Statement by
Mark B. McClellan, MD, Ph.D.
Centers for Medicare and Medicaid Servcies
Department of Health and Human Services

Value-Based Purchasing for Physicians under Medicare
House Ways and Means Subcommittee on Health

Thursday, September 29, 2005

Madam Chairman Johnson, Congressman Stark, distinguished Subcommittee members, thank you for inviting me to testify on value-based purchasing for physicians under Medicare.

I want to take this opportunity to thank the physician community for their heroic efforts on behalf of evacuees of hurricanes Katrina and Rita. Physicians rushed to provide care for those in need without even considering payments or program requirements. Providers who were personally affected by the hurricanes as well as those in areas sheltering evacuees have provided extensive medical services under the most challenging conditions. We have acted expeditiously to provide effective support for these efforts. We’ve done this through administrative adjustments to our Medicare and Medicaid payment rules. And we’ve implemented a new Medicaid waiver that provides for immediate, temporary Medicaid coverage as well as financial support for needed medical services that fall outside of standard Medicaid benefits, all using existing systems in the affected states so that they can be implemented quickly and effectively. Within just ten business days CMS reviewed and approved waivers for the states housing the vast majority of evacuees, including Texas, Arkansas, Mississippi, Alabama, Georgia, Florida, Idaho, and the District of Columbia. And we are working closely with all other states that need financial support. Through these efforts, we are helping all evacuees get the care they need as they get back on their feet, we are making sure that the health care providers get reimbursed for providing that care, and we are making sure that the states hosting the evacuees are covered for any substantial expenses that they incur.

In addition, the physician community is one of our key partners as we work to implement the Medicare Modernization Act (MMA). As you well know, we are rapidly approaching the implementation date for Medicare’s new prescription drug coverage. As physicians have known for many years, adequate access to medications is more important today than ever before. Physician organizations have worked closely with us to help inform their membership about the new benefits coming in Medicare to help their patients get access to up to date care. Physicians all over the country are helping beneficiaries take advantage of the new coverage, for example by providing materials in their offices about the basics of Medicare’s prescription drug coverage, and letting them know where to go to get the information and support they need to make a confident decision. The new Medicare drug coverage will be available on time, nationwide, at a lower cost and with more benefits available than many people had expected. As a result, on average seniors will save many hundreds of dollars next year in their total out of pocket costs. I truly appreciate the time and effort physicians are taking to ensure their Medicare patients have access to the medications they need.

As I testified in July, continued improvement of the Medicare program requires the successful participation of physicians and we need to ensure they are adequately compensated for the care they provide to people with Medicare. But how we pay also matters. In addition to providing adequate payments, Medicare’s payment system for physicians should encourage and support them to provide quality care and prevent avoidable health care costs. After all, physicians are in the best position to know what can work best to improve their practices, and physician expertise coupled with their strong professional commitment to quality means that any solution to the problems of health care quality and affordability must involve physician leadership.

Updates to the current payment system for physicians’ services are projected to be negative for the next seven years. Such continued negative updates raise real concerns about this payment system in terms of assuring access to quality care for Medicare beneficiaries. At the same time, simply increasing spending by adding larger updates into the current volume-based payment system that is already experiencing increases of 12 to 13 percent or more per year would have an adverse effect from the standpoint of Medicare’s finances or beneficiary premiums and cost-sharing, and does not promote better quality care.

However, it is clear, under our current system, there is much potential for physicians to improve the value of our health care spending. Under the current system, there are substantial variations in resources and in spending growth for the same medical condition in different practices and in different parts of the country, without apparent difference in quality and outcomes, and without a clear basis in existing medical evidence. A study published in 2003 looked at regional variations in the number of services received by Medicare patients who were hospitalized for hip fractures, colorectal cancer, and acute myocardial infraction. The researchers found that patients in higher spending areas received approximately 60 percent more care, but that quality of care in those regions was no better on most measures and was worse for several preventive care measures. Further, there are many examples of steps that physicians have taken to improve quality while helping to keep overall costs down.

Because it is critical for CMS payment systems to support better outcomes for our beneficiaries at a lower cost, CMS is working closely and collaboratively with medical professionals and Congress to consider changes to increase the effectiveness of how Medicare compensates physicians for providing services to Medicare beneficiaries. I am engaging physicians on issues of quality and performance with the goal of supporting the most effective clinical and financial approaches to achieve better health outcomes for people with Medicare. We are committed to developing reporting and payment systems that enable us to support and reward quality, to improve care without increasing overall Medicare costs. When clear, valid and widely accepted quality measures are in place, pay-for-performance is a tool that could enable our reimbursement to better support efforts to improve quality and avoid unnecessary costs.

Currently, hospitals and physicians are paid under separate systems. Under these systems, physicians do not receive credit for avoiding unnecessary hospitalizations by providing better care to their patients. However, in our physician group practice demonstration project, physicians could receive performance based payments derived from savings from preventing chronic disease complications, avoiding hospitalizations, and improving quality of care.

The evidence is increasing that when we provide an incentive for reporting and achieving better quality, health care providers respond by using payments to take a range of steps from the simple to the high-tech to make it happen. This should not really be surprising – our health professionals are dedicated, and they want to do everything in their power to get the best care to their patients. So when we support better quality, we enable them to do what they do best.

We’ve seen this approach work first-hand with hospital payments where we have tied the annual hospital payment update to quality measure reporting. It has had a positive impact on the availability of quality information, with about 70 percent of hospitals reporting quality data.

Reporting clinically valid quality measures is an important step toward making it easier to achieve major improvements in quality – if you cannot measure it, it is hard to take steps to improve it. We have been working hard in close collaboration with health professionals and other stakeholders to promote the development of better measures.

Voluntary Reporting of Quality Measures Can Be Implemented Soon
Thanks to the leadership and hard work of many physician organizations, we have made considerable progress creating consensus around a set of primary care quality measures. In addition, we have made substantial progress to develop quality measures for the majority of physician specialties. We now have 66 quality measures for 29 specialties. Those 29 specialties represent about 80 percent of Medicare physician spending. We are also developing the infrastructure so that the reporting of these measures on existing physician claims could begin as soon as 2006. While we are still analyzing the issues, we are working out the details so that voluntary reporting can be accomplished under existing statutory authorities.

CMS Works with Partners to Develop, Endorse, and Implement Quality Measures
The ability to evaluate and measure quality is an important component in delivering high quality care. For several years, CMS has been collaborating with a variety of stakeholders to develop and implement uniform, standardized sets of performance measures for various health care settings. In recent months, thanks to the leadership of many physician organizations, these efforts have accelerated even further.

Our work on the quality measures has been guided by the following widely-accepted principles. Quality measures should be evidence-based. They should be valid and reliable. They should be relevant to a significant part of medical practice. And to assure these features, quality measures should be developed in conjunction with open and transparent processes that promote consensus from a broad range of health care stakeholders. It also is important that quality measures do not discourage physicians from treating high-risk or difficult cases, for example, through a risk adjustment mechanism. In addition, quality measures should be implemented in a realistic manner that is most relevant for quality improvement in all types of practices and patient populations, while being least burdensome for physicians and other stakeholders.

More than two years ago, CMS initiated a process with the National Committee for Quality Assurance (NCQA), the American Medical Association’s (AMA) Physician Consortium for Performance Improvement, and other stakeholders to develop measures that would be appropriate for the ambulatory setting. As part of this endeavor, CMS took the lead in supporting the National Quality Forum (NQF) endorsement of ambulatory care measures developed by the NCQA and the Physician Consortium. The NCQA is a private, not-for-profit organization dedicated to improving health care quality by providing information about health care quality to help inform consumer and employer choice. The NQF is a private, not-for-profit membership organization created to develop and implement a national strategy for healthcare quality measurement and reporting. The result of this activity has been the recent endorsement by the NQF of 36 ambulatory quality measures.

Examples of three ambulatory quality measures are the results of the hemoglobin A1C and LDL and blood pressure tests for diabetic patients. The clinical evidence suggests that patients who have a hemoglobin A1C test below 9 percent, an LDL less than or equal to 100 mg/dl, and blood pressures less than or equal to 140/90 mmHg have better outcomes. These measures are evidence-based, reliable and valid, widely accepted and supported, and were developed in an open and transparent manner. Evidence indicates that reaching these goals can lead to fewer hospitalizations by avoiding complications from diabetes such as amputation, renal failure, and heart disease.

Two quality measures endorsed by NQF for heart failure patients include placing the patient on blood pressure medications and beta blocker therapy. Here too, these therapies have been shown to lead to better health outcomes and reduce preventable complications. Together, diabetes and heart failure account for a large share of potentially preventable complications.

In addition to primary care quality measures, other specialties are developing measures. For example, measures of effectiveness and safety of some surgical care at the hospital level have been developed through collaborative programs like the Surgical Care Improvement Program (SCIP), which includes the American College of Surgeons. Preventing or decreasing surgical complications can result in a decrease in avoidable hospital expenditures and use of resources. For example, use of anti-biotic prophylaxis has been shown to have a significant effect in reducing post-operative complications at the hospital level. This measure is well developed and there is considerable evidence that its use could not only result in better health but also avoid unnecessary costs. These post-operative complication measures, which are in use in our Hospital Quality Initiative, are being adapted for use as physician quality measures. Application of this type of post-operative complication measure at the physician level has the potential to help avoid unnecessary costs as well as improve quality.

We also are collaborating with other specialty societies, such as the Society of Thoracic Surgeons (STS), to implement quality measures that reflect important aspects of the care of specialists and sub-specialists. The STS has already developed a set of 21 measures at the hospital level that are risk adjusted and track many common complications as outcome measures. STS is also conducting a national pilot program to measure cost and quality simultaneously, while communicating quality and efficiency methods across regional hubs with the objective of reducing unnecessary complications and their associated cost. The STS measures have been adapted to a set of five quality measures for physicians, such as for a patient who receives by-pass surgery with use of internal mammary artery.

Many other specialties have also taken steps to develop evidence-based quality measures. On July 14, 2005, I sent a letter to many specialty societies, summarizing some of the work to date and requesting an update on their efforts to develop quality and performance measures.

I want to thank the AMA and specialty societies for their very positive response to this effort. Six months ago few specialties had quality measures. Today the majority of specialties have quality measures. Many specialties have created quality task forces and are participating in the quality measurement process. As a result, a total of 66 quality measures now exist covering 29 specialties. These specialties represent about 80 percent of Medicare physician spending. Activity is underway to prepare the other measures for NQF endorsement. The latest version of all 66 quality measures is attached to this statement. Thirty of the measures have been endorsed by NQF or are part of the NQF process.

CMS has had productive exchanges with most medical specialty organizations. I would encourage organizations that have not entered into discussions with us to initiate a dialogue as soon as possible so we can work together to develop clinically valid measures. In certain areas, compliance with evidence-based practice guidelines has the potential to be a quality measure.

The process we have used with the medical profession to develop quality measures beyond ambulatory care should greatly expedite and facilitate the development, acceptance and implementation of quality measures for additional specialties and services. By working in collaboration with the societies, there has been considerable progress in the measure development process. This preparation will facilitate the NQF endorsement process. However, measures that have not yet gone through the NQF endorsement process are still of great value. Physician reporting of these measures will help foster their acceptance in the medical community and help prepare physicians for their eventual adoption. Moreover, since there is likely to be reporting of the quality measures for a period of time before payment based on performance, NQF consensus is not required to begin reporting of such measures. The rapid progress to develop quality measures for the majority of specialties is a clear indication that quality measures are gaining acceptance as an important element in achieving better performance in our health care system.

Our experience with hospital quality measures is that after a measure is endorsed additional work with stakeholders is necessary to assure successful implementation. The Hospital Quality Alliance played an important role in implementation of the hospital quality measures by facilitating hospital adoption and understanding of technical concerns. The Ambulatory Care Quality Alliance (AQA) can serve a similar role to help with physician adoption of the ambulatory quality measures. The AQA is a consortium led by the American Academy of Family Physicians, the American College of Physicians, America’s Health Insurance Plans and the Agency for Healthcare Research and Quality, CMS and other stakeholders, including the AMA and other physician groups, as well as representatives of private sector purchasers and consumers.

CMS is Developing a System to Simplify Reporting of Quality Measures
The development, endorsement, and consensus process is not sufficient to implement measures successfully. Detailed specifications are needed about such items as the associated diagnosis codes and the rules for reporting (e.g., the ordering vs. performing physician). There is also a key issue about how a payor like Medicare can obtain information on the quality measures. For this reason, while the rapid development of quality measures is ongoing, CMS also has been working on the technical methods for supporting effective, simple, and the least burdensome reporting and payment based on these measures.

In the years ahead, it is expected that electronic record systems can be developed that would provide information that is needed to measure and report on quality while fully protecting patient confidentiality. However, while electronic health records would greatly facilitate the accurate and efficient use of information on quality measures and quality improvement, progress on supporting quality improvement should not be delayed until electronic health records are widely used. Indeed, taking steps now to promote quality reporting and improvement also could promote the adoption of and investment by physicians in electronic records, which would facilitate more efficient quality reporting and quality improvement activities. In the short term, there is considerable evidence that information on a broad range of quality measures can be obtained adequately via information transmitted on existing claims. In particular, with adequate guidance for appropriate coding practices by physician offices, the so-called G-codes, HCPCS codes established by Medicare and reportable on existing claims forms, can be the vehicle to report the information on claims. While HCPCS codes generally represent services furnished, the G-codes would report information on the quality measures, and could potentially be a basis for payment based on the report of such information.

We are in the process of converting all the quality measures into a series of G-codes that could then be reported by a physician on a claim in a way that is simple and does not burden physicians. This reporting mechanism has several advantages. It allows collection of information on the quality measures via an existing system familiar to the physician community. It makes reporting of the information simple for physicians. Furthermore, it allows collection of the quality measures to begin very soon – possibly as early as 2006.

Many changes in Medicare involve changes in the systems used by our contractors to pay claims. We are currently assessing whether changes in our contractor systems might be necessary to implement the reporting of information on the quality measures on claims. We are also assessing implementation issues under a scenario where reporting and subsequent performance could result in a payment differential for physicians.

Many believe that a trial period of a year or two might be appropriate where physicians would report on the quality measures, including quality measures that have broad endorsement and support but that have not yet fully completed a formal consensus process. The bill you introduced, Madam Chairman, H.R. 3617, would begin with reporting and move to performance. Some believe that Medicare could establish a payment differential where physicians who report on the measures get a different payment from physicians who do not report. Many believe that after a trial period for reporting Medicare would then move to a system where the payment differential would be based on performance for the measures.

In many ways, where we are today with reporting quality measures for physicians is analogous to where we were before the MMA enacted section 501, the 0.4 percentage point payment differential for reporting of 10 quality measures. Prior to MMA, mechanisms had been established so that hospitals could voluntarily report information on the quality measures. When MMA was enacted, hospitals quickly responded and most of those institutions that had not previously reported the measures did so. Today there are a total of 20 hospital quality measures. Hospitals can voluntarily report information on the additional 10 measures and such reporting does not have a payment consequence. About 70 percent of hospitals are already reporting on 17 of the measures. Information on all measures reported by hospitals is available on the CMS Hospital Compare website.

The bottom line is that quality measures or indicators have been developed or are well along in the development process for most physicians’ specialties. We are currently developing G-codes to report information on these measures on existing claims. We are sorting through systems issues for our contractors. While we still have much work to do, at this point, we believe that we can make rapid progress in very short order so that broad initial reporting of measures that are very relevant to the quality and cost of care for our beneficiaries could begin as soon as 2006.

CMS Works to Ensure Resources are Utilized Appropriately
In many cases, quality measures may help us get more value for our health care dollars. We need to build on this by examining appropriate resource use. The well documented wide variation in resource use among areas for treating the same medical condition raises questions about whether Medicare is getting good value in all areas.

In my June 24 letter to you and Chairman Thomas, I indicated that we supported and were preparing to implement MedPAC’s March recommendation to Congress that: “The Secretary should use Medicare claims data to measure fee-for-service physicians’ resource use and share results with physicians confidentially to educate them about how they compare with aggregated peer performance.”

Measures of physician resource use have been used and are being developed by a number of public and private entities. The most widely used measure of physician resource use is total expenditures per case. Total expenditures include all resources involved in furnishing the case, including physicians’ services, laboratory services and other diagnostic tests, hospital services, other facilities, drugs, durable medical equipment, etc.

The measures of resource use are generally applied to episodes of care. The beginning of an episode may be defined by a new diagnosis or treatment, such as hospitalization. Such episodes usually end after claims related to the episode are not present for a defined period of time. Such episodes could include heart attacks or broken hips. For surgeons, coronary bypass surgery and hip replacements would be considered episodes. Episodes also may occur for a full year in the case of chronic diseases, such as diabetes, heart failure, and chronic pulmonary disease.

We are working to implement the MedPAC recommendation using information derived from claims data. We are developing resource use measures that target particular tests and procedures that may be over- or under-used, as overuse is inefficient and under-use raises quality concerns. We also are developing pilot projects that will use software programs created by a number of private sector entities. These programs group services into episodes using claims data. The episodes are then assigned to physicians so average resource use can be computed. We plan to pilot test resource use for a few selected conditions in two states. We are assessing measurement issues such as case-mix/severity adjustment and identification of appropriate comparison groups. Our goal would be to share results with physicians confidentially to educate them about how they compare to peers and ultimately to incorporate measures related to services, resources, and expenditures into the payment system as envisioned in your bill, H.R. 3617.

Madam Chairman, thank you again for this opportunity to testify on improving how Medicare pays for services. We look forward to working with Congress and the medical community to develop a system that ensures appropriate payments while also promoting the highest quality of care, without increasing overall Medicare costs. As a growing number of stakeholders now agree, we must increase our emphasis on payment based on improving quality and avoiding unnecessary costs to solve the problems with the current physician payment system. Thanks to the leadership of many private-sector organizations working together, and especially thanks to the leadership of physicians, we have made rapid progress in developing quality measures and indicators as well as in building an infrastructure to allow the reporting of such measures. I would be happy to answer any of your questions.

Last Revised: November 17, 2005