GAB Decision 742
April 15, 1986
California Department of Social Services;
Docket No. 85-172
Ballard, Judith A.; Teitz, Alexander G. Settle, Norval D.
The California Department of Social Services (State) appealed a
decision by the Office of Human Development Services (Agency) that the
State was ineligible for fiscal year 1983 funds under section 427 of
title IV-B of the Social Security Act (Act). Section 427 of the Act
provides that a state may receive additional funds for child welfare
services, beyond the amount available under section 420 of the Act, if
the state meets certain requirements for protecting children in foster
care. The Agency initially approved the State's request for additional
funds for fiscal year 1983 based on a written certification by the State
that it met the requirements of section 427. Subsequently, the Agency
sought to perform a compliance review to validate the State's
self-certification. After determining that the State had established
policies and procedures for implementing the 427 requirements, the
Agency intended to survey a sample of foster care case records to
determine whether these policies and procedures were operational. These
case records were maintained in 19 county offices providing foster care
services in the State. When the State refused the Agency's request to
make a sample of 200 foster care case records /1/ available at no more
than three(2) centrally located sites within the State, the Agency found
the State ineligible for the fiscal year 1983 funds on the ground that
it was unable to document the State's eligibility. As discussed below,
we find that the Agency reasonably required the State to move the case
records to a limited number of sites for purposes of the case record
survey, and, accordingly, uphold the Agency's determination that the
State was ineligible for the section 427 funds on the basis of the
State's failure to comply with this requirement.
.
Issue Presented
The Agency relied on section 422(b)(8) of the Act for authority to
require
the State to move some of its case records, from the counties in
which they
are maintained, for purposes of the case record survey. That
section
provides:
Each plan for child welfare services under this part shall . .
.
provide that the (state) agency administering or supervising
the
administration of the plan will furnish such reports, containing
such
information, and participate in such evaluations as the Secretary
may
require.
The Agency contended that the State had failed to participate in
an
evaluation required by the Secretary. We agree, and the State did
not
dispute, that a case record survey is an evaluation authorized
by
section 422(b)(8). This does not mean, however, that the Agency
may
lawfully impose any requirement it wishes in connection with the
case
record survey. Even a requirement imposed by the Agency in an
area
where it has broad discretion is subject to review. If the
Agency
exercised its authority under section 422(b)(8) reasonably in
requiring
that the State make foster care case records available at no more
than
three sites for purposes of the case record survey, however, the
Board
will not substitute its judgment for that of the Agency. (See
Charles
River Health Care Foundation, Decision No. 252, February 8, 1982,
p.
38)
Agency's Position
The Agency contended that unless the foster care case records were
made
available at no more than three sites, it could not use the
sampling
methodology which it had chosen to employ for(3) the case record
survey.
In Program Instruction 82-06, dated June 3, 1982, the Agency
indicated
that the case record review (survey) would be conducted using
a
sequential sampling technique. (PI 82-06, p. 3) /2/ In a decision
which
recognized the validity of sequential sampling but found that the
Agency
did not use it properly in that case, the Board described the
Agency's
sampling methodology for the section 427 program as follows:
In determining whether a state is satisfactorily operating a
case
review system, the Agency samples part of the state's foster care
case
records. The purpose of sampling is to avoid the time and expense
of
evaluating all the foster care cases in that fiscal year,
by
mathematically extrapolating performance in the universe of cases
from
an evaluation of the sample.
For section 427 reviews, the Agency uses . . . a special type
of
sampling known as "sequential sampling." Sequential sampling
differs
from standard sampling in that the sample size is not fixed, but
rather
is determined by the results of observed cases as they
accumulate. Its
goal is to allow the sampling of a smaller number of
cases than the
fixed sample size of standard sampling. Sequential
sampling thus
creates further cost savings for the Agency and aids the states
by
avoiding the need to present large numbers of foster care case
records.
Using this methodology, the reviewers start sampling a
collection of
records chosen in a scientifically random manner. The
Agency uses a
prepared "decision table" that prescribes how many cases of the
total
sampled must have failed in order to conclude that a state fails
the(4)
section 427 review, as well as how many cases of the total sampled
thus
far must have passed in order for the Agency to stop sampling
and
conclude that a state is eligible for section 427 funding.
Delaware Department of Services for Children, Youth and Their
Families,
Decision No. 699, November 1, 1985, pp. 2-3.
In the instant case, the Agency contended that if the case
records
remained at the 19 sites at which they are presently maintained,
it
would not be feasible to use sequential sampling. (Agency's brief
dated
January 31, 1986, p. 4) If all 200 records in the sample were
collected
at one site, they would be reviewed one at a time and the
results
entered into a decision table in sequential order until a
determination
could be made that no further review was necessary.
However, according
to the Agency, if the records were located at more than
one site, the
review team at each site would have to begin reviewing case
records
simultaneously, telephoning their results to a single source,
and
receiving instructions concerning whether to continue reading
records.
The Agency indicated that if the results from one site were
reported
faster than the results from another site, the State might
erroneously
be found ineligible if the earlier-reporting site had a lower
rate of
compliance with the section 427 requirements. In addition, the
Agency
contended that the reliability of the review would be decreased if
the
records were at multiple sites, since two reviewers would be required
at
each site, as compared with six reviewers at one site. The decrease
in
reliability would presumably be due to differences among reviewers
in
interpreting the case records or in applying the section
427
requirements in each case. In the Agency's view, these
difficulties
justified its decision to limit the number of sites at which
records
would be reviewed to no more than three. (Agency's brief dated
January
31, 1986, Tab A, p. 3) The Agency also asserted that "the
Regional
Office had neither the manpower nor the travel funds to travel to
each
county which had cases in the sample." (Id., Tab C, letter dated
January
17, 1986, p. 1)
State's Position
In response to the Agency's identification of the technical
difficulties
discussed above, the State suggested that case records from the
smaller
counties could be reviewed out of sequence and recorded only when
they
were slated to be counted, thus avoiding the bias introduced
by
reporting results early from a county with a lower rate of
compliance
with the section 427 requirements. (State's reply brief,
dated March
21, 1986, p. 6) The State did not deny that using 38 reviewers (2
at
each of 19 sites) could result in a lack of consistency in how
records
were read. (5)
The State took the position, however, that even assuming that
sequential
sampling at 19 sites was not feasible, the Agency was not
justified in
finding the State ineligible based on its refusal to move case
records
since a standard sampling technique which would not have
required
records to be moved could have been used instead. The
State
acknowledged that standard sampling would be more costly
and
time-consuming than sequential sampling. (State's reply brief
dated
March 21, 1986, p. 4) It argued, however, that the potential risk to
the
State in moving the case records outweighed the considerations
of
administrative economy which prompted the Agency's decision to
use
sequential sampling. The primary risk which the State saw in moving
its
case records was that services to the foster care children whose
records
were moved might be interrupted or delayed or worse. /3/ Examples
given
by the State of problems resulting from the absence of these
records
included a delay in taking action on a placement, a
detrimental
placement, a mix-up in assistance provided because of a change in
social
workers, a missed medical treatment, continuance of a trial date, or
a
missed court date resulting in loss of personal jurisdiction over
the
child. (Id., p. 5)
.
The State also speculated that records might be lost in transit.
(State's
brief dated November 26, 1986, p. 6) In addition, the State
stated that it
would be difficult for it to preserve the confidentiality
of case records
moved to another location. (Id.)
Discussion
We are not persuaded that the concerns expressed by the State
regarding
the effect on foster care services of absent case records
are
sufficiently serious to render the requirement for moving the
records
unreasonable. The Agency indicated that the case records would
not be
out of the county offices more than 7 to 10 days if the review
were
conducted at three sites. (Agency's brief dated January 31, 1986,
p.
7) The State could thus review the case records before they were
moved
and make(6) photocopies, to be retained in the county offices, of
those
portions of the records which it appeared might be needed for
the
short-term (e.g., for a court hearing scheduled in the near
future).
Even if the State wished to take the precaution of copying the
entire
record in each case, in view of its concern that records might be
lost
in transit, the number of cases involved is not so large as to make
this
unduly burdensome. As indicated previously, the Agency stated that
it
would require at most 200 case records for the case record survey.
If
the State selected as the three sites at which the cases would
be
reviewed, the three county offices having the largest number of
case
records, less than 200 case records would have to be moved.
(Since
there is no elaboration in the record regarding the requirement that
the
three sites be "centrally located," we assume that moving the records
to
the three largest county offices would be consistent with
that
requirement.) Moreover, since the concerns expressed by the State
about
interrupted or delayed services apply only to active cases, the
records
for any inactive cases in the sample would not have to be
photocopied.
The State also argued that copying such voluminous amounts of
material
creates the risk that papers will be lost or misfiled.
(State's reply
brief dated March 21, 1986, p. 6) However, if due care were
exercised,
the risk of loss or misfiling as a result of the photocopying
would not
be greater than the risk presented in the ordinary course of
business.
/4/ Moreover, we see no basis for the State's concern regarding
breaches
of confidentiality since the records could be transported in
sealed
boxes and would be reviewed at county offices by the same people
who
would have reviewed them if they had been reviewed at the county
offices
at which they were maintained.
.
Thus, we conclude that the movement of case records to no more than
three
sites would not pose any significant problems for the State.
We
conclude, moreover, that movement of the records was required in
order
to achieve the cost savings that the use of sequential sampling in
lieu
of standard sampling makes possible. The State appears to be
correct
that the potential for improperly finding the State ineligible
because a
site with a lower compliance rate reported its survey results
earlier
could be eliminated simply by recording the results in the
proper
ordered sequence, rather than in the order reported. (This would
be
possible if each case record(7) in the sample is assigned a number in
a
random sequence.) It is likely, however, that this would result in
more
records being reviewed in order to reach a decision
regarding
eligibility or ineligibility than would be necessary if the records
were
reviewed one-at-a-time in the random sequence. This technique
would
also not eliminate the practical difficulties of accumulating reports
by
telephone from 19 different sites. Moreover, we think that the
Agency
has a legitimate concern in attempting to minimize the number
of
reviewers in order to assure consistency of results. While the
same
reviewers could travel to all 19 sites (a procedure the State did
not
suggest), this would clearly be more costly in terms of travel and
the
reviewers' time than if fewer sites were used, and would likely
result
in review of a greater number of records than otherwise necessary.
/5/
Accordingly, the Agency's decision to require the movement of the
case
records for purposes of conducting the case record survey
was
reasonable.
Applicable Regulations and Policy Statements
However, the State argued further that applicable regulations and
policy
statements precluded the Agency from requiring it to move case
records
for purposes of the case record survey. It cited in support of
its
position 45 CFR 74.24, captioned "Access to records," which provides
in
pertinent part:
(a) Records of grantees. The DHHS and the Comptroller
General of the
United States, or any of their authorized representatives,
shall have
the right of access to any books, documents, papers, or other
records of
the grantee which are pertinent to the DHHS grant, in order to
make
audit, examination, excerpts, and transcripts.
The State argued that this regulation, through its use of the
word
"access," contemplated "entry" to conduct a review but not "delivery"
of
records to a different location. (State's reply brief, dated March
21,
1986, p. 2) However, nothing in the cited regulation precludes
a
requirement for the delivery of records under some other authority.
As
discussed(8) previously, section 442(b) (8) of the Act gives
the
Secretary broad authority to conduct evaluations, which in this
case
reasonably encompasses a requirement to move case records from the
site
at which they are ordinarily maintained.
The State also contended that PI 82-06 expressed "a federal policy
that
movement of . . . (the) case records is discretionary with the
State."
(State's reply brief dated March 21, 1986, p. 3) That
program
instruction provides in pertinent part:
The State should be encouraged to assemble the case records in
a
central location for review because the sampling procedure requires
that
they be read in the same order they are selected, regardless of
what
region of the State they are from. (If this sampling method is
chosen,
but the State cannot assemble cases in one location, the Regional
staff
should consult with the Central Office Children's Bureau.)
(PI 82-06, p. 3) We are not persuaded that this states a policy to
allow
the states to determine the number of sites at which records will
be
reviewed. While this language indicates that states will not
be
required to assemble all case records at one location, it clearly
leaves
open the question of how many additional sites will be
permitted.
Moreover, Attachment E of PI 82-06 states:
The least burdensome and most efficient review procedure is
achieved
through the assembly of the entire random sample of case records at
a
single site. This may not be possible in some states and it will
be
necessary to use two or three sites. (p. 5)
Thus, the Agency's decision to permit the State to have the
records
reviewed at no more than three sites is consistent with the
program
instruction.
Single Audit Act
The State also argued that the requirement to move the case records
for
purposes of the case record review was inconsistent with the
Single
Audit Act of 1984, 31 U.S.C. 7501, et seq., which it asserted
"reflects
the legislative intention that the responsibility for audit costs
shift
in the direction of respondent, and not vice versa." (State's
brief
dated November 26, 1985, p. 7) In the instant case, although the
State
could have claimed 50% of the cost of moving (and copying) the
case
records as an administrative cost under(9) title IV-B of the Act,
the
remaining cost would have to be paid from State funds. The
legislation
referred to by the State provides in pertinent part:
A Federal agency that performs or contracts for audits in
addition to
the audits conducted by recipients pursuant to this chapter
shall,
consistent with other applicable law, arrange for funding the cost
of
such additional audits. Such additional audits include . . .
program
evaluations.
31 U.S.C. 7503(e). The Agency argued, however, that this does
not
require that a federal agency pay the entire cost of an audit other
than
audits performed under the Single Audit Act, but merely that it pay
the
maximum it can pay "consistent with other applicable law."
(Agency's
brief, dated January 31, 1986, p. 9) We need not decide which is
the
correct interpretation, however, since, as the State recognized,
this
provision applies to audits with respect to a state or
local
government's fiscal years after December 31, 1984 (31 U.S.C.
7507(a)),
and is thus not applicable to the instant case, which involves a
review
of the State's compliance with the section 427 requirements in
fiscal
year 1983.
Compliance Dispute
The State also alleged that this case involved a compliance dispute,
and
was therefore governed by 45 CFR 201.6, which implements section
1396c
of 42 U.S.C. (State's letter dated August 27, 1985, p. 2) The
Board
does not have jurisdiction over compliance disputes. Section
201.6
provides in pertinent part:
Further payments to a State are withheld in whole or in part if
the
Administrator, after reasonable notice and opportunity for hearing
to
the State agency administering or supervising the administration of
an
approved plan, finds:
(1) That the plan no longer complies with the provisions of the
. .
. Act; or
(2) That in the administration of the plan there is failure to
comply
substantially with any such provision.
A question of noncompliance . . . may arise from . . . the
failure of
a State in practice to comply with a Federal requirement. . .
.
* * *(10)$TIf the Administrator makes a finding of
non-compliance .
. . the State agency is notified that further payments will
not be made
to the State . . . until the Administrator is satisfied that
there will
no longer be any such failure to comply.
45 CFR 201.6(a) and (e). In a letter to the parties, the Board
noted
that section 201.6 provides for the withholding of further payments to
a
state whereas the Agency here seeks to recover funds from the State,
and
asked the State on what basis it maintained that the case involved
a
compliance dispute. (Letter dated September 6, 1985, p. 2) In
response,
the State stated merely that this was a compliance dispute "(since)
the
sanctions were imposed for noncompliance with the requirement to
move
files for the purpose of case reviews. . . ." (State's brief
dated
November 26, 1985, p. 8)
We find, in the absence of any cogent explanation from the State,
that
this case is not properly considered as involving a compliance
dispute
since the Agency seeks to recover funds spent by the State prior to
the
determination appealed from. Moreover, even if the nature of
the
sanction is disregarded, the Agency could reasonably have
determined
that the substantial noncompliance necessary to trigger the
provisions
of 45 CFR 201.6 was not present in this case. Although the
Agency seeks
to recover all section 427 funds for the fiscal year in
question, this
is based on the State's noncompliance with respect to a highly
focused
procedural matter that does not implicate in any way the
State's
administration of the section 427 program. (See
Massachusetts
Department of Public Welfare, Decision No. 438, May 31, 1983)
Thus, this
is not a compliance dispute outside the Board's jurisdiction.
/6/
Effect of Decision
The State requested that if the Board decided this case against it,
the
State be afforded an opportunity to comply with the
contested
requirement. (State's reply brief dated March 21,(11) 1986,
p. 7) The
Agency had previously stated that "(if) California were to
promptly
agree to cooperate in the review at no more than 3 sites, ACYF
would
withdraw the disallowance pending the results of its review."
(Agency's
brief dated January 31, 1986, p. 15, n. 5, emphasis added) The
Agency
did not indicate at what point agreement by the State would no longer
be
considered prompt. We see no reason why the State should be
penalized
for pursuing its appeal as provided for by regulation.
Accordingly, if
the State notifies the Agency within seven days of its
receipt of this
decision that it intends to move the case records as required
by the
Agency, the disallowance would appropriately be withdrawn. This
would
not preclude the Agency from issuing a new disallowance if it
found,
based on the case record survey, that the State was ineligible for
the
section 427 funds.
Conclusion
For the foregoing reasons, we conclude that the Agency reasonably
required
the State to move a sample of its foster care case records to
no more than
three sites for purposes of the case record survey.
Accordingly, we uphold
the Agency's determination that the State was
ineligible for the section 427
funds on the basis of the State's failure
to comply with this
requirement. /1/ Although it appears from
the
record that a sample of 200
case records was selected for review
(Agency's appeal file, Ex. D, p.
1), an Agency official variously
suggested in submissions to the Board that
180 or 200 case records would
be required. (Agency's brief dated
January 31, 1986, Tab A, p. 2;
Agency's letter dated March 3, 1986,
enclosure) The difference is
apparently due to the number of case records
thought desirable for an
"oversample" to be used in the event that the
required sample of 150
case records included some cases that did not meet the
criteria for
inclusion in the sample. An Agency issuance, PI 82-06,
dated June 3,
1982, indicates at Attachment E, p. 5, that an oversample
of only 10
case records is adequate. In view of what PI 82-06 provides,
the Agency
should consider whether an oversample of 50 cases is
necessary.
/2/ The PI provides that if the state conducts its own case
record
review (survey), it may use the sampling methodology of its choice.
(PI
82-06, p. 3) There is no indication in the record here that the
State
wanted to conduct its own review. The PI further provides that
the
Agency will validate the state's review by reviewing 50% or 40 of
the
sample case records reviewed by the State, whichever is higher.
(PI
82-06, Attachment E, p. 5) Since, according to the Agency, 503
records
(419 plus 84 for oversampling) would be required for a standard
sample
(Agency's brief dated January 31, 1986, Tab A, p. 2), the State
would
have to move more records for the validation of a review conducted by
it
using standard sampling than it would if the review was conducted
by
federal reviewers using sequential
sampling. /3/ The
State
indicated that all of the case records were active case records
for
children still in foster care. (State's brief dated November 26,
1986,
p. 4) However, since the case record survey was to include all
children
who entered foster care prior to April 1, 1982 and whose cases were
not
closed by September 30, 1983 (see, Agency's appeal file, Ex. C,
letter
dated April 10, 1984, p. 1), some cases may have been closed by
July
1984, when it had been anticipated that the survey would be
conducted.
(Agency's brief dated January 31, 1986, Tab C, letter dated April
10,
1984, p. 2) /4/ It does
not appear that the State's burden would
be eased if the Agency agreed to
accept photocopies of the case records
rather than the originals. Thus,
there is no reason to question the
reasonableness of the Agency's decision to
accept only originals. (See
Agency's appeal file, Ex. A, p.
2) /5/ The Agency stated that
if
the same reviewers traveled to all 19 sites to review the case
records,
each site would have to have "a proportionate share" of the
random
sample of case records so that, for example, case records from
Los
Angeles County would be randomly distributed to the 18 other
sites.
(Agency's brief dated January 31, 1986, Tab A, p. 3) The case
records
would not have to be moved in this fashion if the State's suggestion
for
recording results in the random sequence were adopted, however.
/6/
The Agency characterized its action as a disallowance as opposed to
a
compliance action. (Agency's brief dated January 31, 1986, p.
11)
However, in our view, the Agency's determination is more
properly
considered as a voiding because the Agency found that the State
was
ineligible for the funds ab initio. (See Ohio Department of
Public
Welfare, Decision No. 472, October 31, 1983, p. 7) In either case,
the
Agency's determination is clearly distinguishable from a
compliance
action since the sanction for both disallowances and voidings
is
retrospective in impact.