Academy comments to CMS on the proposed rule regarding "Medicare Program: Revisions to Payment Policies Under the Physician Fee Schedule for Calendar Year 2006"

Mark B. McClellan, M.D., Ph.D.
Administrator, Centers for Medicare and Medicaid Services (CMS)
Department of Health and Human Services
Attention: CMS-1502-P
P.O. Box 8017
Baltimore, MD 21244-8017

Dear Dr. McClellan:

I am writing on behalf of the American Academy of Family Physicians, which represents more than 94,000 family physicians and medical students nationwide. Specifically, I am writing to offer our comments on the proposed rule regarding “Medicare Program: Revisions to Payment Policies Under the Physician Fee Schedule for Calendar Year 2006,” as published in the Federal Register on August 8, 2005. We especially draw your attention to our comments on the Sustainable Growth Rate and payment reforms that promote higher quality ambulatory care.

CMS proposes the following changes to its practice expense methodology:

The Academy has historically supported a bottom-up approach to the calculation of direct practice expenses and elimination of the non-physician work pool. As such, we support CMS’s proposals in this regard and believe they are long overdue, even though family physicians will benefit only marginally from the proposed changes (i.e., CMS estimates allowed charges by family physicians will only be 0.2% greater after the proposed changes are fully implemented in 2009).

We agree with CMS that these changes will produce a more accurate, more intuitive, and more stable practice expense methodology. The methodology will be more accurate, because, as CMS notes, the practice expense inputs are better refined and more current than those historically used by CMS. Also as CMS notes, we believe it will be more accurate because the bottom-up methodology assumes that the costs of the clinical staff, supplies and equipment are the same for a given service, regardless of the specialty that is performing it. This assumption does not hold true under the top-down direct cost methodology, where the specialty-specific scaling factors create widely differing costs for the same service.

The bottom-up methodology has always been more intuitive to us than the top-down method employed by CMS, and the elimination of the non-physician work pool will make it even more intuitive. As noted in the proposed rule, under the proposed changes, any revisions made to the direct inputs would now have predictable results, since changes in the direct practice inputs for a service would proportionately change the practice expense RVUs for that service without significantly affecting the practice expense RVUs for unrelated services. Finally, stability should be improved, because direct practice expenses should only change for a service if it is further refined or when prices are updated, while indirect practice expenses should change only when there are changes in the mix of specialties performing the service or with the use of any future new survey data for indirect costs.

With respect to the indirect practice expenses, we continue to question why CMS uses physician work RVUs, rather than physician time, in its formula for allocating indirect expenses. We continue to believe physician time makes more sense than physician work in this regard. By definition, physician work is a product of time and intensity. We would contend that physician time is more likely than physician work to drive indirect expenses since we are not convinced such expenses vary with physician intensity. For example, consider two physician services, done in an office, that involve the same amount of time but different levels of intensity. Does it make sense to say that the cost of the utilities varies with the intensity of the service when, for example, the cost of the electricity is, in fact, a function of the time the lights are on while the services are being done? We do not think so. Therefore, as we did when CMS first proposed its formula for indirect practice expense RVUs, we would recommend that CMS use physician time rather than physician work in the allocation of indirect expenses.

Lastly, as concerns the four-year transition for these changes, we understand CMS’s desire to ameliorate the impact of the other proposed changes, especially for those specialties that will be significantly, negatively impacted by the changes. We also appreciate CMS’s desire to give everyone affected ample opportunity to identify any anomalies in the practice expense data, to make any further appropriate revisions, and to collect additional data, as needed prior to the full implementation of the proposed changes. However, we would note that Congress enacted the original legislation mandating resource-based practice expenses in 1994 and that CMS delayed its initial implementation by a year before entering a four-year transition to resource-based practice expenses under its current methodology. Thus, it has already been a decade of delays and transitions since Congress directed CMS to implement resource-based practice expenses, and we find it ironic that CMS proposes to further draw out that transition. We would encourage CMS to shorten or eliminate the transition and finally complete the process of implementing resource-based practice expenses.

Since 2000, CMS has excluded cast and splint supplies from the practice expense database for the Current Procedural Terminology (CPT) codes for fracture management and cast/strapping application procedures, since these supplies could otherwise be separately billed using Healthcare Common Procedure Coding System (HCPCS) codes Q4001 through Q4051. CMS now proposes to eliminate the separate HCPCS codes for these casting supplies and to again include these supplies in the practice expense database. By bundling the cost of the cast and splint supplies into the practice expense component of the applicable procedure codes under the fee schedule, physicians will no longer need to bill Q-codes in addition to the procedure codes to be paid for these materials. This change would affect the practice expense RVUs for the following CPT codes: 23500 through 23680, 24500 through 24685, 25500 through 25695, 26600 through 26785, 27500 through 27566, 27750 through 27848, 28400 through 28675, and 29000 through 29750.

We appreciate that this proposal makes coding and billing for fracture management and casting/strapping easier by reducing the number of codes that physicians must submit in such situations. We also appreciate that CMS has invited the relevant medical specialties to review direct practice expense inputs for the codes in question and provide CMS with feedback regarding the appropriateness of the type and amount of casting and splinting supplies and about the amount of casting supplies needed for the 10-day and 90-day global procedures. We would encourage CMS to vet the feedback it receives with the Practice Expense Review Committee of the American Medical Association/Specialty Society Relative Value Scale Update Committee, so the resulting inputs enjoy the same level of scrutiny and cross-specialty refinement that all of the other direct practice expense inputs have.

In Table 19 of the proposed rule, “Equipment Items Needing Specialty Input for Pricing and Proposed Deletions,” CMS indicates that it is seeking pricing information on hyperbaric chambers from family medicine, internal medicine, and emergency medicine, based on the fact that these three specialties are the primary providers of CPT code 99183, “Physician attendance and supervision of hyperbaric oxygen therapy, per session.” CMS currently assigns a price of $125,000 to this equipment.

While family physicians may be a common provider of this service according to Medicare data, we do not believe that they are common purchasers of such equipment. According to Medicare data, over 90% of hyperbaric oxygen therapy is done in the hospital setting, and only 8% occurs in the physician office setting. We believe that very few, if any, family physicians own a hyperbaric chamber. Accordingly, CMS may want to consult with the hospital community about the cost of this equipment. Alternatively, CMS may want to contact the American College of Hyperbaric Medicine (ACHM) for any information it has in this regard. CMS can access contact and other information about the ACHM on-line at www.hyperbaricmedicine.org.

In the proposed rule, CMS proposes to make some adjustments to its malpractice RVU methodology. These adjustments include excluding data for any specialty that performs less than five percent of a particular service or procedure from the malpractice RVU calculation for that service or procedure. The purpose of applying the minimum threshold was to identify and remove from the data specialties listed infrequently as performing a certain procedure, under the assumption that the infrequent instances of these specialties in CMS’s data represent aberrant occurrences and that removing the associated risk factor from the malpractice RVU calculation would improve accuracy and stability of the RVUs.

We have reviewed the proposed adjustments and have no objections as a result. As noted in the proposed rule, for the most part, they have little or no affect, either on individual codes or on specialties. For instance, the overall impact of removing the risk factor for specialties that occur less than five percent of the time in the CMS data yields no impact on the malpractice RVUs for over 5,280 codes, and there is an impact of less than 1 percent on the malpractice RVUs for over 1,300 additional codes. Only 16 codes decrease by at least 0.1 RVUs, with the biggest decrease being a negative 0.28 impact on the malpractice RVU for CPT code 17108, “Destruction of skin lesions, from a current RVU of 0.82 to a proposed RVU of 0.54.” Conversely, there are only 219 codes for which RVUs increase by at least 0.1, the largest increase being a positive 0.81 RVU increase for CPT code 61583, “Craniofacial approach, skull,” from a current RVU of 8.32 to a proposed RVU of 9.13. Among codes whose malpractice RVUs would increase under CMS’s proposal, 646 have increases of less than 1 percent. The impact by specialty also appears minimal. For family medicine, there would be no impact on allowed charges. For all other physician specialties, the impact ranges from -0.1% to +0.2%.

One adjustment that we did not note in the proposed rule was the use of more current and more accurate premium data. We continue to have concerns that CMS’s use of “actual” premium data is progressively less comprehensive over time. As we understand it, CMS is using actual data for 2001 and 2002 and projected data for 2003. The actual data for 2001 represents 46 states plus Washington, DC, and Puerto Rico. The data for 2002 represents only 33 states. And, the 2003 data was projected because none of the State Departments of Insurance had market share information at the time the premium data was collected and because private insurers were not amenable to releasing premium data to CMS. Thus, out of three years of data for 50 states plus DC and Puerto Rico, CMS is using “actual” data for only about half the combination of time and states.

We continue to find this paucity of actual premium data troubling. We find it hard to fathom why CMS, working, for example, with the National Association of Insurance Commissioners, can’t obtain market share data for 2003 now that we are over halfway through 2005. The reluctance of liability insurers to disclose premium data to CMS is understandable. However, the insurers are not the only ones with such information. The physicians who pay those premiums also know what they are, and we understand that the RUC has offered to work with CMS to collect premium data from physicians in an effort to get more current data. We would continue to encourage CMS to pursue this offer and other avenues for collecting more current actual data, rather than relying, as it apparently does, on an amalgam of actual and projected data.

Medicare regulations at 42 CFR 410.78 currently require the use of an interactive two-way audio and video telecommunications system to furnish a telehealth consultation. In response to a request from the Medical College of Georgia (MCG), CMS is considering whether or not to modify its definition of an interactive telecommunications system to permit one-way video in place of an interactive two-way video system. In this case, the one-way video permits the consulting physician at the distant site to see the patient, but the patient and presenting physician or practitioner cannot see the consultant.

The use of an interactive audio and video telecommunications system permitting two-way real-time interaction between the physician or practitioner at the distant site and the beneficiary and tele-presenter (if necessary) at the originating site is a substitute for the face-to-face examination requirements of a consultation under Medicare, according to CMS. CMS is concerned that the use of one-way video may not be clinically adequate for the evaluation of certain types of patients. Since telehealth services are intended as a substitute for services that traditionally require a face-to-face interaction between a physician (or practitioner) and a patient, CMS believes that the use of a two-way video communication is much less of a departure from this standard than a one-way video communication, because the face-to-face interaction between a physician and a patient allows two-way interactive communication, both verbally and physically. CMS is concerned that, without two-way video, communication of many subtle but important nuances of the interaction between the physician at the distant site and patient or clinical staff at the originating site would be lost, leading to reduced diagnostic accuracy and the possibility of unfavorable medical outcomes.

However, CMS recognizes that in some situations, such as the one described by the MCG (i.e., a timely neurological evaluation for determining suitability for tPA treatment in acute ischemic stroke patients), the benefit of using available one-way video in a timely fashion may outweigh the benefits otherwise attributed to two-way video. Therefore, CMS is reviewing its definition of an interactive telecommunications system and requests specific public comments regarding the added clinical value of two-way interactive video as compared to one-way video for the purpose of furnishing telehealth services. CMS is also interested in receiving comments as to whether an interactive audio and one-way video telecommunications system that permits the physician at the distant site to examine the patient in real-time is clinically adequate for a broad range of specialty consultations.

From an AAFP perspective, telemedicine is the integration of information technology and telecommunication to enhance healthcare delivery. We believe that telemedicine can enrich the delivery of medical care at remote sites such as rural areas and, by creating ready access to information, can provide rural physicians with current medical information that may not be available in an isolated setting. We believe that reimbursement should be made for physician services that are reasonable and necessary, safe and effective, medically appropriate and provided in accordance with accepted standards of medical practice. The technology used to deliver the services should not be the primary consideration; the critical test is whether the service is medically reasonable and necessary.

From our perspective, two-way video may add value to the telehealth consultation by allowing the patient and presenting physician or practitioner to see the body language and non-verbal communication presented by the consulting physician. However, we would not advocate for denial of telehealth services provided using interactive audio and one-way video telecommunication equipment.

In fact, an interactive audio and one-way video telecommunications system that permits the physician at the distant site to examine the patient in real-time is clinically adequate for a broad range of specialty consultations. As noted above, we believe it is the nature of the service, not the nature of the technology that is paramount. Using the analogy of a face-to-face examination put forth by CMS, we would note that a telehealth consultation using interactive audio and one-way video telecommunication equipment is comparable to a physician treating a blind patient. CMS would not deny a face-to-face consultation on the basis that the patient could not see the consulting physician; we fail to see why it would deny a telehealth consultation on the same basis. Accordingly, we would encourage CMS to revise its definition of an interactive telecommunications system to include interactive audio and one-way video telecommunication equipment.

Medicare has a longstanding policy of reducing payment for multiple surgical procedures performed on the same patient, by the same physician, on the same day. In those cases, full payment is made for the highest priced procedure and each subsequent procedure is paid at 50 percent. Effective January 1, 1995, the multiple procedure policy, with the same reductions, was extended to nuclear medicine diagnostic procedures (CPT codes 78306, 78320, 78802, 78803, 78806 and 78807).

In the current proposed rule, CMS proposes to extend the multiple procedure payment reduction to technical component (TC) only imaging services and the TC portion of global imaging services for certain imaging modalities (i.e., ultrasound, computed tomography (CT), computed tomographic angiography, magnetic resonance imaging (MRI), and magnetic resonance angiography) that involve contiguous body parts within a family of codes. For example, the reduction would apply to an MRI of the abdomen done on the same date as an MRI of the pelvis but would not apply to a CT of the chest done on the same date as a CT of the head. From CMS’s perspective, when multiple images are acquired in a single session, most of the clinical labor activities (e.g., greeting, positioning, and escorting the patient, providing education and obtaining consent, retrieving prior exams, setting up the IV, and preparing and cleaning the room) and most supplies (except film) are not performed or furnished twice. Thus, CMS believes it is making duplicate payments for TC of multiple diagnostic imaging services, particularly when contiguous body parts are viewed in a single session. The discount would not apply to the professional component (i.e., the physician interpretation and report) of such procedures.

We concur with CMS’s proposal and encourage its adoption. We agree that CMS is making duplicate payments for certain practice expenses when multiple images are acquired in a single session, especially when contiguous body areas are involved. As the Medicare Payment Advisory Commission (MedPAC) noted in its March 2005 Report to Congress, in which it made a similar recommendation, Medicare’s payment rates are based on each service being provided independently. When that is not the case, certain efficiencies may accrue of which CMS should take advantage. This proposal attempts to do just that.

We would also urge CMS to consider applying a reduction to the professional component in such situations as well. Just as with the technical component, there are certain efficiencies when a physician is reading images of contiguous areas of the same patient on the date. For instance, the interpreting physician only has to review the patient’s history once to know what he or she is seeking, and often, some portion of the scan is an overlap (i.e., a scan of the pelvis often includes a portion of an abdominal scan). Also, usually there is only one dictation for the multiple scans. Accordingly, there is less physician work involved than would be the case if the scans were interpreted independently at different points in time, and CMS should consider applying the multiple procedure reduction to the professional component, too.

Currently, Medicare regulations at 42 CFR 410.23(a)(2), which govern conditions for and limitations on coverage of screening for glaucoma, define the term "eligible beneficiary'' to include individuals in the following high risk categories:

Based on its review of the current medical literature, CMS believes that the evidence is adequate to conclude that Hispanic persons age 65 and older are at high risk and could benefit from glaucoma screening. Therefore CMS proposes to revise the definition of an eligible beneficiary to include Hispanic Americans age 65 and over. In view of the possibility that it may be appropriate to include other individuals in the statutory definition of those at "high risk'' for glaucoma, CMS is requesting comments on this issue and is specifically requesting that anyone providing it with specific recommendations on this issue also provide documentation in support of them from the peer-reviewed medical literature.

Like CMS, we have reviewed the current medical literature on this subject. Unlike CMS, we found insufficient evidence to recommend for or against screening adults for glaucoma. We also note that the U.S. Preventive Services Task Force (USPSTF) found insufficient evidence to recommend for or against screening adults for glaucoma. The USPSTF’s recommendation, including clinical considerations and references from the peer-reviewed literature can be found online at www.ahrq.gov/clinic/uspstf05/glaucoma/glaucrs.htm#clinical. The clinical considerations note that increased intraocular pressure, family history, older age, and being of African American descent place an individual at increased risk for glaucoma. However, there is no reference to Hispanic persons age 65 and over being at high risk. Based on this information, we would not support CMS’s proposal to extend its definition of eligible beneficiary to include Hispanic persons age 65 and over or any other category of individual not already covered by the current definition.

In the proposed rule, CMS notes that it is currently forecasting an update of -4.3% for 2006, and it attributes much of the blame for this reduction to substantial growth in Medicare spending, especially:

CMS encourages comments regarding possible changes to the SGR methodology, including the legal theories that support them. CMS is particularly interested in:

As regards the SGR, we oppose expenditure targets, including the SGR, which are linked to the Gross Domestic Product, in favor of a system based on a fair representation of physicians' costs of delivering care. The cuts in physician payment projected by CMS result from the inherently flawed SGR payment update formula. The SGR, which is linked to the Gross Domestic Product, penalizes physicians and other practitioners by failing to accurately reflect volume increases resulting from new coverage decisions and initiatives promoted by the Federal government. Only physicians are subject to arbitrary cuts due to factors beyond their control. Every other category of health care provider receives positive updates, based on a measure of inflation in their practice costs. That is why we and other physicians are working to enact a bill to replace the SGR with a new payment system that reflects increases in medical practice costs. We note that MedPAC also calls for repealing the SGR formula and basing the conversion factor on the Medicare Economic Index (MEI).

Until a complete revision of the reimbursement formula is accomplished, there is an administrative adjustment that CMS can make immediately. Specifically, CMS should immediately remove, retroactive to the inception of the SGR, the physician-administered drugs from the SGR. These in-office medications are not reimbursed under the Medicare physician fee schedule and should never have been part of the formula used to calculate the conversion factor for physician services. Moreover, the Medicare Modernization Act restructured how these medications are paid for. CMS’s continued inaction, in the face of a growing Medicare ambulatory care reimbursement crisis, is of great concern.

Indeed, we fail to see how CMS can include such drugs in the SGR calculation, given the statutory language surrounding SGR. Specifically, section 1848(f)(2) of the Social Security Act specifies that the SGR for all physicians' services shall be equal to the product of:

(A) 1 plus the Secretary's estimate of the weighted average percentage increase (divided by 100) in the fees for all physicians' services in the applicable period involved,
(B) 1 plus the Secretary's estimate of the percentage change (divided by 100) in the average number of individuals enrolled under this part (other than Medicare+Choice plan enrollees) from the previous fiscal year to the applicable period involved,
(C) 1 plus the Secretary's estimate of the projected percentage growth in real gross domestic product per capita (divided by 100) from the previous applicable period to the applicable period involved, and
(D) 1 plus the Secretary's estimate of the percentage change (divided by 100) in expenditures for all physicians' services in the fiscal year (compared with the previous fiscal year) which will result from changes in law and regulations, determined without taking into account estimated changes in expenditures resulting from the update adjustment factor determined under subsection (d)(3)(B) or (d)(4)(B), as the case may be, minus 1 and multiplied by 100.

As far as we can see, there is no reference to drugs administered incident to a physician’s service anywhere in this formula. There is a reference to “physicians’ services,” but section 1861(q) defines those as “professional services performed by physicians, including surgery, consultation, and home, office, and institutional calls (but not including services described in subsection (b)(6) [i.e., services of interns or residents in training under an approved teaching program] ).” Further, section 1861(s) of the Social Security Act, which defines “medical and other health services,” draws a distinction between “physicians’ services” (1861(s)(1)) and “services and supplies (including drugs and biologicals which are not usually self-administered by the patient) furnished as an incident to a physician's professional service, of kinds which are commonly furnished in physicians' offices and are commonly either rendered without charge or included in the physicians' bills” (1861(s)(2)(A)).

Since the statute states that “physicians’ services,” not “medical and other health services,” should be included in the SGR computation, we believe that CMS cannot consider drugs furnished in a physician’s office that are not usually self-administered in the SGR. Not only does CMS have the obvious statutory authority and guidance to not include these in-office drugs in the SGR, but they have the obligation to remove them from the SGR. That being the case, and since they have been erroneously included in the SGR since the inception of the formula, these drugs should be removed retroactive to the first use of the formula, and CMS should adjust its calculations accordingly.

With respect to payment reforms that promote higher quality ambulatory care, we support Congressional actions to achieve the following regarding physician reimbursement in federally-funded health care programs:

Having a usual source of care is extremely influential in the care people receive. For instance, people who have a usual source of care are also more likely to receive preventive care services, independent of having insurance. (DeVoe JE, Fryer GE, Phillips RL, Green LA. Comparing receipt of preventive care among adults based on insurance status and/or a usual source of care. Am J Public Health 2003; 93(5)). Likewise, we know that for people with chronic health conditions, there are reductions in expenditures with no significant differences in self-rated health status when people have a family physician as their usual source of care (2000 Medical Expenditure Panel Survey). Finally, several studies have confirmed improved coordination of care and reduced expenditures by providing people with a primary care physician as a usual source of care. In fact, Bodenheimer et al. found that 18 of 27 studies concerning just three chronic conditions (congestive heart failure, asthma and diabetes) demonstrated reduced costs or lower use of health care services when the chronic care model was fully implemented, almost exclusively in primary care settings (Bodenheimer T, Wagner EH, Grumbach K., Improving primary care for patients with chronic illness: The chronic care model, part 2. JAMA 2002; 288(15):1909-1914).

Yet, there is no direct compensation to family physicians and other primary care physicians for the considerable time and effort of assuring that each patient’s care is organized correctly and is integrated in a way that makes sense to the patient, while remaining cost-effective to the Medicare program. As noted, such coordination and management ensure that patients’ health is stabilized and that expensive progression of their illness and costly hospitalizations are minimized. A blended model of payment combining fee-for-service reimbursements plus a per-beneficiary, per-month stipend for care management, paid directly to each patient’s designated physician (“personal medical home”), is a promising option that would enable family physicians to redesign their offices to deliver high quality preventive and chronic care with improved outcomes for Medicare beneficiaries.

MedPAC recommended in its March 2005 report that Congress create Medicare pay-for-performance programs for physician services. The Academy would support a Medicare pay-for-performance program for physicians that:

AAFP currently has over 57,000 members in active practice, the vast majority of whom are in small and medium size practices, not large groups. We anticipate that this will be the typical construct of family medicine well into the future. Most people in this country receive the majority of their care from physicians in small and medium size ambulatory care settings. Currently about a quarter of all office visits in the U.S are to family physicians, and the average family practice has about a quarter of patients who are Medicare beneficiaries. Implementing value based purchasing or pay-for-performance in the Medicare program has tremendous implications for millions of patients and for the specialty of family medicine, and AAFP is therefore committed to involvement in the design of a new pay-for-performance program that meets the needs of patients and physicians.

The AAFP supports moving to pay-for-performance in the Medicare program with the goal of continuously improving care of patients. As we recently stated in a joint letter to Congress with our colleague organizations ACP, the American Academy of Pediatrics, and the American College of Obstetricians and Gynecologists, “we believe that the medical profession has a professional and ethical responsibility to engage in activities to continuously improve the quality of care provided to patients… Our organizations accept this challenge.” We have committed to work toward transformation of medical practice, to strengthen the infrastructure of medical practice to support pay-for-performance, and to engage in development and validation of performance measures. While several specific issues remain that must be addressed in implementing pay-for-performance in Medicare, AAFP has a framework for a phased in approach for Medicare.

AAFP is involved in several efforts that are fundamental to moving toward a pay-for-performance system.

First of all, we know that the development of valid, evidence-based performance measures is imperative for a successful program to improve health quality. The AAFP participates actively in the development of performance measures through the Physician Consortium for Performance Improvement. We believe that multi-specialty collaboration in the development of evidence-based performance measures through the consortium has yielded and will continue to yield valid measures for quality improvement and ultimately pay-for-performance.

The AAFP was the first medical specialty society to join the National Quality Forum (NQF). And along with ACP, AHIP and AHRQ, the AAFP is a founding organization of the Ambulatory care Quality Alliance (AQA). However, it is important to distinguish between the role of the NQF and that of AQA. With its multi-stakeholder involvement and its explicit consensus process, the NQF provides essential credibility to the measures that it approves – measures developed by the Physician Consortium, NCQA and others. The AQA’s purpose is to determine which of the measures approved through the NQF consensus process should be implemented initially (the starter set), and which should then be added so that there is a complete set of measures, including those relating to efficiency, sub-specialty performance, and patient experience. Having a single set of measures that can be reported by a practice to different health plans with which the practice is contracted is critical to reducing the reporting costs borne by medical practices. Measures that ultimately are utilized in a Medicare pay-for-performance program should follow this path.

Health information technology effectively utilized in the physician’s office is necessary to the success of quality improvement and pay-for-performance programs. We have learned from the Integrated Healthcare Association’s (IHA) experience in California that physicians and practices that invested in EHRs and other electronic tools to automate data reporting were both more efficient and achieved better quality results, and did so at a more rapid pace than those that lacked advanced HIT capacity. The AAFP created the Center for Health Information Technology (CHiT) in 2003 to facilitate adoption and optimal use of health information technology by family physicians with the goal of improving the quality and safety of medical care and increasing the efficiency of medical practice. We now estimate that over 20 percent of family physicians are utilizing EHRs in their practices, which is twice the number from this time last year. Through a practice assessment tool on the CHiT web site, physicians can assess their readiness for EHRs. We know from the HHS-supported EHR Pilot Project conducted by the AAFP that practices that had a well defined implementation plan and analysis of workflow and processes had greater success in implementing an EHR.

We also know that cost can be a barrier to IT adoption and have worked aggressively with the vendor community through our Partners for Patients Program to lower the price point. The AAFP’s Executive Vice President serves on the Certification Commission for Health Information Technology (CCHIT) which certifies EHRs. The AAFP sponsored the development of the Continuity of Care Record standard, now successfully balloted through the American Society for Testing and Materials (ASTM). We initiated the Physician EHR Coalition, now jointly chaired by ACP and AAFP, to engage a broad base of medical specialties to advance EHR adoption in small and medium size ambulatory care practices.

The AAFP quality initiatives span efforts to emphasize measures like quality improvement, office redesign, and integration of the chronic care model. For example, through our Practice Enhancement Program, teams of physicians and their office staff participate in an intense educational experience accompanied with pre and post course work to acquire the practical tools, skills and knowledge to implement the planned care model into their everyday practices. In another example, through the web-based METRIC (Measuring, Evaluating and Translating Research into Care) program, family physicians assess their systems in practice, review charts and enter patient data, receive feedback on their performance, implement a quality improvement plan, re-measure and reassess. Two module topics currently are available: diabetes and coronary artery disease. The AAFP takes seriously the responsibility to work with our members to continuously improve their clinical care and office infrastructure to better meet the needs of their patients.

While these innovations are exciting and hold great promise, the environment in which physicians practice is challenging at best. And it will come as no surprise that family physicians, while they enjoy caring for their patients, are not enthusiastic about the Medicare program. This program has a history of disproportionately low payments to family physicians, largely because it is based on a reimbursement scheme that is designed to reward volume and to discourage innovations in the provision of care. In general, the prospect of annual cuts in payment is discouraging. The regulatory approach is punitive, and physicians live in fear of violating rules they don’t even know about. In the current environment, physicians know that they will face a 4.3-percent cut in January 2006, and that without Congressional action to repeal the Sustainable Growth Rate formula and create a structure for sustainable financing, they face steadily declining payments into the foreseeable future, even while their practice costs are increasing. To overlay a pay-for-performance program in Medicare, therefore, poses a unique set of challenges and it must be done thoughtfully and carefully because of its size and complexity.

Our consistent message is that if it is not done well, a value-based purchasing or pay-for-performance program will not only fail to improve health care quality but could unravel the preparation and progress that medical specialty societies have carefully undertaken.

“Doing it well” means phasing in a value-based purchasing program that provides incentives for structural and system changes, that encourages reporting of data on performance measures and ultimately rewards continual improvements in clinical performance. Yet, moving the Medicare program in this direction cannot be accomplished in an environment of declining physician payment; Congress and CMS must take steps to stabilize physician payment through positive updates, as proposed by MedPAC. Furthermore, because of its financing structure with Part A and Part B, we believe it is important that CMS report on Medicare program savings, whether they be in Part A or Part B, resulting from Part B quality improvement efforts so that physicians are not penalized into the future.

The following is a proposed framework for phasing in a Medicare pay-for-performance program for physicians that is designed to improve the quality and safety of medical care for patients and to increase the efficiency of medical practice.

Phase 1: All physicians would receive a positive update in 2006, based on recommendations of MedPAC, reversing the projected 4.3-percent reduction. Congress should establish a floor for such updates in subsequent years.

Phase 2: Following completion of development of reporting mechanisms and specifications, Medicare would encourage structural and system changes in practice, such as electronic health records and registries, through a “pay-for-reporting” incentive system such that physicians could improve their capacity to deliver quality care. The update floor would apply to all physicians.

Phase 3: Assuming that physicians have the ability to do so, Medicare would encourage reporting of data on evidence-based performance measures that have been appropriately vetted through mechanisms such as the National Quality Forum and the Ambulatory Care Quality Alliance. During this phase, physicians would receive “pay-for-reporting” incentives; these would be based on the reporting of data, not on the outcomes achieved. The update floor would apply to all physicians.

Phase 4: Contingent on repeal of the SGR formula and development of a long term solution allowing for annual payment updates linked to inflation plus funds to provide incentives through pay-for-performance programs, Medicare would encourage continuous improvement in the quality of care through incentive payments to physicians for demonstrated improvements in outcomes and processes, using evidence-based measures such as the provision of preventive services, performing HbA1c screening and control, prescribing aspirin to diabetics, etc. The update floor would apply to all physicians.

This sort of phased-in approach is crucial for appropriate implementation. While there is general agreement that initial incentives should foster structural and system improvements in practice, decisions about such structural measures, their reporting, threshold for rewards, etc. remain to be determined. The issues surrounding collection and reporting of data on clinical measures are also complex. For example, do incentives accrue to the individual physician or to the entire practice, regardless of size? In a health care system where patients see multiple physicians, to which physician are improvements attributed?

The program must provide incentives – not punishment – to encourage continuous quality improvement. For example, physicians are being asked to bear the costs of acquiring and using health information technology in their offices, with benefits accruing across the health care system – to patients, payers, insurance plans, etc. Appropriate incentives must be explicitly integrated into a Medicare pay-for-performance program if we are to achieve the level of infrastructure at the medical practice to support collection and reporting of data.

The AAFP appreciates the opportunity to share our enthusiasm for, yet caution about, a Medicare pay-for-performance program. We also appreciate this opportunity to comment on matters related to the Medicare Fee Schedule. As always, the American Academy of Family Physicians looks forward to working with CMS in its continued efforts to ensure access to appropriate physician services.

Sincerely,


Michael Fleming, M.D.
Board Chair