Department of Health and Human Services
DEPARTMENTAL APPEALS BOARD
Appellate Division
SUBJECT: Iowa Department of Human Services
DATE: May 3, 1991
Docket No. 90-103
Decision No. 1248
DECISION
The Iowa Department of Human Services (Iowa) appealed the determination
of
the Health Care Financing Administration (HCFA) disallowing
$4,220,199 in
federal financial participation (FFP) claimed under title
XIX of the Social
Security Act (Act) for inpatient hospital services
provided from October 1,
1987 through March 12, 1989. HCFA disallowed
this amount based on its
determination that Iowa's claims were not made
in accordance with state plan
provisions governing reimbursement rates
for inpatient hospital
services. The disallowance represents the
difference between the amount
claimed by Iowa and the amount which HCFA
determined would have been claimed
if Iowa had followed its state plan
in determining reimbursement rates.
HCFA found specifically that Iowa's inpatient hospital rates were
not
calculated in accordance with certain language in the approved
state
plan which HCFA determined described how to calculate a
statewide
average cost per discharge. As discussed in detail below, we
find that
Iowa included this language in its plan by mistake and that the
language
was inconsistent with other parts of the state plan submission
which
reflected the method on which Iowa's claims were actually based
and
which Iowa had always intended to use. Thus, the plan presents, in
its
entirety and on its face, an ambiguity or conflict concerning
the
correct method. Furthermore, it does so not in relation to
a
substantive federal requirement, but where Iowa could have chosen
either
method. Based on all the evidence in the case, we conclude that
Iowa's
actual calculation was the method called for by the approved state
plan
and, accordingly, we reverse the disallowance in full.
I. Factual Background
Until October 1, 1987, Iowa reimbursed inpatient hospital costs
using
prospectively determined per diem rates, i.e., payments based on
the
length of stay of each hospital's patients. Effective October 1,
1987,
however, Iowa changed to a prospective payment system based on
diagnosis
related groups (DRGs). 1/ This type of system categorizes
patients
according to similar diagnoses and reimburses hospitals based on
the
assumption that similar resources are used for treatment of all
patients
in a DRG category. A number of different methodologies can be
used to
implement a DRG system. Iowa's DRG system provided generally
for
determination of a base-year rate per discharge for each hospital.
That
rate, adjusted for inflation, was to be multiplied by a DRG weight
for
each patient to determine the amount of the payment to the
hospital. At
issue here is the method for calculating the statewide
average cost per
discharge (SWACPD), a component of the base-year rate for a
hospital.
Iowa's decision to implement a DRG methodology was based
on
recommendations in a report issued by the Center for Health
Policy
Studies. This report specifically recommended calculation of
a
"statewide average per discharge base amount" by dividing
Medicaid
inpatient expenditures (net of certain payments not relevant here
and
adjusted for inflation) by "the total number of discharges."
Iowa
exhibit (ex.) 1, p. 3-14. Following the issuance of the report,
Iowa
hired a consultant to implement a DRG methodology. The
consultant
convened a task force, which included representatives of the
Iowa
hospital industry, to consider the report's recommendations.
Transcript
of December 11, 1990 hearing (Tr) 198; Iowa ex. 4; Iowa ex. 30, p.
1.
Working with the task force, the consultant developed a proposed
DRG
system which she presented at a seminar sponsored by HCFA in April
1987.
While the consultant modified some of the report's recommendations
based
on consultation with the task force, she used essentially the
same
method for calculation of the SWACPD. Materials distributed by
the
consultant at the HCFA seminar indicated that Iowa would use
a
"statewide base" equal to total expected payments (minus
certain
payments) "divided by the number of discharges and adjusted
by
inflation." Iowa ex. 6, p. 4.
In August 1987, Iowa amended its Administrative Code to change
the
reimbursement methodology for inpatient hospital services to a
DRG
system. These administrative rules, which were adopted following
the
receipt of written comments, provided for calculation of the SWACPD
by
first subtracting certain costs not relevant here from total
Medicaid
inpatient expenditures. They further provided that --
[t]he remaining amount is adjusted to reflect inflation,
and
divided by the statewide total number of Medicaid discharges .
.
. .
Iowa ex. 10, pp. 4 - 5. 2/ Iowa's fiscal agent testified that, from
the
inception of the DRG program in October 1987, he calculated
Iowa's
inpatient hospital rates in accordance with the administrative
rules.
Tr 158, 160 - 161. HCFA did not dispute that the SWACPD was
calculated
in accordance with the administrative rules, although it
questioned
whether these calculations followed the administrative rules in
other
respects not relevant here. 3/
In December 1987, Iowa submitted a proposed state plan to HCFA
which
included the following provision --
The base year cost per discharge . . . is determined by
dividing
the sum of the hospital's total routine service cost and
the
ancillary service cost attributed to Title XIX by the number
of
Title XIX discharges from that hospital. Fifty percent (50%)
of
this hospital-specific cost per discharge is added to
fifty
percent (50%) of the statewide case-mix adjusted average
cost
per discharge to determine the hospital's Title XIX cost
per
discharge.
Iowa ex. 11, p. 3.
After reviewing the proposed plan, HCFA wrote to Iowa on January 28,
1988
requesting "clarification in the plan's language regarding several
features
of the methodology." As pertinent here, HCFA noted that "it is
not
clear how the statewide case-mix adjusted average costs per
discharge are
calculated." Iowa ex. 13, p. 2.
Iowa responded in a February 18, 1988 letter to HCFA which stated
in
pertinent part --
The statewide case-mix adjusted cost per discharge is
determined
by adding all Iowa hospital-specific costs per discharge
and
dividing by the total number of Iowa hospitals. The
resultant
statewide average cost per discharge is then multiplied by
the
hospital's case-mix index in order to become case-mix
adjusted.
The final blended rate is the average of the
hospital-specific
cost per discharge and the statewide case-mix
adjusted average
cost per discharge.
Iowa ex. 14, p. 2.
On April 8, 1988, HCFA advised Iowa by telephone that if revised
plan
pages incorporating the contents of the February 18 letter were
not
received by April 13, HCFA would begin disapproval proceedings
unless
the letter was withdrawn. Iowa ex. 15. On April 13, Iowa
sent HCFA
substitute pages for its proposed plan which repeated the language
in
its letter. Iowa ex. 16, p. 2. This version of the plan was
approved
by HCFA retroactive to October 1, 1987. The language in
question
appeared in paragraph 3 of Attachment 4.19-A. In the letter
approving
the plan, HCFA noted that Iowa had provided an adequate explanation
in
response to its question regarding "calculation of the
hospital-specific
costs per discharge." Iowa ex. 17.
Iowa also included with its substitute pages a number of
"assurances"
required by 42 C.F.R. 447.253 and 447.255. 4/ These
assurances stated
in pertinent part that the amount of the "average proposed
payment rate
. . . [u]nder the plan amendment . . . is expected to be
$2,382.06."
Iowa ex. 16, last page. The same assurance was included
with the
proposed state plan submitted to HCFA in December 1987. Iowa
ex. 11,
last page. The record shows that this amount was determined
using a
SWACPD which was calculated by dividing by the number of
discharges
rather than by the number of hospitals as provided in paragraph 3
of the
state plan. Iowa brief dated 8/6/90, p. 21; Letter from Allen
to
Ballard dated 11/21/90, p. 1; HCFA ex. 2; Iowa ex. 25 and ex. 29.
The assurances included with the substitute pages as well as the
earlier
plan submission also stated that "[i]n accordance with 42 CFR
447.205,
the . . . [State agency] has given public notice concerning the
change
in reimbursement effective October 1, 1987 (copy attached)."
Iowa ex.
16 and 11, next to last page. While the record does not
include the
copy of the public notice referred to in the assurance, it would
have
consisted of either the administrative rules adopted by Iowa or a
notice
published in the newspaper referring to the administrative rules
since
those were the only means through which Iowa gave public notice.
Tr 17.
5/
The claims submitted by Iowa were calculated using a SWACPD obtained
by
dividing total Medicaid hospital costs statewide (net of certain
other
costs not relevant here) by the total number of Medicaid
discharges.
HCFA, however, found that the state plan required Iowa to divide
the
total of all hospital average costs per discharge by the total number
of
hospitals. The method employed by Iowa, unlike the method which
HCFA
claimed was called for by the state plan, gives proportional weight
to
the cost per discharge of those hospitals with the most
Medicaid
patients. Tr 292. Iowa provided unrebutted evidence that
the method
used by HCFA resulted in a payment rate which would reimburse
hospitals
for only 64% of charges. Iowa ex. 26, pp. 3 - 4.
Iowa asserted that it had always intended its state plan to provide
for
the method which it used to calculate its claims. Iowa took
the
position that the language in the approved state plan was submitted
by
mistake; however, it admitted that it could not fully explain why
this
occurred. Iowa brief dated 8/6/90, p. 23; Tr 23 - 24. 6/ The
State's
consultant -- who drafted the language in the original plan
proposal,
the February 18, 1988 letter and the substitute pages -- stated
that she
prepared the February 18 letter without referring to the
administrative
rules. Instead, she relied on other materials on DRG
methodologies
which she had compiled, from which she derived the language in
the
letter which she then adopted in the substitute pages. The
consultant
then gave these documents to her supervisor, the administrator of
the
Division of Medical Services, who was not a reimbursement specialist
and
who did not review the method in question. Tr 206, 212 - 213.
Both the
consultant and her supervisor testified that they had no
recollection of
making a conscious decision to adopt a DRG methodology which
departed
from the administrative rules. Tr 80; 205 - 206.
The next step in the State agency's internal review process was review
by
the "manual editor" in the Bureau of Policy Coordination, who was
responsible
for reviewing state plan submissions against existing
policies such as the
administrative rules. This individual stated that
she had reviewed the
original plan submission, which she regarded as
very technical, to the best
of her ability and sent it on for further
review by her supervisor, by the
division director, and by the director
of the State agency, in accordance
with the State agency's normal
procedure for review of state plan
submissions. 7/ However, the editor
did not review the February 18
letter because it was not identified as
part of the state plan. The
editor further stated that she did not
review the substitute plan pages
thoroughly because she was advised that
the changes had been agreed upon with
HCFA and had to be submitted
immediately. She also indicated that it
was unclear what further review
procedures were followed for the substitute
pages. Iowa ex. 24, pp. 1 -
2; Tr 102, 109, 113, 117 - 118, 122 -
123.
II. Legal Authorities
In order to qualify for federal funding, a state's claim for the cost
of
medical services must be in accordance with an approved Medicaid
state
plan which fulfills certain requirements. Section 1903(a) of the
Act.
Section 1902(a)(13) (as amended in 1980) requires that state
plans
provide for payment of hospital 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
states with flexibility in developing methods of provider
reimbursement.
Regulations implementing the Boren Amendment provide that HCFA
will
approve a state's rate-setting method, so long as the state
provides
required satisfactory assurances. The state plan must,
however, specify
comprehensively the methods and standards used by the state
to set
payment rates, and the state must pay for services using
rates
determined in accordance with those methods and standards. 42
C.F.R.
447.252, 447.253(g).
A number of Board decisions address the issue of rate-setting under
the
Boren Amendment. In California Dept. of Health Services, DAB No.
1007
(1989), the Board stated in pertinent part:
While the State is given considerable discretion in
proposing
the terms of its rate setting methodology in its plan, once
that
plan has been approved by the Agency, the Agency may
reasonably
expect, consistent with the statute and regulations, that
the
State's claims will be consistent with the plan. . . . Thus,
the
Agency should not have to be placed in the position of
making
new substantive judgments about the appropriateness of a
State's
rate every time a State decides for reasons of its own that
it
would be better to depart from the prescribed terms of its
plan
for whatever reason the State may choose. The chosen
vehicle
for departures from the plan methodology under both the
statute
and the regulations would be a change in the State's plan.
California, p. 6. Consistent with this principle, the Board has
held
that a state is not excused from complying with the terms of its
state
plan even if the methodology followed by the state was consistent
with
its administrative rules and would have been approved by the
federal
agency if submitted as a state plan. Tennessee Dept. of Health
and
Environment, DAB No. 950 (1988). See also, Massachusetts Dept.
of
Public Welfare, DAB No. 867 (1987).
However, the Board has found factors such as a state's
consistent
administrative practice and the approvability of its
methodology
relevant where the language of a state plan is ambiguous and
a
determination must be made as to whether the state has in fact
followed
its state plan. In that case, assuming that no federal
requirement is
at issue and the state therefore has discretion in fashioning
its
reimbursement system, the Board looks to whether the
state's
interpretation gives reasonable effect to the language of the plan as
a
whole. The Board also considers the intent of the provision, which
may
be evidenced by the state's administrative practice. The Board
will
defer to a state's interpretation of its own plan if it is reasonable
in
light of the purpose of the provisions and the program requirements.
In
the specific context of rate-setting, the Board will give even
greater
weight to what the state intended since, under the Boren Amendment,
HCFA
must approve a state's proposed methods so long as the state gives
the
requisite assurances. South Dakota Dept. of Social Services, DAB
No.
934 (1988).
The Board has also addressed the question of the legal effect to be
given
to state plan language which is submitted by mistake. In Florida
Dept.
of Health and Rehabilitative Services, DAB No. 1100 (1989), the
Board
considered whether Florida was bound by language in a request for
extension
of a waiver of certain title XIX program requirements which
Florida alleged
was submitted by mistake. 8/ The waiver extension
request, which was
approved by HCFA, stated that "[t]he State does not
cover children 18 or
under, who are living at home." Florida contended
that, due to a
clerical error, the phrase "and who are not otherwise
categorically eligible"
was omitted from the end of this sentence. The
addition of this phrase
changes the meaning to provide coverage of
otherwise eligible children 18 or
under who are living at home. In
considering whether federal funding
claimed for such children was
allowable, the Board looked at the following
factors:
o Whether the context otherwise reflected a
different intent. The
Board found that there
was an internal inconsistency in the
document as a
whole in that the supporting cost data for the
waiver
request included the children in question and
the waiver request
itself included a discussion of
services for children.
o Whether the waiver request would have been
approved if it had
explicitly provided for coverage
of the children in question. The
Board found
that the federal agency would have readily approved
the
provision since it was entirely a state's
decision as to which
eligible groups it would
cover.
o Whether Florida intended to eliminate the children
in question
from the program. The Board found
no evidence of such an intent
apart from the
disputed language itself, although the Board stated
that there was insufficient detail to support Florida's
contention
that a clerical error was involved.
o Whether the federal agency understood the waiver request
to
exclude the children in question. The Board found
that the agency
had not really examined the description of
the eligible groups in
the waiver request and that this
was not a factor in approving the
request.
o Whether Florida operated its waiver program consistently
to
include the children in question. The Board found
that it did.
o Whether there was any credible reason why Florida would
want to
have excluded the children in question. The
Board found that the
federal agency had offered no such
reason.
Based on these factors, the Board determined that it was unreasonable
to
construe Florida's waiver request as excluding the children in
question
and, accordingly, reversed the disallowance. As the discussion
below
indicates, we find that this appeal presents an issue similar to
that
presented in Florida.
III. Discussion
The problem in this case arises because of the language in paragraph 3
of
the approved state plan which explains that the "statewide case-mix
adjusted
cost per discharge is determined by adding all Iowa
hospital-specific costs
per discharge and dividing by the total number
of Iowa hospitals." HCFA
read this to require the division of the sum
of the cost per discharge for
each hospital (obtained by dividing each
hospital's adjusted Medicaid costs
by the number of its Medicaid
discharges) by the number of hospitals.
In making its claims, however,
Iowa divided total Medicaid costs for all
hospitals by the number of
discharges for all hospitals.
Iowa argued that there was an ambiguity in paragraph 3 in that
the
commonsense meaning of the term "statewide average cost per
discharge"
required dividing by the number of discharges. Moreover,
Iowa contended
that the calculation specified in paragraph 3 was in conflict
with the
assurance which stated a proposed payment rate because that amount
was
determined using a SWACPD calculated by dividing by the number
of
discharges. Iowa asserted that the language in the approved plan
on
which HCFA relied was submitted by mistake and that the history of
the
development of Iowa's DRG system showed that the calculation Iowa
used
was the one which was always intended. Iowa therefore asked
for
reformation of the state plan to reflect its real intent. 9/
As discussed below, we conclude that Iowa followed its state
plan
notwithstanding the language in paragraph 3 on "dividing by the
total
number of Iowa hospitals." This language is not dispositive
because
other parts of the state plan submission are inconsistent with
it.
Thus, Iowa's intent becomes relevant in determining which of
the
conflicting provisions governs. Since the record clearly
establishes
that Iowa intended to follow the method reflected elsewhere in
its state
plan submission rather than the method specified in paragraph 3,
Iowa's
compliance with its state plan must be measured against the former
and
not the latter method.
The State Plan Submission Was Internally Inconsistent.
If the question in this case were how to interpret the language in
Iowa's
original state plan submission, which referred to a SWACPD
without defining
it, Iowa would clearly prevail; the commonsense meaning
of that term is that
the total of the adjusted costs for all of the
hospitals should be divided by
the number of discharges (resulting in a
figure representing the amount of
cost averaged over all of the
discharges in the state). However, we
cannot agree with Iowa that the
mere fact that Iowa retained the term SWACPD
in paragraph 3 of the
substitute pages is sufficient to make that paragraph
ambiguous. The
term SWACPD can be reconciled with the substitute
language calling for
dividing by the number of hospitals if SWACPD is read as
an average of
the hospital-specific "cost per discharge" amounts. 10/
On the other
hand, the fact that Iowa did not modify the term SWACPD to
indicate more
clearly what average was achieved by dividing by the number of
hospitals
(which it could have done, for example, by simply putting quotes
around
"cost per discharge") provides some support for Iowa's position
about
what it intended.
In any event, the language of paragraph 3 is not dispositive here
since
Iowa established that other parts of its state plan submission
were
inconsistent with that language. One source of this inconsistency
was
Iowa's assurance regarding the expected payment rate under its
DRG
system. As noted previously, the figure stated, $2,382.06, was
obtained
using a SWACPD calculated by dividing by the number of discharges
rather
than by dividing by the number of hospitals.
HCFA argued that this assurance did not create an internal
inconsistency
since it could not be determined from the face of the provision
how the
SWACPD was calculated. However, the figure used in the
assurance in
fact resulted from calculating the SWACPD by dividing by the
number of
discharges. Thus, the assurance conflicts with paragraph 3
regardless
of whether the conflict is revealed only by an analysis of
the
computation underlying the $2,382.06 figure. 11/
Paragraph 3 is also inconsistent with the assurance which stated
that
public notice of the proposed plan amendment had been given
and
indicated that a copy of the notice was included with the
plan
provision. There was uncontradicted testimony that the notice
referred
to was given either by publication of Iowa's administrative
rules
pertaining to its DRG system or by publication of a newspaper
notice
which referred to these rules, although Iowa was unable to
establish
which document was actually included with the plan. Since
the
administrative rules provided for calculation of the SWACPD by
dividing
by the number of discharges rather than the number of hospitals,
this
assurance also created an internal inconsistency in the state
plan
submission.
HCFA also argued, however, that these assurances did not create
an
inconsistency which permitted the consideration of Iowa's intent
because
the assurances were not a part of the state plan. HCFA cited in
support
of its position the Board's decisions in Delaware Dept. of Health
and
Social Services, DAB No. 1166 (1990) and California Dept. of
Health
Services, DAB No. 1007 (1989). These decisions are
distinguishable from
the instant case, however. In Delaware, the Board
found that where the
state submitted assurances before it submitted a plan
amendment adopting
a new reimbursement system, the amendment was not
effective until the
date the amendment itself was submitted because the
assurances alone did
not provide a detailed description of the payment
methodology, as
required by the applicable regulations. In California,
the Board found
that even if the state had submitted assurances which placed
HCFA on
notice that it was using a different methodology to compute
certain
rates than that specified in the state plan, such assurances
alone,
"unaccompanied by a proposed plan amendment," could not have served
to
amend the state plan methodology. California, p. 8. The
holding in
these cases that assurances alone do not constitute a state plan
does
not preclude our finding here that assurances which conflict with
the
plan with which they were submitted raise a question as to the
meaning
of the state plan itself.
Thus, this case is distinguishable from Tennessee, supra, where there
was
no dispute as to what the state plan required. As in Florida,
supra,
there were internal contradictions in the state's submission.
Accordingly, it
is appropriate to examine Iowa's intent to determine how
the SWACPD is to be
calculated under the state plan.
The Language in Paragraph 3 of the Approved Plan Was Submitted
by
Mistake.
The record in this case bears out Iowa's contention that it never
intended
to calculate the SWACPD by dividing by the number of hospitals,
and that the
language to this effect was submitted by mistake. As
indicated
previously, every iteration of the DRG methodology prior to
submission of the
original plan proposal called for dividing by the
number of discharges.
This included the Center for Health Policy
Studies report which recommended
Iowa's adoption of a DRG methodology,
the materials used by Iowa in the
overview of its DRG methodology
presented at the HCFA seminar, and Iowa's
proposed and final
administrative rules. Moreover, while the original
plan proposal did
not specify how the SWACPD was to be calculated, the
commonsense meaning
of this term requires division by the number of
discharges. (In
addition, the original plan proposal specifies that
the
hospital-specific cost per discharge is to be calculated in
this
manner.)
It was not until Iowa's February 18, 1988 letter to HCFA that
Iowa
specified that the SWACPD was to be calculated by dividing by the
number
of hospitals. We conclude that this language was submitted in
error
since there is simply no indication in the record that Iowa
ever
considered any method of calculating the SWACPD other than the
method
adopted in the administrative rules and consistently used by
Iowa.
Under these circumstances, it is entirely plausible that the
consultant
who prepared the February 18 letter and the substitute
pages
inadvertently took the wrong language from her own materials on
DRG
methodology, as she testified.
Moreover, simply because the same language was submitted to HCFA
twice
before the state plan was actually approved does not prove that
there
was no mistake in its submission, as HCFA argued. The method
which
called for dividing by the number of discharges was widely publicized
at
various points without any evidence that it was ever questioned. On
the
other hand, the February 18 letter and the substitute pages were seen
by
only a limited number of people before they were transmitted to
HCFA.
It appears from the record that, other than the nominal signatory,
the
only person to review the February 18 letter was the
consultant's
supervisor. Given the technical nature of the method, he
could have
easily have failed to detect the error in the language in
question. The
substitute pages were reviewed by the manual editor as
well as the
consultant's supervisor. While the editor was supposed to
review state
plan submissions for consistency with the administrative rules,
she
could have failed to find any inconsistency due to her perception
that
the substitute pages had to be submitted on an expedited basis.
12/ The
manual editor's failure to detect any error can also be
explained by the
fact that she may have limited her review because she knew
that the
language in the substitute pages had already been approved by
HCFA.
HCFA also pointed out that a subsequent plan amendment which changed
other
aspects of the rate computation repeated the language in paragraph
3
regarding calculation of the SWACPD. HCFA brief dated 3/8/91,
citing
HCFA ex. 6. However, this does not establish that Iowa intended
this
language to be a part of its state plan in the absence of any
evidence
that Iowa was aware when it submitted the plan amendment that the
method
specified in paragraph 3 was inconsistent with how it
actually
calculated its claims. 13/
Accordingly, we conclude that the language in paragraph 3 was submitted
by
mistake and that Iowa intended its state plan to provide for
calculating the
SWACPD as specified in the administrative rules and
reflected in the expected
payment rate stated in the assurances.
The Method Used By Iowa Was One Which HCFA Could Have Approved.
Notwithstanding Iowa's intent, its state plan can reasonably be read
to
provide for the method used to calculate its claims only if the
method
is one which could have been approved by HCFA. This is clearly
the case
here.
HCFA indicated that it would have approved the method which Iowa used
to
calculate the SWACPD had the method been specifically articulated
in
Iowa's state plan. Tr 316 - 318. 14/ Moreover, it appears that
there
are no fixed methodologies for implementing a DRG system
of
reimbursement. Tr 295. Thus, as long as Iowa provided an
assurance
that the rate determined for each hospital using Iowa's method
was
"reasonable and adequate to meet the costs" incurred by an
efficiently
and economically operated facility in providing the required care
and
services, HCFA was required under the Boren Amendment to approve it.
15/
HCFA Did Not Rely on the Language of Paragraph 3 in Approving the
State
Plan.
Iowa argued that, under the circumstances presented here, paragraph 3
of
its state plan should be reformed to provide for calculating the
SWACPD
by dividing total Medicaid hospital costs statewide by the number
of
discharges. HCFA noted, however, that reformation of a contract
"is
premised on a mutual mistake," and is not the proper remedy if
the
language of the contract "accords with the intention of one
party."
HCFA brief dated 3/8/91, p. 13, citing, among other authorities,
A.
CORBIN, 3 CONTRACTS, sections 608 and 613 (1960). HCFA took
the
position that reformation of Iowa's state plan was not
appropriate
because there was no mistake on the part of HCFA, which "assumed
that
Iowa meant to compute the statewide average cost per discharge
in
accordance with [paragraph 3 of] its approved State Plan." HCFA
brief
dated 10/17/90, p. 11.
The Board has noted that a "state plan is in the nature of a
contract
between the State and the federal government." Tennessee Dept.
of
Health and Environment, DAB No. 950 (1988), p. 3. However,
the
principle that reformation of a contract is available only where
there
is a mutual mistake was developed largely for commercial contracts
and
does not directly apply here. What makes this situation
distinguishable
is that states are required by statute to develop the methods
and
standards for reimbursing providers. Thus, unlike an
ordinary
commercial contract, the terms of which are negotiated by the
parties,
the portion of the state plan in question here was drawn up by Iowa
and
had to be approved by HCFA as long as it included certain
basic
elements. This case may therefore be more analogous to the
situation
where a bidder discovers a mistake in his bid before award, in
which
case the bidder is permitted to correct the bid if there is clear
and
convincing evidence of both the mistake and the intended bid.
See,
e.g., 65 Comp. Gen. 202 (1986); 48 C.F.R. 14-406 (Federal
Acquisition
Regulation). Based on this analogy, Iowa should be
permitted to replace
paragraph 3 as approved by HCFA with language providing
for dividing by
the number of discharges since there is clear and convincing
evidence
that the language in paragraph 3 was submitted by mistake and that
Iowa
intended to calculate the SWACPD by dividing by the number
of
discharges.
Even if the law on reformation of contracts applied here, however,
the
requirement for a mutual mistake does not appear to be
absolute.
Corbin's treatise, cited by HCFA, acknowledges that the general
rule is
that reformation must be denied where the mistake was not
mutual;
however, Corbin indicates that the better view is that reformation
is
permissible in the absence of mutual mistake as long as the plaintiff
is
not trying to enforce a contract that the defendant never agreed
to.
See CORBIN, supra, sections 603 and 614. We conclude that this test
is
satisfied here. As discussed above, HCFA would have agreed to a
method
which called for dividing by the number of discharges. Moreover,
its
approval of the substitute pages was not conditioned on the language
of
paragraph 3. While HCFA originally noted the absence from the
original
plan proposal of language specifying how the SWACPD was to
be
calculated, HCFA stated after receiving the substitute pages that
it
would approve the plan since Iowa had now explained how to
calculate
hospital-specific costs per discharge. Inasmuch as it was
the
calculation of the SWACPD, not the hospital-specific costs
per
discharge, which was explained in the substitute pages, HCFA appears
not
to have focused on the precise language of paragraph 3.
Instead, it is likely that HCFA simply relied on the expected payment
rate
stated in the assurances in approving Iowa's state plan, given that
the Boren
Amendment requires approval of a state's rate-setting method
as long as the
state makes the requisite assurances. HCFA found the
expected payment
rate stated in the assurances satisfactory. Since HCFA
did not rely on
the language which Iowa mistakenly submitted,
reformation is appropriate
here.
Iowa Had a Valid Policy Reason for Using the Method in Question.
Another consideration which supports a determination that paragraph 3
of
the state plan should be reformed is the fact that the method used
by
Iowa to calculate its claims gives proportional weight to the cost
per
discharge of those hospitals with the most Medicaid patients.
Whether
to give such weight is a policy decision for a state under the
Boren
Amendment. The approach used here was adopted by Iowa through a
process
of consultation with the provider community, which had
representatives
on the State agency's task force. That there was a
valid policy reason
for using the method in question lends credence to Iowa's
position that
this was the method which it had intended to incorporate in its
state
plan. 16/
IV. Conclusion
For the foregoing reasons, we conclude that Iowa's claims were
calculated
in accordance with its approved state plan, which, taken as a
whole, provided
for calculation of the SWACPD by dividing total Medicaid
hospital costs
statewide by the number of discharges. Accordingly, we
reverse the
disallowance in full.
_____________________________ Cecilia Sparks Ford
_____________________________ Norval D. (John) Settle
_____________________________ Judith A. Ballard Presiding
Board
Member
1. Iowa retained a per diem rate system for
Medicaid-certified
physical rehabilitation units, however.
2. This provision was the same in both the proposed
and final
versions of the administrative rules. Compare Iowa ex. 7 and
ex. 10.
3. In particular, HCFA found that the use of a budget
neutrality
adjustment to calculate the claims in question here was not
authorized
by the state plan. Iowa conceded that HCFA was
correct. Letter from
Allen to Ballard dated 11/21/90, p. 3. The
disallowance would have been
higher if Iowa had not applied the budget
neutrality adjustment since
the adjustment reduced the DRG rates actually
paid by Iowa. Certain
other questions HCFA had about the SWACPD
calculations, however, were
satisfactorily answered at the hearing.
See, e.g., Tr 161-173, 247,
252-254.
4. Section 447.253 provides in pertinent part that in
order to
receive HCFA approval of a state plan change in payment methods
and
standards, the Medicaid agency must make certain assurances
satisfactory
to HCFA. Section 447.255 requires the Medicaid agency to
submit certain
information with these assurances, including the amount of the
estimated
average proposed payment rate for each type of provider.
5. Section 447.205 states that the public
notice which the
Medicaid agency is required by section 447.253 to provide
when it is
proposing significant changes to its methods or standards for
setting
payment rates for inpatient hospital services must appear as a
public
announcement in a state register similar to the Federal Register or
in
the newspaper of widest circulation in the state or in each city in
the
state above a certain population.
6. Iowa noted that there were several versions
of the proposed plan
language on word processing disks, including a version
which called for
dividing Medicaid inpatient expenditures by the number of
discharges
rather than the number of hospitals. Iowa ex. 20, p.
1. Iowa suggested
that the substitute pages which were ultimately
approved were therefore
submitted as the result of a clerical error.
Iowa brief dated 8/6/90,
pp. 6 - 7; Iowa ex. 30, p. 4. However, the
existence of multiple
versions of the substitute pages does not account for
the fact that the
same language in the approved state plan appeared in Iowa's
February 18,
1988 letter to HCFA.
7. Iowa's procedures also called for a review by the
Governor. Iowa
ex. 24, p. 1.
8. The Board noted in Florida that the nature of a
waiver request
was essentially unilateral, resulting from the action of the
federal
agency alone. The Board nevertheless stated that "since the
terms of an
approved waiver do have a direct effect on the operation of a
state's
program under its state plan, we would agree that a state is bound
to
follow the provisions of an approved waiver request." Florida, p.
10.
Accordingly, we reject HCFA's argument that the case before us
is
distinguishable from Florida on this basis.
9. Iowa also contended that, even if the language in
the approved
plan was binding, HCFA improperly computed the amount of
the
disallowance involving calculation of the SWACPD, and that the amount
of
any disallowance should not exceed $463,000. Iowa brief dated
11/2/90,
p. 5. However, we need not address this argument in light of
our
conclusion that there was no basis for the disallowance.
10. HCFA did not dispute that the numerator can also
be read as
total Medicaid costs for all Iowa hospitals; thus, we focus only
on the
meaning of the denominator.
11. HCFA argued that Iowa had admitted (at Tr 235,
249) that the
expected payment rate would not change even if the SWACPD was
calculated
differently. When considered in context, however, the
testimony on
which HCFA relied does not constitute such an admission.
Moreover, what
creates the inconsistency in the state plan submission is not
the amount
itself but the calculations which produced the amount. See
Tr 250-252,
257-259.
12. HCFA asserted that there was no reason for Iowa
to have
submitted the substitute pages by April 13, 1988 since Iowa could
also
have avoided disapproval proceedings merely by withdrawing the
February
18, 1988 letter. However, it is significant that Iowa felt
that it was
operating under an April 13 deadline even if it was not since
that
explains why Iowa's normal review procedures were not followed.
13. The record also shows that a state agency
employee who assumed
the consultant's responsibilities became aware that the
language of
paragraph 3 as it appeared in the state plan amendment differed
from a
version of the substitute pages which called for dividing by the
number
of discharges and which he initially believed to be the
approved
version. Iowa ex. 24, p. 2, and p. 5 of attachment to
affidavit.
However, there is no evidence that this individual realized that
the
language in paragraph 3 of the state plan amendment called for
a
calculation that was significantly different from the one Iowa
actually
used.
14. HCFA also stated that the method in the
administrative rules for
determining the SWACPD was "ultimately approved. . .
." HCFA brief
dated 3/8/91, p. 20. However, the record does not
show the date of such
approval or the precise language approved.
15. It is questionable whether the methodology which
HCFA claimed
was required by paragraph 3 could have been approved under
this
standard, since it resulted in a payment amount which reimbursed
only
64% of a hospital's customary charges.
16. In addition, there is no dispute that all of the
hospitals in
question were privately-owned (with the exception of one
hospital
controlled by the Board of Regents of the University of Iowa) and
that
Iowa actually paid the amounts in question here for covered
services
provided to eligible individuals. Thus, Iowa would not receive
any
windfall if it were reimbursed for the full amount of FFP which
it
claimed. Compare Massachusetts, supra (Board found that
rate
adjustments were sought by the state solely to increase federal
Medicaid