Medicare

Floor Speech

Mr. GRASSLEY. Mr. President, as we begin the 112th Congress I want to discuss one of my continuing concerns with the Medicare Program. For the last 10 years, I have served most recently as ranking member and previously as the chairman of the Senate Committee on Finance, which has jurisdiction over Medicare. During this time I have led efforts to reform the Medicare payment system and realign incentives in Medicare to promote higher quality and more efficient care. Today, I would like to address one of the flaws in the Medicare payment system: the inaccuracy of the Medicare geographic adjustment factors used for physician practice expense and the adverse impact they have on rural Medicare beneficiaries' access to care. This flaw has for many years resulted in unfairly low payments to high quality areas like my own home State of Iowa and many other rural States.

Medicare payment varies from one area to another based on the geographic adjustments known as the geographic practice cost indices or GPCIs. These geographic adjustments are intended to equalize physician payment by reflecting differences in physician's practice costs. But they do not accurately represent those costs in Iowa or other rural States. They have failed to do the job. They penalize rather than equalize Medicare reimbursement in rural States and discourage physicians from practicing in areas like New Mexico, Arkansas, Missouri, and Iowa because of their unfairly low Medicare rates. Iowa is widely recognized as providing some of the highest quality care in the country yet Iowa physicians receive some of the lowest Medicare reimbursement in the country due to these inequitable geographic disparities.

I introduced legislation to correct these unwarranted geographic payment disparities in the 110th Congress, the Medicare Physician Payment Equity Act of 2008. In the 111th Congress, I introduced the Medicare Rural Health Access Improvement Act of 2009. And when the Senate Finance Committee conducted its markup of health reform legislation in the fall of 2009, I offered an amendment to reform the practice expense geographic adjustment, PE GPCI, that has caused unduly low payments in rural areas due to the inaccurate data and methodology that is used. My amendment provided more equity and accuracy in calculating this adjustment, and it provided a national solution to the problem. It was accepted unanimously by the Senate Finance Committee, and it was included in the Senate health reform bill, the Patient Protection and Affordable Care Act, PPACA, that was enacted last year.

The goal of my amendment was to assure that the statutory mandate of the Social Security Act is met and that the most recent and relevant data is used for these geographic adjusters. The language of section 3102(b) is very specific. It requires a transitional 2-year period of limited relief to reduce the impact of the current, inequitable practice expense formula in rural areas while a broader analysis of the methodology and evaluation of the data is conducted by the Department of Health and Human Services, HHS. The Secretary is mandated to limit the impact of the existing adjustments by reflecting only one-half of the geographic differences in employee wages and rents in the PE GPCI adjustment for 2010 and 2011 and to hold harmless those localities that would otherwise see a reduction as a result of this adjustment. Most importantly, the provision requires that a longer term solution be implemented in 2012, at which time the Secretary must make appropriate adjustments to the formula to ensure accurate geographic practice expense adjustments.

This 2-year transition in 2010 and 2011 was provided to allow time for a focused, in-depth study by the Centers for Medicare and Medicaid Services, CMS, on the data and methodology used to support a revised PE GPCI formula that would be implemented by January 1, 2012. However, to date CMS has failed to make any significant changes in the sources of the data or the methodology used in calculation of the practice expense adjustment. Although CMS has acknowledged its obligations for an additional study as called for by section 3102(b), they continue to claim that their ``analysis of the current methods of establishing PE GPCIs and [their] evaluation of data that fairly and reliably establish distinctions in the cost of operating a medical practice in the different fee schedule areas meet the statutory requirements'' of section 3102(b), Federal Register, November 29, 2010, Page 73254. I strongly disagree.

When the current Medicare payment system was established, Congress decided that geographic adjustments would be appropriate to equalize physician payment by reflecting differences in physicians' practice costs, and it established the geographic practice cost indices, GPCIs, for physician work, practice expenses, and malpractice premiums. Congress also mandated that HHS use the most recent data available relating to practice expenses in calculating the geographic adjustments for physician practice costs.

However, CMS has long relied upon proxy data sources that bear little to no relevance to actual practice costs, such as using Housing and Urban Development, HUD, apartment rental data to calculate physician office rent. This doesn't have any connection with the cost of office space, let alone a physician's office. Also, the current formula only counts employee wages in four occupations: nurses, clerical personnel and medical technicians but it should reflect employee wages more accurately by also taking into account physician assistants, office administrators, and other more highly compensated specialists commonly employed in practices today. The third category, of ``other'' expense, is considered to be a national market and not adjusted. It should include expenses like office furniture and information technology that cost the same, no matter where you live, but it doesn't. And the weights used by CMS in their methodology are outdated and fail to represent physician practice expenses accurately.

Unfortunately, the more accurate calculation of practice expense costs that was intended to be achieved by my amendment also has been jeopardized by a special interest provision that was added to PPACA behind the closed doors of the majority leader during the Senate floor consideration of health reform. It addresses geographic disparities in Medicare payment but it helps just 5 States at the expense of the other 45 States. It is what I call the ``Frontier Freeloader'' provision. It improves Medicare reimbursement in these frontier States by establishing floors for the hospital wage index and the physician practice expense GPCI. A frontier State is defined as one with 50 percent or more frontier counties, defined as counties with a population per square mile of less than six.

This special deal will ensure that higher payments go to just five rural States in 2011--North Dakota, South Dakota, Montana, Wyoming and Nevada--at the expense of every other State. But the Frontier Freeloader is even more egregious because Iowa and other States like Arkansas and New Mexico that don't benefit from this provision are paying for it! So, taxpayers in your State and mine all the other 45 States--will kick in to pay for this unfair $2 billion Frontier Freeloader carve-out for five States that ends up harming all the other rural States. And that is just the cost for the next few years. The frontier States deal does not sunset, and it is not time-limited. It will continue to benefit so-called ``frontier States'' forever while taxpayers in your State and mine continue to pay the bills. It's another example of how the lack of transparency and the deals made behind closed doors to garner votes last year led to bad policies. And it became law when the President signed the health care reform bill.

I introduced legislation to eliminate the inequitable frontier freeloader provision in the last Congress and to improve Medicare beneficiaries' access to care in all rural States. The Medicare Rural Health Care Equity Act of 2010 would have eliminated this special Medicare reimbursement rate for frontier States and provided additional funds from its repeal to improve reimbursement in all rural States. Iowa provides some of the highest quality care in the country but it does not meet the definition of a frontier State. Certainly Iowa should have been helped since Medicare reimbursement for hospitals and physicians is lower in Iowa than in most of these so-called ``frontier'' States. Medicare also pays much lower rates in other rural States, like Arkansas and New Mexico, but they don't benefit from the Frontier Freeloader because they don't meet the definition of a frontier State. We should improve physician payments for all rural States, not just a select few. And it's unfair to improve hospital payments for just a few States. My legislation would have eliminated those special payments for just five States, and I will be reintroducing that legislation again soon.

The Institute of Medicine, IOM, has been asked by HHS to evaluate the accuracy of the existing geographic adjustment factors and whether the current measures and data are representative of the costs. I have prepared a statement for consideration by the IOM committee charged with this review, the Committee on Geographic Adjustment Factors in the Medicare Program. I urge the IOM to address the inaccuracy of the current geographic adjusters used for physician practice expense, the methodology and data used in their calculation, and the adverse effect of the existing practice expense geographic adjustment factor on rural access to care. I also urge IOM to review the frontier States provision and provide HHS and Congress with recommendations on specific factors that could be used to determine physician practice costs in those States in lieu of the inequitable frontier States floor.

It is my hope that the IOM will carefully consider these comments as it proceeds with its review and develops recommendations and a report to be submitted to HHS and the Congress later this year. I ask unanimous consent that my statement to the IOM be printed in the Record.

There being no objection, the material was ordered to be printed in the Record, as follows:

Statement of Senator Chuck Grassley
(Institute of Medicine, Committee on Geographic Adjustment Factors in the Medicare Program, Jan 5, 2011)

As the senior senator from Iowa and the Ranking Member of the United States Senate Committee on Finance in recent years, I appreciate the opportunity to provide this statement to the Institute of Medicine (IOM) on a study that the IOM has undertaken at the request of the Secretary of the Department of Health and Human Services (HHS) regarding the accuracy of the geographic adjustment factors used for Medicare payment.

For the last ten years, I served either as Ranking Member or as the Chairman of the Senate Committee on Finance, which has jurisdiction over Medicare. During this time, I led congressional efforts to establish more accurate geographic adjusters for Medicare physician payment and to realign incentives in Medicare to promote higher quality and more efficient care. This IOM committee has been asked to evaluate the accuracy of the geographic adjustment factors and to provide their recommendations as to whether the current measures and data are representative of the costs. I would like to address the inaccuracy of the current Medicare geographic adjustment factors used for physician practice expense, the methodology and data used in their calculation, and the adverse effect of the existing practice expense geographic adjustment factors on rural access to care. I offer these comments for consideration by the committee as it proceeds with its review and develops its recommendations and report to HHS and Congress later this year.

MEDICARE'S FLAWED GEOGRAPHIC ADJUSTMENT FACTORS

Medicare's payment system for physicians is flawed in many ways. One of those flaws is the unjustified geographic disparities in payment that has for many years given unfairly low payments to high quality areas like my home state of Iowa and other rural states. Geographic equity in Medicare payment has been a longstanding issue of major concern to me. The new health care reform law, the Patient Protection and Affordable Care Act (PPACA), includes a provision I authored that makes some much needed changes in the calculation of the geographic adjustment factors that is intended to provide more equitable payments to physicians in rural areas and to improve access to health care for Medicare beneficiaries in rural states.

Medicare payment differences from one area to another based on the geographic adjustments known as the Geographic Practice Cost Indices (GPCIs) are intended to equalize physician payment by reflecting differences in physician's practice costs but they do not accurately represent those costs in Iowa or other rural states. They have been a dismal failure, in fact. They discourage physicians from practicing in rural areas because they create unfairly low Medicare reimbursement rates.

I introduced legislation to correct these unwarranted geographic payment disparities in the 110th Congress, the Medicare Physician Payment Equity Act of 2008, as well as the Medicare Rural Health Access Improvement Act of 2009 in the 111th Congress. In the fall of 2009, I also offered an amendment in the Senate Finance Committee markup of health reform legislation to reform the practice expense geographic adjustment that has caused unduly low payments to physicians in rural areas due to the inaccurate data and methodology that is used.

My amendment was intended to provide more equity and accuracy in calculating this adjustment as well as to provide a national solution to the problems that have arisen from the current unwarranted disparities in Medicare payment due to these geographic adjustments. The amendment was accepted unanimously by the Senate Finance Committee during markup of Senate health reform legislation in September 2009. Section 3102(b) of the Patient Protection and Affordable Care Act (PPACA) that passed the Senate and became law is based on this amendment. It requires HHS to improve the accuracy of the Practice Expense Geographic Practice Cost Index (PE GPCI) data and methodology and to examine the feasibility of using actual data or reliable survey data on office rents and non-physician staff wages. These two PE GPCI inputs, which are the only inputs adjusted to reflect local costs, currently do not measure physician costs. Instead, they rely upon proxies. The current input adjustments are not credible because of their reliance on proxy data sources rather than actual physician practice costs. As a result, some physicians are paid more and others are paid significantly less for the very same service with the same time, effort, and expertise needed to furnish that service to a Medicare beneficiary.

I urge the committee to note the wide differences in physician payment under the GPCIs as currently constructed. At the beginning of calendar year 2010, before the transitional adjustments required by PPACA, a 38.894% difference in Medicare physician payment on average existed between the highest paid and the lowest-paid Medicare Part B payment locality (Alaska and Puerto Rico) for the same Medicare service. The PE GPCI disparity for this same period was even greater, ranging from 1.441 (San Francisco) for the highest to 0.694 for the lowest (Puerto Rico) and 0.821 for the second lowest (the rest of Missouri), with 1.0 being the average. The PE GPCI for Iowa was 0.870. This means that physicians in San Francisco received a PE GPCI adjustment that was 144 percent of the average, while Iowa physicians received an adjustment of just 87 percent.

Survey findings of the American Medical Association (AMA) and others challenge this significant range in payment disparity by showing little measurable distinction in physician practice expenses throughout the country. The AMA PPIS is based on actual physician data, rather than the proxy data upon which CMS relies. Geographic distinctions in physician practice expense payment in rural areas should be supported by accurate and reliable data and calculations. I urge the committee to address this discrepancy between credible surveys, based on real physician cost data, and the PE GPCI range established by CMS.

Section 3102(b) requires a transitional two-year period of limited relief to reduce the impact of the current, inequitable practice expense formula in rural areas while a broader analysis of the methodology and evaluation of the data is conducted by HHS. The Secretary is mandated to limit the impact of the existing adjustments by reflecting only one half of the geographic differences in employee wages and rents in the PE GPCI adjustment for 2010 and 2011 and to hold harmless those localities that would otherwise see a reduction as a result of this adjustment. The provision requires that a longer-term solution be implemented in 2012, at which time the Secretary must make appropriate adjustments to the formula to ensure accurate geographic practice expense adjustments. These statutory adjustments were intended to moderate the negative effects of the existing inaccurate GPCI disparities on low-paid Medicare regions while allowing time for a focused, in-depth study by the Centers for Medicare and Medicaid Services (CMS) on the inputs, weights, and data used in the PE GPCI to support a revised formula that would be implemented as of January 1, 2012.

Congress agreed at the inception of the current Medicare payment system that, to the extent physicians practicing in the various Medicare payment localities face higher or lower practice expense burdens, reasonable distinctions in Medicare payment would be appropriate, and it established the Geographic Practice Cost Indices (GPCIs) for physician work, practice expenses, and malpractice premiums to do so. To support the PE GPCI, Congress directed the Department of Health and Human Services to ``use the most recent data available relating to practice expenses ..... in different fee schedule areas.'' (Social Security Act, Section 1848(e)(1)(D)). The statutory requirement makes it clear that there must be a nexus between data sources and actual physician practice expenses as represented by the inputs of the PE GPCI.

However, CMS has long relied upon proxy data sources that bear little to no relevance to actual practice costs. Furthermore, the weights used by CMS are outdated and fail to represent accurately the relativity in expenses in this dynamic and ever-changing field. It is my understanding that the PE GPCI, in particular, is currently supported by data that is neither relevant to physician practices nor credible to physicians. Physicians who serve the Medicare population must bear the burden of their true practice costs while the Medicare payment system upon which they rely fails to reflect those same practice expense costs fairly and accurately.

The goal of Section 3102(b) is to assure that the statutory mandate of the Social Security Act is met and that the most recent and relevant data is used for these geographic adjusters. The language of Section 3102(b) is very specific in its directions but so far CMS has failed to make significant changes in the methodology or data used in calculation of the PE GPCI. The final CMS CY 2011 Medicare physician payment rule sets forth the results of CMS' sixth 3-year GPCI review. Although CMS acknowledged its obligations for an additional PE GPCI study under Section 3102(b) of PPACA, they stated that their ``analysis of the current methods of establishing PE GPCIs and [their] evaluation of data that fairly and reliably establish distinctions in the cost of operating a medical practice in the different fee schedule areas meet the statutory requirements'' of Section 3102(b) (Federal Register, November 29, 2010, Page 73254).

The most recent CMS review and analysis does not provide a new analysis and evaluation of data but merely treads old ground, looking at the PE GPCI underlying data and its weights along the lines of what other studies have already examined. For example, CMS continues to rely, with little justification, on Housing and Urban Development (HUD) section 8 apartment rent data as a proxy for physician rent even though Section 3102(b) directs CMS to evaluate ``the feasibility of using actual data or reliable survey data developed by medical organizations on the costs of operating a medical practice, including office rents and non-physician staff wages in different fee schedule areas.'' If no suitable nationwide data on rental rates for physician office space currently exist, the IOM should recommend other approaches for CMS to use in studying this issue to come up with more reliable data than HUD apartment rents.

CMS acknowledged in the final physician payment rule for CY 2011 that there is much ongoing analysis of the PE GPCI data that could form the basis of future GPCI changes. They stated that they would ``review the complete findings and recommendations from the Institute of Medicine's study of geographic adjustment factors for physician payment'' along with other HHS activities and continue to study the issues as required by Section 3102(b) (Federal Register, November 29, 2010, Page 73256). CMS will consider the GPCIs for CY 2012 again in the context of their annual physician fee schedule rulemaking beginning in CY 2011 based on information that is available then.

A significantly more comprehensive analysis and detailed evaluation should be conducted for the PE GPCI study mandated by Section 3102(b) than what has been detailed by CMS in its final CY 2011 Medicare physician payment rule. New studies, data, and other approaches must exist or be developed to facilitate reliability and accuracy in identifying actual physician practice expenses and setting weights among those expenses. That is why a two-year transition was provided: to ensure that CMS would have sufficient time to do additional studies, if needed, and come up with more meaningful data than, for example, continuing to use apartment rental data which bears no relation to the cost of a physician's office. I urge the committee to provide CMS with specific recommendations for more accurate methodology that could be used to determine the PE GPCIs and obtain more reliable actual or survey data sources to be used in these calculations.

THE INEQUITABLE FRONTIER STATES PROVISION

Unfortunately, the more accurate calculation of practice expense costs that was intended to be achieved by Section 3102(b) has been jeopardized by a special interest provision that was added to PPACA behind closed doors during the Senate floor consideration of health reform. The ``frontier states'' provision addresses geographic disparities but helps just five states at the expense of the other 45. It improves Medicare reimbursement in the so-called frontier states by establishing a permanent 1.0 floor for the PE GPCI as well as for the hospital wage index, effective January 1, 2011. A frontier state is defined as one with 50 percent or more frontier counties, defined as counties with a population per square mile of less than six. The frontier states provision ensures that higher Medicare physician payments resulting from a higher PE GPCI adjustment go to just five states in 2011--Montana, Wyoming, North Dakota, South Dakota, and Nevada.

Iowa provides some of the highest quality care in the country but it does not meet the definition of a frontier state. Yet Medicare reimbursement for hospitals and physicians is lower in Iowa than in most of these so-called frontier states. Medicare also pays much lower rates in other rural states that do not meet the definition of a frontier state.

The frontier states provision is even more egregious because taxpayers in all 50 states will help pay the estimated $2 billion cost for a provision that benefits just five states. That amount is the Congressional Budget Office cost estimate of the frontier states provision for the next ten years. A practice expense floor for rural states may be warranted but it should not be an adjustment for just a few select states. This automatic pay increase for frontier state physicians could result in reduced access for Medicare beneficiaries in nearby rural states that do not have the 1.0 PE floor if physicians migrate to those rural areas where Medicare payment has been significantly increased.

Last spring I introduced legislation, the Medicare Rural Health Care Equity Act of 2010, to eliminate the special Medicare reimbursement rates for frontier states. It is imperative to reduce unwarranted geographic disparities and base physician practice expense costs on actual or reliable survey data, not by legislative fiat that improves physician payments for just a few states. Although legislative action would be required to make changes in this regard, I urge the IOM to review this situation and provide recommendations to HHS on whether specific factors should be considered to determine physician practice costs in frontier states if such a floor did not exist.

CONCLUSION

The practice expense geographic adjustment factor has a significant impact on the health care workforce in rural areas, because it plays a major role in the ability to recruit and retain physicians in rural areas who see more patients and work longer hours for correspondingly lower pay. This in turn can result in Medicare beneficiaries in rural areas having reduced access to physicians and other health care practitioners. Twenty percent of the population lives in rural America yet only nine percent of physicians practice there. Shortages of primary care and specialty physicians currently exist in many rural areas yet unwarranted geographic payment disparities make it difficult to improve access for rural Medicare beneficiaries and other patient populations.

The existing inaccurate geographic adjustments by CMS result in unwarranted and unduly low rural reimbursement rates. More current, relevant, and accurate data sources exist and should be used by CMS to make geographic adjustments to Medicare payments, especially in the area of physician practice expense. The current geographic disparities in payment are not based on actual or reliable data, and they put rural Medicare beneficiaries at risk. I urge the committee to recommend that CMS use actual practice cost data rather than the current inaccurate proxies to ensure that Medicare payment reflects true geographic differences in physician practice costs.

Source