Order Code RL31420
Report for Congress
Received through the CRS Web
Medicare’s Home Health Benefit:
The Fifteen Percent Payment Cut
Updated October 17, 2002
Carolyn L. Merck
Specialist in Social Legislation
Domestic Social Policy Division
Congressional Research Service ˜ The Library of Congress
Medicare’s Home Health Benefit:
The Fifteen Percent Payment Cut
Medicare spending for home health care increased from $2.0 billion in 1988 to
about $18 billion 1996. This spending growth reflected increases in the number of
beneficiaries served, in the number of visits per user, and in the number of Medicarecertified home health agencies (HHAs).
In the Balanced Budget Act of 1997 (BBA 97), Congress sought to curtail rising
home health spending by requiring implementation of a prospective payment system
(PPS) under which HHAs would be paid fixed amounts per episode of care for an
individual beneficiary. Because the PPS was not ready for implementation when the
BBA 97 was enacted, Congress mandated interim changes to lower payments to
HHAs, and these, and other measures, including fraud and abuse surveillance,
resulted in a decline of about 50% in Medicare home health spending by 2000.
PPS payments for FY2001 were to be calculated so that, in the aggregate, home
health spending would not exceed total spending under the previous system but with
the per visit and per beneficiary limits of the previous system cut by 15%. Cutting
those two payment limits by 15% would reduce all PPS payments by only about 7%
because lower-cost HHAs would not be affected by the 15% cut in the limits. (That
is, it would have been a 15% cut in limits, not necessarily in payments.) The home
health spending declines that preceded the PPS caused Congress to postpone the 15%
cut to October 1, 2002. In the absence of data on detrimental effects of the new PPS,
Congress took no action to stop implementation on October 1, 2002, of the cut in
payments. The effective rate of the reduction was 7%, but a simultaneous payment
update of 2.1% made payments in October 2002 only 4.9% below the previous rates.
Evaluations of the reductions in Medicare spending and in the supply of home
health care resulting from BBA 97 do not indicate that beneficiary access to care has
been compromised. Although the number of participating HHAs dropped by 35%
from 1996 to 1999, the supply of agencies has remained stable since October 1, 2000,
when the PPS began. Hospital discharge planners report little difficulty in placing
beneficiaries with an HHA. The General Accounting Office (GAO) reported that, in
the first 6 months of 2001, PPS payments to HHAs were about 35% higher than
estimated agency costs for providing care and recommended that Congress allow the
so-called 15% reduction to be implemented. The Medicare Payment Advisory
Commission (MedPAC) recommended repeal of the reduction, noting that spending
and utilization reductions sought by BBA 97 were achieved without it and that HHAs
might be unable to make firm business plans were the threat of payment reductions
to continue. The Administration indicated that retaining the authority to reduce
payments on October 1, 2002, provided a tool for containing spending if need be.
The Congressional Budget Office (CBO) estimated that repeal of the 15%
reduction would have eliminated $5.2 billion in savings for the 5-year period FY2003
through FY2007, and $16.4 billion over the 10-year period FY2003 through FY2012.
If the reduction had been delayed by one year, $800 million in savings would have
been added to program spending.
Recent Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Home Health Issues Leading up to the Balanced Budget Act of 1997 . . . . . 1
The Balanced Budget Act of 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
The “15%” Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Implementation of the PPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Delay in the “15%” Payment Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Size of the “15%” Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Effects of BBA 97 Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
OIG Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
MedPAC Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
GAO Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Industry Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Medicare’s Home Health Benefit:
The Fifteen Percent Payment Cut
Recent Developments. On October 1, 2002, the cuts in Medicare payments
to home health agencies discussed in this report went into effect. The across-theboard reduction in PPS amounts was 4.9%, calculated as an average reduction of
7% offset by a 2.1% increase to update payments for inflation. The reduction was
scheduled by law to take effect on that date; only congressional action could have
stopped or changed its implementation. The reduced payment amounts apply to 60day episodes of beneficiary care that begin on or after October 1, 2002. Both H.R.
4954 (passed by the House on June 28, 2002) and S. 3018 (introduced in the Senate
on October 1, 2002) would eliminate the home health payment cuts.
Home health services covered by Medicare for homebound beneficiaries include
intermittent or part-time skilled nursing care, physical therapy, and speech therapy.
For beneficiaries receiving at least one of these types of care, Medicare also covers
the services of home health aides and medical social workers. Occupational therapy
may be covered during and after receipt of skilled nursing care or physical or speech
therapy. Services must be medically necessary and carried out under a plan of care
prescribed and reviewed by a physician.
Home Health Issues Leading up to the Balanced Budget Act
Medicare spending for home health care in 1988 totaled $2.0 billion; by 1996
it had grown to about $18 billion, for an average annual increase of 31% during that
time. This spending growth reflected both increasing numbers of beneficiaries
served and more than a three-fold increase in the average number of visits per user.
In addition, the number of Medicare-certified home health agencies (HHAs) that
furnish this type of care increased by 9% a year during most of the 1990s.1
During this period of rapid growth in Medicare spending for home health care,
Medicare paid HHAs during a year based on estimated costs, and made final
settlements after the year when actual allowable costs were determined. Final
payments were based on an agency’s reasonable allowable costs limited to a capped
dollar amount per visit by visit type (e.g., skilled nurse, nurse aide, or therapist). The
maximum amount an agency could be paid in a year was the sum of the cap amounts
for individual services times the number of visits of each type the HHA provided.
If an HHA’s total costs for one type of visit were more than the cap, those costs
For detailed information on Medicare’s home health benefit, see 2000 Greenbook,
Background Material and Data on Programs within the Jurisdiction of the Committee on
Ways and Means. Ways and Means Committee print 106-14, October 2, 2000. p. 131-141.
could, at least to some extent, be compensated by payments for another type of visit
for which the HHA’s costs were below the payment cap. Because there was no limit
on the number of visits of any type a beneficiary could receive, this system provided
incentives for HHAs to maximize the number of visits made as long as their actual
costs were less than the payment limits.2
The Balanced Budget Act of 1997
In the Balanced Budget Act of 1997 (BBA 97), Congress included a number of
provisions aimed at curtailing home health spending growth. These provisions
included a requirement for implementation of a prospective payment system (PPS)
under which HHAs would be paid predetermined, fixed amounts per episode of care
for each individual beneficiary served, and the amount of the payment would reflect
the type and intensity of care furnished. Because the PPS was not ready for
implementation at the time BBA 97 was enacted, Congress included in the law an
interim payment system (IPS) which lowered the caps on payments per visit and
introduced a third tier to the payment system, the so-called “per beneficiary” limit on
payments. Thus, under the IPS, payments to HHAs would be the lowest of (a) their
reasonable costs, (b) payments based on the limits per visit, or (c) payments based on
the new “per beneficiary” formula.
The “per beneficiary” amount was calculated for each HHA by multiplying its
average Medicare payment per beneficiary in 1994 (updated for home health cost
inflation) by the total unduplicated count of Medicare beneficiaries the HHA was
serving in the year to which the limits were applied. That produced an aggregate
annual Medicare spending limit for each HHA; it did not limit payments on behalf
of any individual home health patient. For 79% of HHAs, the new “per beneficiary”
formula proved to be more restrictive than payments based on the capped amounts
per visit. HHAs not affected by the per beneficiary limit were those for which the
aggregate amount payable based on payment caps per visit or on actual costs were
lower than the amount generated by the per beneficiary formula.
In addition to changing the home health payment system, BBA 97 eliminated
venipuncture (the drawing of a blood specimen) as the sole home health service
qualifying an individual for home care, and that change contributed to the reduction
in visits and in program spending for home health care.
Other policies that affected the volume of home health services provided after
1996 included the Health Insurance Portability and Accountability Act of 1996 which
provided for the imposition of civil money penalties on physicians who falsely certify
that a beneficiary needs home health care, a provision some say has had a chilling
effect on physician referrals. And, officials at the Centers for Medicare and Medicaid
Services (CMS) note that at about the same time the IPS went into effect, the
The caps were 112% of the national average cost of each type of visit. The caps on the
payments per visit did not affect agencies for which costs per visit were less than 112% of
the national average. BBA 97 reduced the payment cap per visit to 105% of the national
median cost of a visit (later raised to106%).
Administration required claims processors to implement an intensified case review
process as well as stepped-up fraud and abuse detection activities.
The “15%” Reduction. BBA 97 required the PPS to be implemented on
October 1, 1999 (later moved to October 1, 2000). The Act laid out the basic design
of the PPS and required payment amounts to be determined so that, in the first year
of the PPS, total Medicare spending for home health would be the same as it would
have been had the previous system (the IPS) remained in effect but with a 15%
reduction to caps on the amount HHAs could be paid per visit and on agency-specific
limits per beneficiary.
Application of the 15% reduction to the “per visit” and “per beneficiary” limits
would not result in an across-the-board 15% reduction in all PPS payments. This fact
appears to be widely misunderstood. According to program officials, the effect of
reducing the cap on payments per visit and per beneficiary by 15% would be an
effective reduction in payments under the PPS of 7%. The reason the reduction
would be less than 15% is that some HHAs were not affected by the “per beneficiary”
limit (because their payments were lower based on either the “per visit” limit or their
actual costs), and some agencies were not affected by the “per visit” limits (because
their costs were below the limit). Thus, a cut in per beneficiary limits would have no
effect on some HHAs, and a cut in per visit payment caps would have no effect on
other HHAs. The 7% overall reduction in PPS payments was estimated assuming
implementation of the reduction in FY2001, simultaneous with the PPS. Although
the law implies that the reduction should be recomputed using projections of
spending under the IPS and under the PPS for FY2003, officials at CMS chose to use
the estimate that was prepared for FY2001, largely because of the difficulty in
projecting payments under the old system that was replaced by the PPS nearly 2 years
before the effective date of the cuts.
Implementation of the PPS
When the PPS was implemented on October 1, 2000, home health spending had
declined precipitously from previous levels. In 1996, before implementation of the
IPS and other changes, Medicare home health spending had grown to about $18
billion; in 1999, spending had dropped by nearly half, to $9.3 billion. This spending
drop reflected a two-thirds reduction in the number of home visits provided, from
284 million visits in 1996 to 97 million in1999.
In January 1998, the Congressional Budget Office (CBO) projected that the
home health care provisions of BBA 97 would result in savings of almost $75 billion
over 10 years. In March 1999, CBO re-estimated the effects of BBA 97, and the new
projections showed an additional $56 billion in savings. CBO’s revised estimates
included changes in their underlying economic assumptions as well as revised
estimates of the spending-reduction effects of BBA 97.
Delay in the “15%” Payment Reduction
The 15% reduction in the “per visit” and “per beneficiary” tiers of the IPS was
to have been applied at the time the PPS was implemented in order to benchmark
aggregate payments in the first year of the PPS to a system under which total
spending included those reduced payment limits. However, because Medicare home
health spending had declined substantially by the time the PPS was implemented on
October 1, 2000, Congress delayed the reduction in these two payment limits until
12 months after implementation of the PPS (that is, until October 1, 2001) and
required the Secretary of Health and Human Services (HHS) to report no later than
6 months after implementation of the PPS on the need for a 15% or other reduction
in base payment amounts for the PPS. (These provision were included in the
Balanced Budget Refinement Act of 1999 (“BBRA,” P.L. 106-113).) Aggregate
payments under the home health PPS in its first year of operation (FY2001) were set
to equal the amount Medicare would have spent if the IPS had continued in that year.
Congress enacted further changes to the Medicare program in the Benefits
Improvement and Protection Act of 2000 (“BIPA 2000,” P.L. 106-554). The
President signed this legislation on December 21, 2000, less than 3 months after the
home health care PPS had been implemented. Home health spending had remained
low throughout FY2000, but, although the PPS was designed to contain spending, the
effects of the new system were uncertain. Thus, rather than repealing the 15%
reduction, Congress again delayed its application, putting it off until October 1, 2002,
when there would be 2 years of experience with the PPS, and required the
Comptroller General, as head of the General Accounting Office (GAO) (rather than
the Secretary of HHS), to submit, by April 1, 2002, a report analyzing the need for
the 15% or other reduction. (See below.)
The Size of the “15%” Reduction. In statements made during March 2002,
the Administration noted that Medicare spending for home health care may be rising
under the PPS; therefore they were not inclined to seek repeal of the authority to
impose a payment reduction. However, the Administration said that, if implemented,
the reduction would be only 4.9%, not 15%. The 4.9% is based on an estimated
effective payment reduction of 7%, offset by a 2.1% payment increase under the
routine update in payment amounts FY2003 (effective October 1, 2002). Some say
that the routine payment update on October 1, 2002, should not be counted as
offsetting a payment reduction.
The Budget. Because the law called for the so-called 15% reduction to go
into effect on October 1, 2002, budget projections assumed Medicare home health
spending would be reduced starting in FY2003. Thus, if Congress had repealed the
reduction, the budget estimates for Medicare home health spending would have risen.
According to the CBO, repeal of the “15%” reduction would have added $5.2 billion
in net Medicare spending for the 5-year period FY2003 through FY2007, and $16.4
billion over the 10-year period FY2003 through FY2012.3 (The increased spending
BBA 97 shifted certain home health spending from Medicare Part A to Part B and required
would reflect elimination of the savings attributable to implementation of the cut.)
If the reduction had been delayed again by one year, budget estimates would have
included an $800 million spending increase.
The FY2003 Budget Resolution that passed the Senate Budget Committee on
March 21, 2002 (S.Con.Res. 100) contained a non-binding “sense of the Senate”
statement that the Administration and Congress should work to “avoid the 15%
reduction in the prospective payment system for home health care.” (This resolution
has not passed the Senate.) The Budget Resolution passed by the House of
Representatives on March 20, 2002 (H.Con.Res. 353) did not address the issue.
Effects of BBA 97 Changes. The reduction by more than half in Medicare
home health spending from 1996 to 1999 reflects a 20% reduction in the proportion
of beneficiaries who use home health care; a 40% reduction in the average number
of visits per user; a 35% reduction in participating HHAs; and about a 9% reduction
in the number of HHA employees.4
Four organizations have evaluated the effects of the post-1996 Medicare home
health changes. The Office of the Inspector General of HHS (OIG) addressed the
extent to which the reductions that have already occurred in the supply of and
payments for home health services have affected beneficiary access to care.
The Medicare Payment Advisory Commission (MedPAC) noted that the goals
of BBA 97 have been achieved already with regard to containing Medicare spending
for home health care and considered how the lack of Medicare payment stability in
recent years has affected the industry.
The GAO compared payments to HHAs under the PPS with estimated agency
costs for serving Medicare beneficiaries in order to determine payment adequacy
under the PPS.
An evaluation commissioned by the home health industry compared the current
supply of HHAs and staffing levels of HHAs with the supply of HHAs and personnel
available before the IPS and the PPS. It used the change in the supply of available
services as a measure of access to care.
OIG Findings. In July 2001 the HHS OIG published Access to Home Health
Care After Hospital Discharge 2001 (OEI-02-01-00180), and in October 2001
published Medicare Home Health Care Community Beneficiaries (OEI-02-0100070).
that beneficiary premiums for Part B reflect the inclusion of home health spending, phased
in. The CBO spending increase from elimination of the “15%” reduction is net of the
increase in beneficiary Part B premiums attributable to home health care.
Medicare Payment Advisory Commission. Medicare Payment Policy, Report to the
Congress, March 2002. p. 93-98; and Edward and Ester Polisher Research Institute. Impact
of a Further Payment Reduction on the Medicare Home Health Benefit, March 18, 2002.
The first report reflects the first 6 months of experience with the PPS. The OIG
assessed the extent to which hospital discharge planners experienced difficulty in
finding an HHA to serve Medicare beneficiaries, reviewed trends in the number of
HHAs participating in Medicare, and analyzed the length of hospital stays by
diagnosis related group (DRG) to determine if patients experienced longer inpatient
stays due to delays in finding home care.
The OIG reported that about 90% of discharge planners can place all of their
Medicare beneficiaries needing home health care and that there had been no change
in their experience since implementation of the PPS. Also, the data showed that
“there were no large changes in the types of Medicare beneficiaries being discharged
to HHAs in the last 5 years.” Despite there being about 35% fewer HHAs in business
now compared with past years, 83% of discharge planners said there were sufficient
services available and that, on average, they contacted only one or two agencies to
place a patient. Medicare data showed that between 1997 and 2001 the average
hospital stay decreased for all but two of the 13 most common DRGs, indicating that
there was no discernable effect on length of hospital stay.
Sixty-one percent of discharge planners said that the patients for whom they
experience delays in placement are those who need wound care and expensive
supplies, intravenous antibiotics or other expensive drugs that may not be reimbursed
under Medicare (often requiring the patient to pay), and those needing several visits
per day to monitor intravenous medications.
Overall, the OIG found little evidence that there is a problem with access to care
for beneficiaries with a prior hospitalization, but suggested that access should
continue to be monitored.
The second OIG analysis focused on beneficiaries using home health care who
had no hospitalization or skilled nursing facility stay within 15 days of using care and
who therefore had not had the services of a hospital discharge planner. These
“community beneficiaries” constituted 38% of Medicare home health care users in
1997 and 41% in 2000. The OIG interviewed physicians, community organizations
that assist beneficiaries in finding care, and beneficiaries themselves.
HHAs reported to the OIG that community beneficiaries tend to have chronic
rather than acute health problems. Most of the people interviewed reported that care
was accessible and that most patients who needed home health care and were eligible
for it received it. However, anecdotal reports indicated that patients with certain
conditions (e.g., diabetes, wound care, or Alzheimer disease) were unable to obtain
care, although it was not clear that these patients met all the Medicare home health
care eligibility criteria.
MedPAC Findings. In its March 2002 report, MedPAC documented the
substantial increase in the supply of home health care and volume of services
furnished between 1987 and 1997 and the decline in services furnished since that
time. It noted that the purpose of the IPS and the home health provisions of BBA 97
had been to reduce utilization and spending and that those goals have been achieved.5
MedPAC noted that data on the amount HHAs are actually spending to provide
care under the PPS are not yet available, so the adequacy of PPS payments cannot be
determined relative to HHA expenditures. Therefore, MedPAC evaluated payment
adequacy by the entry and exit of HHAs in the program since implementation of the
PPS. Because the number of HHAs participating has remained stable since the start
of the PPS, MedPAC suggested that the payment amounts are neither inadequate nor
MedPAC recommended that Congress repeal the so-called 15% cut because
substantial reductions in home health spending and use have already occurred and
because leaving the reduction in the law but postponing it repeatedly would just
prolong HHA uncertainty about the future of their business. Moreover, at the time
of the MedPAC report, the precise amount of the cut was unclear (that is, whether
CMS would use the FY2001 estimated reduction of 7%, or revise it for FY2003).
The Commission noted that Medicare payment policy for home health has been in
flux for several years and should be stabilized; payment adequacy issues can be
addressed through the annual payment update process. If it is determined in a future
year that payments are excessively high compared with HHA costs, the annual
payment increase could be scaled back; if payments are determined to be too low, the
increase could be enhanced.
GAO Findings. Because as October 2002 approached no data were available
on actual costs incurred by HHAs for providing patient care under the PPS, the GAO
estimated HHA average expenditures for care during the first 6 months of 2001 and
compared those estimated costs with average PPS payment amounts. (Medicare
Home Health Care: Payments to Home Health Agencies are Considerably Higher
than Costs, GAO-02-663, May 2002.) The report concluded that, on average,
payments were about 35% higher than HHA estimated costs for providing care. The
GAO noted that the reasons payments were high compared to estimated costs was
that patients were receiving fewer visits than had been estimated in the development
of the PPS and were being categorized as needing more costly care than had been
estimated. Hence, HHAs “save” costs by providing fewer visits while being paid at
rates applicable to patients needing more intensive services.
On the basis of estimated average costs compared with average payments, the
GAO asked Congress to consider allowing the so-called 15% reduction to payments
to be implemented. However, it warns that a mechanism such as “risk
sharing”should be used to protect patients from underservice and to protect HHAs
from financial loss if they generally serve high cost patients with complex
Ibid. Medicare Payment Advisory Commission.
Under risk sharing, a cap is placed on the amount an HHA may gain from payments in
excess of costs and on the amount an HHA may lose as a result of having expenses that are
greater than payments.
Home health industry representatives criticized GOA’s estimated average costs
as being inaccurate and not reflective of current service delivery patterns and costs.
They said that, although PPS payments might appear more than adequate on average,
the distributional effects of an across-the-board payment reduction would drive some
HHAs out of the Medicare market. Some critics noted that HHAs could be keeping
their costs low in anticipation of the 15% payment cut.
The Administration commented that the GAO’s estimates were consistent with
its own preliminary analysis of data from the first year of the PPS, but that HHA
actual costs could not be known with certainty until Medicare home health cost
reports become available. The Administration opposed risk sharing adjustments,
saying it is difficult to administer and would result in HHAs not receiving final
payments on a timely basis.
Industry Findings. The home health care industry commissioned a study by
the Edward and Ester Polisher Research Institute and released it on March 18, 2002,
(Impact of a Further Payment Reduction in the Medicare Home Health Benefit.7)
This evaluation focused on the 35% reduction in the number of HHAs participating
in the program from before the IPS until implementation of the PPS and on staffing
reductions. It suggested that reduction in the supply of agencies and personnel is
evidence of reduced access to care and that “any additional cut in payments for
Medicare home health services risks exacerbating the access concerns and possibly
further jeopardies the quality of services for the most vulnerable beneficiaries.” It did
not attempt to determine if the reduction in participating agencies has caused eligible
beneficiaries needing care to go without services. The report observed that the
number of participating HHAs has not changed much since implementation of the
PPS, with few agencies leaving the program and few entering. From that
observation, it concluded that PPS payments are inadequate if no new HHAs have
entered the program.
The report stated that, even if the 15% reduction were postponed and not
implemented as scheduled, as long as it remained in the law as a possibility, HHAs
would be unable to plan ahead; if the reduction were implemented, agencies might
have to reduce services to Medicare beneficiaries.
Arguments supporting the October 1, 2002, implementation of the so-called
15% reduction (effectively, a 4.9% cut to PPS rates) were generally based on
preliminary indications that spending for home health care under the PPS was rising,
that payment amounts were generous compared with agency costs, and that the
industry could absorb a 7% reduction when offset by a 2.1% increase under the
annual payment update process. Also, there was little indication that eligible
beneficiaries in need of care had been unable to find it.
Those arguing that the reduction should have been postponed, but not repealed,
continue to say that it is too early to judge accurately the rate of spending under the
Edward and Ester Polisher Research Institute. Impact of a Further Payment Reduction on
the Medicare Home Health Benefit, March 18, 2002. p. 13-14.
PPS and the ability of the industry to absorb the reduction. By retaining the
postponed provision in the law, the Administration would have had available a tool
to contain spending should it rise above acceptable levels in the future.
The arguments in favor of repealing the reduction and not letting it go into effect
were that home health spending reductions sought by BBA 97 were met and
exceeded; the number of participating HHAs has declined to the point that
beneficiary access to care could be in jeopardy if payments were reduced further;
HHAs would be unable to make long-term business or financial plans if the provision
remained in the law; and, if future spending were to rise above acceptable levels, it
could be restrained through curtailing future annual payment updates. Nevertheless,
until data on HHA expenditures compared with payment amounts and patterns of
beneficiary services are available, the distributional effects among HHAs of the
across-the-board reduction remain unknown.