DEPARTMENTAL GRANT APPEALS BOARD
Department of Health and Human Services
SUBJECT: Colorado Department of
Social Services
Docket Nos. 88-34
Decision No. 985
DATE: September 8, 1988
DECISION
The Colorado Department of Social Services (Colorado or State) appealed
a
determination by the Health Care Financing Administration (HCFA)
disallowing
a total of $1,301,669 in federal financial participation
(FFP) claimed under
Title XIX (Medicaid) of the Social Security Act
(Act) for payments to two
intermediate care facilities. HCFA disallowed
$551,155 claimed for
payments to Beverly Manor Nursing Home for services
provided to residents
over age 21 and under age 65 from July 1, 1976
through June 30, 1979, and
$750,514 for payments to Phoenix Center for
similar services provided from
June 30, 1976 to June 30, 1978. The
disallowance was based on audits
finding that the two facilities were
institutions for mental diseases
(IMDs).
Section 1905(a) of the Act excludes from the definition of
"medical
assistance," for which Medicaid funding is available, care or
services
provided to individuals under age 65 who are in an IMD.
HCFA
regulations define an IMD generally as having the "overall character"
of
an institution established and maintained primarily for the care
and
treatment of persons with mental diseases. The IMD exclusion has
been
addressed in a number of Board and court decisions; the key holdings
in
those decisions are that an intermediate care facility (ICF) may be
an
IMD and that evidence gathered by HCFA using unpublished "criteria"
may
support a disallowance if it shows that the .facility was an IMD
under
the regulatory definition. See Connecticut Department of
Income
Maintenance v. Heckler, 471 U.S. 524 (1985); Minnesota Dept. of
Human
Services, DGAB No. 831 (1987).
For the reasons discussed below, we reach the following conclusions
in
this case:
o We reject the State's argument that the disallowance is
invalid
because some of the criteria resulted in the auditors
collecting
evidence lacking probative value. The disallowance is based
on the
regulatory definition of an IMD, as applied to the evidence as a
whole,
not on the criteria themselves.
o We reject the State's argument that the disallowance should
be
dismissed on the basis that HCFA did not act in a timely fashion.
The
audits here occurred within a short period after the services
were
provided and any subsequent delays in the disallowance proceedings
were
reasonable when considered in light of the particular
circumstances
here. Moreover, the State knew during this entire period
that the
status of the facilities as IMDs was in question and that it
should
retain any evidence relevant to the issue of the overall character
of
the facilities.
o We nonetheless reverse the disallowance for Beverly Manor
Nursing
Home after weighing the evidence regarding the overall character of
that
facility. We find that evidence of the nature of the patient
population
does not show that Beverly Manor primarily was engaged in
providing care
and treatment of persons with mental diseases, given the low
percentage
of Beverly Manor patients who may be counted as mentally diseased
under
HCFA's guidelines as revised after the audit. Other evidence,
such as a
statement on the facility's admission policy, establishes only that
some
Beverly Manor patients were receiving psychiatric treatment
services,
not that the facility was primarily providing such services.
o We uphold the disallowance of Phoenix Center because, unlike
the
evidence for Beverly Manor, the evidence regarding Phoenix
Center
overwhelmingly shows that the facility was an IMD, under any
reasonable
reading of the regulation. While we do not consider each
type of
evidence to be of equal weight, some of the evidence is of a
highly
probative type, and, when the evidence is considered as a whole,
it
compels a finding that this facility was primarily engaged in
providing
care and treatment of persons with mental diseases.
Accordingly, we reverse the $551,155 disallowance for Beverly Manor
and
sustain the disallowance of $750,514 for Phoenix Center.
Statutory and Regulatory Background
Title XIX of the Social Security Act provides grants to states for
medical
assistance to eligible low-income persons. Section 1905(a), in
defining
"medical assistance," specifically excludes payments for
services to "any
individual who has not attained 65 years of age and who
is a patient in an
institution for . . . mental diseases" (except
inpatient psychiatric hospital
services for individuals under age 21).
This exclusion has been present in
the Medicaid program from the
program's inception in 1965. The
exclusion rests on a congressional
intent not to finance through the Medicaid
program long-term care for
institutionalized individuals who were
traditionally the responsibility
of state and local governments. S.
Rep. No. 404, 89th Cong. 1st Sess.
Pt. I, 144-47 (1965).
The regulations implementing the exclusion for IMD patients define an
IMD
as --
an institution that is primarily engaged in providing
diagnosis, treatment or care of persons with mental diseases,
including medical attention, nursing care and related services.
Whether an institution is an institution for mental diseases is
determined by its overall character as that of a facility
established
and maintained primarily for the care and treatment
of individuals
with mental diseases, whether or not it is
licensed as such . . .
.
42 C.F.R. 435.1009 (1978-1981).
To help its reviewers identify IMDs, the Social and Rehabilitation
Service
(SRS), HCFA's predecessor agency, established "criteria" in a
series of
transmittals to SRS field staff between 1975 and 1977, as part
of the "Field
Staff Information and Instruction Series" (FSIIS).
In December 1982 (after the disallowance periods here), HCFA
issued
section 4390 of the State Medicaid Manual, which explained to
state
Medicaid agencies HCFA's guidelines for determining if a facility is
an
IMD and amplified the factors to be used cumulatively to determine
the
facility's overall character. Several factors concerned
patient
population and required that patients be classified as mentally
diseased
or physically diseased. In section 4390, HCFA stated that
patients with
a diagnosis listed as a mental disorder in the
International
Classification of Diseases (ICD) should be classified as
mentally
diseased if the institutionalization resulted from that
diagnosis.
HCFA excluded from consideration as mental diseases (despite
their
listing in the ICD) mental retardation and certain diagnoses related
to
the general category of "senility."
In September 1986, HCFA revised section 4390 of the State Medicaid
Manual
to provide additional information. Dkt. No. 88-34, Ex. B.
Among
other things, the revised section states that organic brain
syndrome
should not be .considered a mental disease, even though listed as
a
mental disorder in the ICD, if the facility being reviewed
is
appropriately treating the patients by providing only general
nursing
care, since this condition (like senility) is essentially
untreatable
from a mental health point of view. The revisions to
section 4390 of
the State Medicaid Manual also recognized difficulties in
classifying
patients with a diagnosis of alcoholism and established some
guidelines
based on the Board's holding in Granville House, Inc., DGAB No.
529
(1984).
Prior Board Decisions and Judicial Precedent
Both the Board and the courts have considered the issues raised
in
determining if a facility is an IMD. Below, we summarize
issues
relevant to the instant case addressed in prior Board decisions
and
court cases.
o The Board has held that the term "institution for mental diseases"
in
the statutory funding exclusion may include private intermediate
care
facilities (ICFs) which are primarily maintained for the care
and
treatment of patients with mental diseases, and that evidence
collected
by HCFA using its unpublished FSIIS "criteria" could support
a
disallowance when the evidence shows that the facility was an IMD
under
the regulatory definition. Joint Consideration: Institutions for
Mental
Diseases, DGAB No. 231 (1981). The United States Supreme Court
upheld
the determination that an ICF could be an IMD. Connecticut v.
Heckler,
471 U.S. 524 (1985).
o In the Board's prior cases, the central focus of the Board's
inquiry
was whether the record, taken as a whole, established that the
disputed
facilities were IMDs. In DGAB No. 231, the Board specifically
upheld
HCFA's use of the FSIIS criteria to gather evidence, including
the
percentages of patients admitted because of diagnoses of mental
diseases
under the ICD. The Board found that, while the FSIIS criteria
were not
themselves dispositive, the factual evidence assembled under
the
criteria established the overall character of the facilities as
IMDs.
The Board's findings were upheld by the court in California v.
Bowen,
Civ. No. 5-82-180-EJG (E.D. Ca. May 21, 1986), and, with respect to
six
of nine facilities by the court in Illinois v. Heckler, No.
82C-1349
(N.D. Ill. June 30, 1986). With respect to the remaining
facilities in
Illinois, and in Minnesota v. Heckler, 718 F.2d 852 (8th Cir.
1983), the
respective courts found that the Board had relied too heavily
on
patient-counting factors, such as the number and percentages of
mentally
diseased patients, rather than on the treatment being given by, and
the
general nature of, the facilities.
o In several cases, the Board recognized the substantial
difficulties
in classifying patients as mentally diseased, and the
possibly
disruptive effects of small shifts in patient percentages for
facilities
close to the 50% mark. The Board upheld disallowances in
DGAB No. 231,
finding that statistical evidence relating to the patient
population was
only one of several factors upon which HCFA's determinations
had been
based. The Board reversed a disallowance in Washington Dept.
of Social
and Health Services, DGAB No. 785 (1986), because the patient
population
statistics had not been developed consistently with stated HCFA
policy
on classifying patients, and the other evidence was inconclusive
on
whether the facility was an IMD.
o In several cases, the Board considered the treatment issues raised
by
the Eighth Circuit and the Northern District of Illinois.
In
Massachusetts Dept. of Public Welfare, DGAB No. 413 (1983), the
Board
stated that patient diagnoses and related factors should not be the
sole
factors in an IMD determination and that treatment should be
considered.
But the Board held that the IMD exclusion extended beyond
psychiatric
hospitals or facilities providing active treatment for mental
diseases,
since Congress had apparently recognized that many IMDs provide
little
active psychiatric treatment. In Granville House, Inc., DGAB No.
529
(1984), the Board found that, because the disease of alcoholism has
both
physical and mental components, HCFA must develop rules or guidelines
to
evaluate what types of alcoholism treatment indicate IMD status.
Discussion
Below, we first discuss the general legal arguments raised by Colorado.
We
then evaluate the evidence regarding the character of Beverly Manor
and
Phoenix Center. I. General Issues
a. The auditors' use of some non-probative
criteria
does not invalidate the disallowance.
Colorado argued that, although HCFA's guidelines or criteria "have
become
clearer and more defined over the years, some of the criteria
remain
unreasonable and are therefore arbitrary and capricious."
Colorado's Brief,
p. 9. Colorado argued, for example, that locked wards
are increasingly
prevalent in facilities which treat patients with
organic brain syndrome or
Alzheimer's disease, diagnoses which would not
require psychiatric
care. Since these diagnoses would not be considered
mental diseases
under HCFA's revised guidelines, Colorado said, the
presence of locked wards
for these patients would not be probative in
determining if the overall
character of the facility was an IMD.
Colorado objected to several other
guidelines or criteria.
Even if we agreed with Colorado on its objections to particular
audit
criteria, we would find no basis to invalidate all IMD determinations
in
which the criteria were used. In prior decisions, the Board
has
recognized that evidence collected by applying some of the criteria
is
less probative of the overall character of the facility than
evidence
related to other criteria. See, e.g., Washington Dept. of
Social and
Health Services, DGAB No. 785 (1986), pp. 11-13 (rejecting as a
basis
for IMD determination mere presence of some employees with mental
health
training, provision of psychosocial programs, and relatively
young
average age of patients). But even when the Board has found
evidence
related to one or more criteria to lack probative value, the Board
has
considered whether the evidence, taken as a whole, was sufficient
to
establish that the facility's "overall character" met the
regulatory
definition of a facility "established and maintained primarily for
the
care and treatment of individuals with mental diseases."
The FSIIS criteria were simply internal information-gathering steps
used
to collect evidence about a facility. Upon review, the Board
considers
the evidence as a whole regarding a disputed facility, attributing
more
weight to those types of evidence which are more probative of
the
overall character of the facility. See Minnesota Dept. of
Human
Services, DGAB No. 831 (1987).
.Since determination of IMD status is thus based on the
regulatory
definition, not on the criteria, we reject Colorado's argument
that the
disallowance should be reversed simply because HCFA's criteria may
have
resulted in the auditors collecting some evidence that lacks
probative
value.
b. The passage of time does not warrant dismissal
of
the disallowance.
Colorado argued that HCFA had failed to prosecute this disallowance in
a
timely manner and that the delay prejudiced Colorado's ability
to
produce evidence to defend against the disallowance. Colorado
conceded
that the principle of laches does not generally apply to the
federal
government, but Colorado alleged that the action should be dismissed
as
untimely under the Administrative Procedure Act (APA) provision at
5
U.S.C. 555(b). In addition, Colorado charged that HCFA's delay
in
formulating guidelines or criteria for IMDs gave Colorado
inadequate
notice of HCFA's expectations and prevented Colorado from
properly
administering its Medicaid program.
The APA provides: "With due regard for the convenience and necessity
of
the parties or their representatives and within a reasonable time,
each
agency shall proceed to conclude a matter presented to it." 5
U.S.C.
555(b). Contrary to what the State suggested, this section does
not
provide a basis for dismissing this disallowance. The
section
contemplates an evaluation of the particular circumstances of each
case
and does not contemplate dismissal where there is no showing that
agency
inaction has prejudiced the non-federal party. See, e.g., Irish
v.
Securities and Exchange Commission, 367 F.2d 637 (9th Cir. 1966).
The cases Colorado relied upon are consistent with this view. In
Equal
Employment Opportunity Commission v. Bell Helicopter Co., 426 F.
Supp.
785 (N.D. Tex. 1976), the court's decision was based on an
examination
of the specific evidence involved and on a finding that the
delay
prejudiced the capability of one party to present evidence in
its
defense. In Nader v. Federal Communications Commission, 520 F.2d
182
(D.C. Cir. 1982), the D.C. Circuit did not dismiss an action for
delays,
but merely ordered that the administrative agency submit a schedule
for
its completion.
.Here, the State's arguments focus on the total time period
involved,
without considering the numerous steps taken during that
period. The
record shows that those steps included the following.
The audits
finding potential violation of Agency regulations were performed
within
a few years after the disallowance period. The auditors provided
the
State several opportunities to comment on the audit findings
and
extended the comment periods at the State's request. Disallowances
were
then issued and were appealed to this Board. Board proceedings
were
stayed twice (first, to permit the parties to resolve informally
issues
regarding patient statistics and, later, to wait for the results of
a
hearing in a Board case involving related issues, in which Colorado
also
participated) and, ultimately, dismissed without prejudice pending
the
outcome of related court cases. The State agreed to the stays, as
well
as to the dismissal, although it was informed that HCFA
might
nonetheless press the disallowances at a later point. After the
court
decisions (including a Supreme Court decision on an issue which
would
have resolved the cases if the Court had found for the states),
HCFA
counsel notified the State in March 1987 that it was willing to
discuss
the issues and staff meetings followed. When HCFA issued a
new
disallowance in January 1988, HCFA indicated that it was no
longer
pursuing its original findings on three facilities, but was pursuing
the
findings on Beverly Manor and Phoenix Center.
Some delays did occur in this process, but they were not
unreasonable
considering the complexity of the issues, the consent of the
State to
major delays, the impact of related cases, and the fact that, as
a
result of the related cases and HCFA's changes in its guidelines,
the
issues have been substantially clarified and narrowed.
Moreover, the State did not show that it was prejudiced by the passage
of
time. The State argued that witnesses who might have information
on
Phoenix Center "may not be available for cross-examination at this
late
date or "if they can be located through investigation and effort,
their
memory will have faded over the years." State's .Brief, Dkt. No.
88-34,
p. 7. The State made a similar argument with respect to Beverly
Manor
and also provided an affidavit attesting that Beverly Manor
patient
records had been purged and cannot be reviewed. See Dkt. No.
88-34, Ex.
G.
The record shows, however, that the State had an opportunity during
prior
Board proceedings to identify any witnesses it wished to present
at the
hearing scheduled for March 1983 (and later postponed). The
State did
not indicate that it wished to present any witnesses other
than experts on
what diagnoses should be considered mental diseases and,
possibly, the
auditors (if the parties could not agree on daily census
figures or the State
otherwise wished to challenge the credibility of
the audit findings).
Moreover, the State did not allege here that it
had made any attempt to
locate potential witnesses, in spite of the fact
that many of the potential
witnesses to which the State referred were
State or county employees.
We also note that the statements of facility
personnel and others interviewed
by the auditors are not necessary to
our decision on Phoenix Center (although
we do not consider such
statements to be entirely unreliable, as the State
suggested).
Similarly, there is no indication in the record that the State sought
to
obtain medical records from Beverly Manor prior to 1988 although
the
auditors had at the State's request provided a list of patients
included
in the audit statistics.
The State implied that it would need to question now unavailable
witnesses
from the Colorado Bureau of Investigation (CBI) to determine
the source and
reliability of the information on "admission
authorization" documents relied
on in the Phoenix Center audit. The
State was able to obtain copies of
the admission authorization forms
from CBI's files, however; these forms
contain information on both
diagnoses and treatment, and their accuracy was
certified by the
attending physicians. Moreover, these forms were
originally prepared
for the Colorado Department of Social Services
itself. Dkt. No. 88-34,
Exs. N, O. Thus, the State had available
more information and a better
means of evaluating that information than the
State suggested.
Finally, we reject the State's argument that it was prejudiced by
the
changes in HCFA guidelines over the years and could not have known
at
the time of the earlier Board proceedings what evidence HCFA
would
ultimately consider relevant to IMD status. If anything, the
State
benefited from the refinement of HCFA's guidelines (for example,
from
the deletion of certain diagnoses). In any event, we are not
applying
HCFA's guidelines as a standard the State had to meet; rather,
as
discussed above, the test is the regulatory definition. The State
had
timely notice of that definition and should have known that it
should
preserve any evidence relevant to the overall character of
the
facilities.
We note, too, that although the State is subject to
recordkeeping
requirements applicable to the Medicaid program, HCFA did not
argue here
that the State was required to have affirmative documentation that
these
facilities were not IMDs during the disallowance period. Rather,
HCFA
identified the issue as whether the evidence of record establishes
that
Beverly Manor and Phoenix Center were, during the period in
question,
IMDs under the regulatory definition. HCFA's Brief, Dkt. No.
88-34, p.
4. The State has known about HCFA's evidence since HCFA
issued the
audit reports and has had adequate opportunity to rebut that
evidence
and to state why it does not show that the facilities were IMDs.
II. Facility Specific Issues
a. Beverly Manor Nursing Home
Beverly Manor Nursing Home (Beverly Manor) was licensed by Colorado as
a
110-bed intermediate care facility during the disputed period July
1,
1976 through June 30, 1979. Federal auditors reviewed Beverly
Manor's
claims for payments for services rendered to Medicaid patients,
patient
medical records maintained by the facility, and the facility's
written
admission policies. The auditors interviewed the Acting
Administrator
of the facility. The auditors also reviewed the files of
the Colorado
Department of Health for correspondence, interoffice memoranda,
and
other information related to Colorado nursing homes. Dkt. No.
82-125,
Ex. E, p. 2..The auditors concluded that Beverly Manor's
overall
character was that of an institution primarily engaged in the care
and
treatment of persons with mental diseases. The auditors' conclusion
was
based on six major findings: (1) mental diseases were the
primary
diagnoses of 76% of all patients, including from 54 to 71% of
Medicaid
patients on a daily basis; (2) 56 of the facility's 110 beds were in
a
locked unit; (3) at least 57% of the Medicaid patients had
previously
been inpatients at State mental institutions or hospital
psychiatric
wards; (4) the staff included a consulting clinical psychologist,
a
psychiatric nursing staff, and mental health workers; (5)
admissions
policies and correspondence with the Colorado Department of
Health
indicated that the facility was geared toward serving patients
with
mental diseases; and (6) 72% of the Medicaid patients were under age
65
when admitted. Id., pp. 3-4.
As we discuss below, we find that the record does not establish that
the
facility met the regulatory standard of "primarily engaged in
providing
diagnosis, treatment or care of persons with mental diseases . . .
."
42 C.F.R. 435.1009. In particular, we find that, while the
patient
population statistics and other evidence show that the facility
served
some patients with mental diseases, it does not establish that
the
facility primarily served such patients.
1. Do patient population statistics
indicate
that Beverly Manor was an IMD?
Colorado contended that the auditors' patient statistics were not
accurate
and, thus, were not probative of the facility's status. We
examined the
evidence in the record in light of HCFA's current
guidelines on classifying
patients. Section 4390, Dkt. No. 88-34, Ex.
B. While it is clear
that the facility served some mentally diseased
patients, the evidence simply
does not support a finding that Beverly
Manor primarily served mentally
diseased patients.
We reach this conclusion based on the following:
o The percentage of mentally diseased patients found by the
auditors
was substantially lower than 50% when required adjustments
in
classifying patients are made. HCFA's current guidelines in
section
4390 of the State Medicaid Manual state that organic brain syndrome
and
senility should not be considered mental diseases, and that
patients
with alcoholism should be classified as mentally diseased only when
the
treatment methods indicate treatment as a mental disease (see
the
discussion of the Granville case above). When patients with
these
primary diagnoses are reclassified, there were no days during
the
disallowance period on which the percentage of patients with
mental
diseases as a primary diagnosis exceeded 40% of the total
patient
population. Dkt. No. 82-125, Exs. H, 7. In view of
the lack of any
evidence of alcoholism treatment methods in the record, we
find that the
record does not support classifying as mentally diseased
those patients
with a primary diagnosis of alcoholism at this
facility. (See our
discussion of the treatment evidence below.)
o Even if patients with a primary diagnosis of alcoholism were
included
as mentally diseased, fewer than half of the resident patients had
a
primary diagnosis of a mental disease on 459 out of the 1095 days in
the
period. The percentage was higher on the remaining days, but in
this
facility a few patients could have caused this percentage
increase.
Since there is no evidence that the facility remained above 50% for
a
substantial, continuous period of time, we must assume that
it
fluctuated around the 50% mark throughout the period. In
past
decisions, the Board recognized the difficulty in relying on
patient
population statistics close to the 50% mark; an ICF may properly
treat
some patients with mental diseases, and it is difficult for a state
to
determine at what point the facility would attain the overall
character
of an IMD. See Washington Dept. of Social Services, DGAB No.
785, pp.
9-10.
o The audit team included no medically qualified personnel and thus
the
classification of patients as mentally diseased does not carry as
much
weight as a medical evaluation. While participation of
medically
qualified personnel is not a requirement, as HCFA suggests in its
1987
Financial Management Review Guide, the Board has indicated in
prior
decisions that the qualifications of the audit team may be
significant
in evaluating the accuracy of patient classifications. See,
e.g.,
Minnesota Dept. of Human Services, DGAB No. 831 (1987).
o The auditors looked simply at the primary diagnosis of each
patient,
not at the reason for institutionalization when a patient had
been
diagnosed with both physical and mental disorders, as HCFA's
own
guidelines require. See Dkt. No. 88-34, Ex. B.
Thus, we find that the patient population statistics do not support
the
conclusion that Beverly Manor was an IMD during the disputed
time
period, since on each day less than half of the resident patients
were
mentally diseased. HCFA argued that a facility could have the
overall
character of an IMD even if less than half of the patients were
mentally
diseased, if other factors indicated such a character. As we
discuss
below, we find that the other evidence does not support that
conclusion
in this case.
2. Do other factors indicate that Beverly
Manor
was an IMD?
The evidence related to factors other than patient population
indicates
only that the facility provided a program of psychiatric services
to
some patients, not that it was primarily engaged in providing care
and
treatment to the mentally diseased.
HCFA relied principally on issuances by the facility and a
memorandum
prepared in the licensing and certification process, all of
which
indicated that the facility had a psychiatric program for the
treatment
of some patients. Dkt. No. 82-125, Ex. 1; Dkt. No. 88-34, Ex.
M. But
these documents did not state that these were the only type of
patients
served by the facility, or even that the facility primarily served
those
patients. In sum, these documents indicated only that the
facility was
capable of handling mentally diseased patients and sought to
obtain
referrals of such patients.
For example, while the facility issuances indicated that the facility
had
a psychiatric program, they did not state that this was the primary
treatment
program. These issuances referred specifically to the "locked
North
Unit program," and the description of that program does not
indicate that it
extended to the facility as a whole. Dkt. No. 82-125,
Ex. 1
(attachment). While it is clear that the North Unit was
primarily
involved in the treatment of mentally diseased patients, there is
no
basis to assess the overall character of the facility from the
operation
of one unit.
Similarly, the inter-office communication from the Colorado Department
of
Health, Division of Medical Care Licensing and Certification cited by
HCFA
did not provide sufficient information to support a finding that
the facility
was an IMD. Dkt. No. 88-34, Ex. M. Although the
memorandum stated
that the facility had "concentrated" on patients of
all ages with psychiatric
conditions for the last three years and that
it was "one of two [facilities]
catering to the mentally ill," it is
undisputed that the facility provided
services to the mentally ill. Id.
This assessment did not establish
that the facility primarily served
mentally diseased patients.
Moreover, these documents were prepared for purposes which limit
their
probative value. The issuances from the facility appear to have
been
directed at attracting referrals of this type of patient, and there
is
no reason why they should have listed other capabilities of
the
facility, or other types of patients served. The State's memorandum
was
not directed at the question of the facility's status as a possible
IMD;
the communication was directed at complaints about the facility
from
neighboring residences and about possible violations of patients'
rights
and a lack of supervision for some patients.
HCFA also relied on the presence of some staff trained in mental
health.
As HCFA's own guidelines recognize, the mere presence of staff
members
with mental health training does not render a facility an IMD;
the
relevant question not addressed in this record is whether the number
of
mental health professionals was unusual for an intermediate
care
facility. Dkt. No. 88-34, Ex. B. The auditors found merely
that the
facility employed a psychiatric nursing staff, two full-time
mental
health workers and a consulting clinical psychologist who
apparently
visited weekly. See Dkt. No. 82-125, Ex. E, p. 11; Dkt No.
82-125, Ex.
1. Furthermore, the record does not indicate whether these
employees
served the facility as a whole, or only the psychiatric program at
the
facility.
The other evidence upon which HCFA relied is not probative of
IMD
status. The affiliation agreement with a community mental health
center
is not probative absent evidence that such an agreement was unusual
for
an intermediate care facility which was not an IMD. Similarly, the
mere
presence of locked wards does not indicate that the facility was an
IMD.
A high proportion of beds in locked wards might be an indication of
IMD
status if other factors were present to confirm the indication,
but
there are other types of patients, such as those with organic
brain
syndrome or Alzheimer's disease, for which locked wards can be
used.
Although HCFA argued that the proportion of Beverly Manor beds in
the
locked units (51%) was far in excess of what could be justified by
the
numbers of patients who appear to have diseases such as organic
brain
syndrome or Alzheimer's disease, it is not clear whether the locked
unit
was used exclusively for mentally diseased patients, or even whether
the
locked unit was fully occupied.
The other patient characteristics cited by HCFA are not, by
themselves,
useful factors in determining IMD status. Although the
patients had a
young average age, there may be valid reasons other than
mental disease
why patients younger than 65 might require long term care, and
a
facility may justifiably accept younger patients and even specialize
in
them. See Washington Dept. of Social Services, DGAB No. 785
(1986).
Similarly, the fact that 57% of the Medicaid patients with
primary
diagnoses of mental diseases (37% of total patients) had previously
been
inpatients at State mental institutions or at psychiatric wards of
local
hospitals is also not an indication of IMD status absent
supporting
evidence of the reason why these patients were referred to
Beverly
Manor. Certainly, if the State was trying to avoid its
traditional
responsibility for IMD care by dumping patients out of State
mental
hospitals into Beverly Manor, or if Beverly Manor was being used as
an
alternative mental institution, that would be highly relevant. But
the
evidence does not establish those conditions here.
In sum, the evidence other than patient population statistics presented
by
HCFA clearly confirms that Beverly Manor provided some services of
a
psychiatric nature. But the evidence does not show that Beverly
Manor
had the overall character of an IMD.
Accordingly, we reverse the disallowance of $551,155 claimed for
services
provided to Beverly Manor residents during the audit period.
b. Phoenix Center
Phoenix Center was licensed by Colorado as a 160-bed intermediate
care
facility during the disputed period June 30, 1976 to June 30, 1978
(and
through March 30, 1977 as a 9-bed residential care facility). In
June
1978, Phoenix Center withdrew from the Medicaid program and closed
due
to a bankruptcy action. Federal auditors, from August 1980 to
February
1981, reviewed Phoenix Center's claims for payments for
services
rendered to Medicaid patients, admission authorizations for
Medicaid
patients, correspondence in State and county files concerning
Phoenix
Center, Phoenix Center's personnel policy, newspaper advertisements,
and
other public descriptions of the facility, and a listing of 126
patients
transferred to other facilities by the Jefferson County Mental
Health
Center when Phoenix Center closed. This listing included the
patients'
diagnoses. The auditors also interviewed county health
officials and a
former mental health worker at Phoenix Center.
The auditors concluded that Phoenix Center's overall
characteristics
demonstrated that it was primarily engaged in the care and
treatment of
persons with mental diseases. The auditors' conclusion was
based on six
major findings: (1) mental diseases were the primary diagnoses
of 88% of
all patients listed on the admissions authorizations, 81% of
the
patients transferred by Jefferson County Mental Health Center
when
Phoenix Center closed, and 87% of the patients listed on Medicaid
claim
forms; (2) some of the facility's beds were in a locked unit; (3)
many
of the patients had previously been inpatients at State
mental
institutions or hospital psychiatric wards; (4) the facility had
a
placement agreement with the Jefferson County Mental Health Center
to
house mentally diseased patients, and employed a staff psychiatrist
and
mental health workers; (5) the admissions policies and
correspondence
with the Colorado Department of Health indicated that the
facility
specialized in patients with a primary psychiatric diagnosis and
should
properly be considered a psychiatric treatment facility; (6)
the
facility advertised itself as a psychiatric facility and publicly
stated
that all patients were in need of treatment of a psychiatric nature;
and
(7) 96% of admission authorizations and 91% of Medicaid claims were
for
patients under age 65.
As we discuss below, we find that the evidence in the record compels
the
conclusion that Phoenix Center had the overall character of an IMD
under
any reasonable reading of the definition of an IMD. While we
agree with
Colorado that some of the factors considered by the auditors,
taken
separately, would not necessarily indicate that the facility was an
IMD,
we find the cumulative evidence to be overwhelming.
1. Do the patient population
statistics
indicate that Phoenix Center was an IMD?
Colorado argued that the population statistics on which HCFA relied
were
inaccurate because they were calculated from records which were
not
reliable. Colorado alleged that at least half of the
admission
authorizations the auditors relied on were signed on July 1, 1977,
for
admissions as early as 1969. Colorado pointed out that neither this
set
of records nor the records on Medicaid claims contained information
on
every patient at the facility during the disputed time period, and
that
statistics based on the admission authorizations may be skewed
by
multiple admissions of mentally diseased patients. Colorado
also
alleged that the relocation list may indicate a high percentage
of
mentally diseased patients because "immediately prior to closure
a
facility may experience a heavier concentration of patients,
including
psychiatric patients, who are more difficult to place in
other
facilities, assuming that the more 'typical' patients,
without
behavioral problems, perhaps with more resources, are able to
anticipate
the closure with earlier relocation." Dkt. No. 88-34,
State's Brief, p.
16.
.Despite its general assertion that the records used were not as
accurate
as patient medical records, Colorado did not clearly explain
why the
particular records the auditors examined would necessarily
be
unreliable. Both the admissions authorizations and the relocation
lists
were based on professional medical evaluations. Authorization on
the
basis of a diagnosis of a mental disease means that a
medical
professional has determined that institutional placement is
justified
because of the patients' mental disease. Thus, it is
reasonable to
infer that the diagnosis was current at the time of the
admission, and
reflects on the nature of the facility in which the recipient
is placed.
Similarly, the relocation list was prepared by medical
professionals at
Jefferson County Mental Health Center at the time each
patient was
evaluated for relocation, and indicates that the mental diseases
were
ongoing. In our view, the admission authorizations may be as
probative
of IMD status as an examination of the diagnoses listed in
patients'
medical records, which might mistakenly classify as mentally
diseased a
patient with a historical diagnosis of mental disease who was
being
treated for something else. Admission authorizations are a type
of
document mandated by federal requirements that a state have a
system
designed to ensure that Medicaid recipients are placed in ICFs (or
other
institutional facilities) only if they need that level of care.
See
section 1902(a)(30) of the Act.
While the admission authorization and Medicaid claims records
pertained
only to Medicaid patients, and did not include private pay
patients, the
State did not allege that the Medicaid patients were not
representative
of the patient population as a whole, and we have no reason to
think
they were not. We also note that, although the State had copies
of the
admission authorizations, it pointed only to instances of
multiple
admissions of mentally diseased patients and provided no analysis
to
show that multiple admissions did not occur at the same rate
for
physically diseased patients in this facility.
.Even after considering the possibility of some inaccuracy, however,
we
find that the statistics support a determination that the facility
was
an IMD. For instance, as Colorado conceded, even when the
percentage of
mentally diseased patients calculated from admission
authorizations is
reduced by eliminating patients with primary diagnoses such
as
chronic/organic brain syndrome, alcoholism, Korsakoff's Syndrome
(which
is alcoholism-related) and the category labelled "other mental
diseases"
(this was not defined by the auditors), the remaining percentage
of
mentally diseased patients is still well above 50% of the total
patient
population reviewed. Dkt. No. 88-34, State's Brief (the exact
figure
would be 56%).
Moreover, we do not think all of the reductions proposed by the State
are
warranted. In light of the evidence discussed below that the
facility's
treatment approach to all patients was psychiatric in nature,
the alcoholic
patients should be counted as mentally diseased under the
standard that the
State itself advocated, i.e., the nature of the
services provided.
Including these patients would increase the
percentage of mentally diseased
patients to 64% of the total Medicaid
admissions. Also, we see no
reason to exclude the additional 6% of
patients with "other mental diseases"
simply because the auditors did
not define this category; the auditors
explained that they used the ICD
classifications, and Colorado had access to
the admission authorization
forms used by the auditors.
The State's challenge to the relocation lists is without merit.
Even
assuming that the State is correct as a general rule that a
facility
which was about to close would likely have more patients with
mental
diseases, the State provided no evidence that that was in fact the
case
here. Indeed, the record indicates that the facility owner
had
requested that the facility admit no more psychiatric patients and
start
admitting geriatric patients a few months prior to the date the
facility
closed. Dkt. No. 88-34, Ex. 19, unnumbered p. 9.
Colorado's objections to the statistical evidence were founded in part
on
differences between the source materials and procedures used here and
those
described in the Financial Management Review Guide section HCFA
issued in
April 1987, included in Docket No. 88-34 as Exhibit C. We
agree that
statistics gathered in accordance with HCFA's new guidelines
would have
greater weight, but this does not mean that we must totally
discount the
statistics here.
In sum, we find the patient population statistics support
the
determination that the facility was primarily engaged in
treating
mentally diseased patients. Furthermore, as we discuss below,
we find
that these statistics are consistent with other evidence in the
record
and that the weight of this evidence indicates that the statistics,
if
anything, understated the extent to which patients were placed in
the
facility for treatment of mental diseases.
2. Do other factors indicate that
Phoenix
Center was an IMD?
The record contains evidence related to factors (other than
patient
population statistics) on which HCFA relied in determining that
Phoenix
Center was an IMD. For the most part, Colorado did not dispute
the
presence of these factors, but argued that these factors should not
be
given much weight in determining IMD status. We list here those
factors
which we find most probative of the status of Phoenix Center, and
then
we discuss Colorado's objections.
The auditors found evidence that Phoenix Center considered itself to be
a
psychiatric treatment center and held itself out as such. The
minutes
of the October 26, 1976 meeting of the facility's advisory board
contain
the comments of one participant who stated that the facility was
the
second largest psychiatric facility in the state, yet it still only
had
a nursing home license. Dkt. No. 88-34, Ex. 16. A letter
dated
December 9, 1976 from the Administrator of Phoenix Center to the
federal
Regional Consultant for Psychiatry stated:
Phoenix Center is licensed . . . as an intermediate care
nursing home, and is therefore receiving Medicaid nursing home
rates. However, in reality, Phoenix Center is the second
largest
psychiatric treatment center in the State of Colorado.
An active
treatment program is designed and implemented for
each of the 160
residents . . . . In other words, Phoenix
Center is providing
psychiatric treatment not unlike that
provided by Fort Logan and
Colorado State Hospital. . . . I
would like to see Phoenix
Center reclassified as a mental
health treatment center or
supplemented by the Department of
Institutions for the type of care
Phoenix has to offer. . . .
Dkt No. 88-34, Ex. 14.
.Phoenix Center's admission procedures, both standard and
emergency,
stated that "all admissions must . . . have a primary diagnosis of
a
psychiatric nature." Dkt No. 88-34, Ex. 10 (emphasis added).
The
facility's personnel manual stated that the facility "specializes
in
patients with a primary psychiatric diagnosis." Dkt No. 88-34, Ex.
9.
It advertised in newspapers and telephone listings using the
caption
"psychiatric care" with no reference to any other type of care
provided.
Dkt No. 88-34, Ex. 12. The record also contains a brochure in
which the
facility stated that "its primary focus is psychiatric in nature"
and in
which the facility described the availability of Mental Health
Workers
who serve "as primary therapist for the resident." Dkt. No.
88-34, Ex.
13 (emphasis added).
The auditors also found contemporary evidence that State
officials
considered Phoenix Center to be an IMD. In a survey by
Colorado's
Division of Professional Standards Review, dated March 8, 1977,
the
surveyor noted that most of the patients have "psychiatric
or
neurologically related diagnoses" and stated that the current
patients
"are not true ICF patients." Dkt. No. 88-34, Ex. 10. A
letter dated
May 16, 1978 from the Colorado Department of Institutions,
Division of
Mental Health, to the appellant stated that Phoenix Center "is
illegally
receiving payments under Title XIX because of the fifty percent
rule
(i.e., not more than 50 percent of the residents can be
psychiatric
patients). . . . Phoenix Center appears to be meeting a
need for the
care of the chronically mentally ill who do not require
hospitalization
but who need supervised care. . . ." Dkt. No. 88-34,
Ex. 18.
The notes of a March 24, 1978 meeting of Colorado Department of
Health
officials indicate that the State was aware that many patients
at
Phoenix Center were coming from surrounding State mental hospitals
and
community mental health centers and that the facility had an
environment
not suited to geriatric patients. The notes contain a
comment that
"[s]ome of the patients are too ill to be in a nursing home and
belong
in a psychiatric hospital" and describe the effect of the lack
of
therapy on "these psychiatric residents." Dkt No. 88-34, Ex. 17.
We do not base our decision here on evidence in the record such as
the
existence of a locked ward, staff qualifications, the age of
the
patients, or the affiliation agreement with a community mental
health
center, for the same reasons we expressed in connection with these
types
of evidence in discussing Beverly Manor. Nor do we rely on
statements
made to auditors by a former mental health worker or county
employees,
since these are not as reliable as the facility's
contemporaneous
records, and since these are not necessary to our
decision. We note,
however, that none of this evidence is
inconsistent with the conclusion
that Phoenix Center was an IMD during the
disallowance period.
Colorado argued that the facility's own advertisements and the
statements
of its administrator should not be considered reliable
indications of the
facility's status, since those advertisements and
statements were
self-serving attempts to attract patients from the
marketplace and to bolster
the standing of the facility, which was
troubled. Colorado also pointed
out that the administrator was under
investigation for fraud at the time.
We recognize that the facility's own advertisements may not always
be
reliable indications of the overall character of the facility, but
we
find that, when supported by business records such as facility
advisory
board minutes and correspondence with State and federal agencies,
they
can be probative. Furthermore, the mere fact that the
administrator was
under investigation for fraud does not draw into question
the specific
statements which we find to be probative here. Nothing in
the record
indicates that the character of the facility was a significant
element
of any allegations of fraud.
In sum, we consider the evidence cumulatively and conclude that there
is
compelling evidence that Phoenix Center was an IMD within any
reasonable
reading of the regulatory definition. We base our conclusion
on the
patient data coupled with contemporaneous characterization of
the
facility by the State and the facility itself, expressed in
business
records and public advertisements.
Conclusion
For the reasons discussed above, we reverse the disallowance of
$551,155
in FFP for services provided by .Beverly Manor Nursing Home and
uphold
the disallowance of $750,514 in FFP for services provided by
Phoenix
Center.
________________________________ Cecilia Sparks Ford
________________________________ Norval D. (John) Settle
________________________________ Judith A. Ballard
Presiding
Board