Medicare Home Health: Effect on Spending of Limiting Payment for Non-Patient-Care Costs

Published by the Government Accountability Office on 1999-10-19.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

      United States
GAO   General   Accounting
      Washbgton, D.C. 20648

      Health, Education, and

        October 19. 1999

        The Honorable Fortney (Pete) Stark
        Ranking Minority Member
        Subcommittee on Health
        Committee on Ways and Means
        House of Representatives

        Subject:          Medicare Home Health: Effect on Suendinn of Limitina Pavment for Non-
                          Patient-Care Costs

        Dear Mr. Stark:

        Medicare spending for home health care grew from $3.7 billion in 1990 to $17.3 billion in
         1998. Concerned about rising spending, the Congress required the Secretary of Health and
        Human Services to implement an interim payment system (IPS) to control outlays until a
        prospective payment system (PPS) could be put in place on October 1,200O. PPS rates will
        be set so that Medicare expenditures will be equivalent to what would have been spent under
        the IPS if the IPS limits that will be in effect on September 30,2000, were reduced by 15

        Concerned about the substantial portion of home health agency (HHA) costs that were not
        directly related to patient care, you asked us to model the impact of constraining these costs
        through various limits. In this context, we (1) examined the variation in total and non-
        patient-care costs across agencies and (2) estimated the effect on Medicare payments if
        constraints were imposed on payments for non-patient-care costs. To obtain this
        information, we used data on 4,910, or approximately 75 percent, of the free-standing HHAs
        with information in the Health Cam Financing Administration’s (HCFA) fiscal year 1996
        Health Care Provider Cost Report Infotmation System data file (see encl. I).’ We pexformed
        our work between August and October 1999 in accordance with generally accepted
        government auditing standards.


        Per-visit costs varied widely both by visit type and across free-standing agencies. Home
        health aide visits were the least expensive, and medical social service visits were the most
        expensive. Across agencies, costs per visit for the most expensive agencies were 4 to 10
        times those of the least expensive agencies, depending on the type of visit. Non-patient-care

        ‘HHAs may be either free-standing or facility-based. Most of the facility-based agencies are
        hospital-based. According to fiscal year 1996 data, there were 2,628 facility-based agencies
        and 6,465 free-standing agencies.

                             GAO/EEHS-OO-19R        Medicare    Home Health Payment Limits
    costs constituted a substantial portion of the cost for each home health visit, averaging
    around 44 percent for each visit type. Moreover, the portion of visit costs that were not
    directly related to patient care was higher for more expensive visits.
    In addition, for the sample of free-standing HHAs we analyzed,Medicare payments would
    have been approximately 4 to 13 percent less if payments for non-patient-care costshad been
    heid to various limits based on the cost experience of a subsetof HHAs. For example, if
    Medicare payments for non-patient-care costs had been limited to the median costs of free-
    standing HHAs (the 50th percentile), total payments would have been reduced by 3.9
    percent. If payments for non-patient-care costs had been limited to the cost level of the least
    expensive 20 percent of HHAs (the 20th percentile), total spending would have been 12.6
    percent lower. The current per-visit cost limits already indirectly constrain Medicare
    payments for non-patient-care costs, although not as much as a limit applied directly to non-
    patient-care costs would. It is not known how the savings estimateswould have differed if
    all HHA.~, including the generally higher-cost hospital-based ones, had been included in the
    Home health care is an important Medicare benefit that enables beneficiaries with post-acute
    care needs (such as recovery from joint replacement) and chronic conditions (such as
    congestive heart failure) to receive care at home rather than in other settings. The scopeof
    home health care has changed markedly since Medicare began to cover it. Expansion of
    eligibility has allowed more beneficiaries to qualify for services and permitted more services
    to be provided to users, transforming the benefit from one that originally covered short-term
    care to one that covers long-term care as well. The standardsfor what constitutes necessary
    or appropriate home health care are not well defined, and service provision is inconsistent
    across agencies. Even the most basic unit of service, the visit, is not clearly defined. Costs
    per user vary widely, reflecting differences in patient needs,treatment patterns, agency type,
    and ogeogmphiclocation.            .
    Costs per user also reflect the number, mix, and cost of individual home health visits. Costs
    for each visit type are composed of direct patient care expenses(predominantly labor costs
    for caregivers) and non-patient-care expenses (such as capital, plant operations and
?   maintenance, transportation, and administration). Hospital-based HHAs generally have
    higher total per-visit and non-patient-care costs than free-standing agencies do.
    Prior to payment changes mandated in the Balanced Budget Act of 1997, agencies were paid
    on the basis of their costs up to preestablished per-visit limits equal to 112 percent of the
    national average cost per visit.2 There was a separate limit for each type of visit (skilled
    nursing; physical, occupational, or speech therapy; medical social service; and home health
    aide), bat the knits were applied in the aggregate to each agency’s costs. That is, costs
    above the limit for one visit type would still be paid if costswere sufficiently below the limit

    %e schedule of limits for cost reporting periods be,@nningon or after July 1,1994, and
    before July 1,1996, was based on per-visit limits calculated in 1993 (which were based on
    data from cost reporting periods ending on or after June 30,1989, and before May 3 1, 1991,
    and were adjusted forward to 1993). The Omnibus Budget Reconciliation Act of 1993
    mandated that there be no changes in the HHA limits for cost reporting periods beginning on
    or after July 1, 1994, and before July 1, 1996.

    2                   GAO/HEHSOO-19R            Medicare     Home Health Payment Limits
for other visit types so that, in total. agency costswere below the sum of their volume-
adjusted Per-visit limits.
The IPS was implemented on October 1, 1997. Under the IPS, agencies are paid their actual
costs up to the lower of the per-visit limits applied in the aggregate or an agency-specific
revenue cap that is based on a per-beneficiary amount and the number of beneficiaries
served. The per-visit limits control the payments per visit, while the revenue cap constrains
the visits provided to users. Under the PPS,the payment methodology will change again:
payments will be established in advance of service delivery, and the payment for each user
will vary to account for patient characteristics and other factors that affect the cost of home
health care.


Average agency costs per visit varied considerably by the type of home health visit.
Average visit costs varied even more for each type of visit across agencies. As a result of
the wide range in visit costs across both visit fypes and agencies, the dollar amounts
attributed to non-patient-care expenses were highly variable as well. Moreover, the share of
per-visit costs attributable to non-patient-care expenseswas larger for more expensive visits
of all types.

In fiscal year 1996, the average free-standing agency per-visit costsranged from $43 for a
home health aide visit, the most frequent visit type, to $15 1 for a medical social service visit
(see encl. II). Skilled nursing visits, the secondmost frequently supplied service, averaged
$95. Non-patient-care costs accounted for a substantial proportion of these costs, averaging
around 44 percent for each visit type. The dollar amount of the average agency’s non-
patientcare expenses varied from $19 for home health aide visits to $66 for medical social
service visits.

For each of the visit types, the cost per visit varied substantially across agencies. For
example, one-quarter of the HHAs had skilled nursing costs at or below $76, while the
quarter of agencies with the most expensive skilled nursing visits had visit costs of $107 or
higher (see table 1). Across the f&e-standing agenciesin our sample, per-visit costs for the
most expensive agencies were 4 to 10 times those of the least expensive agencies, depending
on the visit type.

3                    GAO/HEHS-00-19R           Medicare    Home    Health   Payment     Limits
Table 1: Free-Standiw HHA Costs uer Visit. bv Visit Tvoe. Fiscal Year 1996

                                     Distribution of agency costs, by quartile”

 Visit type             25th percentile        50th percentile              75th percentile

 Skilled nursing                     $76                         $91                          $107

 Physical therapy                      86                        105                           131

 Occupational                          84                        105                           134

 Speech therapy                        85                        107                           136

 Medical social                       100                        130                           171

 Home health                           34                        41                               48

%e first quartile is the 25th percentile, the median is the 50th percentile, and the third
quartile is the 75th percentile.
Source: GAO analysis of fiscal year 1996 HHA Medicare cost report data.
The share of the total costs attributed to non-patient-care expenses was higher for more
costly visits. For example, non-patient-care costsfor skilled nursing visits averaged 38
percent of total visit costs for the least expensive quarter of agencies, compared with 47
percent for the most expensive quarter. The range is even greater for home health aide
visits, with the non-patient-care portion averaging 37 percent of total visit costs for the least
expensive quarter of agencies and 49 percent for the most expensive quarter of agencies.

4                    GAO/HEHS-OO-19R          Medicare    Home     Health    Payment     Limits

Caps on Medicare payments for non-patient-care costs, in addition to the existing per-visit
limits, could have lowered Medicare payments to free-standing HHAs by nearly 4 percent to
almost 13 percent, depending on the level of the caps. The per-visit limits, which are based
on national average costs, indirectly restrict payment for non-patient-care     costs, so non-
patient-care payment caps would need to be set below the cost of the median HHA to
substantially affect Medicare expenditures. Although non-patient-care         COW are constrained,
the application of the current per-visit cost limits still allows considerable vtiation    in non-
patient-care costs across agencies.
Aggregate Medicare spending for free-standing HHAs in fiscal year 1996 could have been
3.9 percent lower if non-patient-care costs had been limited to the median costs of free-
standing HHAs, or the 50th percentile amount. At this cap, non-patient-care   payments
would have been no higher than $40 for a skilled nursing visit and $18 for a home health
aide visit (see table 2).
Table 2: Free-Standing     HHA Estimated         Non-Patient-Care       Costs Der Visit. bv Visit Tvne,
Fiscal Year 1996

                                  Distribution     of non-patient-care      costs, by docile”

     Visit type      20th percentile     39th percentile            40th percentile         50th percentile





                                   26                      33                         39                      45

  Speech                           27                      33                         39                      45

                                   34                      42                         49                      57

                                   11                       14                        16                      18

TkciIes   are percentiles at the IO’, 20” . . .90” percentiles.

Source: GAO analysis of fiscal year 19% HHA Medicare                   cost report data.

Aggregate savings could have reached 12.6 percent if the non-patient-care   payment limits had
been equal to the 20th percentile costs of free-standing HHAs (see table 3). At this cap, non-

 5                     GAOIHEHS-OO-19R                Medicare       Home Health           Payment Limits
 Ijatient-care payments would have been limited     to $27 for a skilled nursing visit and $11 for a
 home health tide visit, as shown in table 2.

ITable 3: Estimated Reductions in Medicare Spending for Free-Standing HHAs With Various
I,imts on Non-Patient-Care   ExDenses. bv Percentile. Fiscal Year 1996

                             Payment limit, set at percentile of non-patient-care costs
                         20th percentile      30th percentile 40th percentile 50th percentile          .
Percent reduction             12.6                   8.8              5.9              3.9
in Medicare

&urce: GAO analysis of fiscal year 1996 HHA Medicare cost report data.

In 1996, the per-visit limits for each visit type except home health aide were near or below
the median agency per-visit costs (see table 4). As a result, payments for non-patient-care
costs were already being constrained. This is why the savings associatedwith a non-patient-
care payment cap at the 50 * percentile are not higher.

Table 4: Per-Visit Pavment Limit and Proportion of Free-Standing HHAs With Costs at or
Below the Pavment Limit, bv Visit Tm, Fiscal Year 1996

Source: GAO analysis of fiscal year 1996 HHA Medicare cost report data.

Per-visit limits on HHA payments do not constrain payments for non-patient-           costs,or
limit the variation in these payments, as much as a cap directly applied to payment for such
costswould. Under the per-visit limits, agencies could receive higher payments for non-
patient-care costsif low direct patient care costs resulted in total visit coststhat were below
the per-visit limit. Furthermore, the per-visit limits are applied in the aggregate, so agencies
with costshigher than the limits for one type of visit may still be paid their total costsif their
costsfor another visit type are sufficiently below the limit.
 HCFA had an opportunity to comment on a draft of this letter. HCFA officials affirmed the
 need to implement a PPS that will ensure accurate payments that are based on patient needs
 and that will no longer vary from agency to agency on the b&s of non-patient-care costs.

  6                  GAOIHEHS-OO-19B              Medicare    Home    Health   Payment     Limits

We are sending copies of this report to the Honorable Nancy-Ann Min DeParle, HCFA
Administrator,   and other interested patties, and we will make copies available to others on
request. If you or your staff have any questions regarding this letter, please contact me at
1,202) 5 12-7 119 or Carol Carter, Assistant Director, at (3 12) 220-77 11. Other contributors to
this analysis were Jean Chung, Christine DeMars, and Daniel Lee.

Sincerely yours,

Laura A. Dummit
Associate Director, Health Fiiancing
  and Public Health Issues

Enclosures - 2

                      GAO/HEHS-00-19B           Medicare Home Health Payment Limits
ENCLOSURE I                                                                                    ENCLOSURE I

                                         SCOPE AND METHODOLOGY

We used fiscal year 1996 Medicare cost report data from the Health Care Financing Administration’s
(HCFA) Health Care Provider Cost Report Information System (HCRIS) data file to determine the
variation in home health per-visit costs, identify actual home health agency (HHA) payments, and estimate
payments that could result from limiting non-patient-care costs. We used fiscal year 1996 data because
they were the most complete data available at the time of our analysis. To control for regional wage
differences, we adjusted costs for wage differences across8eo8raphic areas according to the methodology
prescribed in the regulations1 We excluded the 2,628 facility-based agencies from our sample because
their non-patient-care costs are not separately reported in HCRK2 We excluded 1,555 other providers
that fell into the following categories:

    -        agencies with low or no Medicare utilization or no administrative costs;
             agencies we defined as outliers, or those with wage-adjusted average cost per visit greater than
             $800 or less than $10;
            agencies for which the individual cost component amounts did not add up to the reported total
            cost per visit amount; and
            agencies with cost report periods lasting less than 10 months or more than 13 months, since these
            agencies may have distorted costs.

    After all exclusions, the total number of agencies remainin g for our analysis was 4,910. Because some
    agencies do not provide all types of visits, however, the number of agencies included in each step of the
    analysts varied depending on the type of visit being analyzed.

 The Medicare cost report presents costs for each visit type in five categories: direct costs associated with
 providing a visit, capital-related costs,plant operations and maintenance costs,transportation costs, and
 administrative and general costs. The non-patient-care costs were defined as the last four cost categories.

    ‘58 Fed. Reg. 36,751 (July 8, 1993).

 *Although our sample excluded facility-based HHAs, it is important to note that there are instances in
which the Medicare pro,-        has discounted facility-based costsin developing reimbursement rates (for
example, for skikd nursing facility payments) and has calculated separate rates for free-standing and
hospital-based facilities (for example, for dialysis services).

8                                   GAOIHEHS-OO-19R        Medicare   Home Health Payment      Limits
ENCLOSURE I                                                                               ENCLOSURE I

Almost all (97 percent) of the non-patient-care costs in aggregate are in the administrative and gene&
cost category.

                                         GAO/HEHSdKb19R Medicare Home Health Payment Limits
ENCLOSURE                II                                                                                                                                                           ISNCLOSURE II

                                                     ‘I’OTAL AND NON-PATIENT-CARE    COSTS PER VISIT. BY VISIT TYPE AND PERCENTILE         GROIJPINGS

Home health alde
Average cosl/vlsit                                                    4,897         $46      $43       $23     $28      $32          $38          $46                 562               $61              $03
Average     IIon-pallent-care          cosUvislt                                              SIO       $8     110      $13          $16          $21                 $26               530              546
Average     % nonpatlent        care                                                         44%      34%     36%       30%        _-42% -.-46%              ___     46%               60%              60%
Average        % nonpntlent            care    for                                           44%
all vlalt    types                                                        -                                                                   -

 I II                                                                                                                     GAO/HEW+OO-19R          Medicare         Home     Hcnltl~   Fnyment   IhIts
Ordering      Information

The first copy of each GAO report and testimony         is free.
Additional   copies are $2 each. Orders should be sent to the
following   address, accompanied   by a check or money order
made out to the Superintendent     of Documents,      when
necessary.   VISA and MasterCard    credit cards are accepted,      also.
Orders for 100 or more copies to be mailed       to a single address
are discounted     25 percent.

Orders      by maik

U.S. General Accosting               Office
P.O. Box 37050
Washington,  DC 20013

or visiti

Boom 1100
700 4th St. NW (corner          of 4th and G Sts. NW)
U.S. General Accounting           Oflice
Washington,  DC

Orders may also be placed by calling    (202) 512-6000
or by using fax number  (202) 512-6061,     or TDD (202)       512-2537.

Each day, GAO issues a list of newly available      reports     and
testimony.   To receive facsimile   copies of the daily list or any
list from the past 30 days, please call (202) 512-6000        using a
touchtone   phone.   A recorded   mena will provide    information    on
how to obtain these lists.

For information        OP how to access GAO reports     on the INTERNET,
send an e-mail        message with ‘info” in the body to:


or visit    GAO’s     Wotld   Wide     Web Home   Page at:

United States
General Accounting    OfIice
Washington,   D.C. 205484001

Of6cial   Business
Penalty   for Private   Use $300

Address   Correction    Requested