Extract
Recent proposals in the United States to cut Medicare
physician-reimbursement rates for certain orthopaedic procedures, particularly
in the field of hip and knee arthroplasty, have heightened the general
orthopaedic community's interest in Medicare physician-reimbursement
policy.Familiarizing oneself with Medicare physician-reimbursement policies in
particular is important as most private and public insurers base their
payments on Medicare fee schedules and regulations. The 2004 American Academy
of Orthopaedic Surgeons (AAOS) Physician Census revealed that >90% of
orthopaedists in private practice identified insurance or Medicare and
Medicaid reimbursement levels and increasing practice expenses as major
concerns1. Medicare
and Medicaid accounted for an average of 33% of an orthopaedist's practice in
2006, having risen steadily from 26% in
19882.
Recent proposals in the United States to cut Medicare
physician-reimbursement rates for certain orthopaedic procedures, particularly
in the field of hip and knee arthroplasty, have heightened the general
orthopaedic community's interest in Medicare physician-reimbursement
policy.
Familiarizing oneself with Medicare physician-reimbursement policies in
particular is important as most private and public insurers base their
payments on Medicare fee schedules and regulations. The 2004 American Academy
of Orthopaedic Surgeons (AAOS) Physician Census revealed that >90% of
orthopaedists in private practice identified insurance or Medicare and
Medicaid reimbursement levels and increasing practice expenses as major
concerns1. Medicare
and Medicaid accounted for an average of 33% of an orthopaedist's practice in
2006, having risen steadily from 26% in
19882.
This article is a primer on the history of the Medicare Physician Fee
Schedule, its current status, and issues that are currently at the forefront
of Medicare health policy agendas in the United States. We conclude with a
discussion of the proposed changes to the sustained growth rate formula and
the design and implementation of pay-for-performance programs.
The AAOS has led the charge in representing orthopaedic interests in
Washington, DC, and in keeping its membership updated on issues through
bulletins and legislative updates. However, for physicians and researchers
interested in understanding the issues surrounding Medicare, the task of
sifting through and integrating the myriad of academic, governmental, and
organizational resources available is daunting. We aim to organize these
sources into an accessible, unified format. We also present the orthopaedic
surgeon with opportunities to become involved in shaping future policy at a
range of commitment levels.
In 1965, President Lyndon B. Johnson signed the Title XVIII Amendment to
the 1933 Social Security Act, creating Medicare as the first major federal
health insurance entitlement program. The program's original goal was to
reduce the risk of financial disaster for the elderly and their families. It
also provided access to medical services that had at that point been largely
unaffordable for many elderly retired workers and their
spouses3. Medicare
was to be administered by the Health Care Financing Administration, which was
renamed the Centers for Medicare and Medicaid Services (CMS) in 2001. The
Health Care Financing Administration (and now CMS) operated within the
Department of Health and Human Services.
Medicare was originally a health insurance program for people who were
sixty-five years of age or older. Since its passage, other beneficiaries have
been added to the program and they include persons on disability payments from
Social Security or the Railroad Retirement Board, persons with end-stage renal
disease requiring continuing dialysis or kidney transplantation, and all state
and local government employees not covered under Social Security. Medicare
originally had two components: Part A (Hospital Insurance) and Part B
(Supplemental Medical Insurance).
Part A covers hospital inpatient care, skilled nursing facilities, hospice
care, and some home health care. It is funded by a 1.45% payroll tax on both
employees and employers. The United States Department of the Treasury credits
the Medicare Hospital Insurance Trust Fund with any annual excess Medicare tax
revenues over the amount spent for current benefits. This surplus is invested
in special securities that are spent by the government to ease fiscal
pressures on other programs. When Medicare faces a negative cash flow (i.e.,
the benefit payment exceeds the income from payroll taxes and beneficiary
charges), the Treasury, using the prior surplus, must cover the deficit. In
2004, Medicare began using interest earnings from its trust fund to cover
Medicare expenditures in excess of its tax income.
Part B covers inpatient and out-patient physician services; emergency room
care; outpatient clinic and surgical care; physical, occupational, and speech
therapy; radiation therapy; diagnostic tests (e.g., laboratory work and
radiographs); durable medical equipment (e.g., prostheses and oxygen); and
some home health care. Office-administered drugs such as chemotherapeutic
agents, hyaluronic acid, and hematopoietin are also included in Part B. In
cases in which physicians participate in Medicare, Medicare reimburses
providers 80% of the service's allowable (not actual) charge; the patient
provides the remaining 20%.
Part B is partially funded by monthly premiums and a deductible paid by the
recipients. Starting January 1, 2007, monthly Part-B premiums are based on the
recipient's income. For example, if a person earns <$80,000 or a couple
earns <$160,000, the Part-B premium for each person is $93.50. If a person
earns >$200,000 or a couple earns >$400,000, the Part-B premium for each
person is $161.40. Incomes between these extremes have associated progressive
premiums between these two rates. Approximately 75% of the funding for Part B
comes from the general revenue of the United States
Treasury4. Patients
may also purchase so-called Medigap private health insurance that pays for
many health-care services not covered by Parts A and B.
The Balanced Budget Act of 1997 introduced Medicare Part C: Medicare+Choice
(renamed Medicare Advantage in 2003). Part C subsidizes the health maintenance
organization industry by allowing beneficiaries to select a managed care
provider for comprehensive health-care services and paying that provider a
predetermined per capita fee. Examples of such plans are health maintenance,
provider-sponsored, and preferred provider organizations. In order to join one
of these plans, enrollees have to have both Medicare Part A and Part B and
they must continue to pay the Part-B premiums. Most Medicare Advantage plans
have lower copayments than Medicare Parts A and B. The programs also cover
additional services such as preventive care, eye-glasses, dental care, and
hearing aids.
While Part C was originally touted as a cost-containing policy, a number of
studies have shown that health maintenance organizations are being paid more
than the average fee-for-service costs in their area. In fact, in 2005,
Medicare paid Medicare Advantage health maintenance organizations 7.8% more
than the average local fee-for-service costs, translating into an estimated
national total of >$2.7 billion in additional
spending5. By 2006,
to help curb spending, 75% of Medicare plan payments were adjusted on the
basis of the enrollees' risk profiles. By 2007, 100% of plans will receive
risk-adjusted
rates5.
In 2006, the optional Medicare Part D (under the Medicare Prescription Drug
Improvement and Modernization Act, also known as the Medicare Modernization
Act) was implemented to cover outpatient prescription drugs. Beneficiary
premium payments (which cover 25.5% of the total program cost), the United
States Treasury general fund, and state governments provide the funding for
Part D. Part D links enrollee spending to their income. In 2006, the average
base beneficiary premium was $32.20 and the annual deductible was $250.
A major critique of Part D is the so-called doughnut
hole6). Part-D
coverage is generally divided into three phases. Depending on the particular
plan chosen by the enrollee, he or she pays, in the first phase (the initial
coverage period), a deductible and approximately 25% of the drug costs. In
2006, this initial coverage period ended when total drug costs exceeded $2250.
The third phase (catastrophic coverage) began when the enrollee had spent more
than $3600 (excluding premiums) out of pocket in 2006. When catastrophic
coverage begins, the enrollee pays about 5% of the drug costs. In between
these two periods, there is a gap in coverage (i.e., the doughnut hole), in
which most people must pay 100% of their drug costs out of pocket. At the
beginning of each year, the enrollee goes back to the beginning of the first
phase. An estimated 24% to 38% of all enrollees are projected to fall into
this hole6.
The architects of Medicare did not focus on controlling health-care costs
but rather tried to appease health-care providers in order to ensure their
cooperation in its implementation. They modeled the Medicare payment system on
the existing health insurance market dominated by the modified fee-for-service
model of Blue Cross-Blue Shield. Toward this end, Medicare initially
reimbursed physicians using the customary, prevailing, and reasonable payment
system.
The reasonable charge that would be reimbursed by Medicare was
defined as the lowest of the following charges: (1) the physician's
actual charge, (2) the physician's customary charge (i.e.,
the median of an individual physician's charges for a specific service within
a specific time interval), or (3) the prevailing charge (i.e., the
fee in the 90th, and later the 75th, percentile charged by specialty-specific
physicians within a Medicare payment area). Under the customary, prevailing,
and reasonable system, physicians had incentives to raise charges, leading to
a rapid increase in program payments. Furthermore, there arose wide geographic
fee variations, disconnects between reimbursements and resources utilized, and
different payments for the same service depending on the physician's
specialty.
The Omnibus Budget Reconciliation Act of 1989 established a Medicare fee
schedule for physicians that decoupled Medicare's payment rates from the
physicians' charges for services. Rather than continuing to pursue a
charge-based payment system, a resource-based relative value system was
developed. The Health Care Financing Administration awarded William Hsiao, PhD
(Harvard School of Public Health), the contract for evaluating the so-called
relative values of physician work.
The objective of the resource-based relative value system was to assign
each Current Procedural Terminology (CPT) code a relative value unit
(RVU)7. An RVU is a
nonmonetary relative unit of measure that indicates the relative resources
required to perform a medical service. This system permits objective
comparison of the work involved in performing each procedure relative to all
other procedures. Hsiao and colleagues conducted a survey of physicians to
determine the relative work involved in providing approximately 800 unique
physician services7.
On the basis of those data, the researchers extrapolated the work values for
the remaining services.
It is important to understand the distinction between this resource-based
approach and the previous charge-based approach. Under the previous
charge-based approach, reimbursements were linked to physicians' charges.
Under the current resource-based approach, reimbursements depend on the value
of the resources required to perform a particular service. The resource-based
relative value system is unique in a free-market economy in that it does not
factor in the value of services to the consumer (i.e., the
patient)8.
A cornerstone of the research by Hsiao et
al.9 was their
assertion that certain invasive procedures consumed the same resource inputs
as certain evaluation-and-management services (e.g., outpatient clinic visits,
inpatient consultations, etc.); however, under the customary, prevailing, and
reasonable system, invasive procedures were typically compensated at more than
double the rate of evaluation-and-management services. The resource-based
relative value system of Hsiao et al. was designed to rectify this disparity.
Hsiao et al. recognized that their RVU proposal would offer relatively higher
compensation for certain evaluation-and-management services, perhaps inducing
physicians to shift their activities to some extent away from invasive
services9.
Medicare has a considerable impact on the American health-care system not
just because it comprises a substantial portion of health-care provider
payments but also because of the ripple effects of the program's policies on
other health-care delivery systems. The Medicare fee schedule has become the
de facto national health-care reimbursement schedule for all physician
services.
Despite the particular payer-mix of one's patients, almost all physicians
are affected by Medicare policies. A 2003 survey of thirty-three health plans
serving approximately thirty-one million members found that the primary
(largest enrollment) benefit plans of those health plans were influenced by
Medicare's resource-based relative value system
methodology10. In
fact, 39% of the plans consistently used the actual Medicare RVUs.
Beyond the Medicare fee schedule, non-Medicare payers often also adopt
other features of Medicare policy that are in some cases highly unfavorable to
physicians. For example, Medicare's global surgical periods were adopted in
Medicaid (87%), Blue Cross-Blue Shield (80%), managed care organizations
(69%), and other non-Medicare programs
(26%)11. Under the
Medicare global surgical payment policy, payment to the surgeon for a surgical
procedure includes a standard package of preoperative, intraoperative, and
postoperative services. The preoperative period included in the global fee for
major surgery is one day. The postoperative period for major surgery is ninety
days. Having the surgical fee include care for ninety days after the procedure
is now accepted as the industry standard, but it was Medicare that introduced
it and defined this global time-period as ninety
days12.
Since the gradual implementation of the resource-based relative value
system starting on January 1, 1992, a fee schedule has determined Medicare
payments for physician services. Each CPT code (the common billing terminology
that links each procedure with a numeric code) is assigned three types of RVU
values: (1) an RVU value for physician work, (2) an RVU value for practice
expene, and (3) an RVU value for professional liability (i.e., malpractice)
insurance. Simply stated, Medicare reimbursement = physician work + practice
expenses + malpractice insurance fees.
Originally, the formula for determining reimbursement rates was going to
include a factor accounting for the opportunity cost of going into certain
specialties, recognizing that additional training time is needed to provide
certain services and that physicians should be rewarded for foregoing income
during that training
time9. However,
prior to the implementation of the resource-based relative value system,
Congress dropped this opportunity cost component from the
equation8.
The RVU value for physician work was resource-based from the beginning.
Over time, the RVUs for practice expense and malpractice also shifted from
charge-based to resource-based. These three RVU values are summed,
geographically adjusted, and then multiplied by a dollar conversion factor to
determine the physician reimbursement for that procedure:
Medicare reimbursement = [(work RVU × work GPCI) + (PE RVU × PE
GPCI) + (PLI RVU × PLI GPCI)] × CF with RVU indicating relative
value unit; GPCI, geographic practice cost indices; PE, practice expense; PLI,
professional liability insurance (malpractice); and CF, conversion factor.
Each component of this system is reviewed individually. It is important to
emphasize that the RVU values are based on survey data collected from
physicians, underscoring how crucial it is for physicians to complete these
surveys consistently and accurately.
A. Physician Work RVU
This component accounts for an average of 54% of the total RVUs for a
service7. A
resource-based physician work RVU is based on the time, technical skill,
physical effort, mental effort and judgment, and psychological stress
(associated with concerns about adverse outcomes) required to perform that
service. The original valuations of physician work RVUs were based on the
research of Hsiao et
al.13.
The American Medical Association (AMA)-Specialty Society Relative Value
Scale Update Committee (RUC) develops physician work RVUs annually for new and
revised CPT codes. The major national medical specialty societies (e.g.,
orthopaedic surgery, general surgery, anesthesiology, family medicine, and
dermatology) appoint twenty-three of the twenty-nine members of the RUC. Three
of these twenty-three seats rotate on a two-year basis, with two reserved for
an internal medicine subspecialty (e.g., geriatric medicine and oncology) and
one for any other specialty (e.g., spine care). The remaining six seats are
occupied by the chair of the RUC, the cochair of the RUC Health Care
Professionals Advisory Committee Review Board, the chair of the Practice
Expense Review Committee, and representatives of the AMA, the American
Osteopathic Association, and the CPT Editorial Panel. The AMA Board of
Trustees selects both the RUC chair and the AMA representative to the RUC.
Every five years, RUC reviews physician work RVUs for existing codes as part
of the required Medicare Five-Year Review.
B. Practice Expense RVU
This component accounts for an average of 41% of the total RVUs for a
service7. There are
two types of practice expenses: direct and indirect costs. Direct costs are
equipment, supplies, and clinical and administrative staff needed to provide a
particular service for a patient. Indirect costs are office rent, equipment,
utilities, and administrative staff who are not directly involved in a
particular individual patient service.
Prior to 1999, practice expense RVUs were based on so-called historic
charges. Practice expense RVUs were calculated by multiplying the national
average allowed charge for each procedure under the CPR system by the
percentage of every specialty's income used to cover practice expenses. The
Social Security Amendments of 1994 mandated that CMS develop a methodology for
resource-based valuation of practice expenses. The Balanced Budget Act of 1997
required that this resource-based system be phased in over four years
beginning in 1999. The Balanced Budget Act also required CMS to reduce the
1998 practice expense RVUs for invasive services to fund an increase in
practice expense RVUs for office visits.
In 1999, a resource-based approach to valuation was implemented on the
basis of two factors: the AMA's Socioeconomic Monitoring System survey and the
Clinical Practice Expert Panel
data14. First, CMS
estimates the specialty's total practice expense pool (as determined from
Socioeconomic Monitoring System data) for six different practice expense
categories for each specialty. An aggregate specialty practice cost
is thus determined, estimating the total practice expenses for that specialty.
Then, CMS allocates each practice expense pool to specific services provided
by that specialty according to estimates of the relative resources required to
deliver each service (based on Clinical Practice Expert Panel data). For the
services performed by multiple specialties, CMS averages the expenses of all
of the specialties that provide that service. Finally, an adjustment is made
so that total physician payments are budget neutral (i.e., the same
as they would have been under previous payment systems). This is a
top-down approach, which is based on a specialty's aggregate practice
expenses (the top) rather than the practice expenses associated with a
particular procedure (the bottom) performed by that specialty.
Since 1999, rates also differ on the basis of the site of service. A
non-facility value is assigned if the procedure is performed in a physician's
office or an independent neuroimaging or laboratory center. A facility expense
is assigned to procedures performed in a hospital, surgical center, or nursing
home. In general, this new resource-based system reimbursement is higher for
services performed in a physician's office and lower for those performed in a
facility, reducing reimbursement for major surgical procedures.
The current top-down approach to calculating practice expense takes into
account both indirect and direct costs by proportionately dividing total
practice expense among procedures. In June 2006, CMS recommended switching to
a bottom-up methodology for calculating practice expense RVUs. This
approach uses procedure-level data for clinical staff times, supplies, and
equipment needed to perform a particular service. It does not take into
account indirect costs not associated with any one particular procedure, e.g.,
rent and
utilities15.
Starting in 2007, CMS is beginning a four-year transition to implementing the
bottom-up methodology for direct costs and updated survey data for indirect
costs. The upcoming changes in how practice expense RVUs are determined willbe
extremely important as practice expense RVUs account for approximately 41% of
Medicare physician reimbursement.
C. Professional Liability Insurance RVU
This component accounts for an average of 5% of the total RVUs for a
service7. Initially,
malpractice RVUs (like practice expense RVUs) were based on historic charges.
In 2000, the malpractice RVU also shifted from being charge-based to
resource-based. This component reflects the national average premium for
malpractice insurance for a specialty and the specialty's risk factor
(determined by dividing the national average premium for each specialty by the
national average premium for the specialty performing that service with the
lowest average premium). This is a specialty-weighted approach in that
malpractice RVUs are based on the weighted average of the risk factors for all
specialties performing that particular service.
D. Geographic Practice Cost Index
The geographic practice cost index takes into account regional differences
in the costs of practicing medicine, such as the cost of office rent and staff
salaries, which vary widely by region across the United States. There is a
separate regional geographic practice cost index published annually for each
of the three RVU components. These are updated every three years for the
eighty-nine Medicare payment localities (states, counties, or groups of
counties). The Medicare Modernization Act of 2003 mandated that, for the next
three years, work RVUs could not be decreased because of the work geographic
practice cost index. This meant that if a physician lives in an area with a
work geographic practice cost index of <1.0, the geographic practice cost
index adjustment is frozen at 1.0. Fifty-eight of the eighty-nine physician
payment areas have benefited from this exemption. This provision was extended
for one more year and is due to expire on December 31,
200716.
E. Conversion Factor
Updated annually, the conversion factor is a multiplier that converts the
geographically adjusted total RVUs for a particular service into a dollar
payment13,17,18.
A discussion of the methodology for determining the annual conversion factor
updates follows.
The AMA Relative Value Scale Update Committee (RUC) annually evaluates and
recommends work RVUs to CMS for new and revised CPT codes. Since 1993, CMS has
accepted, on the average, >90% of the annual RVU recommendations made by
the RUC. Since 2001, CMS has accepted =95% of the RUC's recommendations for
new and revised
codes19.
When a new or revised CPT code is brought forward by the AMA CPT Editorial
Panel, RUC notifies the appropriate specialty societies. These physician
groups then survey their membership regarding the operative duration, surgical
intensity, and frequency and complexity of preoperative and postoperative
visits through an RUC-approved survey. The specialties then present the
results of their survey to the RUC. It should be emphasized that the primary
data used by the RUC to assign or update work RVUs to a particular CPT code is
a RUC-approved survey, and yet only thirty responses are required to make
policy decisions. The results of these surveys, which are highly subjective
and often associated with very low response rates, are the predominant factor
in determining future Medicare reimbursement policy for physician services.
Furthermore, the RUC is composed of a wide range of specialties that, because
of the Medicare budget neutrality rule, each stand to lose reimbursement value
of their own CPT codes if another specialty's code is awarded an increased
RVU. The RUC provides recommendations to CMS, and CMS can accept, reject, or
modify these recommendations.
The Omnibus Budget Reconciliation Act of 1990 required that the Health Care
Finance Administration comprehensively review all existing work RVUs at least
every five years to determine whether they are overvalued, accurately valued,
or undervalued. However, given the current multitude of CPT codes, the burden
of reviewing each code every five years is impossible. Specialty groups and
the public can submit codes for consideration. To have a code considered for
review, they must offer the RUC compelling evidence that the established RVUs
for the service are incorrect and explain why the recommended value is
correct. Additionally, CMS can propose codes for review by the RUC. While CMS
also must explain its rationale for reviewing certain codes, their
explanations are often not as detailed and rigorously supported as those of
other requesting parties. It is widely believed that CMS particularly requests
reviews of high utilization codes. In the most recent five-year review in
2005, CMS requested reviews of CPT codes 27130 (primary total hip
replacement), 27447 (primary total knee replacement), and 27236 (open
treatment of femoral neck fracture). The primary joint arthroplasty RVUs had
not been reviewed since the inception of the resource-based relative value
system in 1992. Each service examined during this process may be judged
undervalued or overvalued, and its RVUs are then adjusted accordingly.
To gather data to support a change or no change in work RVUs, the
appropriate specialty society must send its membership an RUC-approved survey.
The survey includes a case vignette of a so-called typical patient and
procedure, and asks physicians to determine the average amount of time spent
on each portion of the case. The operating-room time encompasses preoperative
evaluation and positioning, scrubbing, dressing, waiting time, "skin to
skin" time, and the immediate postoperative time. The postoperative time
encompasses hospital visits, discharge summary, and office visits for the
entire global period, which includes the first ninety days after surgery for
most surgical procedures. Physicians are also asked to compare the intensity
and difficulty of two
procedures20.
The AAOS typically sends out >1000 twelve-page surveys for each code
being evaluated. The surveys take approximately twenty-five to thirty minutes
to complete. The RUC requires responses from at least thirty physicians for
each code. The survey results are analyzed by the subspecialty group and are
then presented to the AAOS to formulate a final recommendation to the RUC.
While a random sampling of the AAOS membership receives these surveys, members
may also volunteer to fill out the
surveys21. It is
essential that orthopaedic surgeons fill out these surveys accurately. If the
AAOS cannot collect sufficient data to present to the RUC, certain undervalued
services will not receive an increase in RVUs and other procedures deemed
overvalued by the CMS may face declining
reimbursements22.
Lavernia and Parsley raised concerns about the survey's validity,
identifying a number of potential
biases20. The
surveys are not pretested to shed light on item ambiguities and other sources
of bias and error. Furthermore, they have shown that surgeons tend to
underestimate the time spent in the operating room. The validity and accuracy
of these survey results are crucial as the RVUs for both new and revised codes
are assigned on the basis of the survey data.
In preparing its recommendations to CMS, the RUC decides whether to accept
or to modify the specialty society's recommendation. CMS can then accept or
modify the recommendations of the RUC. For the five-year review in 2000, the
RUC recommended that the RVUs be increased for 469 codes, maintained for 311
codes, and decreased for twenty-seven codes. CMS accepted 98% of these
recommendations23.
The third five-year review commenced in November 2004. CMS' final rule was
published in the Federal Register in November 2006; on January 1,
2007, the new RVUs were
implemented23.
The RUC's power has recently come under scrutiny by the Medicare Payment
Advisory Commission (Med-PAC) Chair Glenn Hackbarth, JD. In March 2006,
Hackbarth commented that CMS has "relied too heavily on physician
specialty societies to identify services that are misvalued" and has
therefore not done a "good job of identifying services that may be
overvalued."24
He worried that overvalued surgical codes and the commensurate decrease in
primary care service reimbursement have resulted in "a pretty
precipitous drop-off" for those individuals choosing primary care as a
specialty25. He
proposed the establishment of an outside panel, armed with resources needed to
independently collect data and develop evidence, to identify overvalued
surgical
procedures1,24.
Overall Medicare expenditures (in billions) for Part B have increased
steadily, from $2.2 in 1970, to $11.2 in 1980, $44 in 1990, $90.7 in 2000, and
$153.5 in 2005, primarily because of an increase in the volume and intensity
of physician
services26. Plan B
benefits as a percentage of the GDP has increased from 0.19% in 1970 to 1.20%
in
200526-28.
The resource-based relative value system slowed this growth. Spending on
physician services grew at an average annual rate of 10.8% from 1985 to
199127. With the
introduction of the resource-based relative value system, the growth slowed to
an average annual rate of 4.7% from 1992 to 2000.
To estimate the change in orthopaedic physician reimbursement for surgical
procedures since the resource-based relative value system was implemented, we
identified the twenty-five most common inpatient orthopaedic procedures
performed in United States hospitals, using the most recently available (2004)
complete National Hospital Discharge Survey inpatient data from the
National Center for Health
Statistics29. These
twenty-five procedures encompassed roughly 66% of all orthopaedic inpatient
procedures performed that year (see Appendix).
We focused on three points in time: 1992 (the year that the resource-based
relative value system was implemented), 1998 (the year that the SGR [sustained
growth rate] formula was implemented and the year before the practice expense
RVUs became resource-based), and 2007 (the current value). Between 1992 and
1998, there was an average increase of 12% (range, —47% to +31%) in
reimbursements for these services. Between 1998 and 2007, there was an average
decline of 7% (range, —23% to +36%) in reimbursements for these
services. Overall, between 1992 and 2007, there was an average increase of 4%
(range, —46% to +63%) in reimbursements for these services (see
Appendix).
Using the consumer price index, we factored in the rate of inflation to
reflect more accurately the impact of reimbursement rate changes. The consumer
price index, published by the United States Bureau of Labor Statistics, is a
measure of the prices paid by urban consumers for a market basket of consumer
goods and services (e.g., food and beverages, housing, apparel, and
transportation). The average consumer price index was 140.3 in 1992 and is
202.416 in 2007. A consumer price index adjustment factor of 1.44
(202.416/140.3) was applied to 1992 rates to convert them into 2007 dollar
amounts. With use of this inflation adjustment factor, the average
reimbursement rates, since 1992, for the most frequently performed orthopaedic
surgical procedures have declined 28% (range, —62% to +13%). Total joint
arthroplasty faced the greatest decline during this period
(Table I).
The 1989 Omnibus Budget Reconciliation Act introduced the pivotal concept
of so-called budget neutrality to Medicare Part B by requiring that increases
or decreases in RVUs for a given year may not increase or decrease Medicare
expenditures for that year by >$20 million compared with expenditures
without those changes. This means that upward revisions of existing codes or
additions of new codes must be offset by reductions in reimbursement or
deletion of other codes. There is therefore an inherent internal conflict
among the RUC membership, as a gain for one subspecialty inevitably means a
loss for another. Furthermore, evaluation-and-management billing counts toward
annual Medicare budget targets. Drives to increase reimbursement for clinic
visits (primarily benefiting primary care providers) are therefore balanced by
decreasing reimbursement for invasive procedures.
The Balanced Budget Act of 1997 introduced the sustained growth rate
formula to control the growth in Medicare physician spending by setting annual
budget targets and adjusting the conversion factor accordingly. The sustained
growth rate is based on four factors: (1) medical inflation (i.e., the
estimated percentage change in fees for physicians' services, including drugs
delivered in physician offices); (2) changes in Medicare fee-for-service
enrollment; (3) estimated inflation-adjusted ten-year average growth per
capita in the United States gross domestic product; and (4) changes in
physician spending resulting from law and regulation.
While payments for services are not withheld if the sustained growth rate
target is exceeded, the next Medicare fee schedule update is decreased if
actual expenditures outpace target expenditures the previous
year30. Volume
controls are unique to Medicare physician services and do not limit growth of
any other Medicare expenditure segment.
The annual update in the conversion factor for physician services is
dependent on three factors: (1) the sustained growth rate, (2) the Medicare
Economic Index, and (3) an adjustment factor to bring the Medicare Economic
Index update in line with the sustained growth rate target. The Medicare
Economic Index is a measure of inflation faced by physicians with respect to
their practice costs and general wage levels. The Medicare economic index
reflects annual changes in prices for such things as the physician's own time,
compensation of nonphysician employees, rent, and medical equipment. The
adjustment factor is calculated on the basis of the estimated difference
between the allowed and the actual expenditure for the preceding year, the
actual expenditures during the preceding year, and the sustained growth rate
for the past year. The annual conversion factor update cannot be >3% above
or 7% below the Medicare Economic Index
annually31.
Budget neutrality has divided the physician community into those whose
practice largely involves performing procedures and those whose practice is
largely office based. The most recent five-year review brought this schism
into the forefront. A coalition of medical specialties, led by the American
College of Physicians, proposed that many evaluation-and-management codes be
revaluated16. The
medical community initially expected an overall work value increase of 20% for
evaluation-and-management codes in 2007. In fact, CMS initially cited a 37%
increase in work RVUs for intermediate office visits for an established
patient. However, in 2006, CMS raised work RVUs for 227 services and lowered
them for only twenty-six. To maintain budget neutrality in 2007, CMS reduced
all work RVUs by approximately 10%, using a so-called work RVU adjuster
factor. For surgeons, this will be somewhat offset by an overall increase in
the evaluation- and-management reimbursement for evaluation-and-management
codes built into the ninety-day global surgical
payment16. However,
this adjustor has reduced the overall increase in evaluation-and-management
work RVUs to just 8%32. Nonsurgeons, who rely heavily on
evaluation-and-management reimbursement, are therefore particularly interested
in seeking ways to identify overvalued surgical procedures.
In late 2002, the estimated sustained growth rate for 2003 was
—4.4%33.
However, in 2003, with the retrospective correction of 1998 and 1999 spending
targets, physician fees increased at an average of 1.4%. The Medicare
Modernization Act of 2003 temporarily averted future decreases with a
requirement that the annual conversion factor update be no less than 1.5% for
the years 2004 and
200534. This
averted a projected 2004 decrease of 4.5% and a 2005 decrease of 3.3% in the
conversion
factor27,35.
In 2006, a 4.4% reduction in the conversion factor was seen initially; this
cut was again averted. The Deficit Reduction Act of 2005, signed on February
8, 2006, reverted the conversion factor back to 2005 levels, retroactively
reimbursing physicians who had been paid under the reduced rate since January
200636. Again, a
2007 projected 5% decrease in the conversion factor was averted by another
congressional
freeze16.
While annual congressional freezes have averted recent decreases in the
conversion factor, in fact, when accounting for inflation, doctors actually
experience a relative decrease in their income annually as the conversion
factor fails to increase. None of these legislative actions made revisions to
the sustained growth rate spending targets nor did they address the underlying
flaws in the sustained growth rate formula. Future sustained growth rate
targets will automatically offset this cumulative increased spending with even
further decreased rates. Thus, decreases in physician reimbursements have only
been postponed. Under the current sustained growth rate, physicians should
expect an overall 37% decrease in reimbursements from 2007 to 2015 without
congressional
intervention37.
The AMA and the Alliance of Specialty Medicine have called for an end to
these stopgap measures by fixing the formula itself to prevent further
cuts38. They are
particularly opposed to the current practice of counting Medicare-covered
outpatient drugs and incident-to services (i.e., services performed by
ancillary personnel under the supervision of a qualified Medicare provider)
toward the annual physician reimbursement expenditure targets. In fact,
Medicare Part-B spending on drugs increased from $2.76 billion in 1997 to
$10.87 billion in
200439. The
sustained growth rate also inappropriately links the Medicare fee schedule to
the gross domestic product, which does not accurately reflect increasing
Medicare patient-care costs. While modifications to the current system would
certainly be helpful, the Alliance is, in fact, calling for repeal and
replacement of the sustained growth rate formula and proposing the Medicare
Economic Index as an
alternative38.
MedPAC (Medicare Payment Advisory Commission), an independent federal body
established by the Balanced Budget Act of 1997 to advise the United States
Congress on issues affecting the Medicare program, has also been advocating
for some fundamental Medicare changes. MedPAC does not support projected
sustained physician fee cuts because they fear that such cuts may decrease
access to services for Medicare
beneficiaries24.
MedPAC studied the sustained growth rate approach to containing Medicare costs
and identified four major flaws: payment is disconnected from the cost of
producing services, the volume control mechanism functions as a national
target without incentives for individual physicians to control volume,
regional variations in volume-influencing behavior are not taken into account,
and all volume increases are treated the same whether they are desirable or
not40.
While the sustained growth rate is perhaps not an acceptable way to control
the growth in Medicare spending, CMS is looking for other methods of
instilling financial discipline. It is for this reason that the new sustained
growth rate reform and/or replacement bills are linked to pay-for-performance
proposals to maximize the value of Medicare physician
spending41.
The pay-for-performance program (also known as P4P or value-based
purchasing) is a focal point of Medicare reform. Under a pay-for-performance
system, the focus is on the quality, rather than on the volume and intensity,
of the services
delivered24. The
overriding goal of pay-for-performance programs is to improve quality and
reduce costs by offering physicians meaningful incentives for achieving
"standard, recognized and attainable
measures."42
For example, identifying and treating osteoporosis in patients with fragility
fractures has been proposed as a measure of quality care.
Given the high procedure volumes, high costs, substantial variation in
practice patterns, and perceptions of overutilization and inappropriate
utilization associated with orthopaedic surgical procedures, it is not
surprising that both government-sponsored and commercial health plan
pay-for-performance pilots have targeted orthopaedic surgery. Most notable is
the CMS 2007 Physician Quality Reporting Initiative, mandated by the Tax
Relief and Health Care Act of
200643. Under the
Physician Quality Reporting Initiative, providers who participate in the
Medicare and Medicaid programs will be eligible to receive up to a 1.5% bonus
payment (subject to a cap) on their Medicare and Medicaid claims for reporting
quality measures, in the form of Category-II CPT codes or G-codes, for
services that have a designated corresponding quality measure. Of the
seventy-four measures defined in the Physician Quality Reporting Initiative
Program, ten are related to musculoskeletal
care43. Many policy
experts view voluntary pay-for-reporting as the first step in a sequential
process that will ultimately lead to mandatory pay-for-performance
programs.
There are many concerns about pay-for-performance
programs38. In
fields such as orthopaedic surgery, the evidence-based guidelines that govern
many pay-for-performance initiatives are still in their developmental stages.
There is concern that efficiency measures will be used as a proxy for quality,
which may amount to little more than cost-profiling of providers. Furthermore,
assessing surgical outcomes is complex as a patient's comorbidities and
compliance with postoperative treatment regimens may influence outcomes as
much as, if not more than, the surgeon's skill. To prevent physicians from
"cherry-picking" the healthiest patients, it will be necessary to
implement risk adjustment mechanisms that take into account the fact that
certain patients are inherently more at risk for complications than are
others. Without such an adjustment, it might be economically prohibitive to
treat patients with multiple comorbidities, thus limiting patient access to
care42.
The Alliance of Specialty Medicine, which includes the AAOS, has voiced its
goals and concerns regarding a pay-for-performance
system44. Any
pay-for-performance program should not be subject to budget neutrality (i.e.,
a separate fund should be established to reward high-performing physicians
rather than punitively taking money away from the rest). It should not be used
as a mechanism to control the volume of services provided. Reporting of
quality or efficiency indicators and health outcomes data should not be
administratively prohibitive to physicians. Measures must be
specialty-specific and developed by the physician community. Any
pay-for-performance requirement should be first phased in and pilot-tested on
a voluntary
basis44. While
quality measures can be used as tools for learning and for improvement in
quality and patient safety, physician groups want to make sure that the data
remain confidential and not subject to legal
proceedings15. In
response to the recent increase in pay-for-performance initiatives among both
government and commercial payers, the AAOS has expanded its infrastructure and
intensified its efforts in the area of performance measurement development.
The AAOS leadership stated their commitment to ensuring that these measures
are "developed by the people who deliver the care" and are
"relevant, evidence-based, valid, practicable, not overly burdensome,
pilot-tested, phased-in and risk
adjusted."45-47
Medicare reimbursement policies are rapidly evolving, and the orthopaedic
community needs to make sure that it helps to shape the future of this
program. For example, the American Orthopaedic Association has warned that,
without decisive, proactive action, the federal government will impose a
pay-for-performance model and establish outcome measures without consulting
the orthopaedic
community48.
While pursuing business training (e.g., an MBA) and/or becoming a public
servant (e.g., running for Congress) is obviously not practical for most
orthopaedic surgeons, there are a variety of ways to help shape the future of
Medicare with much lower levels of commitment. For example, the AAOS is
continuously trying to identify undervalued musculoskeletal services and is
calling for members to help in collecting data to defend reimbursement rates.
When you receive a survey to assess resources needed to perform a service, it
is essential that you take the time to complete it honestly and accurately and
return it promptly. The politics of the current five-year review underscores
the importance of these surveys. Furthermore, orthopaedists can engage in
economic studies quantifying the impact of present and future Medicare changes
on the practice of orthopaedic surgery. Studies examining the costs of
providing certain services are useful in identifying undervalued codes for
future review. Concerted orthopaedic political activity is therefore essential
to effect change.
There is a noteworthy example of orthopaedic surgeons identifying flaws in
the Medicare payment system, validating their claims through research,
presenting their findings to the CMS, and consequently effecting real
change49,50.
In March 2003, a team of orthopaedic surgeons formed a multicenter research
group to collect and analyze data on resource utilization in primary and
revision total joint arthroplasty. They found that revision total joint
replacements consumed substantially more resources than primary total joint
replacements, yet hospitals were paid the same for both primary and revision
procedures. In October 2004, the team presented their findings to CMS,
highlighting the impact this payment discrepancy had on quality and patient
access to care. Subsequently, CMS announced that they would delete DRG
(Diagnosis-Related Group) 209 (major joint and limb reattachment procedures of
lower extremity) and replace it with new separate DRGs for primary and
revision total joint arthroplasties. In their final rule, CMS specifically
acknowledged the efforts of these orthopaedic surgeons and cited their
research
findings51,52.
Additionally, CMS acknowledged the impact these changes would have on
improving quality and access to care for patients who need a total joint
replacement by minimizing payment disincentives for providing high-quality
care.
Orthopaedists joined for another successful political mission in 2006. In
June 2006, as part of the current five-year review, CMS proposed to cut work
RVUs for total hip arthroplasty by 21%, work RVUs for total knee arthroplasty
by 10%, and work RVUs for hip fracture treatment by 18%. The orthopaedic
community unified under the leadership of the nonpartisan Orthopaedic
Political Action Committee, worked together, and presented datadriven
arguments to CMS. On November 1, 2006, CMS announced that it would not
implement these reimbursement cuts. This victory underscores the importance of
an orthopaedic political presence in
Washington53.
Medicare has a broad impact on physician reimbursement regardless of the
particular demographic data of the patients seen by an individual physician.
Therefore, an understanding of the basic evolution and structure of Medicare
is important. In 1992, Medicare instituted a physician fee schedule and
transitioned from a charge-based to a resource-based reimbursement system. The
Medicare physician reimbursement for a given procedure equals the sum of three
geographically adjusted relative value units (work, practice expense, and
malpractice) multiplied by a conversion factor.
Physician reimbursement for the most commonly performed orthopaedic
surgical procedures has increased an average of 4% since the introduction of
this Medicare resource-based relative value scale. However, when adjusted for
inflation with use of the consumer price index, the rates have declined an
average of 28% since 1992.
Medicare's unique principle of budget neutrality for physician
reimbursement dictates that increases in reimbursement for one procedure must
be balanced by a decreased reimbursement for another. The current sustained
growth rate formula, implemented to achieve budget neutrality, is flawed and
is calling for annual decreases in the conversion factor.
Large-scale Medicare changes loom on the horizon, particularly in relation
to the reform or replacement of the sustained growth rate formula and the
institution of value-based purchasing (e.g., pay-for-performance) programs.
Although pay-for-performance programs offer the promise that quality will be
recognized in the budget-neutral Medicare environment, a pay-for-performance
program could decrease physician reimbursements across the board and give
bonuses to those meeting certain performance goals defined by the government.
Any pay-for-performance program should be clinically relevant, evidence-based,
valid, not overly burdensome, pilot-tested, phased in, and risk adjusted.
These policies are being created right now, and there are many
opportunities at a range of commitment levels for orthopaedic surgeons to
become involved in shaping Medicare's future. The need for orthopaedic
involvement in the formation of government policies is particularly crucial in
the case of the proposed pay-for-performance program. Government definitions
of so-called quality-care benchmarks as determinants of payment could be
inaccurate and unreasonable without the input of the orthopaedic community.
More economic studies quantifying the cost of providing orthopaedic services
and identifying undervalued codes are also needed.
Tables presenting common Medicare health policy abbreviations, the
twenty-five most commonly performed orthopaedic procedures in 2004, and
changes in Medicare reimbursement rates for these twenty-five procedures from
1992 to the present are available with the electronic versions of this
article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM).
Note: The authors thank Karen M. Bernstein and Louise A. Borda
at Massachusetts General Hospital and Stephanie Marcus at the Library of
Congress for their assistance with data collection and the preparation of this
manuscript.
Fong T. Assessing four decades of
Medicare, Medicaid. Mod Healthc.
2005;35:
6-7,24,42,1.356
2005
Glass KP, Anderson JR. Relative value
units: from A to Z (Part I of IV). J Med Pract Manage.
2002;17:
225-8.17225
2002
[PubMed]
Wilbur RH. "Resource-based"
practice expense: how we got where we are today. Ann Thorac
Surg. 1997;63:
1821-3.631821
1997
[CrossRef]
Hsiao WC, Braun P, Dunn D, Becker ER,
DeNicola M, Ketcham TR. Results and policy implications of the resource-based
relative-value study. N Engl J Med.
1988;319:
881-8.319881
1988
[PubMed][CrossRef]
Medicare program; fee schedule for physicians' services —
HCFA. Final rule. Fed Regist.
1991;56:
59502-811.5659502
1991
[PubMed]
Medicare Payment Advisory Commission.
Report to Congress: Impact of resource-based practice expense payments
for physician services. Washington, DC: Medicare Payment Advisory
Commission; December2004.
2004
Peck B. What surgeons should know
about... The 2007 medicare physician fee schedule. Bull American Coll
Surg. 2007;92:
8-10.928
2007
Scarrow AM. Physician reimbursement
under Medicare. Neurosurg Focus.
2002;12:
e8.12e8
2002
[PubMed]
Lavernia CJ, Parsley B. Medicare
reimbursement: an orthopedic primer. J Arthroplasty.
2006;21(6 Suppl 2):
6-9.216
2006
[PubMed][CrossRef]
National Center for Health Statistics,
Centers for Disease Control and Prevention, US Department of Health and Human
Services. National hospital discharge survey: 2004 annual summary with
detailed diagnosis and procedure data. Series 13, No. 162. 2006 Oct.
.
Accessed 2007 Mar 10.
http://www.cdc.gov/nchs/data/ceries/sr_13/sr13_162.pdf
Office of the Actuary. Centers for
Medicare and Medicaid Services. Estimated sustainable growth rate and
conversion factor, for Medicare payments to physicians in 2006.
Baltimore, MD: Centers for Medicare and Medicaid Services; 2005Dec21.
2005
Brown CA. The 1997 Medicare fee schedule
and results of the five-year review. Bull Am Coll Surg.
1997;82:
18-22.8218
1997
[PubMed]
Ginsburg PB, Berenson RA. Revising
Medicare's physician fee schedule — much activity, little change.
N Engl J Med. 2007;356:
1201-3.3561201
2007
[PubMed][CrossRef]
Emmer S, Allendorf L. The Medicare
Prescription Drug, Improvement, and Modernization Act of 2003. J Am
Geriatr Soc. 2004;52:
1013-5.521013
2004
[CrossRef]
American Academy of Orthopaedic
Surgeons. Medicare Physician Payment was topic of two-day
committee hearing; Rep. Burgess introduces legislation addressing the flawed
formula and quality measures. AAOS Washington Office Legislative Updates. 2006
Jul 28.
.
Accessed 2007 Feb 20.
http://www3.aaos.org/Govern/Federal/Federaleg/2006LegUpdates.cfm
Fong T. The party's over. Amid Medicare
birthday, talk on quality gets serious. Mod Healthc.
2005;35:
10-1.3510
2005
Wenger DR. Leadership in orthopaedics:
taking a stand to own the bone. J Bone Joint Surg Am.
2006;88: 450;
author reply 450-1.88450
2006
[PubMed]
Bozic KJ, Durbhakula S, Berry DJ,
Naessens JM, Rappaport K, Cisternas M, Saleh KJ, Rubash HE. Differences in
patient and procedure characteristics and hospital resource use in primary and
revision total joint arthroplasty: a multicenter study. J
Arthroplasty. 2005;20(7
Suppl 3): 17-25.2017
2005
[CrossRef]
Bozic KJ, Katz P, Cisternas M, Ono L,
Ries MD, Showstack J. Hospital resource utilization for primary and revision
total hip arthroplasty. J Bone Joint Surg Am.
2005;87:
570-6.87570
2005
[PubMed][CrossRef]
Bozic KJ. CMS changes ICD-9 and DRG
codes for revision TJA. AAOS Bulletin.
2005;53:
9-10.539
2005
Centers for Medicare and Medicaid Services (CMS),
HHS. Medicare program; changes to the hospital inpatient
prospective payment systems and fiscal year 2006 rates. Final rule. Fed
Regist. 2005;70:
47277-707.7047277
2005