the claims were eligible
for
FFP in Medicaid. The question is whether the State has to show, as
to
each and every claim included in the disallowance, that the recipient
of
services was neither enrolled in nor eligible for Medicare.
The State argued that the Agency could not question State
claims
arbitrarily, "and thereby impose an unreasonable burden upon the
State.
. . . The Agency must have a reasonable basis for its decision
to
disallow federal funds." (State brief, p. 17) The State cited
Arkansas
Department of Human Services, Decision No. 671, July 10, 1985,
in
support of this position.
The Agency clearly had a reasonable basis for its disallowance here.
The
State admitted on the first page of its opening brief that "some
Medicare
payable claims were undoubtedly paid," i.e., by Medicaid. In
the
State's response to the draft audit report it admitted "that
Medicare should
have paid a high percentage of these cases." (State's
appeal file, p.
14a; see also Tr., p. 163)
On the other hand, we are dealing with a disallowance based on an
audit
report which we have found is incorrect in amount. In addition,
the
audit tape, on which the disallowance is based, was destroyed.
We
consider next whether either party should alone bear the loss of
the
tape, and if not, how we should treat the joint responsibility.
The Agency provided the State with a computer tape listing the
unallowable
claims, which had a one-year expiration date placed on it by
the OIG
auditors; this tape covered both parts of the
disallowance. 13/ The State contended that after it received the
tape,
it attempted to secure Medicare reimbursement. The State alleged
that
the Agency allowed it to bill only those claims which could be
processed
through a Medicare Part B carrier, and the Agency refused to allow
the
State to bill those claims which had to be processed through a Part
A
intermediary. The State argued that, following its partial billing
of
claims to the carrier, it returned the computer tape to the State
tape
library to await the Agency's decision on a billing of
the
intermediaries. As a result of the expiration date, however, the
tape
was routinely destroyed by the State library at the end of the
year.
The State maintained that an untranscribed telephone conference
between
the parties, on February 14, 1986, confirmed that the
workpapers
previously given the State did not disclose the methodology used
in
creating the tape. However, the Agency did provide the State with
the
computer programs used to generate the unallowable claims. The
Agency
provided the State with twenty-three programs which contained
over
10,200 lines of computer code. Although the Agency witness did
testify
at the hearing as to the methodology used to construct the tape
(Tr.,
pp. 118-144), it was impractical to attempt to reconstruct the
tape.
The Agency argued that the auditors left their final computer tape
with
the State to enable it to identify the disallowed claims and to be
able
to bill Medicare. 14/ Further, the Agency argued that the
State
admitted that it had reformatted the tape and sent it to Blue Shield
for
claims processing. (Agency brief, p. 17) Therefore, aside
from the
computer tape, the State had a list identifying the disallowed
claims.
Moreover, the Agency argued that since the State had the tape for
at
least a year, not only did it extrapolate its own billing tape but
it
also had time to review the audit tape entirely or on a sampling
basis
to verify the disallowed claims.
Although the Agency did provide the State with a copy of the
tape
containing the disallowed claims, it was foreseeable that the
dispute
would take longer than one year to resolve. 15/ But it
was the Agency
auditor who put the expiration date on the computer tape.
The State has admitted that a high percentage of claims should have
been
paid by Medicare. In New York State Department of Social
Services,
Decision No. 284, April 29, 1982, where there were deficiencies in
the
audit, we said:
The remedy for this failure by the
Agency is not a reversal of
the
disallowance since the audit re- port did clearly
identify
unallowable costs and the
question of recoveries goes merely to
the quantum of unallow- able costs which continue to be
charged
to federal funds.
(p. 12)
The situation here is similar, since the audit report did clearly
identify
unallowable costs, so the question of recovery goes merely to
the "quantum of
unallowable costs." A reversal is clearly unjustified,
but we are left
with the question of how to resolve the disputed costs.
The Agency's
suggested disposi- tion is also unreasonable. This is to
require the
State to rebut the disallowance on the basis of an
individualized review of
some 400,000 claims, where the original tape
identifying the claims was
destroyed.
It is clear from the record that substantial errors in
Pennsylvania's
claim are present, and they cannot be ignored. While the
questions
surrounding the loss of the tape by no means relieve the State of
its
fundamental obligation to prove and document its claim, nonetheless
we
find the Federal contribution to the confusion sufficient at least
to
challenge the notion that the State may only meet its burden by the
very
resource-intensive reexamination of all several hundred thousand
claims.
We find that the Federal participation in the confusion, while
not
overwhelming, was sufficient to establish for HCFA the reasonable,
and
minimal, burden of allowing Pennsylvania an opportunity to meet
its
documentation burden, if it can or will, by some reasonable
alternative
methodology; only if Pennsylvania cannot or will not do so should
the
present disallowance amount be upheld as, in effect,
substantively
unchallenged.
IV. The Choice of Remedies.
1. Overview.
We have determined above that while the disallowance was legally
correct
in principle, the amount was obviously overstated. We found
that the
audit report on which the disallowance was based was not accurate,
so
that we will not accept its computation of the amount of
the
disallowance. However, we also found that the State bears a burden
of
proof, modified, in the facts of this case, by a HCFA obligation to
be
reasonable in what it will accept.
We consider below several possible remedies or solutions to the
quantum
problem. Clearly it is not the function of the Board, nor are
we in a
position, to determine the correct quantum of the disallowance
here. It
will of course require the cooperation of the parties, but
there is
Board precedent for such a course of action. In Ohio
Department of
Public Welfare, Decision No. 226, October 30, 1981, in a
situation where
a disallowance was obvi- ously called for, but the quantum
had not been
properly estab- lished, the Board spoke of the shared
responsibilities
of the parties which "track the cooperative nature of
Federal-state
relations in assistance programs generally and of the Medicaid
program
in particular." (p. 3) 2. We uphold the disallowance of
any claims
where the State has been
reimbursed by Medicare.
Perhaps the simplest problem in this complex case is how to deal with
the
$658,514 the State has collected from its billing of Blue Shield,
the
Medicare Part B carrier. (Tr., p. 25) We agree entirely with
the
Agency's statement in its reply brief:
The State . . . has, in fact, collected
a substan- tial sum from
the Medicare
Part B carrier, thereby requiring the State
to
return the FFP erroneously claimed in
connection with these Part
B claims .
. . Certainly, to the extent the State has
made
Medicare recovery, the FFP
disallowance must be sustained . . . .
(p. 17)
The State in its reply brief casually admits its liability in
a
footnote.
The Agency asks the State to account for
the funds recovered in
the Blue Shield
billing. The State agrees that the Agency
is
entitled to repayment of its share of
monies recovered from the
Medicare
carrier.
(p. 8, n. 1)
There is a disclaimer added: "However the precise status of
that
repayment is not known by counsel." (Id.) Whether counsel
should have
known, or at least tried to find out, is immaterial here.
The State
claimed FFP in payments which should have been paid by
Medicare. When
Medicare did in fact pay, the State should adjust for
the FFP it had
claimed in those payments. Some adjustments may have to
be made for
coinsurance and deductibles, but generally the State should
adjust based
on the percentage of FFP it claimed in these payments
originally. This
should not be difficult to establish, and it is
clearly the State's
burden to do so. Therefore, Pennsylvania shall have
30 days from
receipt of this decision (or such longer time as HCFA itself
chooses to
give) to submit evidence justifying a disallowance amount less
than the
federal share of the amount collected; if the State does not do so,
the
disallowance is upheld in this amount.
3. Direct Interchange Billing.
The State for a period of almost a year before the disallowance urged
the
Agency to settle the whole matter by bookkeeping entries
transferring funds
to Medicaid from Medicare. (See, e.g., Agency
Exhibit 1, p. 2)
The State's reasoning behind this
was simple. The State used Medicaid funds for paying for
medical
services which should have been paid for with Medicare funds.
Ergo,
Medicare should reimburse Medicaid for the amount which
Medicaid
overpaid. Leaving aside the difficult question of how to
determine how
much was in fact payable by Medicare, the Agency argued that
the problem
could not be solved by bookkeeping. The Agency claimed that
statutory
and regulatory provisions prevented such a solution. 16/
4. Billing of Intermediaries.
The majority of the disallowed claims which were on the auditors'
tape
have never been submitted to Medicare for payment. Only 16% of
these
claims were submitted for payment to Blue Shield, the Medicare Part
B
carrier, according to the State. (Agency Exhibit 1) The
remaining 84%,
consisting of outpatient services by hospitals, and services
by home
health agencies and rural health clinics, were ordinarily paid for
by
Medicare through so-called Part A intermediaries. These
intermediaries
reimbursed the providers of services to recipients enrolled in
Medicare.
There was no provision for a state billing the intermediaries
directly
for these services. The Agency maintained that direct billing
of the
intermediaries was prohibited by the federal Assignment of Claims
Act,
31 U.S.C. 203. (Tr., p. 43)
In its post-hearing briefing the Agency admitted that it had waived
the
requirements to permit carriers to make direct Medicare Part B
payments
to State agencies. However, the Agency contended that a waiver
to
permit the State to bill the intermediaries directly, and then have
them
pay the State directly, "would require costly systems modifications
and
disruptions of intermediaries' operations." (HCFA Post- Hearing
Brief,
p. 11)
In the absence of billing intermediaries, the State can get payment
from
Medicare for these remaining services only by a lengthy and
costly
individual procedure. In each case where the State had paid a
provider
with Medicaid funds for a service to a person enrolled in Medicare,
the
State would have to tell the provider to bill the intermediary.
The
intermediary would then pay the provider with Medicare funds
(assuming
the claim was otherwise properly payable), and notify the
State. The
provider has now been paid twice, once with Medicaid funds
and once with
Medicare funds, so the State must attempt to recover the
duplicate
payment. The State can bill the provider, or if the State is
still
doing business with the provider, it may instead take an offset
or
credit on other monies the State owes the provider. Obviously
neither
the provider nor the State is going to be happy with this
procedure
where it involves literally hundreds of thousands of claims.
17/
Billing the providers is not only a complex matter but also a
costly
one. The State produced testimony at the hearing that the
average value
of a Part B claim for the audit period (after taking into
account
coinsurance and possible deductibles), was $14.27. It would
cost the
State approximately $4.53 to return the claim to the provider; it
would
cost the provider $15 to $25 to bill the intermediary; and it would
cost
the Medicare intermediary $2.34 to process the claim. It
would
therefore cost a total of $21.79 to $31.79 to bill a claim
averaging
$14.27. (Tr., p. 31) The Agency did not contradict any
of this
testimony. 18/
5. The Pilot Project.
The parties have from time to time discussed a "pilot project" which
would
allow the State to bill the intermediaries directly.
The procedure is described in Agency's Exhibit 33. (See also Tr.,
pp.
46-48)
It is not clear why the pilot program has not been pursued. The
Agency
stated in its supplemental brief that, contrary to repre- sentations
by
the State, "a pilot program is being instituted in Pennsylvania
for
direct billing of the intermediary." (p. 3, n. 3) Although
the Agency
brief stated that the use of the pilot program for retroactive
billing
of the claims at issue had been approved, the Agency counsel
later
withdrew this statement.
The Board inquired again about the pilot project in its list of
questions
for the parties to consider in post-hearing briefing. (Letter
to counsel
dated September 3, 1986) The Agency replied that the pilot
program "is
an available option." However, continued the Agency,
"counsel for
appellant has indicated that it is impractical to use the
pilot program for
recovery of the claims at issue." (p. 13)
We find nothing in the record to support such a conclusion as to
the
State's position. In its post-hearing brief the State's reply to
the
Board's inquiry was that "Both parties are still negotiating on
the
pilot project." 19/
6. Statistical Sampling.
In our guidance to the parties on post-hearing briefing, we asked
whether
it was possible "to determine the amount of liability by a
statistically
valid sample of a limited number of claims." (Letter,
September 3, 1986,
question 6) The State stated simply that it did not
know. The
Agency did give the question serious consideration. Its
reply began
with the word "No," but its following explanation indicated
that such
sampling was complicated but possible. (Post-hearing brief,
p. 12)
We agree with the Agency that the sample would also have to take
into
account that in Blue Shield's sample of 5,000 claims over 13%
had
already been paid by Medicare, being duplicate claims. (Id., pp.
12-13)
The Board has in the past repeatedly found the use of valid
statistical
sampling to be a reasonable way to calculate the amount of
a
disallowance in cases dealing with a large number of individual
claims.
See University of California--General Purpose Equipment, Decision
No.
118, September 30, 1980; California Department of Health Services -
San
Joaquin Foundation, Decision No. 182, May 29, 1981; Nebraska
Department
of Public Welfare, Decision No. 422, April 29, 1983;
California
Department of Social Services, Decision No. 816, December 5,
1986. In
fact, as contrasted to reviewing hundreds of thousands of
items, an
intensive review of a sample actually can produce a more
precise
extrapolated result. We urge the parties to explore the use
of
statistical sampling to settle at least part of the difficult
quantum
problems in this case.
The State has the initial burden of presenting some reasonable method
for
determining a fair amount, since it obviously owes the Agency a
substantial
sum. While we said we will not hold the State to a
claim-by-claim proof
of allowability (unless the State does not act
reasonably and timely), we
will require it to devise some reasonable
method of computing what is
due. The Agency, on the other hand, should
not arbitrarily object to
any method the State suggests. For reasons
discussed above, it has the
burden of fairly considering the State's
proposed methodology.
Conclusions:
We have upheld above in principle the disallowance for FFP claimed
for
those either enrolled in Medicare or eligible for enrollment. For
these
reasons, we have determined that Pennsylvania shall have 30 days
(or
such longer time as HCFA allows) to submit to HCFA a detailed
written
proposal containing a methodology for establishing a disallowance
amount
for Medicare enrollees and eligibles. The methodology should
contain
reason- ably expeditious timelines to avoid unnecessary delay.
HCFA
should promptly review the proposal and accept or reject it, bearing
in
mind that the proposal should be rejected only if HCFA determines
that
it would not result in a reasonably accurate approximation of the
proper
disallowance amount. 20/ If HCFA rejects the proposal,
Pennsylvania
may return to the Board within 30 days after receiving the
rejection for
our (expedited) review
of that action. If Pennsylvania does not timely submit the
required
proposal, or if we uphold a HCFA determination that Pennsylvania had
not
proposed an adequate methodology, then the present disallowance
amount
is upheld in full.
We also uphold the disallowance as to any amounts collected by the
State
in its Blue Shield billing, subject to adjustment as described on p.
22
above.
________________________________
Cecilia
Sparks Ford
________________________________
Norval
D. (John) Settle
________________________________
Alexander G. Teitz Presiding
Board
Member
1. The disallowance stated that it covered the
period from July 1,
1982 through December 31, 1983. However, at the
hearing in this case
there was Agency testimony that the claims reviewed ran
through January
15, 1984. (Transcript of Hearing (Tr.) pp.
136-37) There is no
indication or claim that there was any prejudice to
the State merely
because the audit and the disallowance incorrectly refer to
an ending
date of December 31, 1983. The State had a full opportunity
to contest
the entire disallowance and it did so. We therefore review
the
disallowance for the entire period through January 15, 1984.
2. This case was delayed because substantial
extensions of time for
briefing were needed by both parties; the reason
generally given for the
requests was that the parties were engaging in
settlement discussions.
The parties' schedules also delayed setting the
hearing date.
3. Medicare ordinarily pays for 80% of covered Part B
services; the
deductible is $75 per recipient per year.
4. An "accrete" state is one where the Social
Security
Administration (SSA), by agreement with the state, makes the
disability
determination of whether individuals are eligible for Medicaid
based
upon the individual's entitlement to SSI benefits. (Title XVI of
the
Act, section 1634)
5. There is an exception in the regulation to provide
FFP in
Medicaid for services rendered to persons eligible for Part B
Medicare
but unenrolled, if they were also eligible for Medicaid during a
maximum
retroactive period of three months prior to the filing of a
Medicaid
application. However, the Agency stated that it was
unreasonable to
assume that of the 7,855 Medicaid recipients eligible for but
not
enrolled in Part B, all such individuals were new Medicaid
applicants
during the period audited, or that all such individuals
actually
received Part B services within three months of a Medicaid
application
during the period audited. (Agency brief, p. 14) But,
it is also
unreasonable to assume that none of the 7,855 recipients could
fall in
this category. On remand the State should have the opportunity
to show
which claims, if any, were entitled to FFP under this
exception. The
burden of making such a showing falls on the State, as
in all cases of
entitlement to FFP, as discussed below.
6. The BENDEX is a monthly updated system designed to
allow the
state to, in part, verify an individual's receipt of Social
Security
benefits under Title II of the Act, and verify an individual's
earnings.
The SDX file is a system updated monthly for use by the State to
enable
it to examine an individual's record for evidence of cash
assistance
eligibility. (Agency's response brief, p. 7)
7. The State argued in its brief that 60 percent of
the claims were
rejected; however, at the hearing, one of the State's
witnesses
testified that 40 percent of the claims were rejected, and the
State
agreed that the 40 percent figure should be used. (Tr., p. 203)
The
Blue Cross billing paid only 25.1% of claims filed, according to
a
sample; the difference is due to counting claim lines in one
billing,
and separate claims in the other. See also n. 9.
8. The time factor in rebilling was not significant,
since the State
obtained a waiver of timely claiming requirements for this
particular
Blue Shield billing. Minor discrepancies could account for a
small
percentage of rejections. These could include, for example,
duplicate
payments to the same or other providers, non-covered services, and
cases
where the carrier could not obtain proper information to process
the
claims. (See Agency Exhibit 27)
9. Blue Shield said it received 58,219 claims from
the State; the
5,000 random selection was approximately 10%. The State
said it
submitted approximately 115,000 claims. The discrepancy is
caused by
the difference in counting claims in Medicaid and Medicare; the
State
counts each Medicaid claim-line item as a separate claim. (Tr.,
p. 98)
There was no attempt made in either the State sample of 1,000 claims,
or
the Blue Shield sample of 5,000, to have a random selection which
could
be statistically valid.
10. Persons entitled to disability benefits under
Title II (Social
Security) must have their disability continue for 24 months
without
interruption in order for them to be also eligible for
Medicare. In a
State sample of 100 cases, rejected by Blue Shield for
non-enrollment,
the Agency found that nine disabled individuals were
ineligible for
Medicare during the period audited because their Title II
eligibility
had ceased during the 24 months. (HCFA reply brief, p.
22)
11. The same letter is at page 25a of the State's
Appeal File. It
contains underlining which was apparently not in the
original.
12. Since we find that the audit was flawed, we do
not need to
consider the State's argument that the auditors failed to comply
with
audit standards established by the Comptroller General. There is
in
fact nothing in the record to indicate that the auditors failed
to
follow standards specifically binding on them.
13. This case was complicated even further by the
State's apparent
failure to recognize that the auditor's tape covered both
parts of the
disallowance, i.e., Medicaid payments for those enrolled in
Medicare,
and those not enrolled but eligible. In its original brief the
State
said that "the audit tape erroneously generated a listing of
individuals
designated as enrolled in Medicare who were not." (State
brief, p. 14)
The Agency pointed out in its response that the claims
identified by the
auditors which were submitted to the carrier included
claims from both
groups. (HCFA reply brief, n. 9) In its reply,
the State said it would
"conduct an analysis of this issue." (State
reply brief, n. 3) Nothing
further was done. On the record before
us we must conclude that the
tape covered both classes of disallowed
claims. The audit identified
111,345 Medicaid claims for those eligible
but not enrolled, and 345,609
claims for those enrolled in Medicare.
(State appeal file, p. 10a) One
of the auditors testified that the tape
was not printed to hard copy
"because 485,000 claims was going to be an awful
lot of paper." (Tr.,
p. 145)
14. At the hearing, the Agency's witness testified
that it had the
State make a copy of the tape for the Agency. When the
Agency attempted
to have the tape converted to its computer system, the tape
was
destroyed, thus leaving with the State the only copy, which had
a
one-year expiration date. (Tr., pp. 178-179)
15. A state must retain relevant records where an
audit is in
progress until the audit is resolved. (45 CFR 74.21)
There is
apparently no similar requirement for federal agencies, and the
auditors
themselves regarded the tape as property of the federal
government.
(Agency supplemental brief, p. 6) However, normal
procedures of OIG
"require review of an audit tape at the time of the
expiration date to
ascertain the audit status and to decide whether a tape
should be
maintained or not." (Id.) Obviously this was not
done.
16. Such a bookkeeping exchange, even if otherwise
feasible, would
not take into account cases where the State's claim for
Medicare
reimbursement was invalid for a reason other than a failure to
detect
the Medicare enrollment of the recipient. For example, to have
Medicare
reimburse Medicaid in all cases where the disallowance found
payment
should have been made by Medicare would not take into account the
number
of cases where Medicare payment had already been made. For
instance,
the sample conducted by Blue Shield found that 13.1% of the claims
were
for either a duplicate claim or a duplicate charge. (Ex. 27, p. 2;
Tr.,
p. 87)
17. The Agency's post hearing brief, in answer to the
Board's
questions, stated that the auditors had identified 739,299 claims
for
Part B enrollees, and only 58,219 claims had ever been billed
to
Medicare, through Blue Shield, the Part B carrier. (p. 12)
18. In cross-examination the Agency did bring out
that the State
would not have to pay the $15 to $25 estimated cost to the
provider, nor
the cost to the intermediary. The State replied that
while it would not
pay directly, it would pay for it in "increased bad
feelings and
relationships with providers." (Tr., p. 45) Since
providers are paid
on a fee or rate basis that is based on or related to cost
in some
manner, the extra costs to the providers would presumably end up
being
paid eventually in some form by either Medicaid or Medicare.
19. There is of course the problem of the State
having sufficient
information to bill the intermediaries, with the loss of
the audit tape.
Also, the method would seem impractical except where the
State has an
ongoing relation- ship with the providers so that it can collect
the
duplicate payments by offset. However, the State's position was
that it
was prepared to proceed. (Tr., pp. 46-48)
20. If Pennsylvania decides to propose use of
statistical sampling,
they may wish to note that HHS auditors have estab-
lished guidelines,
which we have observed in other cases, de- signed to
assure that valid
methodology (e.g., proper sample sizes and appropriate
corrections for
sampling error) which may be