Copies: To order copies of the Federal Register containing this document, send your request to: New Orders, Superintendent of Documents, P.O. Box 371954, Pittsburgh, PA 15250-7954. Specify the date of the issue requested and enclose a check or money order payable to the Superintendent of Documents, or enclose your Visa or Master Card number and expiration date. Credit card orders can also be placed by calling the order desk at (202) 512-1800 or by faxing to (202) 512-2250. The cost for each copy is $10. As an alternative, you can view and photocopy the Federal Register document at most libraries designated as Federal Depository Libraries and at many other public and academic libraries throughout the country that receive the Federal Register .

This Federal Register document is also available from the Federal Register online database through GPO access, a service of the U.S. Government Printing Office. The Web site address is http://www.access.gpo.gov/nara/index.html.

Information on the physician fee schedule and pricing files can be found on our homepage. You can access this data by typing the following: http://cms.hhs.gov/physicians/pfs or you can access this data by using the following directions:

1. Go to the CMS homepage ( http://www.cms.hhs.gov ).

2. Place your cursor over the word “Professionals” in the blue area near the top of the page. Select “Physicians” from the drop-down menu.

3. Scroll down and under “Payment/Billing” select “Physician Fee Schedule'.

The Physician Fee Schedule pricing information is contained in two public use files.

(1) National Physician Fee Schedule Relative Value File—This file contains all CPT/HCPCS (excluding codes beginning with B, E, L, K, and O), their short descriptions and a status indicator, which denotes whether or not the service is priced under the physician fee schedule. The file also contains the components used in the calculation of the annual pricing amount (that is., the RVUs, GPCIs, and conversion factor), anesthesia conversion factors, and the payment policy indicators used to price the claims with surgical modifiers. This file does not contain the calculated pricing amounts.

(2) Physician Fee Schedule Payment Amount File National/Carrier—This file contains the CPT code and the Medicare price for all services priced under the Physician Fee Schedule. These data can be downloaded for (a) the entire country, or (b) for a selected carrier (in most cases carriers correlate with states). There is no option of requesting data for selected HCPCS codes. The zip file, which is downloaded, contains a file named PF04pc.doc, which explains the data contained in each column. This file also contains a description of pricing localities used in the Physician Fee Schedule. Due to the size of the national file (as well as many of the carrier-specific files), these data are provided in a comma-delimited format, which can be used to populate database applications. Generally speaking, these data are too large for Excel, however if a carrier specific file has 3 or fewer localities, Excel can be used.

Another file that providers may find useful is the Zipcode to Carrier Locality File. This file will map ZIP Codes to CMS carriers and localities and map Zip Codes to their State and determine whether the ZIP Code has a rural designation as determined by CMS. You can access this file by typing the following: http://cms.hhs.gov/providers/pufdownload/default.asp#alphanu or you can access this data by using the following directions:

1. Go to the CMS homepage ( http://www.cms.hhs.gov ).

2. Place your cursor over the word “Professionals” in the blue area near the top of the page. Select “Physicians” from the drop-down menu.

3. Scroll down and under “Payment/Billing” select “Medicare Payment Systems.”

4. Scroll down and under Coding Files select “Zipcode to Carrier Locality File.”

In addition, because of the many organizations and terms to which we refer by acronym in this proposed rule, we are listing these acronyms and their corresponding terms in alphabetical order below:

Since January 1, 1992, Medicare has paid for physicians' services under section 1848 of the Social Security Act (the Act), “Payment for Physicians” Services.” This section provides for three major elements: (1) A fee schedule for the payment of physicians' services; (2) limits on the amounts that nonparticipating physicians can charge beneficiaries; and (3) a sustainable growth rate (SGR) for the rates of increase in Medicare expenditures for physicians' services. The Act requires that payments under the fee schedule be based on national uniform relative value units (RVUs) that are based on the resources used in furnishing a service. Section 1848(c) of the Act requires that national RVUs be established for physician work, practice expense, and malpractice expense. Section 1848(c)(2)(B)(ii)(II) of the Act provides that adjustments in RVUs may not cause total physician fee schedule payments to differ by more than $20 million from what they would have been had the adjustments not been made. If adjustments to RVUs cause expenditures to change by more than $20 million, we must make adjustments to ensure that they do not increase or decrease by more than $20 million.

In the July 2000 proposed rule, (65 FR 44177), we listed all of the final rules published through November 1999. In the August 2001 proposed rule (66 FR 40372) we discussed the November 2000 final rule relating to the updates to the RVUs and revisions to payment policies under the physician fee schedule.

In the November 2001 final rule with comment period (66 FR 55246), we made revisions to resource-based practice expense RVUs; services and supplies incident to a physician's professional service; anesthesia base unit variations; recognition of Physicians' Current Procedural Terminology (CPT) tracking codes; and nurse practitioners, physician assistants, and clinical nurse specialists performing screening sigmoidoscopies. We also addressed comments received on the June 8, 2001 proposed notice (66 FR 31028) for the 5-year review of work RVUs and finalized these work RVUs. In addition, we acknowledged comments received in response to a discussion of modifier-62, which is used to report the work of co-surgeons. The November 2001 final rule also updated the list of services that are subject to the physician self-referral prohibitions in order to reflect CPT and Healthcare Common Procedure Coding System (HCPCS) code changes that were effective January 1, 2002. All these revisions ensure that our payment systems are updated to reflect changes in medical practice and the relative value of services. This final rule also conformed our regulations to reflect statutory provisions of Medicare, Medicaid, and State Child Health Insurance Program (SCHIP) Benefits Improvement and Protection Act of 2000 (Pub. L. 106-554) (BIPA) concerning: the mammography screening benefit; biennial screening pelvic examinations for certain beneficiaries; expanded coverage for screening colonoscopies to all beneficiaries; annual glaucoma screenings for high-risk beneficiaries; coverage for medical nutrition therapy services for certain beneficiaries; expanded payment for telehealth services; payment for certain Indian Health Service for some services under the physician fee schedule; and revision of the payment for certain physician pathology services.

In the December 31, 2002 final rule with comment period (67 FR 79966), we refined resource-based practice expense RVUs and made other changes to Medicare Part B policy. These included: The pricing of the technical component for positron emission tomography (PET) scans, Medicare qualifications for clinical nurse specialists, a process to add or delete services to the definition of telehealth, the definition for ZZZ global periods, global period for surface radiation, and application of endoscopic reduction rules for certain codes. In addition, this rule: Updated the codes subject to physician self-referral prohibitions, expanded the definition of a screening fecal-occult blood test, and modified our regulations to expand coverage for additional colorectal cancer screening tests through our national coverage determination process. We also made revisions to the SGR, the anesthesia conversion factor (CF), and the work values for some gastroenterologic services. We finalized the calendar year (CY) 2002 interim RVUs and assigned interim RVUs for new and revised procedure codes for CY 2003, clarified the enrollment of therapists in private practice and the policy regarding services and supplies incident to a physician's professional services, and made technical changes to the definition of outpatient rehabilitation services.

This final rule also revised the regulations at § 485.618 to allow registered nurses (RNs) to provide emergency care in certain critical access hospitals (CAHs) in frontier areas (an area with fewer than six residents per square mile) or remote locations (locations designated in a State's rural health plan that we have approved).

As required by statute this final rule also announced that the physician fee schedule update for CY 2003 was −4.4 percent, the initial estimate of the SGR for CY 2003 was 7.6 percent, and the CF for CY 2003 was $34.5920, effective March 1, 2003. However, on February 28, 2003 (68 FR 9567), after enactment of the Consolidated Appropriations Resolution of 2003 (Pub. L. 108-7), we published a final rule that revised the estimates used to establish the SGRs for fiscal years 1998 and 1999 and announced a 1.6 percent increase in the CY 2003 physician fee schedule CF for March 1 to December 31, 2003. The CF from March 1 to December 31, 2003 is $36.7856 and the anesthesia CF for this period is $17.05. All other provisions of the December 31, 2002 final rule were unchanged by the rule published February 28, 2003.

Under the formula set forth in section 1848(b)(1) of the Act, the payment amount for each service paid under the physician fee schedule is the product of three factors—(1) a nationally uniform relative value for the service; (2) a geographic adjustment factor (GAF) for each physician fee schedule area; and (3) a nationally uniform conversion factor (CF) for the service. The CF converts the relative values into payment amounts.

For each physician fee schedule service, there are three relative values—(1) an RVU for physician work; (2) an RVU for practice expense; and (3) an RVU for malpractice expense. For each of these components of the fee schedule, there is a geographic practice cost index (GPCI) for each fee schedule area. The GPCIs reflect the relative costs of practice expenses, malpractice insurance, and physician work in an area compared to the national average for each component.

The general formula for calculating the Medicare fee schedule amount for a given service in a given fee schedule area can be expressed as:

The CF for CY 2004 appears in section IX. The RVUs for CY 2004 are in Addendum B. The GPCIs for CY 2004 can be found in Addendum D.

Section 1848(e) of the Act requires us to develop GAFs for all physician fee schedule areas. The total GAF for a fee schedule area is equal to a weighted average of the individual GPCIs for each of the three components of the service. In accordance with the statute, however, the GAF for the physician's work reflects one-quarter of the relative cost of physician's work compared to the national average.

Approximately 7,500 codes represent services included in the physician fee schedule. The work RVUs established for the implementation of the fee schedule in January 1992 were developed with extensive input from the physician community. A research team at the Harvard School of Public Health developed the original work RVUs for most codes in a cooperative agreement with us. In constructing the vignettes for the original RVUs, Harvard worked with expert panels of physicians and obtained input from physicians from numerous specialties.

The RVUs for radiology services were based on the American College of Radiology (ACR) relative value scale, which we integrated into the overall physician fee schedule. The RVUs for anesthesia services were based on RVUs from a uniform relative value guide. We established a separate CF for anesthesia services, and we continue to recognize time as a factor in determining payment for these services. As a result, there is a separate payment system for anesthesia services.

Section 1848(c)(2)(C) of the Act required that the practice expense and malpractice expense RVUS equal the product of the base allowed charges and the practice expense and malpractice percentages for the service. Base allowed charges are defined as the national average allowed charges for the service furnished during 1991, as estimated using the most recent data available. For most services, we used 1989 charge data aged to reflect the 1991 payment rules, since those were the most recent data available for the 1992 fee schedule.

Section 121 of the Social Security Act Amendments of 1994 (Pub. L. 103-432), enacted on October 31, 1994, required us to develop a methodology for a resource-based system for determining practice expense RVUs for each physician service. As amended by the BBA, section 1848(c) required the new payment methodology to be phased in over 4 years, effective for services furnished in 1999, with resource-based practice expense RVUs becoming fully effective in 2002. The BBA also required us to implement resource-based malpractice RVUs for services furnished beginning in 2000.

In response to the publication of the August 15, 2003 proposed rule, (68 FR 49030), and the December 2002 interim final rule, (67 FR 79966), we received approximately 2,433 comments. We received comments from individual physicians, health care workers, and professional associations and societies. The majority of comments addressed the physician fee schedule proposals related to the dialysis G codes, “incident to” therapy services, and the geographic practice cost indices locality payment discussion issue.

The proposed rule discussed policies that affected the RVUs on which payment for certain services would be based. Certain changes implemented through this final rule are subject to the $20 million limitation on annual adjustments contained in section 1848(c)(2)(B)(ii)(II) of the Act.

After reviewing the comments and determining the policies we would implement, we have estimated the costs and savings of these policies and added those costs and savings to the estimated costs associated with any other changes in RVUs for 2004. We discuss in detail the effects of these changes in the Regulatory Impact Analysis in section XIII.

For the convenience of the reader, the headings for the policy issues correspond to the headings used in the August 15, 2003 proposed rule. More detailed background information for each issue can be found in the December 2002 interim final rule with comment period and the August 2003 proposed rule.

Section 121 of the Social Security Act Amendments of 1994 (Pub. L. 103-432), enacted on October 31, 1994, required us to develop a methodology for a resource-based system for determining practice expense RVUs for each physician's service beginning in 1998. In developing the methodology, we were to consider the staff, equipment, and supplies used in providing medical and surgical services in various settings. The legislation specifically required that, in implementing the new system of practice expense RVUs, we apply the same budget-neutrality provisions that we apply to other adjustments under the physician fee schedule.

Section 4505(a) of the Balanced Budget Act of 1997 (BBA) (Pub. L. 105-33), enacted on August 5, 1997, amended section 1848(c)(2)(B)(ii) of the Act and delayed the effective date of the resource-based practice expense RVU system until January 1, 1999. In addition, section 4505(b) of the BBA provided for a 4-year transition period from charge-based practice expense RVUs to resource-based RVUs.

Further legislation affecting resource-based practice expense RVUs was included in the Medicare, Medicaid and State Child Health Insurance Program (SCHIP) Balanced Budget Refinement Act of 1999 (BBRA) (Pub. L. 106-113) enacted on November 29, 1999. Section 212 of the BBRA amended section 1848(c)(2)(B)(ii) of the Act by directing us to establish a process under which we accept and use, to the maximum extent practicable and consistent with sound data practices, data collected or developed by entities and organizations. These data would supplement the data we normally collect in determining the practice expense component of the physician fee schedule for payments in CY 2001 and CY 2002. (In the 1999 final rule (64 FR 59380), we extended, for an additional 2 years, the period during which we would accept supplementary data.)

Effective with services furnished on or after January 1, 1999, we established a new methodology for computing resource-based practice expense RVUs that used the two significant sources of actual practice expense data we have available—the Clinical Practice Expert Panel (CPEP) data and the American Medical Association's (AMA) Socioeconomic Monitoring System (SMS) data. The methodology was based on an assumption that current aggregate specialty practice costs are a reasonable way to establish initial estimates of relative resource costs for physicians' services across specialties. The methodology allocated these aggregate specialty practice costs to specific procedures and, thus, can be seen as a “top-down” approach.

A brief discussion of the major steps involved in the determination of the practice expense RVUs follows. (Please see the November 1, 2001 final rule (66 FR 55249) for a more detailed explanation of the top-down methodology.)

• Step 1 —Determine the specialty specific practice expense per hour of physician direct patient care. We used the AMA's SMS survey of actual aggregate cost data by specialty to determine the practice expenses per hour for each specialty. We calculated the practice expenses per hour for the specialty by dividing the aggregate practice expenses for the specialty by the total number of hours spent in patient care activities.

• Step 2 —Create a specialty specific practice expense pool of practice expense costs for treating Medicare patients. To calculate the total number of hours spent treating Medicare patients for each specialty, we used the physician time assigned to each procedure code and the Medicare utilization data. We then calculated the specialty specific practice expense pools by multiplying the specialty practice expenses per hour by the total physician hours.

• Step 3 —Allocate the specialty specific practice expense pool to the specific services performed by each specialty. For each specialty, we divided the practice expense pool into two groups based on whether direct or indirect costs were involved and used a different allocation basis for each group.

(i) Direct costs—For direct costs (which include clinical labor, medical supplies, and medical equipment), we used the procedure specific CPEP data on the staff time, supplies, and equipment as the allocation basis.

(ii) Indirect costs—To allocate the cost pools for indirect costs, including administrative labor, office expenses, and all other expenses, we used the total direct costs combined with the physician fee schedule work RVUs. We converted the work RVUs to dollars using the Medicare CF (expressed in 1995 dollars for consistency with the SMS survey years).

• Step 4 —For procedures performed by more than one specialty, the final procedure code allocation was a weighted average of allocations for the specialties that perform the procedure, with the weights being the frequency with which each specialty performs the procedure on Medicare patients.

For services with physician work RVUs equal to zero (including the technical components of radiology services and other diagnostic tests), we created a separate practice expense pool using the average clinical staff time from the CPEP data and the “all physicians” practice expense per hour.

We then used the adjusted 1998 practice expense RVUs to allocate this pool to each service. We have removed services from the nonphysician work pool if the requesting specialty predominates utilization of the service. Also, for all radiology services that are assigned physician work RVUs, we used the adjusted 1998 practice expense RVUs for radiology services as an interim measure to allocate the direct practice expense cost pool for radiology specialties to the most appropriate SMS specialty.

Since many specialties identified in our claims data did not correspond exactly to the specialties included in the SMS survey data, it was necessary to crosswalk these specialties to the most appropriate SMS specialty.

Because we believe that most physical therapy services furnished in physicians' offices are performed by physical therapists, we crosswalked all utilization for therapy services in the CPT 97000 series to the physical and occupational therapy practice expense pool.

The nonphysician work pool is a special methodology that we used to determine practice expense RVUs for many services that do not have physician work RVUs. While the nonphysician work pool is of benefit to many of the services that were originally included, we have allowed specialties to request that their services be removed from the pool. Because the nonphysician work pool includes a variety of services performed by many different specialties, we use the “all physician” average practice expense per hour in place of a specialty-specific practice expense per hour.

As discussed in the August 15, 2003 proposed rule, we are continuing to study the alternatives that are available and any modifications to the nonphysician workpool would be published in proposed rulemaking.

Comment: Several specialty societies expressed support for the ongoing study of this complex issue and appreciate that any modifications to the nonphysician workpool would be published as proposed rulemaking for review and comment prior to implementation. A biopharmaceutical company commented that we should move forward to develop a new methodology that better recognizes actual resource consumption so that we can develop a preferable alternative.

Response: We are appreciative of the support and will continue to study this issue.

As required by the BBRA, we established criteria to evaluate data collected by organizations to supplement the data normally used in determining the practice expense component of the physician fee schedule. We have required supplementary survey data to be submitted by August 1 to be considered for computing practice expense RVUs for the following year. We proposed to change the required submission date to March 1, which would allow us to publish our decisions regarding survey data in the proposed rule and provide an opportunity for public comment on survey results. We also proposed to extend for an additional 2 years the period for accepting survey data that meets the criteria set forth in the November 2000 final rule (as modified in the December 31, 2002 final rule). The deadline for submission of the supplemental data to be considered in CY 2005 and CY 2006 would be March 1, 2004 and March 1, 2005, respectively.

Comment: Specialty societies expressed appreciation for our proposal to extend the deadline for submission of surveys. Commenters also approved of our proposal to change the due date for submission of supplemental practice expense survey data to March 1, so that the implications of the use of the survey data could be discussed in the proposed rule.

Response: We will implement the change in the submission dates for supplementary surveys as proposed. The deadline for submission of the supplemental data to be considered in CY 2005 and CY 2006 would be March 1, 2004 and March 1, 2005, respectively. We will revise § 414.22(b)(6)(ii) to reflect this change.

The College of American Pathologists (CAP) submitted supplemental survey data for independent laboratories for consideration for CY 2004. Our contractor, The Lewin Group, evaluated the data and has recommended acceptance.

Comment: Based on our proposal to revise the date for submission of supplemental survey data, CAP requested that we delay incorporation of this survey data until next year's proposed rule. CAP also expressed an interest in being able to evaluate the combined effects of the use of the new survey data along with the technical change for pathology services before the changes are implemented. Therefore, CAP requested that we also extend the moratorium on calculating the technical component as the difference between the global and professional component practice expense RVUs by one additional year, as discussed in the August 15, 2003 proposed rule. This request for a delay in incorporating the new survey data, as well as extending the moratorium was supported by the AMA and several specialty societies.

Response: We agree with the comments that suggest extending by one year the moratorium on calculating the technical component practice expense RVU as the difference between the global and professional component RVUs for pathology services. We also agree with comments suggesting that we not incorporate the CAP survey into the practice expense methodology until next year. We will evaluate the CAP survey in next year's proposed rule at the same time we show the effect of the above described change for pathology services.

While we typically assign all staff, equipment and supply costs for services with professional and technical components (PC and TC) to the technical portion of the service, in the proposed rule we discussed limited instances where it is appropriate to assign direct inputs to a PC service. We proposed to modify the practice expense methodology to allow direct inputs to be added to PC services when these inputs are clearly associated with the professional service, including when the PEAC makes such recommendations. Specifically we proposed to add the PEAC recommended staff times to the PC of the following cardiac services: CPT codes 93508, 93510, 93511, 93514, 93524, 93526, 93527, 93528, 93529, 93530, 93531, 93532, 93533 and 93624.

Comment: The RUC, the AMA, the American College of Physicians and societies representing cardiologists, cardiac rhythm specialists, interventional radiologists, nuclear medicine, chest physicians, radiation oncologists, radiologists, endocrinologists and dermatologists expressed support for this change in methodology. Commenters were also in agreement with the specific CPT codes mentioned in the proposed rule, but requested that direct inputs also be added to the PC of CPT codes 93619, 93620 and 93642, which were reviewed at the January PEAC meeting. The RUC comment indicated that additional codes might be identified at future PEAC/RUC meetings.

Response: We will finalize the proposed assignment of direct practice expense to the proposed 14 cardiac services and will add the PEAC recommended inputs to the PC of CPT codes 93619, 93620 and 93642, as requested by the commenters.

We use Medicare utilization data in the development of specialty-specific practice expense RVUs that are then weight averaged to determine a single practice expense RVU per code. Prior to 2003, we used the most recent complete year of utilization data to determine the practice expense RVUs. In the December 31, 2002 final rule (67 FR 79982), we adopted a policy of using the 1997 through 2000 Medicare utilization in the practice expense methodology. For new codes created since 2000, there are no Medicare utilization data. In the August 15, 2003 rule we proposed to follow a similar practice to the one described above and use specialty-specific Medicare utilization data for codes created after 2000 at the first opportunity they become available to us. Since we will not have any utilization data at the time we first establish practice expense RVUs for a new code, we proposed that we continue, whenever possible, to make an assumption about the specialty that will likely provide the service or to use the “all physician” average when we do not have sufficient information to assign any given specialty.

Comment: The specialty societies representing internal medicine, rheumatology and pulmonary medicine supported our proposal to use 1997 through 2000 Medicare utilization data for all codes that were in existence at that time and to use specialty-specific Medicare utilization data for codes created after 2000 when utilization data first become available, using the “all physician” average when we do not have sufficient information to assign a given specialty. These commenters, as well as several others, suggested that the RUC and the specialty societies could provide information on the specialties that will likely perform a new service to minimize the potential changes to the practice expense RVUs that will occur when we substitute actual for estimated utilization. However, a specialty society representing gastroenterology expressed concern that we are moving forward with plans to shift the basis of our methodology for compiling data to a five-year basis. The commenter urged us to not make changes until extensive impact comparisons are conducted that can be evaluated by physician community.

Response: We will implement our proposal to use specialty-specific Medicare utilization data for codes created after 2000 at the first opportunity they become available to us. We will also continue, whenever possible, to make an assumption about the specialty that will likely provide the service or to use the “all physician” average when we do not have sufficient information to assign any given specialty. Information about the specialty we assign to a code that has no utilization data can be found in the utilization data files we make available on the CMS web site following final rule publication. With respect to the comment about shifting to a 5-year basis of utilization data for the practice expense methodology, we are making no change in policy for codes that existed in the 1997 to 2000 period. We are using only the later year utilization data for codes that have been created since that time. Any information from the RUC that could assist us in this process would be welcomed.

Comment: A specialty society representing colon and rectal surgeons agreed with our general utilization methodology, but disagreed that averaged 1997-2000 utilization data should be used for all codes that were not in existence for the entire period. The commenter argued that the frequency for these codes might be artificially low because the coding was new and that this may impact the relativity between new and old codes in the same family with similar inputs. The society suggested that any code that did not exist during the entire 1997-2000 period default to 2002 or most recent data.

Response: As we have explained, the Medicare utilization is important to the practice expense methodology because it determines which specialty scaling factors will be applied to the estimated practice expense input values in determining the practice expense RVUs for each service. The proportion of the volume billed by each specialty is more important to determining the practice expense RVU for a given service than the total volume. If the code is low in volume but the proportion of the code's volume billed by each specialty is generally consistent over time, there will be little or no difference in a code's practice expense RVUs, whether we use its initial year of utilization or a later year to determine its value.

Comment: Commenters representing dermatology as well as a pharmaceutical company expressed concern regarding the decrease in payment for photodynamic therapy, CPT code 95657. The commenters noted our discussion in the proposed rule indicating that this reduction in the practice expense RVUs is occurring because of updates to the Medicare utilization data used in the practice expense methodology. As a result of the updated utilization data, the practice expense methodology now uses the dermatology scaling factor (0.54) for supplies instead of the all physician average (1.29), and this change leads to the reduction in payment for the code. The commenters urged us to reconsider the proposal and at least to reinstate physicians' ability to bill separately in 2004 for the light-activating agent under the appropriate J code and also to remove the drug from the practice expense portion of the procedure.

Response: One of the functions of the utilization data in our practice expense methodology is to assign all procedures to the specialty-specific cost pools of the specialty or specialties performing them. Each cost pool has its own scaling factor. This scaling factor is used to scale the aggregate CPEP procedure-level costs for a specialty to the aggregate costs for the same specialty as determined by the SMS practice expense data. As we indicated in the proposed rule, we do not have utilization data upon which to determine the practice expense RVUs for a new code at the time it is created. As a default, we have assigned many new codes the “all physician” scaling factor until we have the data to move these codes into the appropriate specialty cost pools. Because it allows us to apply the appropriate specialty scaling factor, the use of the updated utilization data in the practice expense methodology can lead to increases or decreases in the value of a code, even though its practice expenses remain unchanged. In this case, the supplies scaling factor for dermatology is lower than that for “all physicians,” leading to a decrease in practice expense RVUs when the dermatology scaling factor was applied to the CPEP data of the photodynamic therapy service.

We believe the initial practice RVUs for photodynamic therapy were too high, because the later information on Medicare utilization indicates that we should have used the dermatology scaling factor which would have produced a lower practice expense value. As we indicate above, we are working to minimize changes that will occur in the practice expense RVUs for a service by making an initial assumption about which specialty will likely bill us for a service. However, we believe our policy for new codes should be consistent with how we determine the practice expense RVUs for existing codes, even if updates to the Medicare utilization data lead to increases or decreases in the practice expense RVUs.

Though we believe that it is appropriate to use the updated utilization that results in a reduction in payment for CPT code 96567, we will pay separately for the light activating agent beginning January 1, 2004. However, we are also further considering whether Medicare should pay separately for certain topical drugs in certain circumstances. Any change in policy would be discussed in future rulemaking.

Comment: Specialty societies representing radiation oncology, as well as individual commenters, expressed concern about the decrease in payment for the intensity modulated radiation therapy (IMRT) treatment service, CPT code 77418. The commenters stated that this was due to a “quirk” in the utilization data relating to new codes and requested that this code be priced by the non-physician work pool methodology.

Response: We will calculate the practice expense RVUs for the IMRT treatment service, CPT code 77418, using the nonphysician workpool methodology. This will be consistent with the way we currently calculate the practice expense for all other radiation therapy services with no physician work RVUs.

Comment: The specialty society representing radiation oncology also noted that there was a reduction in the practice expense RVUs for the intensity modulated radiation therapy planning procedure, CPT code 77301. A remote cardiac monitoring service questioned why the use of new utilization data could decrease the value of a code such as HCPCS code G0249 for the provision of test material and equipment for home INR monitoring.

Response: Both CPT code 77301 and HCPCS code G0249 were new codes for which we did not have utilization data and which were initially assigned the “all physician” scaling factor. As described above, now that we have the utilization data, the services have been placed in the specialty-specific cost pools based on how the service is billed to Medicare, which have lower scaling factors than the “all physician.” This shift has led to the reduced practice expense RVUs for CPT code 77301. If we had placed this code in the radiation oncology cost pool to begin with, it would have had the reduced practice expense payments for the past two years as well. HCPCS code G0249 will actually have increased practice expense RVUs in 2004 due to the effect of the repricing of supplies.

Comment: We received one comment that questioned how updated utilization data could have such a huge and direct effect on specific codes. The commenter requested clarification from us on the workings of the utilization data within the practice expense methodology so that the public will understand how utilization data will affect new technologies in the future.

Response: As explained above, one of the functions of the utilization data in our practice expense methodology is to assign all procedures to the specialty-specific cost pools of the specialty or specialties performing them. If we do not know the specialty, we have used “all physician” scaling factors. The “all physician” scaling factors could be higher or lower than the specialty-specific scaling factor and produce different RVUs for the code. For instance, CPT code 77301-26 is a PC service that has no direct cost inputs. Thus, its practice expense RVUs are affected only by the indirect cost scaling factor. To develop the 2003 practice expense RVUs for this code, we adjusted indirect costs allocated to this code by the “all physician” indirect cost scaling factor of 0.57. However, for 2004, we have Medicare utilization data from 2002 for this procedure code. Radiation oncologists and radiologists respectively billed Medicare for 67 percent and 30 percent of the total volume of services provided to Medicare patients in 2002. The weighted average scaling factor for all the specialties that bill Medicare for this procedure code is 0.48. Since we are adjusting indirect costs by 0.48 instead of 0.57, the final practice expense value is lower.

The PEAC, a subcommittee of the RUC, has, since 1999, been providing us with recommendations for refining the direct practice expense inputs (clinical staff, supplies, and equipment) for existing CPT codes.

In the December 31, 2002 proposed rule, we responded to the PEAC recommendations for the refinement to the CPEP direct practice expense inputs for over 1200 codes, including refinements to codes from almost every major specialty. In addition, the recommendations included standardized times for office-based clinical staff for services provided during a patient's hospitalization and for discharge day management services, as well as pre-service clinical staff times for 323 neurosurgery procedures. We reviewed and accepted all of the recommendations. We received the following comments on these revisions.

Comment: We received comments from specialty societies representing dermatology, dermatolgic surgery and Mohs surgery expressing concern regarding the decrease in practice expense RVUs for skin biopsy procedures, CPT codes 11100 and 11101 and the destruction of benign or premalignant lesion services, CPT codes 17000 and 17003. The commenters questioned whether the reductions reflect errors in the validated practice expense inputs used in the practice expense calculations.

Response: We have checked the practice expense inputs and found that these match the clinical staff, supply and equipment inputs as recommended by the RUC. The reduction in practice expense RVUs was caused by the refinement of these inputs, which, in turn, was based on the presentation made to the PEAC by the dermatology specialty society. We will, therefore, not make any further revisions to the practice expense inputs for these services in this final rule.

In the August 15, 2003 proposed rule we included the PEAC recommendations from meetings held in September of 2002 and January 2003 as well as recommendations on the refinements to the clinical staff time for all 90-day global services. In addition, the PEAC convened a workgroup to make recommendations on the refinement of all the 116 remaining evaluation and management codes. We reviewed the submitted PEAC recommendations and proposed to accept them.

Comment: The American Osteopathic Association expressed appreciation that we supported the recommended changes for the osteopathic manipulative treatment codes and commended us for accepting the PEAC recommendations for the clinical staff times for 90-day global codes. The American College of Obstetricians and Gynecologists stated that our acceptance of the PEAC recommendations is an example of exceptional cooperation and collaboration in meeting the healthcare needs of Americans served by the Medicare program. The American Academy of Dermatology applauded our acceptance of the year's PEAC recommendations. The AMA and the American College of Radiology stated that they appreciate our recognition of the significant resources specialty societies have devoted to the practice expense refinement process and is thankful that our practice expense staff avail themselves of specialty society input. The American College of Surgeons also supported our acceptance of the PEAC recommendations, including the decision to permit exceptions to the standard pre-service times for some surgical procedures. The College other specialty societies also expressed appreciation for our commitment to the refinement process.

Response: We, in turn, are appreciative of these positive comments. We believe that it is only because of the cooperative working relationship between the specialty societies, the AMA and CMS that there has been such a high level of success in tackling practice expense refinement.

Comment: The American College of Physicians as well as other specialty societies representing surgeons, otolaryngologists, podiatrists, geriatric psychiatrists, obstetricians and gynecologists, cataract and refractive surgeons, neurosurgeons, dermatologists, rheumatologists, radiologists and radiation oncologists supported our inclusion of the PEAC recommendations in the proposed rule because this would better enable specialty societies to address their impact and make comments prior to publication of the final rule.

However, specialty societies representing chest physicians and thoracic physicians disagreed with our decision to change our previous practice of including the PEAC recommendations in the final, rather than the proposed rule, because this meant that the recommendations from the March PEAC meeting were not included for this year. The society argued that changing this long-standing policy without announcing it in the Federal Register is inappropriate. The comment also contended that the specialty societies agreed to the inputs at the PEAC meeting; therefore, negative comments would not be forthcoming.

Response: We discussed this issue at the January PEAC meeting and indicated that we were considering including the PEAC recommendations in the proposed rule and that the March recommendations would most likely not be included. We made this decision because, now that the PEAC is refining such a large number of codes, the revisions to the inputs were not only changing the practice expense RVUs of the refined codes, but also the values of services that were not refined. Therefore, we believed it was prudent that revisions be subject to comment before the revisions were implemented.

Comment: The specialty society representing podiatry identified some discrepancies between the PEAC recommendations and the inputs in the CPEP database for CPT codes 10060, 11000, 11055, 11056, 11057 and 11752 and requested that these be corrected.

Response: We have made the corrections as requested.

Comment: The American Society of Transplant Surgeons (ASTS) commented that it is not appropriate to apply either the PEAC-approved standard clinical staff times or RN/LPN/MTA staff blend for 90-day global procedures to the transplant recipient or living donor services. ASTS stated that it had been unaware that the PEAC was applying the standard to all 90-day services unless a case was made to the PEAC that the times should be increased. ASTS argued that there are substantial atypical staff times required for transplant recipients due, in large part, to the intensive education required for the transplant patient. The commenter noted that the three new CPT codes for living donor hepatectomies, CPT codes 47140-47142, were given increased pre-service clinical staff time by the RUC and have an RN as the staff type. ASTS requested that the current clinical staff times be retained and that an RN be assigned rather than the blended staff type to the following transplant services: CPT codes 32851, 32852, 32853, 32854, 33935, 33945, 47135, 47136, 48554, 48556, 50320, 50360, 50365, 50380, 50547.

Response: It does seem reasonable that at least some of these services would have increased pre-times as do the living donor hepatectomies recently reviewed by the RUC. Therefore, we will restore the original CPEP clinical staff pre-times and use the RN staff type for the above services on an interim basis for the coming year. We anticipate that the society will bring all of these codes to the PEAC for review for either the January or March meeting to ensure that the times for the codes receive the same scrutiny as did the new transplant codes. It should be noted that a few of the codes have lower original CPEP pre-time than the PEAC standard of 60 minutes; for those codes we did not change the PEAC standard time. We also are not revising the post-procedure clinical staff times for these codes, because the current times are in line with the post-service times assigned to the new living donor hepatectomy codes recently reviewed by the RUC.

Comment: A commenter noted that high dose rate (HDR) brachytherapy CPT codes 77781, 77782, 77783 and 77784 were not listed in Addendum C of the proposed rule. Since these codes were approved by the PEAC and forwarded to CMS, ACR questioned why these codes were not listed.

Response: The CPEP data base files had been revised to reflect the PEAC recommendations for these codes. It was an oversight that they were not included in Addendum C.

Comment: The American College of Surgeons listed several possible errors in the CPEP database:

CPT code 11450—missing 1 minute of staff time

CPT codes 10080, 10081, 11770, 12032, 12035, 12046, 12047, 21550, 21920, 37609, 38300, 45300-45327, and 46600-46615—missing correct number of gloves.

CPT codes 45900, 45905, 45910, 47382, 49320, 49321, 49322, 49422, 49429—supplies listed incorrectly—have nonfacility inputs when PEAC recommended none in office setting.

Response: We thank the College for checking the database so carefully. We have made the suggested corrections, with the following notes: For CPT codes 10080, 10081 and 11770, the PEAC recommendation listed 5 gloves, not 6. For CPT codes 45300-45327 and 46600-46615, we adjusted the quantity of unsterile gloves to reflect that there are 2 pair in the minimum visit supply package; in addition, CPT codes 45321 and 45327 were not priced in the nonfacility setting.

Comment: The American Society of Colon and Rectal Surgeons noted a few errors in the CPEP supply database. The supply inputs had not been changed to match the accepted new recommendations for CPT codes 45900, 45905, 45910, 47382, 49320, 49321, 49322, 49422 and 49429.

Response: We have made the corrections to the supply database and thank the specialty for bringing this to our attention.

Comment: The American Speech-Language-Hearing Association (ASHA) questioned the proposed 28 percent reduction in the practice expense for CPT code 92507, Treatment of speech, language, voice, communication, auditory processing and/or aural rehabilitation status. The reduction is attributable to a decrease in clinical staff time. ASHA contended that the PEAC recommendation was based on a vignette for a child receiving such therapy, but that the time involved with a typical adult patient receiving this treatment is much longer. ASHA stated that a more reasonable time for clinical staff for this service is 69 minutes compared to the proposed 46 minutes.

Response: We understand that the scenario for performing this service for a child might be very different than for an adult because an adult can participate in a more protracted therapy session. Because it is not clear to us at this time what would be the typical scenario, we will, on an interim basis, average the clinical staff time needed during a speech therapy session for a child with that suggested by ASHA for an adult. We will, therefore, assign 58 minutes of clinical staff time to this service, with the expectation that ASHA will present CPT code 92507 for further discussion and review at the PEAC.

Comment: We received several comments in response to our acceptance of PEAC recommendations for evaluation and management (E/M) codes that reduced payment rates for six nursing home services (CPT codes 99301-99303 and 99311-99313) and two home visit codes (CPT codes 99348 and 99350). This payment reduction is primarily due to a decrease in the clinical staff time assigned to these services.

The American Academy of Family Physicians (AAFP) supported our acceptance of the PEAC recommendations for the E/M nursing facility services. The commenter noted that current practice expenses are higher for services provided in the non-SNF nursing facility than those provided in the SNF facility. The commenter contended that the direct practice expense inputs should not vary based on the type of nursing facility setting and supported the elimination of the current differential in the practice expense RVUs between the SNF and non-SNF facility setting.

However, the American Medical Directors Association (AMDA) representing long term care physicians, the American Geriatrics Society (AGS) and a health care management company, Health Essentials, all disagreed with our decision to accept the E/M nursing facility PEAC recommendations and asked us to reconsider our decision to implement them in 2004. The request to delay implementation was echoed by the American Academy of Home Care Physicians and AGS relating to the two E/M home visit codes.

The home care physicians argued that the PEAC recommendations for the two home visit codes are flawed because these codes have not yet been surveyed by the specialty performing this service. The commenters also contended that their views were not represented when the PEAC considered the refinements of the E/M home visit codes. Similarly, the AMDA noted that the PEAC workgroup responsible for formulating the recommendations for the nursing facility codes did not include long term care physicians. The AMA also commented on this issue and expressed concern that the PEAC recommendations did not include the views of all the relevant medical specialties and requested that we delay implementation of these E/M code recommendations to allow impacted medical specialties an opportunity to present new information to the PEAC.

In addition, the AMDA expressed concern regarding the current work RVUs for nursing home visit services.

Response: At the time the PEAC recommendations were forwarded to CMS, we agreed with the views expressed by the AFPP as to the reasonableness of the practice expense recommendations for the E/M codes for the nursing facility and home visits. However, we are also of the opinion that the relevant medical specialties should be given the opportunity to have their views considered by the PEAC. Consequently, we will not go forward with these E/M recommendations in 2004. This will allow time for the PEAC to reconsider the eight E/M codes with input from representatives from the nursing home and home visit specialties. We will use current CPEP practice expense inputs to price these codes for 2004.

With regard to the concern expressed about the work RVUs for the nursing home visits, in the 2004 final rule we will solicit recommendations on codes to be reviewed during the next 5-year review of work and we suggest that the society recommend review of these codes.

Comment: A specialty society representing gastroenterologists commented that the increased clinical staff pre-time added to certain colorectal procedures needs to be applied equally to gastroenterologists who provide those services.

Response: We have a single payment for each procedure regardless of the specialty performing the service. Therefore, gastroenterologists will be paid the same as colorectal surgeons when performing those services for which we allowed increased pre-service clinical staff time.

Comment: The American College of Radiology submitted several corrections to the CPEP database for those instances where the database differed from the PEAC recommendations that we accepted. The College stated its appreciation for the opportunity to review the practice expense data file for completeness and accuracy and applauded our efforts to ensure that the database captures correct and complete practice expense data.

Response: We thank the College for the time and effort expended in checking this detailed data. We have made revisions to 19 codes: We changed the quantity of sodium chloride injection for CPT codes 78306, 78315, 78460, 78461, 78464, and 78465; adjusted the quantity of films for CPT code 76812; added missing supplies to CPT codes 77408, 77409, 77411, 77412, 77414, 77416, 76830 and 77290; removed equipment that had been deleted from CPT codes 78478 and 78480; and corrected a typographical error in the pre-service clinical staff time for CPT codes 73218 and 75555.

We use the practice expense inputs (the clinical staff, supplies, and equipment assigned to each procedure) to allocate the specialty-specific practice expense cost pools to the procedures performed by each specialty. The costs of the original inputs assigned by the Clinical Practice Expert Panels (CPEP) were determined by our contractor, Abt Associates, based primarily on 1994 and 1995 pricing data from supply catalogs. In addition, for many items on the equipment and supply list, the associated costs were based on the recommendations of a CPEP panel member, rather than on actual catalog prices. Subsequent to the CPEP panels, equipment and supply items have also been added to the CPEP data, with the costs of the inputs provided by the relevant specialty society.

We contracted with a consultant to assist in obtaining current pricing information and also to recommend revisions to improve the uniformity and consistency of the CPEP supply database. On the basis of these recommendations, in the August 15, 2003 proposed rule, we proposed updates to the cost information for supplies in the database. In addition, we proposed the following database revisions:

We proposed that supplies be assigned to one of 14 categories.

We proposed combining items which appeared to be duplicative and modifiying descriptions using a key first word when possible for easier identification of items. For example, “mayo stand cover” and “drape, sterile Mayo” have both been changed to “drape, sterile, for Mayo stand.”

The current CPEP database contains over 72 unit descriptions associated with supplies (for example, item, gram, and cup). To provide consistency and ensure that inputs in the database accurately reflect the quantity of an item used, we proposed to standardize the unit description of items. We also proposed to specifically identify items intended for single use through the use of “uou” (unit of use) following the unit. These changes were reflected in Addendum D of the proposed rule.

There were also items that had not been identified or for which pricing information was not found that were included in Table 1 in the August 15 proposed rule. Items that we proposed to delete from the database were also identified in this table. We requested that commenters, particularly the relevant specialty groups, provide us with the needed pricing information with appropriate documentation. We also stated if we did not obtain verified pricing information for an item, it would be eliminated from the database.

Comment: The RUC expressed appreciation for the enormity of the repricing project and stated that the proposed approach was well organized and comprehensive. The American Association of Orthopedic Surgeons also agreed that the assignment of supply categories would be helpful in future refinement activities. The American College of Physicians, the American College of Surgeons, and the American Urological Association expressed support for our proposal to create a numbering system and to standardize the descriptions of supply items to increase accuracy of use. The American Academy of Dermatology also supported this standardization of proposed “unit of use” as long as its application does not assume that “one size fits all” as some supplies may go from milliliter to liter in usage. The American Society of Cataract and Refractive Surgery and the Outpatient Ophthalmic Surgery Society thanked us for the repricing proposal because this will ensure that we are using the more accurate and up-to-date supply costs, thus reimbursing physicians more fairly. The American College of Radiology recognized the need to update supply and pricing information in the practice expense database and commended us for committing to this extensive project. The American College of Surgeons also agreed that the update of prices for supplies will improve the accuracy of the direct practice expense data. The Society of Nuclear Medicine commended us for committing to this extensive project. The American Urological Association also appreciated this effort and acknowledged it as a huge undertaking.

Response: We appreciate the positive feedback and would like to thank all the staff of the specialty societies who worked with our contractor to obtain the most representative prices for all of the supplies in the CPEP input database.

Comment: A specialty society representing podiatrists agreed with removal of hallux implant and the broach kit from the list of supplies to be included under practice expense as both are separately billable and the broach kit is also reusable. The commenter did not agree with removal of the sterile ankle tourniquet since this is packaged as a single use item. The comment included pricing information at $42.87 each (with documentation) for this supply.

Response: We will delete the hallux implant and the broach kit from the CPEP supply data. We will retain the ankle tourniquet using the pricing information supplied by the society.

Comment: Several commenters expressed concern about the reduction in nonfacility practice expense for the interstitial laser coagulation of the prostate procedure, CPT code 52647. A manufacturer of endo-surgery equipment stated that the main reason for this decrease was the decrease in the price assigned to the laser fiber used in this procedure. We had proposed a price of $290 for this item, but the commenter submitted documentation that indicated that the laser fiber should be priced at $850 for CPT code 52647. In addition, the commenter noted that we had proposed in Table 1 to delete the laser fiber because it was reusable; however, this was incorrect as the laser fiber used in this procedure could not be reused and should not be deleted from our supply list.

Response: When the laser fiber was repriced, we believed the item included in the supply list for CPT code 52647 was the same as a “laser tip,” which was priced at $290. We thank the commenters for clarifying the issue. We agree that the laser fiber used in this procedure is a disposable supply that we will retain in our CPEP supply data at the $850 price documented by the commenter.

Comment: Commenters representing cardiac arrhythmia specialists and a remote cardiac monitoring system recommend that we not delete the transtelephonic monitor as a supply even though we are correct that the patient and physician re-use this supply during the course of the pacemaker's life. The specialty society commenter requested that the expense of this supply, which costs $190, should be spread out over approximately 5 years.

Response: The transtelephonic monitor as described would be considered a piece of equipment, rather than a reusable supply. However, unless the equipment costs over $500, we consider it as an indirect cost and it is not included as a direct input. Therefore, we will delete the item from our list of direct practice expense inputs as proposed.

Comment: A specialty society representing chest physicians agreed that the oximetry sensory probe, CPAP nasal pillow and flow sensor are reusable and should be deleted from the list of CPEP supply inputs. The society also agreed that albuterol is separately billable and should also be deleted. Another commenter, representing sleep medicine, agreed that the nasal pillow should be deleted. However, the commenter representing chest physicians and a commenter representing thoracic physicians disagreed with the proposal to delete methacholine chloride because there is no “J” code to use when billing, thus forcing physicians to used an unlisted service code. The commenters also contended that the aerochamber should not be deleted because, although reusable, it has a life of only about six months and should be costed out accordingly. In addition, the commenters disagreed that the inhaler is separately billable because a multi-use canister is utilized for this test; therefore, the amount used from the canister for each test should be included in the practice expense.

Response: We will delete the oximetry sensory probe, CPAP nasal pillow and flow sensor and albuterol from the list of CPEP supply inputs. We will also delete the aerochamber, because an item that is reusable over a six-month period cannot be classified as a disposable supply. The commenter is correct that there is not a HCPCS “J” code for methacholine chloride. Therefore, we will keep this in the supply database as requested so that physicians can avoid the burden of submitting paper claims. We also will keep the inhalant in the database using the quantity of 1 gram per procedure at $0.788.

Comment: Specialty societies representing radiologists and interventional radiologists disagreed with the classification of the Arrow mechanical thrombectomy device as reusable. The commenter contended that this device is single-use because the difficulty in cleaning the intra-luminary surface areas could lead to a risk of contamination if the device is reused. Moreover, reprocessing the thrombectomy device may result in fatigue-related failure.

The societies also disagreed with our contention that a Seldinger needle is reusable; rather a Seldinger needle is single-use and should not be removed as a supply item. It is the commenter's understanding that hospitals are not in the practice of resterilizing Seldinger needles.

While generally favoring reorganization of CMS' supply listing for ease of use and not directly opposed to supply categories, one of the commenters was concerned over the potential loss of granularity of cost data associated with the use of supply categories and would oppose the averaging of costs for the supply categories unless it is appropriate to average from a cost and clinical standpoint. A similar comment was sent by the radiology specialty society.

Response: We will retain the thrombectomy device and the Seldinger needle as disposable supplies in our CPEP input database. With regard to the classification of supplies, the commenter misunderstands the purpose of assigning a classification to each supply. This will not be used for pricing purposes in any way. Rather, the classifications can be useful as a way to sort the long list of supplies in the database to make it easier to find a particular item.

Comment: The contractor responsible for helping us with the repricing of supplies informed us that a supply assigned to the endometrial ablation procedure, CPT code 58353, was listed as a catheter tray when it should be described as a thermal ablation balloon catheter at a price of $727. In addition, our contractor supplied us with prices for several new supply and equipment items mainly for otolaryngology, that were not priced in the proposed rule but were included in the PEAC recommendations.

Response: We will make the appropriate changes in the CPEP supply and equipment databases.

Comment: Commenters representing pediatricians, pulmonary physicians and family physicians pointed out that the new price we had assigned to the safety syringe and needle did not cover the actual cost of purchasing the entire needle stick device that is required by the Occupational Safety and Health Administration.

Response: Our repricing contractor researched this issue for us and agreed that the price we were proposing was too low for the appropriate item. Based on documentation for a 10 ml Syringe with SafetyGlide Needle, the safety syringe and needle will be priced at $.435 each, instead of the $.28 that was proposed.

Comment: A surgical society commenter pointed out that we listed an achalasia balloon in Table 1 in the proposed rule and indicated that it was a supply used with CPT codes 45905 and 45910. The commenter stated that both of these codes were refined in January and that they were not priced in the office setting; therefore the balloon should no longer be listed as a supply used with these services.

Response: Our CPEP database currently has these codes priced only in the facility setting. However, these services had previously been priced in the office and Table 1 was apparently developed before the last of the PEAC recommendations were entered. The achalasia balloon no longer appears on the CPEP supply database.

Comment: We received comments from the American College of Physicians and another medical society representing allergy and immunology with concerns about reductions in reimbursement for the five venom immunology CPT Codes (95145-95149). The commenters believe the reductions are due, in part, to the use of incorrect supply costs for venom extracts that we priced at $5.18 per ml. The commenters provided documentation of current prices of five different venoms from two of the largest manufacturers of venom extracts. They proposed a price-averaging methodology utilizing the small and large quantities of venoms that are available from the two suppliers. A price of $12.22 per milliliter of venom antigen results from using this methodology, and the commenters suggest that this price be used in valuing four of the five CPT Codes for venom immunology, with the exception of CPT Code 95147. When a patient requires three stinging insect venoms, as for CPT 95147, the commenters believe the 3-Vespid mix is typically used. Again, the commenters suggested the same price-averaging method noted above using cost information from the two vendors, which results in a price of $23.49 per ml. This 3-vespid mix price could also be used to value CPT Codes 95148 (four venoms) and 96149 (five venoms) with the single venom, priced at $12.22, added once to CPT code 97148 and twice to CPT Code 97149.

Response: We were pleased to receive the comments, as well as the requested documentation, on the price for various venom extracts, because the venom pricing information was not included in the PEAC recommendations forwarded after the September 2002 meeting for these CPT Codes. This lack of data necessitated the use of a generic stinging insect venom price of $5.18 per ml. We accept the pricing information supplied by these specialty societies, although we do not agree with their proposed averaging of prices from both the small (5ml and 6ml) and larger (10ml and 12ml) quantities of venoms. We believe it is more appropriate to average the venom prices using the larger (10ml and 12ml) quantities because of the volume that is used in an accepted venom immunotherapy program, which consists of a build up period of about four months followed by monthly maintenance therapy. The following prices result from this approach: $10.70 per ml of venom and $21.26 for the 3-Vespid Mix. Venom pricing for the five CPT codes would be as follows: CPT Code 95145 (one venom) at $10.70, CPT Code 95146 (two venoms) at $21.40, 95147 (three venoms using 3-vespid mix), would be $21.26; CPT Code 97148 (four venoms), $21.26 + $10.70 = $31.96; and the venom antigen price for CPT Code 97149 (five venoms) would be $42.66 ($21.26 + $10.70 + $10.70).

Comment: JCAAI also supplied pricing information for the multi-tine device that was requested in Table 1 of our proposed rule. As was suggested above, the commenters again proposed we average costs for high and low volume purchases, excluding bulk pricing, to obtain the price for each test.

Response: We appreciate the pricing information forwarded by JCAAI and selected a purchase quantity that is in the middle of the suggested range. For percutaneous allergy testing, CPT code 95004. This purchase quantity represents testing 200 typical patients, each receiving 40 tests. We have added this Multi-tine per test price, $0.233,to the CPEP database for CPT codes 95004 and 95010.

Comment: The American Speech-Language-Hearing Association (ASHA) provided pricing information for the following items accompanied by the requested documentation: Aphasia assessment treatment forms—$2.84 (for a diagnostic aphasia examination form and aphasia diagnostic profile), communication books/treatment notebook—$1.50 and eartip insert—$0.65 each or $0.39 each (two sources). The American Academy of Otolaryngology—Head and Neck Surgery (AAO-HNS) submitted a price for the eartip insert of $0.23 each and suggested that the communication books/treatment notebook be deleted. The (AAO-HNS)also submitted a price for cottonoids at $0.875 each and for the phenol applicator kit at $15.95 each.

Response: We will use the submitted price for the aphasia forms and will price the eartip insert at $0.423, which is the average of the three prices submitted. The notebook, which is assigned to the speech-language therapy code, would be used over a course of treatment, and is not a disposable supply that is used or priced for a single service. Therefore, we will delete this item from our CPEP supply data. For the phenol applicator kit, we will use the price of $15.152 per kit that represents an average price for a 6-kit and a 24-kit quantity purchase. Because these kits contain the phenol that is used in the procedures, phenol has been deleted as a separate supply from the 11 CPT codes that are assigned the kit. AAO-HNS used a 10-pack quantity to assign a price to each cottonoid, but we are using a 200-pack quantity that reflects the high usage of this item. Therefore, we are using $0.773 as the price for each cottonoid.

Comment: Specialty societies representing radiation oncology and radiology disagreed that the fiducial screws used with the intensity modulated radiation therapy procedure should be deleted from the CPEP input supply list. The society argued that the screws are typically used for this procedure and that they are not separately billable.

Response: We will retain the fiducial screws in the list of supplies assigned to the intensity modulated radiation therapy procedure.

Comment: The American Society of Colon and Rectal Surgeons offered description changes for two services, CPT codes 46917 and 46924. The society recommended that the descriptor for the laser tip for both codes be changed to “laser tip, bare (single use)” at $150. The commenter also requested that an ablation laser generator at $59,890 be added to both codes and the existing laser, diode laser, and laser generator be deleted.

Response: A note from our contractor who is working on our repricing effort verified the above changes and we have revised our supply and equipment databases to reflect them.

Comment: The American Association of Orthopaedic Surgeons agreed with the proposed supply deletions listed in Table 1 of the proposed rule that are used in orthopaedic surgery. In addition, the association agreed with the concept of standardization of unit descriptions. However, the comment contends that the term “unit of use (uou)” is unclear and that we should consider alternative terms and abbreviations that would be more intuitive.

Response: The supply items in Table 1 that were listed for orthopaedic surgery are broach kit, hallux implant, sterile hand table drape, sterile cuff tourniquet, cephalosporin and sterile ankle tourniquet. As stated above, we will be deleting the broach kit and hallux implant and will also delete the hand table drape, cuff tourniquet and cephalosporin. As also noted above, we will retain the sterile ankle tourniquet in the supply database because the comment from the podiatry society argued that this item was not typically reused.

With regard to the comment on the use of “unit of use,” we selected the “unit of use” (uou) term to indicate any item that is packaged for single use, even if the item is not completely used up. This most often occurs with items that are packaged sterile. For example, “bacitracin (0.9gm uou)” refers to one 0.9gm foil package. The quantity entered would be 1 and not a smaller amount such as 0.3. Once this foil package is broken, it is considered “used up” and therefore the unit of use is 0.9gm. Specifically, any item with a “unit of use” designation is meant to be indicated in whole number “unit of use” quantities, not partials ( e.g., entered as 1, 2, 3, etc, and not 0.5, 1.5, etc.).

Comment: A commenter representing sleep medicine stated that our proposed price of $25 is significantly below prices for standard CPAP masks used in the polysomnography service, CPT code 95811. The commenter submitted prices from two manufacturers that average to $88.

Response: It appears that the commenter has submitted prices for a reusable CPAP mask that would not be included in our CPEP data as a disposable supply. Therefore, we will price the disposable mask at $25.135, as proposed.

Comment: We received a comment from the American Physical Therapy Association (APTA) that contended there is a rank order anomaly caused by the increased price for the electrode used for CPT code 97033, iontophoresis. APTA noted that the price of a “pair” of electrodes was $16 in 2001 but has increased to $23.98 under our current supply repricing initiative. APTA has asked that we review the proposed cost of this item as a means to moderate the rank order anomaly.

Response: We appreciate the comments offered by APTA and have reviewed the cost of the supplies assigned to the iontophoresis service. We determined that the electrodes for this service are packaged and priced as “kits” that contain the complete set of electrodes needed to provide one iontophoresis treatment. Therefore, only one electrode “kit” is needed for this code, as opposed to the two electrode “pairs” currently in our supply database. Consequently, we have changed the supply list for iontophoresis in our database to reflect that there is one kit, not two electrodes, at the proposed price of $11.99. We believe that this should correct the rank order anomaly.

The following table, “Table 1 Items Needing Specialty Input,” lists those items on which we had requested specialty input, comments we received and the actions we are taking.

We proposed to assign, on an interim basis, the following practice expense inputs to CPT code 99183, Physician attendance and supervision of hyperbaric oxygen therapy, per session, when performed in the office setting:

Staff: Respiratory Therapist for 135 minutes (for a 2 hour treatment); Supplies: Minimum Visit Supply Package, 180 liters of oxygen, 187 cubic feet of air; Equipment: Hyperbaric chamber.

Comment: A freestanding hyperbaric oxygen center expressed appreciation that we priced this procedure in the non-facility setting. The commenter also requested that we add certain staff time and some supplies to the practice expense inputs assigned to this service.

The additional supplies requested include disinfectant for cleaning the hyperbaric chamber after each patient, two otoscope covers to check patients' ears pre and post treatment, and a denture cup and urinal. An additional 24 minutes of clinical staff time (using the standard staff blend) was also requested for preparing the room, greeting and gowning the patient, patient education, taking vital signs before and after treatment, positioning the patient and cleaning the room.

Response: We believe that the request for the above additional practice expense inputs is reasonable. Currently, we have assigned clinical staff time only for assisting during the procedure itself; additional time was calculated using the times used by the PEAC for the tasks listed. Therefore, we are adding these inputs to those already assigned to the hyperbaric oxygen service. We have also requesting that the PEAC review these inputs at a future meeting and the RUC has stated that the PEAC will be reviewing this CPT code at the January or March 2004 meeting.

Comment: A commenter from another freestanding hyperbaric center expressed concern that the proposed physician fee schedule payment for CPT 99183 is approximately 25 percent of the payment in the hospital setting. The commenter lists additional costs that should be considered such as special cleaners and solvents for cleaning the chamber, the costs of adherence to quality standards and costs for laundering patients' clothing, sheets and blankets. The commenter also stated that the hyperbaric chamber costs more than the $125,000 we have assigned the item.

Response: As mentioned above, we have added disinfectant solution for cleaning the chamber. We will be proposing the repricing of all equipment in our CPEP database next year, which should ensure that the price for the hyperbaric chamber reflects the typical cost. The cost of laundering and much of the quality assurance costs are considered indirect and are not reflected in our direct cost database. However, if the PEAC does refine this code as planned, we will review any recommendation submitted.

We proposed to eliminate the special practice expense pool for maxillofacial prosthetic services and to use otolaryngology as the crosswalk for oral surgeons and maxillofacial surgeons as a more appropriate approximation of the specialties' practice expense per hour.

Comment: The American Association of Oral and Maxillofacial Surgeons expressed appreciation for our work on this issue over the past three years and heartily concurred with the decision to crosswalk maxillofacial prosthetics to otolaryngology. The American Academy of Otolaryngology-Head and Neck Surgery also supported our proposed crosswalk.

Response: We will implement the crosswalk of maxillofacial prosthetics to otolaryngology as proposed.

We proposed revising the practice expense inputs for holter monitoring codes to remove items that were not needed to perform the services. Specifically, we proposed deleting the ECG electrodes and laser paper, as well as the electric bed, computer and holter monitor from CPT codes 93225 and 93231 and deleting the razor, nonsterile gloves, alcohol swab and tape, as well as the electric bed and exam table from CPT codes 93226 and 93232.

Comment: A commenter representing an independent diagnostic testing facility and another representing cardiologists expressed support for the proposed revisions to the holter monitor codes.

We also received a comment from the RUC stating that the direct practice expense inputs for these above holter monitoring services will be reviewed by the PEAC at the January 2004 meeting.

Response: We will make the proposed changes to the holter monitoring codes on an interim basis and will be glad to review the recommendations from the PEAC when we receive them next year.

Comment: We have received requests from several commenters that we value certain procedures currently priced only in the facility setting in the non-facility setting as well. A manufacturer commented that there is a need to price the hysteroscopic endometrial ablation procedure, CPT code 58563, in the office to ensure Medicare patient access to this alternative to hysterectomy in the least intrusive and least costly setting. Several individual gynecologists have expressed concern about the absence of a nonfacility rate for this service because the facility payment does not cover the costs of performing this procedure in the office.

A manufacturer of endoscopic and surgical supplies and equipment expressed concern that several urology services which had previously been priced in the non-facility setting, are no longer priced in that setting. The commenter contended that the procedures can be performed safely in the office and that patients will be forced to go to a hospital or ambulatory surgical center for these procedures if the office payment does not reflect the direct costs incurred by the physician. The services in question are three cystourethroscopy procedures, CPT codes 52224, 52275, 52276, and two destruction of penile lesion procedures, CPT codes 54057 and 54065.

A consultant representing non-hospital based providers of LDL apheresis, CPT code 36516, requested that we price this procedure in the nonfacility setting and provided some cost data for this code. The commenter stated that this procedure is commonly provided outside of hospitals. A medical technology company requested that we price the percutaneous implantation of neurostimulator electrodes procedure, CPT code 64561, in the nonfacility setting. This service had previously been priced in the office.

Response: We are aware that technological advances make it now possible for more procedures to be safely performed in a physician's office. However, CPT code 58563 has recently been reviewed by the PEAC, and neither the gynecology specialty society nor the PEAC recommended pricing this code in the office setting. Likewise, the urology procedures and the neurostimulator service were reviewed this year by the PEAC and the apheresis services last year by the RUC, and the PEAC and the RUC recommended that these services not be priced in the office setting based on the presentation made by the specialty societies. We would not rule out working further with the commenters on these requests, but we believe that it would not be appropriate to take such an action in this final rule. We will be willing to discuss this issue further to determine whether any action should be proposed in the future.

Comment: The RUC comment identified the following anomalies in the CPEP database for the clinical staff time for a few codes with 000 day global periods:

In 1997, CPT created new codes to differentiate between open and percutaneous abscess drainage. Unlike their open procedure counterparts, all of the percutaneous codes were assigned a global period of 000 days with no follow-up visits assigned. However, CMS crosswalked the direct inputs from the open codes, which have a different global period, to the percutaneous codes, including the time assigned for post-procedure office visits. The percutaneous abscess drainage codes identified are CPT codes 32201, 44901, 47011, 48511, 49021, 49041, 49061, 50021, 58823. The comment stated that each of these codes is currently priced in the facility setting only. Because these procedures are predominately performed in the inpatient setting, the comment further recommended that we assign zero direct practice expense inputs for these codes.

The commenter also pointed out that CPT code 67875, Closure of eyelid by suture, has an assigned global period of 000 and includes no post-procedure visits in the work relative value. However, the original CPEP process appears to have assigned the code clinical staff time, supplies, and equipment related to a follow up visit.

Response: We agree with the RUC that these 0-day global codes should not have any direct costs assigned for post-procedure follow up visits. Therefore, we are deleting from the database all the inputs related to such visits.

Comment: Several commenters have expressed concern with the unexplained reduction in nonfacility practice expense RVUs for HCPCS code G0166, External counterpulsation.

Response: We have examined the practice expense data files and have discovered an error in the database. This has now been corrected.

Comment: A specialty society representing dermatology commented that the practice expense RVUS for laser treatment of psoriasis procedures, CPT codes 96920-96922, appear overvalued.

Response: The practice expense has increased for these codes because we did not have a price for the laser tip used in these procedures until this year. The laser tip is now priced at $240. We have made adjustments to ensure the practice expense RVUs reflect the correct pricing of supplies as well as the specialty performing the service.

Comment: One specialty society that represents gastroenterologists commented that we cut the payment rate for the colonoscopy procedure, CPT 45385, by 10 percent in the nonfacility setting without explanation or justification.

Response: The decrease in payment for this code is due to the decreased practice expense inputs now assigned to the service. The PEAC submitted recommendations for the direct practice expense inputs for this service that were based on a presentation made by two other gastroenterological specialty societies, and we have accepted these recommendations because we believe them to be reasonable. The code was included on Addendum C, “Codes for Which We Received PEAC Recommendation on Practice Expense Direct Cost Inputs,” in the proposed rule.

Comment: Several commenters representing pediatricians, family physicians and chest physicians stated their concern with the proposed decrease in the practice expense RVUs for immunization services, CPT codes 90471 and 90472, which were removed from the non-physician work pool and priced under the top-down methodology starting in 2003.

Response: We will return the two immunization services to the nonphysician work pool. As discussed above, we are increasing the price assigned to the needle stick prevention device that is in the supply list for the immunization codes. However, the practice expense RVUs for these codes would still be less than the current values. As discussed above, the price for the needle stick prevention device is still fluctuating as new manufacturers enter the market. In addition, it is still not clear exactly which device is optimal for the protection of medical staff. Therefore, until these issues are settled, we will price these immunization services in the nonphysician work pool. This will prevent any sharp decrease in payment for these codes, as well as for payments for the HCPCS G-codes for administration of influenza, hepatitis and pneumococcal vaccines, which are crosswalked to the payment for CPT code 90471.

Comment: We received a comment from Venable, a diathermy manufacturer, who voiced concerns about previous decreases in both the work and the practice expense RVUs for the diathermy procedure, CPT code 97024. According to the commenter, the PEAC recommendations we accepted for 2002 included a substantial reduction in clinical labor time, the elimination of supplies, and the undervaluing of the diathermy equipment, including the assignment of inadequate time for equipment use. Citing our current CPEP price of $3,120 as too low, the commenter noted the cost of the diathermy machines they manufacture range from $19,000 to $30,000 and noted the actual time of a typical treatment is 20 minutes, and not 15, as currently listed. A previous comment from the electrophysiology specialty section of the American Physical Therapy Association (APTA) stated that the average cost of diathermy ranges between $10,000 and $15,000.

Response: We believe the practice expense recommendation we accepted from the PEAC in 2001 for the clinical labor and supplies is appropriate. We would note here that the resultant PEAC recommendation for clinical labor was just one minute less than that proposed by the American Physical Therapy Association at the 2001 PEAC meeting. We continue to support the PEAC's decision to eliminate the supplies for some of the modality procedures, including diathermy, since these services are typically performed with other therapy procedures where the supply costs are captured. However, we agree with the commenter that the current pricing of the diathermy equipment in our CPEP database appears too low, and we will price the diathermy, on an interim basis, at $10,000 for the 2004 fee schedule. In addition, we will assign the requested 20 minutes as the typical time the diathermy equipment is in use for each service. We are planning to propose a repricing of all of the equipment included in our database next year and will revisit the pricing of the diathermy equipment at that time.

In response to the commenter's work RVU concern, next year's final rule will solicit recommendations of codes to be considered for review under the five-year review of work that is to occur in 2005.

Comment: A commenter representing prosthetic urology focused on reductions in payment for several 90-day global prosthetic urology procedures. The commenter contended that these procedures were affected by the adoption of the standard clinical staff times for 90-day global procedures that did not reflect the extra staff time required for patient training during post-procedure visits. In addition, almost half of the prosthetic urology services were established in 2002 and this appeared to have a negative effect on these codes. The commenter strongly recommended that the standard clinical staff times not be applied to the prosthetic urology codes and that we reinstate the “benchmark” clinical staff times.

Response: The commenter is correct that the major cause of the decrease in practice expense RVUs for these services is the use of the standard clinical staff time for 90-day global services. We do not have “benchmark” clinical staff times to reinstate for any of these services. Rather, the current staff times are from the original CPEP panel estimates that have not been reviewed by any multi-specialty panel, such as the PEAC. We accepted the PEAC recommendation to apply the standard clinical staff time to all 90-day global services that had not been reviewed by the PEAC as having exceptions to the standard times. All specialties, including urology, had ample opportunity to present any codes for which they believed the standards did not apply; these urology codes were not brought to the PEAC for review. We do not believe we have a sufficient factual basis for changing the clinical staff times for these services in this final rule. However, we would consider any recommendations for revising the pre- and post-service clinical staff times in the future. As to the effect of using the most recent utilization data in calculating the practice expense RVUs for the new prosthetic urology services, please see the discussion on “Utilization Data” earlier in this section.

Comment: A specialty society representing emergency medicine, an emergency medicine practice management association and an emergency medicine physician practice management organization all commented that the adjustment made in the November 2, 1998 final rule (63 FR 58821) to use the “all physician” practice expense per hour to calculate two indirect cost pools does not make up for the uncompensated care costs of emergency medicine physicians. The practice management association questioned our previous claim that this adjustment was made as a proxy for uncompensated care and asserted it was rather a generic measure to address the low practice expense per hour for emergency medicine. The specialty society commented that it would be difficult to design a supplementary survey to capture the needed data on the levels of uncompensated care.

Response: It is amply clear from reading our entire response in the November 2, 1998 final rule that we considered the adjustment to the indirect costs to serve as a proxy for the uncompensated care experienced by emergency medicine physicians. We believe that, if this adjustment is seen by the specialty as insufficient, the best recourse is for the specialty to undertake a supplementary practice expense survey. By working with our contractor, the Lewin Group, the specialty society should be able to modify the survey in such a way that more accurate data on uncompensated care could be obtained. The data from such a survey could then take the place of the current adjustment to the practice expense per hour for emergency medicine because a proxy for uncompensated care would no longer be needed.

Comment: We received comments from a provider of extracorporeal photopheresis therapy, CPT code 36522, requesting a refinement of the practice expenses of this service in the office setting. Believing this service to be undervalued, the commenter supplied a comprehensive listing of the direct inputs, for the labor, equipment and supplies deemed necessary for the provision of this in-office service. Of particular note among the various suggested supply items was the inclusion of a photopheresis procedural kit.

Response: We want to thank the photopheresis provider for the practice expense suggestions. At this time, we do not have sufficient information regarding the typical resources needed to proceed with a comprehensive refinement of the practice expenses for the in-office provision of photopheresis. However, in reviewing the commenter's various practice expense proposals, we were struck by the obvious absence of the photopheresis procedural kit in our supply database. Consequently, this kit has been added to our CPEP database on an interim basis. We note that there are general similarities between the commenter's proposed inputs for clinical labor and equipment and our current data. We would anticipate a future discussion regarding this service in order to fully refine the practice expense direct cost inputs for photopheresis.

The Act requires that payments vary among Medicare physician fee schedule (MPFS) areas according to the extent that resource costs vary, as measured by the Geographic Practice Cost Indices (GPCIs). Section 1848(e)(1)(C) of the Act requires us to review, and, if necessary, adjust the GPCIs at least every 3 years. This section of the Act also requires us to phase in the adjustment and implement only 1/2 of any adjustment if more than 1 year has elapsed since the last GPCI revision. The GPCIs were first implemented in 1992. The first review and revision was implemented in 1995, the second review was implemented in 1998, and the third review was implemented in 2001. As explained in the August 15, 2003 proposed rule, the fourth GPCI review and revision was scheduled for implementation in 2004. However, because the work and practice expense GPCIs rely primarily on special tabulations of U.S. Census data not yet available, review and revision of only the malpractice GPCI component would occur for implementation in January 2004.

The malpractice GPCI is the most volatile of the three indices with relatively large variations existing between geographic payment localities. We proposed using actual 1999 through 2002 malpractice premium data and forecasting the malpractice premium rates for 2003. We were unable to include proposed malpractice GPCIs based upon this revised malpractice premium data in the August 15, 2003 proposed rule because we were still in the process of collecting the data. We stated that the revised malpractice GPCIs published in this year's final physician fee schedule regulation would be considered interim and subject to public comment.

For purposes of the 2004 update to the malpractice GPCIs we collected actual malpractice premium data for years 1999 through 2001. For 2002 we were able to obtain actual malpractice premium data for 32 states plus Puerto Rico. Where actual malpractice premium data were obtained, premiums were collected from the 20 physician specialties with the largest share of total Medicare RVUs for 2002. Premiums were collected from those insurers with the largest market share and those insurers that when summed with other large insurers comprised at least 50 percent of the state market share for claims-made policies with a $1 million individual case limit and $3 million aggregate case limit.

For those 18 states plus the District of Columbia for which we were unable to obtain actual 2002 premium data, we estimated the 2002 premium based upon an examination of growth rates from 1999 to 2001.

Malpractice premium data were not available for 2003. Two statistical approaches were examined to forecast 2003 malpractice premiums, simple extrapolation and projections based upon the average of historical year-to-year changes (mean rate of change). In most instances, the forecast 2003 premiums were similar using either approach. There was a tendency for the linear extrapolation method to yield slightly more extreme values (positive and negative) so the more conservative, mean rate of change approach was chosen.

Comment: Several commenters expressed concern about the continued use of proxy data, especially HUD residential rent data and nonphysician professional wage data, in the GPCI methodology.

Response: This final rule does not update the work or practice expense GPCIs. Any questions related to the use of proxy data in the calculation of the work and practice expense GPCIs will be responded to as part of future rulemaking.

Comment: One commenter stated that there should be no geographic differences under the physician fee schedule. This commenter felt that the data sources utilized for the construction of the locality specific GPCI indices do not accurately reflect legitimate differences in physician practice costs and that the current methodology did not appropriately reflect the variation that might be caused by case mix, availability of health care resources, and individual practice styles.

Response: Section 1848(e)(1)(A) of the Act requires that payments vary among areas as resources costs vary as reflected by the GPCIs. We agree that there will be some variation in case mix and practice styles between different specialties and individual practitioners. The physician fee schedule was established in 1992 to eliminate the large unjustifiable payment differences that existed among services, specialties, and geographic areas by establishing a national uniform payment system that can vary only as area resource costs vary as measured by the GPCIs. The GPCI component weights represent the average physician expense weights across all physician specialties and are intended to reflect the average costs across all services and specialties in a geographic area and not to reflect exactly the costs of each individual practitioner.

Comment: One commenter stated that there should be no geographic payment differentials because these payment differentials operate as a disincentive for practitioners to practice medicine in rural areas.

Response: Section 1848(e)(1)(A) of the Act requires that payments vary among areas as resources costs vary as reflected by the GPCIs. It should be recognized that the current methodology associated with the calculation of GPCIs partially benefits practitioners in rural areas. This is because the law requires that only one-quarter of area cost differences in physician work, the largest of the three fee schedule components, be recognized. Thus, about 40 percent of fee schedule payments are by statute not adjusted for area cost differences. When combined with the index of 1.000 for medical equipment, supplies, and miscellaneous (which represents about 13 percent of total physician resource costs) this means that there is a national fee schedule for about 53 percent of the average physician payment. That is, only about 47 percent of overall physician payments are adjusted for area resource cost differences. In addition, 34 states have a single statewide GPCI wherein all physicians, whether urban or rural, are paid the same. All of these factors shift payments from higher cost, usually urban, areas to lower cost, usually rural, areas.

Comment: One commenter felt that we should not use projected 2003 premium data and instead should actually collect 2003 premium data.

Response: Currently, 2003 premium data is not available. This is why we will utilize projected 2003 premium data in this update. We plan to utilize more current premium data as it becomes available.

Comment: Although several commenters expressed their support for the use of more current malpractice premium data, a few commenters had concerns about the use of 2001 through projected 2003 premium data and felt that we should use only projected 2004 premium data in place of the three year average.

Response: Since the malpractice index has proven to be the most volatile of the indices in past updates, with significant changes from year-to-year, we will not base the malpractice GPCI upon just one year of projected data. In order to protect against aberrant premiums for any given year, we will utilize a three-year average. We will use 2001 through projected 2003 premium data for the three-year average.

The current methodology projects 2003 malpractice premiums based upon actual malpractice premiums for 1999 through 2002. Since we will continue to collect updated malpractice premium data, we do not think it is appropriate to project through 2004 absent actual 2003 malpractice premium data.

Comment: One commenter suggested that due to the volatility associated with malpractice insurance premium data, we should collect premium data and re-scale the Malpractice GPCI annually.

Response: We agree that, because malpractice insurance premiums are volatile, the Malpractice GPCI is also the most volatile of the three indices. We also agree with the commenter's suggestion regarding annual collection of malpractice premium data. We plan to undertake this collection for 2003 premium data in early 2004. If premium data suggest a re-scaling is warranted, we may revise the GPCIs more frequently than every three years.

Comment: Several commenters requested that we make available to the public the malpractice premium data that was utilized in the calculation of the revised malpractice GPCIs.

Response: Since some of the data upon which the GPCIs were constructed is based upon the reporting of individual malpractice insurance companies, there are some confidentiality issues associated with making the malpractice premium data public. We will attempt to make available any information that is appropriate on our Web site at http://www.cms.hhs.gov.

Comment: The American Medical Association's Relative Value Update Committee (RUC) has requested that CMS work with the RUC's Professional Liability Insurance Workgroup to explore the utilization of premium data that might be collected by the RUC.

Response: We agree with the RUC request and look forward to working with the RUC to obtain more current professional liability premium data.

Acquiring data on malpractice insurance rates and using that data to adjust Medicare payments for future malpractice insurance prices is a difficult task. Malpractice insurance rates are quite volatile due to a variety of factors. Some of these factors are changes in State insurance laws, business decisions of malpractice insurance carriers, and changes in how medicine is practiced.

The volatility of malpractice premium data was quite evident in the data we collected in conducting our review of malpractice GPCIs. Based on these data and the comments received on the August 15, 2003 proposed rule, we have modified some of our GPCI calculations and assumptions.

We are very concerned about implementing sharp changes in malpractice GPCIs for 2004, which directly impact physician fee schedule payment amounts. At the same time, we recognize the importance of updating malpractice GPCIs to ensure local differences in physician costs are included in payment amounts. To be sensitive to both of these considerations, we decided to apply a modulating factor of .5 to the changes in the malpractice GPCIs. In other words, as part of our review and analysis of the malpractice GPCIs, we reduced the difference between the new and previous malpractice GPCIs by 50 percent.

As directed by the statute, we will implement 1/2 of this change in the first year (CY 2004) and 1/2 of this change in the second year (CY 2005). During this two-year phase-in, we will continue to monitor local malpractice markets, work with the State Departments of Insurance, and collaborate with the RUC to obtain the most current and best malpractice premium data available. As better data are obtained, we will review, propose changes, and revise the malpractice GPCIs as appropriate. The transitional 2004 and full 2005 GPCIs can be found at Addendum D and Addendum E, respectively. These malpractice GPCI revisions necessitate a budget neutrality adjustment, as required by law. Therefore, we adjusted the 2004 through 2006 malpractice GPCIs by 1.0021.

In the August 15, 2003 proposed rule we requested comments on the composition of the current 89 Medicare physician payment localities to which the GPCIs are applied.

Comment: We received numerous comments from professional medical associations, beneficiaries, and practitioners requesting that the specific counties in which they practice medicine or receive medical care be removed from their current locality assignment.

Response: We will continue to examine alternatives for reconfiguring the current locality structure. We expect to further consider this issue as part of future rulemaking.

The November 1, 2001 final rule (66 FR 55269) included a discussion of CPT Category III codes (also known as CPT tracking codes) and stated that carriers have discretion for coverage and payment of services described by these CPT tracking codes unless we have made a national coverage determination (NCD). We have received requests to create national payment amounts for some CPT tracking codes even if there has been no NCD. Based on these requests, we proposed to change our policy regarding payment for CPT tracking codes and create national payment policy and determine national payment amounts for CPT tracking codes when there is a significant programmatic need for us to do so. This policy change would not change the contractor's discretion over coverage for the CPT tracking codes, but could establish a payment level to be used if the contractor finds that coverage is warranted. In addition, carriers would not be required to establish a payment amount for a tracking code until they receive a claim for the code.

Comment: Several commenters expressed concerns about this proposal. They believe that establishing a national payment rate for these codes risks premature creation of payment levels of reimbursement and creates an expectation for the future value of the code. The commenters also stated that establishment of a national price could also subvert the RUC process because such pricing could influence subsequent RUC valuation or our acceptance of the RUC's recommendations. Other commenters were supportive of the proposal, with some suggesting that we work with the specialty societies and the RUC in determining appropriate payment rates. One commenter suggested that an alternative to the proposal would be to use the existing refinement panel process because these refinement panels are multispecialty and feature the relevant specialty expertise. One commenter also requested we establish RVUs for specific tracking codes in the final rule.

Response: We understand the reservations and concerns of the commenters. As we indicated in the proposed rule, we would determine national payment amounts for CPT tracking codes only when there is a significant programmatic need for us to do so. If there is a need to establish payment amounts for a tracking code, we would appreciate the assistance of the relevant specialty societies and the RUC and such pricing would be subject to public comment. However, in some instances, interim values might need to be established if timing does not permit us to obtain prior input from the medical community.

We will finalize our proposal to create national payment policies and determine national payment amounts for CPT tracking codes when there is a significant programmatic need for us to do so. We note that, as discussed in the August 15, 2003 proposed rule, this policy change would not change the contractor's discretion over coverage for CPT tracking codes, but would establish a payment level if the contractor finds that coverage is warranted.

The definitions for excision of benign lesions (CPT codes 11400 through 11446 inclusive) and excision of malignant lesions (CPT codes 11600 through 11646 inclusive) were substantively changed for 2003. These codes are now reported based on the excised diameter (actual skin removed) rather than on the size of the lesion. Based on these changes to the code descriptors, we proposed to make the work RVUs the same for removal of all skin lesions with the same excised diameters that are from the same area of the body, whether the lesions are benign or malignant. For example, the work RVUs for the removal of benign skin lesions from the trunk, arms or legs with excised diameter 1.1-2.0 cm, CPT code 11402, would be the same as the work RVUs for CPT code 11602, which is the removal of malignant skin lesions from trunk, arms or legs with excised diameter of 1.1-2.0 cm.

Comment: The specialty society representing dermatology objected to this proposal and contended that the excision of malignant lesions generally goes deeper and is more time-consuming than the excision of benign lesions and that malignant lesion excision also requires greater skill and embodies greater risk. The society stated that this proposal ignores a multi-specialty effort by a CPT Integumentary Workgroup, the CPT Editorial Panel and the RUC to revise the code descriptors and to assign work RVUs to these services. This view was supported by a joint comment from the heads of several surgical specialties. The RUC also urged us to delay finalizing this proposal until the RUC has the opportunity to provide further recommendations related to these services. In addition, the specialty societies representing podiatry, general surgery, colon and rectal surgery, osteopathy, ophthalmology, plastic surgery, otolaryngology as well as the AMA, the Mayo Foundation and individual physicians also urged us to withdraw this proposal. Medical Group Management Association requested the policy rationale for equating the work RVUs for the benign and malignant code pairs. The specialty society representing family physicians agreed with and supported our position that there is no difference in physician work involved in excising a benign or malignant lesion. However, the commenter did not support our proposal to implement such RVU changes unilaterally and stated that we should utilize the CPT and RUC process.

Response and Final Decision: We still believe that the physician work for these services is sufficiently similar not to warrant differences in the work RVUs. However, we will maintain the 2003 work RVUs as interim values for 2004 to allow opportunity for the specialty to resurvey these services. Note: That due to the adjustments to work RVUs to match the MEI weights, the work RVUs in Addendum B may differ from the values in 2003.

As explained in the August 15, 2003 proposed rule, technological advances have made cardiac telemetry equipment, typically used in hospitals, available in the home setting. Coverage of this technology is currently at the discretion of the local Medicare contractors because there is no national coverage determination for this service. We proposed to establish new HCPCS codes to specifically describe this service along with proposed RVUs and PE inputs for payment as follows:

GXXX1—Electrocardiographic monitoring for diagnosis of arrhythmias, utilizing a home computerized telemetry station and trans-telephonic transmission, with automatic activation and real time notification of monitoring station, 24-hour attended monitoring, per 30-day period of time; includes recording, monitoring, receipt of transmissions, analysis, and physician review and interpretation. (global)

We proposed 0.52 physician work RVUs and 0.24 malpractice RVUs for this service and proposed crosswalking the practice expense inputs from CPT Code 93268 Patient demand single or multiple event recording with presymptom memory loop, 24-hour attended monitoring, per 30 day period of time; includes transmission physician review and interpretation.

GXXX2— Electrocardiographic monitoring for diagnosis of arrhythmias, utilizing a home computerized telemetry station and trans-telephonic transmission, with automatic activation and real time notification of monitoring station, 24-hour attended monitoring, per 30-day period of time; recording (includes hook-up, recording and disconnection).

We proposed 0.07 malpractice RVUs and crosswalked the practice expense inputs from CPT Code 93270, Patient demand single or multiple event recording with presymptom memory loop, 24-hour attended monitoring, per 30 day period of time; recording (includes hook-up, recording, and disconnection).

GXXX3— Electrocardiographic monitoring for diagnosis of arrhythmias, utilizing a home computerized telemetry station and trans-telephonic transmission, with automatic activation and real time notification of monitoring station, 24-hour attended monitoring, per 30-day period of time; monitoring, receipt of transmissions, and analysis

We proposed 0.15 malpractice RVUs and crosswalked the practice expense inputs from CPT Code 93271, Patient demand single or multiple event recording with presymptom memory loop, 24-hour attended monitoring, per 30 day period of time; monitoring, receipt of transmission, and analysis.

GXXX4—Electrocardiographic monitoring for diagnosis of arrhythmias, utilizing a home computerized telemetry station and trans-telephonic transmission, with automatic activation and real time notification of monitoring station, 24-hour attended monitoring, per 30-day period of time; physician review and interpretation.

We proposed 0.52 physician work RVUs and 0.02 malpractice RVUs and also crosswalked the practice expense inputs, from CPT code 93272 Patient demand single or multiple event recording with presymptom memory loop, 24-hour attended monitoring, per 30 day period of time; physician review and interpretation only.

Comment: Commenters representing cardiac arrhythmia specialists and cardiologists recommended that we withdraw the proposal to create new G codes for monitoring heart rhythms. The commenters stated that this request was not made by the medical community nor from the manufacturers of these heart rhythm monitoring systems. The commenters contended that the proposal appears to address specifically one manufacturer and specifies a particular mode of transmission and patient location, even though there are other new systems of this type that are not captured by this proposal.

The commenters recommended that we allow this technology to be utilized on a local level before implementing a national coding solution. The commenters further supported that when this new technology warrants a national coding solution, a CPT coding application should be initiated and then the code should be sent to the RUC for review. The commenting specialties stated their willingness to provide medical input into the evaluation, coding and reimbursement for this new technology. Two commenters also stated that the descriptors and the proposed reimbursement do not reflect the monitoring systems that have been developed. Other commenters also requested that we withdraw or reconsider our proposal, as it did not follow the established process for creating and valuing new codes. One specialty society representing clinical endocrinologists supported the establishment of these HCPCS codes, while another commenter, a cardiac monitoring company, provided a general outline of how the various cardiac monitoring technologies can best be used for maximum quality and value. Another commenter suggested that until efficiency of the new technology is demonstrated this proposal should be postponed.

Response: Our intention in proposing these G codes was to recognize and nationally price all currently available real time cardiac telemetry monitoring technology. It was not intended to address only one system currently in use. Based on the concerns raised by commenters, we will not proceed with these proposed HCPCS codes because we want to ensure that any HCPCS codes developed encompass the various technologies that are being utilized for such monitoring.

Flow cytometry is a technique to analyze single cell suspensions from blood, bone marrow, body fluids, lymph nodes, and other tissues. The technique, currently coded as CPT code 88180, Flow cytometry, each cell surface, cytoplasmic or nuclear marker, quantifies cell surface, cytoplasmic, and nuclear antigens. The August 15, 2003 proposed rule discussed our concerns that the current coding scheme (payment on a per marker basis) may encourage the performance of more markers than may be medically necessary because the pathologist determines what markers to perform and when to perform them. We indicated that we understood the laboratory community would be reviewing this issue and considering whether to recommend changes to the current coding for the procedure. We also requested recommendations on appropriate values for the procedure should we wish to develop a future proposal.

Comments: Commenters, both individuals and organizations, asked that we not put forth a proposal for payment of flow cytometry. The College of American Pathologists (CAP) has proposed coding revisions to both the immunology and anatomic pathology section of CPT and is working with other groups to establish practice guidelines for flow cytometry. CAP asked that we not establish new “G” codes for 2004, but work with CAP and allow the CPT and RUC evaluation process to be used to determine appropriate coding and relative value units for flow cytometry.

Decision: We agree with the commenters. We will work with CAP, the CPT and the RUC to develop appropriate coding and payment policies for flow cytometry.

In the August 15, 2003 rule, we proposed to make CPT codes 90918, 90919, 90920, and 90921 for the monthly capitation payments (MCP) invalid for Medicare. We also proposed to create 3 new G codes in place of each CPT code with payments varying with the number of visits provided within each month to an end stage renal disease (ESRD) patient. Under our proposal, there would be separate codes when the physician provides 1 visit per month, 2-3 visits per month and 4 or more visits per month. The code for 1 visit per month would have the lowest payment while a higher payment will be provided for 2 to 3 visits per month and the highest payment for 4 or more visits per month. These new codes would be reported once per month for services performed in an outpatient setting that are related to the patient's ESRD. These physician services would continue to include the establishment of a dialyzing cycle, outpatient evaluation and management of the dialysis visits, telephone calls, and patient management provided during a full month. These codes would not be used if a hospitalization occurred during the month.

The proposed codes are as follows:

GXXX5—End Stage Renal Disease (ESRD) related services per full month, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

GXXX6—End Stage Renal Disease (ESRD) related services per full month, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

GXXX7—End Stage Renal Disease (ESRD) related services per full month, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

GXXX8—End Stage Renal Disease (ESRD) related services per full month, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

GXXX9—End Stage Renal Disease (ESRD) related services per full month, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

GXX10—End Stage Renal Disease (ESRD) related services per full month, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

GXX11—End Stage Renal Disease (ESRD) related services per full month, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

GXX12—End Stage Renal Disease (ESRD) related services per full month, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

GXX13—End Stage Renal Disease (ESRD) related services per full month, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

GXX14—End Stage Renal Disease (ESRD) related services per full month, for patients 20 years of age and over; with 4 or more face-to-face physician visits per month.

GXX15—End Stage Renal Disease (ESRD) related services per full month, for patients 20 years of age and over; with 2 or 3 face-to-face physician visits per month.

GXX16—End Stage Renal Disease (ESRD) related services per full month, for patients 20 years of age and over; with 1 face-to-face physician visit per month.

We based the proposed payments on the assumption that many physicians would provide 4 or more visits to their ESRD patients and a smaller proportion would provide 2-3 visits or only 1 visit per month. Using Medicare utilization data from 2002, we proposed the following relative value units for the new G codes that would make Medicare's aggregate payments for ESRD related services under the physician fee schedule approximately equal to current payments for procedure codes 90918 to 90921:

As part of the proposed rule we also solicited comments on how to further revise our payment methodology to improve quality of care and outcomes. We requested information that could help us design future demonstrations that would study both dimensions of care (quality and utilization) and help ensure that payment is based on appropriate patient-specific care that has been shown to lead to improved outcomes for this complex patient population.

Comment: We received many comments from physicians, the RUC, specialty societies, dialysis centers and nephrologists, as well as other individuals and organizations who expressed concerns with our proposal to alter the way physicians are reimbursed for services provided to End Stage Renal Disease (ESRD) patients and who urged us to withdraw the proposal. The RUC and the AMA, as well as other specialty organizations, expressed disappointment that we developed this proposal without consultation from the medical community and outside the usual CPT and RUC process. The Renal Physicians Association (RPA), the American Society for Nephrology (ASN), the American Association for Kidney Patients (AAKP), and the National Kidney Foundation (NKF) all supported the principle of optimizing nephrologist-dialysis patient interaction, which is included in the proposal. However, the RPA contended that the proposal as currently constituted is unworkable, may negatively impact some dialysis patients and is being put on an unreasonably precipitous implementation schedule. The AAKP outlined similar concerns but believed that increased nephrologist-dialysis patient interaction will lead to improved outcomes and also urged that an advisory committee be established to assist in the effort to further improve quality and coordination of care for dialysis patients. The Medicare Payment Advisory Commission (MedPAC) agreed that the current payment method lacks accountability and quality incentives, and thus encouraged CMS to address these issues. However, MedPAC also expressed concern that without baseline data it was unclear how we could determine and measure the impact of the proposed changes on quality and access. MedPAC further stated that the adjustments to payment should be made subsequent to the collection of information on resource costs and clinical guidelines. Together with these adjustments, further incentives should be added to the monthly payment to reward and improve the quality and access of dialysis-related physician care, which is consistent with MedPAC's June 2003 recommendations. Below are the specific issues raised by commenters:

Many comments concerned the large variation in proposed payments to physicians who see a patient only once a month, compared to the proposed payment for seeing a patient either two or three times during the month or four or more times during the month. In addition, commenters stated there is more work involved in managing care of the ESRD patients between visits.

Response: Based on our review of the comments, we agree that a significant amount of physician work for patients with ESRD occurs outside of the face-to-face visit with the patients. Since there may be significant physician work associated with providing physician services to ESRD patients between visits, we agree that there should be less difference in the payment levels than we proposed. By raising the minimum payment level, we are accounting for the extensive patient care coordination and other non-face-to-face management required by ESRD patients. However, we continue to believe that more physician work is associated with more frequent face-to-face visits with the patient, and any variation in the payment amounts should reflect this difference.

First, we determined the appropriate relative relationship among different codes. For instance, we believe that approximately 25 percent more physician work is involved with providing two to three visits than with a single visit, and 50 percent more physician work is associated with providing four or more visits. By paying a single amount regardless of how often the patient is seen, we believe our current policy pays too much if the patient is seen fewer than four times per month. Thus, we revised our payment to be consistent with different levels of physician work associated with providing monthly management of dialysis patients. We are setting our aggregate revised payments equal to aggregate current payments. Consistent with these assumptions, we determined the following RVUs:

We used the above principles to establish our monthly capitation payments (MCP) for patients 20 or older. For patients younger than 20, we are using the same relationship that exists among the current MCP codes for different age groups for the new codes that we are creating. For example, the current MCP code for a patient under 2 (CPT code 90918) has work RVUs that are approximately 2.5 times the work RVU for a patient 20 or older (CPT code 90921). Thus, Medicare's work RVU for each code for a patient 2 years or younger will be 2.5 times the amount of the corresponding service provided to a patient 20 or older. These values can be considered as interim and we plan to seek the advice of the RUC in evaluating these codes once the policy has been implemented. There are efforts underway (for example, in their 2004 workplan, the OIG has indicated they will conduct a review of ESRD monthly capitation payments and physician services) which will provide data on the type frequency and content of physician encounters as suggested by MedPAC. However, we believe a change should be made in the interim to improve care and accountability. The use of these new codes will also enable us to collect data about the frequency of physician visits.

We received comments regarding the impact of these proposed changes on the sustainable growth rate (SGR) calculations. Commenters expressed concern that, if physician behavior changes and physicians increase the number of visits provided per month, actual expenditures would surpass the target projection, resulting in a future payment reduction for all of medicine.

Response: Section 1848(c)(2)(B)(ii)(II) of the Act requires that changes to RVUs cannot cause the amount of expenditures to increase or decrease by more than $20 million from the amount of expenditures that would have been made if such adjustments had not been made. As indicated above, we have established RVUs for the new monthly capitation codes so that Medicare's aggregate payments for these services are equal to what we would have paid in the absence of these changes. We are not expecting any impact on payment for other physician fee schedule services. However, we will continue to review this issue as we work with the medical community to further refine Medicare policy for treating patients needing dialysis services.

Many comments were received regarding home dialysis because patients who dialyze at home typically see their physicians less frequently than other ESRD patients. One commenter suggested that home dialysis patients be excluded from the proposed change and that we continue to pay the current MCP rate for services to these patients.

Response: We have created four G codes for the management of home dialysis patients in each of the age groups and will pay for the home dialysis patients at the same rate as codes G0309, G0312, G0315, and G0318 respectively. Although the codes for home dialysis patients will pay physicians slightly less than the former MCP, physicians will still have a relative incentive to increase the use of home dialysis. We believe this is consistent with Section 1881(b)(3)(B) of the Social Security Act which states “With respect to payments for physicians’ services furnished to individuals determined to have end stage renal disease, the Secretary shall pay 80 percent of the amounts calculated for such services on a comprehensive monthly fee or other basis (which effectively encourages the efficient delivery of dialysis services and provides incentives for the increased use of home dialysis) for an aggregate of services provided over a period of time (as defined in regulations).”

The new G codes for the monthly management of home dialysis patients will be as follows:

G0320—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients under two years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0321—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients two to eleven years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0322—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients twelve to nineteen years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0323—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients twenty years of age and older.

The American Society of Nephrology also commented that “reimbursement should be constructed so that home dialysis patients should see their nephrologist at least monthly, with further visits on an as needed basis.” We will not specify the frequency of required visits at this time but expect physicians to provide clinically appropriate care to manage the home dialysis patient.

If home dialysis patients are hospitalized during the month, four new G codes have been created: G0324, G0325, G0326, and G0327. These codes will be used to report daily management of home dialysis patients for the days the patient is not in the hospital. CPT codes 90922, 90923, 90924, and 90925 will be considered inactive for Medicare because they are now redundant as other codes are to be used by physicians billing for services to ESRD patients.

The new G codes are as follows:

G0324—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients under two years of age.

G0325—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients between two and eleven years of age.

G0326—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients between twelve and nineteen years of age.

G0327—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients twenty years of age and over.

For example, if a home dialysis patient is in the hospital for 10 days (counting the calendar day of admission and the calendar day of discharge) and is cared for 20 days in his or her home, then 20 units of the code for the appropriate aged patient is billed.

If a home dialysis patient receives dialysis in a dialysis center or other facility during the month, the physician is still paid the management fee for the home dialysis patient and cannot bill the codes in the range of G0308 through G0319 or CPT codes 90935 or 90937, even though the physician may see the patient during his/her center dialysis.

We received comments about the role of nonphysician practitioners. It was not clear to the commenters whether visits by these practitioners could count as face-to-face encounters by the MCP physician. The commenters also asked about billing by physicians (for example, a “rounding” physician or fellow) other than the physician who is billing the monthly capitation rate.

Response: Physicians may utilize nonphysician practitioners: nurse practitioners, physician assistants, and clinical nurse specialists, who are able under the Medicare statute to furnish services that would be physician services if furnished by a physician and who are eligible to enroll in the Medicare program, to deliver some of the visits during the month. The rules for the use of these physician extenders would be consistent with the rules for split/shared evaluation and management visits: The nonphysician practitioners and physician must be in the same group practice or employed by the same employer/entity; and the physician must perform some portion of the service in a face-to-face encounter, in this case one or more visits during the month with the patient. In this situation, to bill the service under the physician's UPIN/PIN, the physician and not the physician extender should be the practitioner to perform the visit with the complete assessment of the patient and to establish the patient's plan of care. If the nonphysician practitioner is the practitioner who performs the complete assessment and establishes the plan of care, then the MCP service should be billed under the UPIN/PIN of the nurse practitioner, physician assistant, or clinical nurse specialist.

It is also possible for the physician to use another physician to provide some of the visits during the month, but the physician who provides the complete assessment, establishes the patient's plan of care and provides the ongoing management should be the physician who submits the bill for the monthly service. The non-MCP physician must have a relationship with the billing physician such as a partner, employees of the same practice, or supervising physician and fellow doing sub-specialty training.

Each practitioner should document in a shared medical record services he/she personally performed. Only one practitioner can bill for the management of the ESRD patient in any month. In addition, when a nonphysician practitioner or a “rounding physician” sees a dialysis patient for management of ESRD, they cannot bill an evaluation and management service for the same patient unless there is a separate, substantial and documented service evaluating the patient for care unrelated to the patient's dialysis.

Commenters indicated that the lack of geographic considerations would negatively impact physicians and patients in rural and some urban settings where physician visits require significant travel time. Extended travel time can make it difficult for physicians to see patients as often as patients can be seen when the physician's office is near the dialysis facility.

Response: We believe that the policy to allow nurse practitioners, physician's assistants, clinical nurse specialists, and other physicians to deliver some of the visits to patients as well as changes in the payment to more accurately reflect non-visit services and the relative value of additional visits will ameliorate these access issues.

Commenters noted that the proposed rule did not provide enough detail regarding alternative billing procedures if hospitalization occurs during the month.

Response: For ESRD patients (other than home dialysis patients) who are hospitalized during the month, the physician may bill the code that reflects the number of face-to-face visits during the month on days when the patient was not in the hospital (either admitted as an inpatient or in observation status).

Comment: Many commenters asked for clarification regarding the documentation requirements, if any, associated with the new codes.

Response: We have chosen not to include specific documentation guidelines in this rule. Instead, physicians should document what is clinically relevant, including but not limited to the patient's current status and complaints, a clinically appropriate physical examination, assessment of the patient's treatment for ESRD that includes assessment of the adequacy of the dialysis treatment, the status of the patient's vascular access, assessment and treatment of the other conditions associated with ESRD, such as anemia, electrolyte management, and bone density, as well as changes to the patient's management.

Comment: A comment was received that HIPAA transaction and code set rules may not be met if these new codes were implemented.

Response: G codes are part of the HCPCS coding system and are in compliance with the HIPAA transaction and code set rules.

Comment: Commenters asked for additional clarification on whether visits counted toward the MCP can be provided in settings other than the dialysis facility.

Response: The visits for management of ESRD patients may occur in the physician's office, in an outpatient hospital or other outpatient setting or even in the patient's home as well as in the dialysis facility.

Comment: Commenters inquired how physicians would deal with visits and related billing for traveling patients who receive their treatment away from their usual site of treatment.

Response: If the physician manages the care of a patient who is receiving treatment away from the patient's usual site of treatment, the physician who bills for managing the care of the patient is still paid according to the number of times the physician has a face-to-face visit with the patient. If the patient is to be away for an extended period of time, the patient would be managed by the physician who has face-to-face visits with the patient, and that physician would be the one billing for the patient's care management.

Comment: Commenters representing the American Osteopathic Association, the American Academy of Family Physicians, the National Coalition for Quality Diagnostic Imaging Services, the American Society for Echocardiography and Focus on Therapeutic Outcomes, Inc., provided information on quality initiatives their respective organizations have undertaken or suggestions for relating quality to payment. The National Kidney Foundation recommended the use of technology and other forms of communication to care for ESRD patients and to support constant attention to quality. In addition, the Society for Interventional Radiology commended our efforts to increase the use of arteriovenous fistulae for vascular access in dialysis patients as part of its National Vascular Access Improvement Initiative, but indicated there might be a need to clarify certain policies. The American Association of Kidney Patients (AAKP) also recommended the establishment of a commission or advisory group with representation of the kidney community that could be charged with recommending proposals to tie reimbursement to outcomes. AAKP stated that although the proposed changes are important, these changes remain a change in process of delivery of care that may improve actual outcomes, rather than a change in actual outcomes, that is, in rehabilitation, morbidity, mortality, and quality of life. MedPAC agreed with CMS that the proposed change to provide incentives for additional nephrologist-dialysis patient interactions may not be the ideal method to improve patient outcomes and to achieve this goal, CMS should partner with the ESRD community and work toward a long-term solution. MedPAC suggested that we investigate and incorporate physician clinical practice guidelines into our payment approach, and measure physician quality directly. MedPAC also suggested that we examine whether physician resources vary based on patient complexity, stating that to the extent that resources do vary, a case-mix adjustment—similar to the one MEDPAC recommended for payment to dialysis facilities in its June 2003 report—would be desirable.

Response: We appreciate the information and suggestions provided by the commenters and will take these into consideration. We plan to investigate the use of new technology to improve the management of ESRD patients as part of our overall focus on quality.

Final Decision —We will create the following G Codes to be used for ESRD patients other than home dialysis, based on the age of the patient and number of visits:

G0308—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

G0309—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

G0310—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients under 2 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

G0311—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

G0312—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

G0313—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 2 and 11 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

G0314—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 4 or more face-to-face physician visits per month.

G0315—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 2 or 3 face-to-face physician visits per month.

G0316—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients between 12 and 19 years of age to include monitoring for the adequacy of nutrition, assessment of growth and development, and counseling of parents; with 1 face-to-face physician visit per month.

G0317—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients 20 years of age and over; with 4 or more face-to-face physician visits per month.

G0318—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients 20 years of age and over; with 2 or 3 face-to-face physician visits per month.

G0319—End Stage Renal Disease (ESRD) related services during the course of treatment, for patients 20 years of age and over; with 1 face-to-face physician visit per month.

In addition we have created the following G codes for home dialysis patients:

G0320—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients under two years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0321—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients two to eleven years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0322—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients twelve to nineteen years of age to include monitoring for adequacy of nutrition, assessment of growth and development, and counseling of parents.

G0323—End stage renal disease (ESRD) related services for home dialysis patients per full month; for patients twenty years of age and older.

G0324—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients under two years of age.

G0325—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients between two and eleven years of age.

G0326—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients between twelve and nineteen years of age.

G0327—End stage renal disease (ESRD) related services for home dialysis (less than full month), per day; for patients twenty years of age and over.

Comment: We received several comments concerning the pricing of CPT code 93701, electrical bioimpedance. One commenter, a carrier medical director, requested that this service be considered a technical component service as there is no physician work (professional component) required to produce the results. The commenter referenced the RUC recommendation of 0.00 work that was not accepted by CMS in November 2001. Other commenters stated that pricing of this service should be revisited and the American College of Cardiology recommended work component of 0.25 RVUs be accepted. Commenters also questioned the valuation of the practice expense component, particularly in light of the escalating costs associated with this service.

Response: In next year's final rule we will be accepting recommendations for codes to be considered under the five-year review of work that will occur in 2005. The commenters will be able to respond to that solicitation, and submit this CPT code, as well as any other services they believe need to be reviewed to ensure they are appropriately valued. We are currently in the process of reviewing and obtaining updated pricing for equipment contained in the practice expense data files and proposed changes to pricing for equipment will be included in next year's proposed rule. We would suggest that the commenters review this information when published to ensure that the cost of the equipment is accurately reflected in the database.

Comment: One commenter, a manufacturer, suggested that the work RVUs of certain codes for the ablation of liver tumors (CPT codes 47380, 47370 and 47382) appeared to be undervalued.

Response: As discussed in the previous response, in next year's final rule we will be accepting recommendations for codes to be considered under the five-year review of work that will occur in 2005. The commenter will be able to respond to that solicitation and submit these codes, as well as any additional services they believe need to be reviewed to ensure they are appropriately valued.

Comment: Two commenters requested that HCPCS codes G0173 and G0251, which are used for reporting stereotactic radiotherapy and stereotactic radiosurgery under the hospital outpatient prospective payment system, be activated for payment under the physician fee schedule.

Response: We are reluctant to establish payment for these services under the physician fee schedule at this time absent specific information on freestanding centers providing this service. We would welcome information and data from these commenters, and other individuals and providers, on the provision of these services in freestanding centers so that we can fully evaluate this issue.

Comment: The AMA and several specialty organizations expressed concern about the establishment of the numerous G codes that were contained in the proposed rule. The commenters state that continual development of G codes, without consultation with the CPT Editorial Panel, the RUC or the physician community undermines the annual review process that CMS has established in the final rule. Further, the commenters argue that the establishment of G Codes undermines the requirements of the Health Insurance Portability and Accountability Act (HIPAA) for coding standardization and an open process for establishing codes.

Response: As we have stated in previous rulemaking, it is sometimes necessary to develop G codes to accommodate changes in legislation, regulation, coverage, and payment policy. We appreciate the input of the medical community and to the extent possible, will work with the CPT Editorial Panel, the RUC and the physician community prior to establishment of these codes.

Comment: The American Society of Interventional Pain Management commented on the differences in payment allowances for various pain management services and other non-pain management services furnished in conjunction with pain management services in various settings, including the physician's office, the OPD and the ASC.

Response: In accordance with the law, we have established payment rates for office-based procedures, using the non-facility practice expense relative value units. However, the office does not represent a practice site where these services are usually performed.

Medicare payment under the physician fee schedule for the physician work is the same in all practice settings. However, the practice expenses are reimbursed differently depending on the practice site. Practice expenses associated with procedures performed in the outpatient departments (OPDs) or ambulatory surgical centers (ASCs) are paid under the OPD or ASC payment system respectively. Practice expenses associated with procedures performed in the physician's office are paid through the physician fee schedule payment system.

In the August 15, 2003 rule we proposed to adopt the definition of diabetes used to determine beneficiary eligibility for Medical Nutrition Therapy (MNT) for purposes of coverage for outpatient diabetes self-management training when the beneficiary has a diagnosis of diabetes. Specifically, we stated that the criteria currently set forth at § 410.141(d), would be replaced with definition of diabetes used for medical nutrition therapy at § 410.130 which reads as follows:

Diabetes means diabetes mellitus consisting of two types. Type 1 is an autoimmune disease that destroys the beta cells of the pancreas, leading to insulin deficiency. Type 2 is familial hyperglycemia that occurs primarily in adults but can also occur in children and adolescents. It is caused by an insulin resistance whose etiology is multiple and not totally understood. Gestational diabetes is any degree of glucose intolerance with onset or first recognition during pregnancy. The diagnostic criterion for a diagnosis of diabetes for a fasting glucose intolerance test is greater than or equal to 126 mg/dL.

A technical error in the proposed rule on page 49070, placed the revised eligibility requirements in § 410.141(f). The eligibility requirements will replace those currently in § 410.141(d).

Comment: We received comment noting that the language for the actual regulatory language had the wrong section letter.

Response: As noted above, this was a technical error.

Final Decision: The following language will replace what was in the proposed rule. “Section 410.141 is amended by replacing paragraph (d) to read as follows: § 410.141 Outpatient diabetes self-management training. (d) Beneficiaries who may be covered. Medicare Part B covers outpatient diabetes self-management training for a beneficiary who has been diagnosed with diabetes.”

Comment: The comments were very supportive of our efforts to streamline this requirement. Several commenters recommended that the definition of diabetes be revised to include patients who might not be classified as Type 1, Type 2, or gestational diabetes in the definition. Most commenters recommended the use of a fasting glucose test of greater than or equal to 126 mg/dL. One commenter suggested the measurement be taken on two occasions. Most commenters also recommended the addition of a random glucose test of greater than 200 mg/dL, with one commenter adding with symptoms of uncontrolled diabetes. Several commenters suggested use of an abnormal glucose tolerance test (GTT). One commenter also suggested the use of a 2 hour post-glucose challenge of greater than or equal to 200 mg/dL test on two different occasions. The American Association of Clinical Endocrinologists (AACE) also suggested that coverage of medical nutrition therapy be expanded to those with impaired fasting glucose.

Response: The definition of diabetes used in the MNT regulation was based on language found in the 2000 Institute of Medicine report entitled, “The Role of Nutrition in Maintaining Health in the Nation's Elderly. We did not have any other generally recognized definition of diabetes at that time and did not intend to limit our definition of diabetes. Regarding the laboratory tests, the characteristics of the commenters' suggestions are generally the same. The base measurement that is already in our MNT regulation, a fasting glucose of 126 mg/dL, is a common measure. Three commenters also noted the use of 200 mg/dL for a random glucose test. The major variation between the commenters was that one suggested multiple measurements. Also, we note that patients with an impaired fasting glucose level do not necessarily meet any of the popular definitions of diabetes.

Final Decision: We agree that in some ways our proposed definition may not include some patients diagnosed with diabetes. We also agree that our clinical laboratory measurements used to determine the presence of diabetes should be expanded. The definition provided by AACE appears to meet the clinical concerns of the medical community and our concerns that no individuals have their treatments delayed unduly if they have obvious symptoms of uncontrolled diabetes. Therefore, we are adopting their clinical definition. We will also broaden our general language to include diabetes of other types. Our final language will be, “Diabetes is diabetes mellitus, a condition of abnormal glucose metabolism diagnosed using the following criteria: A fasting blood sugar greater than or equal to 126 mg/dL on two different occasions; a 2 hour post-glucose challenge greater than or equal to 200 mg/dL on 2 different occasions; or a random glucose test over 200 mg/dL for a person with symptoms of uncontrolled diabetes.” We will also make a conforming amendment to 410.130 for MNT. However, we are constrained from covering MNT for anyone who is not diagnosed with diabetes by the section 1861(s)(2)(V) of the Act that limits coverage of MNT to beneficiaries with diabetes or renal disease.

In almost all settings, our regulations specify that outpatient therapy services can be delivered only by qualified physical therapists, occupational therapists, physical therapy assistants, occupational therapy assistants, and speech-language pathologists as defined by § 484.4. Section 1862(a)(20) of the Act requires that any therapy services furnished incident to a physician's professional services must meet the standards and conditions that would apply to such therapy services if they were furnished by a therapist, with the exception of the licensing requirement. While there are currently no national standards for qualifications of individuals providing outpatient therapy services incident to physicians' services, we believe that standards similar to those in § 484.4 are appropriate. In the proposed rule, we stated that we are considering adopting the existing qualification and training standards (with the exception of licensure) in § 484.4 for individuals providing therapy services independently and incident to physicians' services. While we did not propose a change at this time, we requested comments from the public, particularly physicians and staff who would be affected, on adoption of the existing standards in § 484.4, for services of independent therapists and “incident to” services, as well as comments regarding alternatives that we might use to ensure that qualified staff are providing “incident to” therapy services.

We received comments from major therapy organizations and individual therapists representing therapy services, physician organizations and individual physicians and associations and individuals representing other health care professionals, such as athletic trainers, kinesiotherapists and exercise physiologists. A wide spectrum of views was expressed by these commenters. Commenters representing therapists were supportive of establishing consistent training standards in all settings, while physicians favored reliance on the individual physician for quality control. The non-therapist health care providers were concerned about their role in providing therapy services and cardiac rehabilitation and pulmonary service providers were concerned that their services might be affected.

We will review and consider these comments as we determine whether to make a future proposal. Meanwhile, contractors may continue to develop local medical review policies that are consistent with the statute, applying to physical therapy, occupational therapy and speech-language pathology services the same standards and conditions that would apply to such therapy services if they were furnished by an independent therapist, with the exception of the licensing requirement.

In the December 2002 final rule, we modestly increased the work of anesthesia services. These changes were based on the analysis submitted by the RUC of its review of the work of 19 high volume anesthesia codes. The RUC had provided us with its analysis but did not furnish us with a definitive recommendation. The increase in anesthesia work resulted in an increase in the national anesthesia conversion factor. (We increased the physician work component of the anesthesia conversion factor by 2.10 percent to reflect a 9.13 percent increase in anesthesia work applied to 23 percent of anesthesia allowed charges represented by the 19 codes. As a result of this increase, we applied a 1.6 percent increase to the anesthesia CF.) The American Society of Anesthesiologists expressed concern about the completeness of the review of anesthesia codes under the five-year review. Therefore, in February 2003 we asked the RUC to continue its review of anesthesia work values so that we could develop a final recommendation for a change in the anesthesia CF involving all anesthesia codes. In the proposed rule we stated we were waiting on the RUC's response to our request.

The RUC has spent a considerable amount of effort of studying this issue. The RUC's anesthesia workgroups consisted of a range of physician specialists, including various surgical specialists, who have knowledge about the anesthesia services studied. As a result of their review, the RUC approved and presented the following recommendations to CMS:

1. The RUC position is that the 5-year review has been completed.

2. The RUC anesthesia workgroup analysis only applies to the 19 anesthesia codes and associated 19 surgical codes.

3. The Workgroup recommendations to the RUC stated that there are structural differences between the anesthesia coding system and the remainder of the physician coding system, which contributes to the difficulties in making extrapolations to the entire set of anesthesia services. Among other things, the workgroups and the RUC were concerned that the anesthesia codes cover too large a number of surgical codes making it necessary to examine surgical codes within the anesthesia code, and the 19 selected anesthesia codes may not be the most representative codes.

The ASA disagrees with the RUC's recommendations and asked that we extrapolate from the 19 surveyed procedures to all anesthesia codes.

When we developed the 2002 final physician fee schedule rule on the second five-year review, one of our concerns was that the RUC's initial findings were not presented as specific recommendations. We wanted to pursue approaches consistent with RUC recommendations. Therefore, in early 2003, we asked the RUC to more clearly present their recommendations.

Based on our review of the history and analysis of this issue and the final recommendation of the RUC, we have decided not to extrapolate from the surveyed procedures to the entire universe of anesthesia procedures; we will make no further adjustments to anesthesia work under the second five-year review.

There are differences in Medicare payment policies between a teaching anesthesiologist involved with two concurrent cases with residents and a teaching CRNA involved with two concurrent cases with student nurse anesthetists.

Currently, if a teaching anesthesiologist is involved with two concurrent cases with anesthesia residents, the medical direction rules apply. Payment for the physician's medical direction is based on 50 percent of the allowance otherwise allowed if the anesthesiologist performed the anesthesia case alone.

For anesthesia services furnished prior to July 1, 2002, we allowed full payment if a non-medically directed certified registered nurse anesthetist (CRNA) supervised a single case involving a student nurse anesthetist. No payment was made if the teaching CRNA supervised two cases involving student nurse anesthetists. In August 2002, we released the Medicare Carriers Manual Transmittal 1766 relating to the involvement of a non-medically directed teaching CRNA with two student nurse anesthetists. The American Association of Nurse Anesthetists (AANA) noted that their standards for approved nurse anesthesia training programs allow the teaching CRNA to supervise two concurrent cases involving student nurse anesthetists. The new policy allows the teaching CRNA to be paid, for his/her involvement with two concurrent cases with student nurse anesthetists, but not at the full fee level. If a teaching CRNA is involved with two concurrent cases with student nurse anesthetists, payment may be based on the base unit plus the time that the teaching CRNA is present with the student nurse anesthetist. To bill the base unit, the teaching CRNA must be present with the student nurse anesthetist throughout the pre- and post-anesthesia care. This payment per case is usually higher than the 50 percent paid to the teaching anesthesiologist for medically directing resident cases.

In the proposed rule, we asked for comments on the appropriateness of applying the CRNA teaching/resident policy to teaching anesthesiologists.

Comment: The American Association of Nurse Anesthetists commented that it was unclear how the new rule for teaching anesthesiologists would operate with the medical direction rules, particularly if there were more than two concurrent anesthesia cases.

Response: The new policy for teaching anesthesiologists would apply only when there are two concurrent cases, and the cases involve residents. The medical direction payment policy would continue to apply, as it has previously, for three or four concurrent anesthesia cases regardless of the qualified individual (for example, CRNA, resident, or anesthesiologist assistant) who is administering and monitoring anesthesia under the physician's medical direction.

Comment: The ASA requested that the teaching anesthesiology payment regulations be revised so that the teaching anesthesiologists be paid in a similar manner to teaching surgeons. Under the teaching physician rules, the teaching surgeon can be paid the full fee for each of two overlapping surgeries involving residents. The ASA understands that such a proposal would require a revision to Medicare regulations and would require rulemaking.

The ASA requested that, at least, in the interim, we allow teaching anesthesiologists to be paid similarly to teaching CRNAs for two concurrent cases. However, ASA specifically requested that this policy be used in addition to the current medical direction payment policy. In other words, the ASA wants the teaching anesthesiologist to be able to choose case-by-case, whether to seek payment similar to the teaching CRNA (that is, full base units and time units based only on actual presence with the resident) or based on the medical direction rules (that is, 50 percent of the full base and time units).

According to the ASA, a number of anesthesiology department heads believe the nurse anesthesia payment rule is not appropriate to the teaching of already-licensed physicians. They question the need for the teaching physician to participate in the pre- and post-op anesthesia care (to obtain full base units), they think that participation of the teaching anesthesiologist in the key portions of the procedure is far more important than the number of minutes present with the resident (which is the relevant consideration under the teaching physician policy for a single case with a resident).

We have decided to allow teaching anesthesiologists to bill, similarly to teaching CRNAs, for their involvement in two concurrent cases involving residents. This will apply to anesthesia services furnished on or after January 1, 2004.

The anesthesiologist can bill base units and actual time, based on the amount of time the physician is present with the resident during each of two concurrent cases. To bill base units, the physician must be present with the resident during the pre- and post-anesthesia care included in the base units. If the physician is not present with the resident during the pre- and post-anesthesia care, the physician may bill the case as a medically directed case.

The anesthesiologist must document his/her involvement in cases with anesthesia residents. The documentation must be sufficient to support the payment of the fee and available for review upon request. We have revised § 414.46 to incorporate this change.

CPT Code 96155 (Health and behavior intervention, each 15 minutes,face-to-face; family (without the patient present))

This code describes a visit with a patient's family without the patient being present and was first included in the November 1, 2001 final rule. It was incorrectly listed as an active code for which payment could be made under the physician fee schedule. Our longstanding payment policy is that we do not pay for visits with family where the patient is not present. Payment for such visits is included in the pre- and post-service work of a visit where the patient is present. Consistent with this policy, this code is not payable under the physician fee schedule.

Comment: A few commenters urged us to continue to list this code as an active code under the fee schedule as they do not agree with our policy. The commenters do not agree with our assertion that payment for such visits is included in the pre- and post-service work of a visit when the patient is present and believe that not covering the service could result in diminished quality of care. One commenter disagreed that this was a technical correction since this code is currently being paid for under the fee schedule.

Response: As we indicated in the proposed rule, this was erroneously listed as an active code, contrary to longstanding Medicare policy. To be consistent with our policy, no payment may be made for this service under Medicare, and the code will be assigned a status indicator of “N”.

Section 1834(d)(3) of the Act requires that the payment amount for a screening colonoscopy be set at the level for a diagnostic colonoscopy. We have established RVUs for an incomplete diagnostic colonoscopy (CPT code 45378-53) However, an incomplete screening colonoscopy (HCPCS G0105 with modifier ‘53’ or HCPCS G0121 with modifier ‘53’) is currently carrier priced. To make payment for screening colonoscopy consistent with payment for a diagnostic colonoscopy, effective January 1, 2004, Medicare will make payment for an incomplete screening colonoscopy, HCPCS G0105 with modifier ‘53’ and HCPCS G0121 with modifier ‘53’, at the same rate as an incomplete diagnostic colonoscopy (CPT 45378-53). The Medicare carriers will no longer manually price the practitioner payment for an incomplete screening colonoscopy.

Comment: Several commenters noted that section 1871 of the Act requires a 60-day public comment period. Such period traditionally starts with the date the proposed rule is published in the Federal Register . However, for the Physician Fee Schedule Proposed rule, CMS began the start of the 60-day comment period on August 8, the date the proposal was put on display at the Federal Register , rather than August 15, the date the proposal was published in the Federal Register . The commenters request that CMS revert to the traditional start of the comment period, that is, the date of publication in the Federal Register . One commenter suggested that CMS should accept electronically submitted comments when the comment period begins earlier than the publication date.

In addition, several commenters urged CMS to resolve the process issues associated with publishing the proposed and final rule. They indicated that the delayed publication of the proposed rule, combined with missing information from addendums and impact tables, makes review and analysis problematic. The commenters also expressed concern that CMS has insufficient time to evaluate public comments and this is contrary to the spirit of the Administrative Procedures Act.

Response: CMS is keenly aware of the tight time frame between publication of the proposed and final rules. We make every effort to respond to requests from physician specialty groups and providers to include items in the proposed rule that affect payment levels, such as assigning RVUs to new CPT codes and revising RVUs for existing codes. It is difficult to both address numerous concerns and publish the proposed rule in a timely fashion. We will continue to make every effort to publish the proposed rule as early as possible. However, despite the short time frame for issuing the final rule, we take the review and analysis of comments very seriously. CMS devotes the necessary staff resources to ensure that every comment is properly considered.

Furthermore, the statute does not provide that the comment period commences with publication in the Federal Register . Section 1871(b)(1) of the Act states that before issuing a regulation in final form, “the Secretary shall provide for notice of the proposed regulation in the Federal Register and a period of not less than 60 days for public comment thereon.” While the proposed rule did not actually appear in the Federal Register until August 15, 2003, it was filed and went on public display at the Federal Register several days earlier on August 8, 2003. Accordingly, the contents of the proposed rule were, in fact, publicly available for the full 60-day comment period.

Section IV.B of this final rule describes the methodology used to review the comments received on the RVUs for physician work and the process used to establish RVUs for new and revised CPT codes. Changes to codes on the physician fee schedule reflected in Addendum B are effective for services furnished beginning January 1, 2004. The tables and discussions in this section concerning the work RVUs do not reflect the effect of the adjustment to work RVUs to match the MEI weights as discussed in section VI. The referenced work RVUs may differ from the work RVUs in Addenda B and C that reflect this adjustment.

Our December 31, 2002 final rule (67 FR 79966) announced the final work RVUs for Medicare payment for existing procedure codes under the physician fee schedule and interim RVUs for new and revised codes. The RVUs contained in the final rule applied to physician services furnished beginning March 1, 2003. We announced that we considered the RVUs for the interim codes to be subject to public comment under the annual refinement process. In this section, we summarize the refinements to the interim work RVUs published in the December 2002 final rule and our establishment of the work RVUs for new and revised codes for the 2004 physician fee schedule.

Although the RVUs in the December 2002 final rule were used to calculate 2003 payment amounts, we considered the RVUs for the new or revised codes to be interim. We accepted comments for a period of 60 days. We received substantive comments from many individual physicians and several specialty societies on approximately 10 CPT codes with interim work RVUs. Only comments on codes listed in Addendum C of the December 2002 final rule were considered.

To evaluate these comments we used a process similar to the process used in 1997. (See the October 31, 1997 final rule (62 FR 59084) for the discussion of refinement of CPT codes with interim work RVUs.) We convened a multispecialty panel of physicians to assist us in the review of the comments. The comments that we did not submit to panel review are discussed at the end of this section, as well as those that were reviewed by the panel. We invited representatives from the organization from which we received substantive comments to attend a panel for discussion of the code on which they had commented. The panel was moderated by our medical staff, and consisted of the following voting members:

• One or two clinicians representing the commenting organization.

• One primary care clinician nominated by the American College of Physicians/American Society of Internal Medicine.

• Four carrier medical directors.

• Four clinicians with practices in related specialties, who were expected to have knowledge of the service under review.

The panel discussed the work involved in the procedure under review in comparison to the work associated with other services under the physician fee schedule. We assembled a set of 300 reference services and asked the panel members to compare the clinical aspects of the work of the service a commenter believed was incorrectly valued to one or more of the reference services. In compiling the set, we attempted to include—(1) services that are commonly performed whose work RVUs are not controversial; (2) services that span the entire spectrum from the easiest to the most difficult; and (3) at least three services performed by each of the major specialties so that each specialty would be represented. The intent of the panel process was to capture each participant's independent judgment based on the discussion and his or her clinical experience. Following the discussion, each participant rated the work for the procedure. Ratings were individual and confidential, and there was no attempt to achieve consensus among the panel members.

We then analyzed the ratings based on a presumption that the interim RVUs were correct. To overcome this presumption, the inaccuracy of the interim RVUs had to be apparent to the broad range of physicians participating in each panel.

Ratings of work were analyzed for consistency among the groups represented on each panel. In addition, we used statistical tests to determine whether there was enough agreement among the groups of the panel and whether the agreed-upon RVUs were significantly different from the interim RVUs published in Addendum C of the December 2002 final rule. We did not modify the RVUs unless there was a clear indication for a change. If there was agreement across groups for change, but the groups did not agree on what the new RVUs should be, we eliminated the outlier group and looked for agreement among the remaining groups as the basis for new RVUs. We used the same methodology in analyzing the ratings that we first used in the refinement process for the 1993 physician fee schedule. The statistical tests were described in detail in the November 25, 1992 final rule (57 FR 55938).

Our decision to convene multispecialty panels of physicians and to apply the statistical tests described above was based on our need to balance the interests of those who commented on the work RVUs against the redistributive effects that would occur in other specialties.

We also received comments on RVUs that were interim for 2003, but for which we did not submit the RVUs to the panel for review for a variety of reasons. These comments and our decisions on those RVUs commented upon are discussed in further detail below.

The table below lists those interim codes reviewed under the refinement panel process described in this section. This table includes the following information:

• CPT Code. This is the CPT code for a service.

• Description. This is an abbreviated version of the narrative description of the code.

• 2003 Work RVU. The work RVUs that appeared in the December 2002 rule are shown for each reviewed code.

• Requested Work RVU. This column identifies the work RVUs requested by commenters.

• 2004 Work RVU. This column contains the final RVUs for physician work. (These work RVUs may differ from the work RVUs in Addenda B that reflect the adjustment to work RVUs to match the MEI weights.)

CPT code 17310  Chemosurgery (Mohs micrographic technique) including removal of all gross tumor, surgical excision of tissue specimens, mapping, color coding of specimens, microscopic examination of specimens by the surgeon, and complete histopathological preparation including the first routine stain (e.g. hematoxylin and eosin, toluidine blue); each additional specimen after the first 5 specimens, fixed or fresh tissue, any stage (List separately in addition to code for primary procedure).

Prior to 2003, this code was reported once for all specimens over five generated during a particular stage of Mohs surgery. Beginning in 2003, the code is used to report each specimen over five during a particular stage of Mohs surgery. The RUC recommended maintaining 0.95 work RVUs for this code as an interim value. We disagreed and assigned a work value of 0.62 work RVUs to this code pending further recommendations from the RUC. We believed this value was appropriate for the new descriptor since it allows reporting of CPT code 17310 for each specimen rather than reporting once for all specimens. It also places this code in the correct rank with the other Mohs surgery services, CPT codes 17304-17307, and with the codes for pathology consultation during surgery, CPT codes 88331 and 88332.

Commenters disagreed with the rationale we had used to arrive at the interim work value and indicated that we used inappropriate time/intensity data and failed to include surgery work, focusing only on pathology work. Commenters also stated that the intent of this code has not changed and that CMS had ignored past policy which recognizes CPT code 17310 as an add-on service and thus allows the separate billing of services for each additional specimen beyond the first five. Based on these comments, we referred this code to the multispecialty validation panel for review.

Final decision: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we have assigned 0.95 work RVUs to CPT code 17310.

CPT Code 38204  Management of recipient hematopoietic progenitor cell donor search and cell acquisition.

We disagreed with the RUC recommendation of 2.00 work RVUs for CPT code 38204. We believed we are already making payment for any physician work associated with this service as part of our payment for other bone marrow transplant codes (that is, CPT codes 38205, 38206, 38240, 38241, and 38242) and have significant concerns about how this code would be used in actual practice. Therefore, we assigned CPT code 38204 a status indicator of “B,” meaning that we will not make separate payment for this service.

Comments: Some commenters urged us to reconsider the RUC recommendation. In addition, the RUC submitted a comment disagreeing with our contention that the physician work associated with this code is included in other transplant codes. The RUC also asserted that discussions of this issue at the RUC meetings provided substantive information on how this code would be used.

Response: We continue to believe that the work of this service is contained in other transplant codes and are maintaining the status indicator of “B.” Therefore, we will not make separate payment for this service.

As explained in the December 31, 2002 final rule, the work RVUs for these codes were revised by CMS to finalize outstanding issues related to the five-year review of the gastroenterology codes. For CPT code 43219, we maintained the work RVU of 2.80. Review of information supplied by specialty societies did not provide compelling evidence that the work RVUs should be changed. Based on a review of the physician time data and a comparison to other stent placement codes, we assigned 4.35 work RVUs to CPT code 43256 and 2.94 work RVUs to CPT code 44383, in order to place these services in proper rank order to the other stent placement codes.

Comment: Some commenters felt that we improperly intervened in assigning work RVUs to these services albeit to correct rank order anomalies. Based on these comments we referred these codes to the multispecialty validation panel for review.

Response: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we are retaining work RVUs of 2.80 for CPT code 43219, 4.35 for CPT code 43256 and 2.94 for CPT code 44383.

CPT code 45335  Sigmoidoscopy, flexible; with directed submucosal injections any substance.

The RUC recommended work RVUs of 1.46 for CPT code 45335 based on a comparison to CPT code 45330, with incremental work RVUs added to reflect increased pre-, intra-, and post-service work. We disagreed with the RUC recommendation and compared this service to the analysis and recommendation provided by the RUC for CPT code 43201, which is also a new submucosal injection code. Based on the increased risk of complications (resulting in higher intra-service intensities) and the fact that several sites are being injected instead of one, we assigned a work RVU of 1.36 to CPT code 45335.

Comment: Some commenters expressed concern about the rejection of the RUC recommendation for this service and believed that we had misinterpreted the RUC findings.

Response: Upon further review and consideration of the RUC recommendation we will accept the RUC recommended work RVU of 1.46 for this service.

CPT Code 45340  Sigmoidoscopy, flexible; with dilation by balloon, each stricture.

The RUC recommended a work RVU of 1.96 for this code, which includes 1.00 RVU for the incremental work based on the need for conscious sedation to perform this procedure. (Other flexible sigmoidoscopies do not require conscious sedation.) In the December 31, 2002 rule we stated that we did not believe it is appropriate to assign a work RVU for CPT code 45340 that is based on the presumption that a portion of the work value is for the provision of conscious sedation. Rather, we compared the RUC recommendations for work and physician time for other endoscopic dilation codes to the incremental times for CPT code 45340 and assigned a work RVU of 1.66 to CPT code 45340.

Comment: Some commenters urged us to accept the RUC recommendation, noting that our characterization of RUC recommendations on conscious sedation was inaccurate. The commenters stated that the RUC has concluded that there is an increase in the amount of physician work relating to conscious sedation, but has been unable to identify a specific numerical value for that additional increment. The RUC is in the process of determining the universe of codes that include conscious sedation as an inherent part of the service provided by the operating physician to ensure these services are appropriately valued. Based on these comments we referred this code to the multispecialty validation panel for review.

Response: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we have assigned 1.89 work RVUs to CPT code 45340.

CPT Code 51798  Measurement of post-voiding residual urine and/or bladder capacity by ultrasound, nonimaging.

The RUC recommended 0.38 work RVUs based on a urology survey that reported that this procedure is performed 75 percent of the time by the physician and also based on a comparison of this procedure to CPT code 76857, Ultrasound, pelvic (nonobstetric, B-scan and/or real time with image documentation; complete. We disagreed. This code is replacing a HCPCS level two code that was assigned 0.00 work RVUs because it is typically performed by a nurse or other clinical staff. We believed that CPT code 51798 is, therefore, also a nonphysician service and assigned 0.00 work RVUs to this service.

Comment: Some commenters requested that we reconsider our decision to assign 0.00 work RVUS to this service. The commenters argued that our reason for disagreeing with the RUC recommendation is based on a stated belief that there is no physician work involved, not on actual survey data as presented by the American Urological Association (AUA) and accepted by the RUC. Commenters urged that CMS work with AUA to review this decision or include this code as part of the multi-specialty validation panel for refinement of work RVUs. Based on these comments, we referred this code to the multispecialty validation panel for review.

Response: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we will retain 0.00 work RVUs for CPT code 51798.

CPT Codes 58545-58554  Laproscopic hysterectomy/myonectomy procedures.

We accepted the RUC recommendations for work RVUs for these services.

Comment: Some commenters stated that new values have been established for these services based on new survey data and that the RUC has new recommendations for these services. In their comments on the December 31, 2002 rule, the RUC included these new work RVU recommendations and urged us to review these during the refinement process.

Response: We are in agreement with the RUC recommended values for these services. However, to provide an opportunity for public comment we are including these in the RUC Recommendations for New and Revised codes for 2004 (table xx) and will consider the RVUs interim for 2004.

CPT code 75954  Endovascular graft placement for repair of iliac artery (e.g. aneurysm, pseudoaneurysm, ateriovenous malformation, trauma) radiological supervision and interpretation.

The RUC agreed with the specialty societies and recommended a value of 2.93 work RVUs based on comparing this code to CPT codes 75952, Endovascular repair of infrarenal abdominal aortic anuerysm or dissection, radiological supervision and interpretation, (work RVU of 4.5) and 75953, Placement of proximal or distal extension prosthesis for endovascular repair of infra renal abdominal aortic aneurysm, radiological supervision and interpretation, (work RVU of 1.36). The recommended RVU was midway between the RVUs of the reference procedures. We did not agree with the RUC recommendation. Based on the specialty societies' description of the work of CPT code 75954 (which is virtually identical to the description of the work for CPT code 75953) and in order to maintain correct rank order in this family of codes, we assigned a work RVU of 1.36 to CPT code 75954.

Comment: Some commenters expressed concern about the rejection of the RUC recommendation, particularly since the recommendation was based on data presented by several specialty societies. The commenters stated that the data reflected the proper rank order of this service and indicated that physicians in those specialties that perform ileac aneurysm endorepair may be in a better position to judge the relationship of this code to other imaging services. Based on these comments, we referred this code to the multispecialty validation panel for review.

Response: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we have assigned 2.25 work RVUs to CPT code 75954.

In the December 2002 final rule, this CPT code replaced HCPCS code G0195, which had a work RVU of 1.50 in 2002. The Healthcare Professionals Advisory Committee (HCPAC) recommendation of a work RVU of 0.00 for CPT code 92610 was accepted by CMS.

Comment: Some commenters representing the long term care industry expressed concern with the reduction in work for this service. The rule provided no explanation of the HCPAC recommendation of 0.00 work RVUs for this service and the commenters requested that this issue be addressed.

Response: As requested by the commenters, a discussion of the HCPAC recommendation of 0.00 work RVUS was provided as part of the multispecialty validation panel, which was attended by the commenters.

We did not accept the RUC recommendations for work RVUs for these services (0.99 for 92613, 0.88 for 92615 and 1.10 for 92617) and assigned each of these CPT codes a work RVU of 0.00. We stated that these three services refer only to a separately identified physician review and interpretation of the fiberoptic endoscopic evaluation and that we consider this physician interpretation and report bundled into an E/M service. We stated that the physician who does not perform the testing should only bill for the patient when performing an E/M service, not as the supervisor of another professional performing and reviewing the initial fiberoptic endoscopic evaluation. The interpretation is an integral part of the testing itself and, if a nonphysician professional has the credentials and experience to perform this testing, then that professional should also provide the interpretation of the findings.

Comment: Some commenters urged us to reconsider the RVUs and payment policies related to these services and to accept the RUC recommendations for these codes. The commenters asserted that the physician's detailed frame-by-frame analysis of the video recorded procedure needed to develop the diagnosis and report following this testing is not related to an E/M service. Rather, this is similar to other services where there is a report and interpretation by the physician that is separate from an E/M service. The commenters further stated that the RUC valued each procedure code and physician interpretation and report code separately, based on the coding structure created by CPT. As a result, the interpretation and reporting is separated from each test, and the RUC recommendations do not combine the interpretation with the testing. If the code were to combine the work of interpretation and the testing then the code descriptor would need to be modified and work RVUs revalued. As a final point, commenters disputed our assertion that a nonphysician professional with the credentials and experience to perform this testing should also provide the interpretation of the findings. Based on these comments we referred this code to the multispecialty validation panel for review.

Response: As a result of the statistical analysis of the 2003 multispecialty validation panel ratings, we have assigned 0.71 work RVUs to CPT code 92613; 0.63 work RVUs to CPT code 92615; 0.79 work RVUs to CPT code 92617.

In the December 31, 2002 final rule (67 FR 79966), we also responded to the RUC recommendations on the practice expense inputs for the new and revised CPT codes for CY 2003. There were no comments received on these and therefore we are finalizing our proposals.

As we indicated in the August 15, 2003 proposed rule, RUC recommendations for RVUs for 23 new CPT codes for 2003 were received too late for incorporation in the December 31, 2002 final rule. We proposed interim RVUs for these codes and, as with all interim values, these were subject to comment. In their comments on the December 2002 final rule, the AMA-RUC requested that we consider their late recommendations for these codes during refinement. Several specialties also requested that we consider the late RUC recommendations. We had considered addressing these as part of the refinement process, but determined that we should follow the process used for all RUC recommendations and solicit public comment on the valuation of these services. Therefore, we are including the RVUs for codes listed in the table below, along with the codes that are new and revised for 2004, as interim for 2004. Following is a discussion of those codes for which did not accept the RUC recommendation.

We continue to have the same concerns as outlined in the December 31, 2002 final rule (67 FR 80007) with respect to moving these codes off of the laboratory fee schedule. We are maintaining a status indicator “I” for these services making them not valid for Medicare purposes.

CPT Codes 93784  Ambulatory blood pressure monitoring, utilizing a system such as magnetic tape and/or computer disk, for 24 hours or longer; including recording, scanning analysis, interpretation and report, 93786  Ambulatory blood pressure monitoring, utilizing a system such as magnetic tape and/or computer disk, for 24 hours or longer; recording only, 93788  Ambulatory blood pressure monitoring, utilizing a system such as magnetic tape and/or computer disk, for 24 hours or longer; scanning analysis with report, and 93790  Ambulatory blood pressure monitoring, utilizing a system such as magnetic tape and/or computer disk, for 24 hours or longer; physician review with interpretation and report.

The RUC recommendations for these codes were received too late for inclusion in the 2003 final rule. We had established the following work RVUs for these services during 2002 in response to a national coverage determination: CPT code 93784-0.17 work RVUs; 93786-0.00 work RVUs; 93790-0.17 work RVUs and had indicated that CPT code 93788 was not covered. We stated we would maintain these work RVUs until we receive a RUC recommendation.

Comment: Some commenters urged us to consider the RUC recommendations during the refinement process and also questioned the noncovered status of CPT code 93788. CPT codes 93786 and 93788 are two separate codes for the technical component and the coding format is identical to the coding used for Holter monitoring, which also has two codes for the TC of the service. Commenters also requested that CPT code 93788 be listed as a covered service.

Response: We are accepting the RUC recommendation of 0.38 work RVUs for CPT codes 93784 and 93790 and 0.00 work RVUs for CPT code 93786. We have reviewed the issue of noncoverage of CPT code 93788 and based upon the information provided by the commenters will recognize CPT code 93788 for coverage and payment under the physician fee schedule. We are also accepting the RUC recommendation of 0.00 for CPT code 93788.

We received the following comments on HCPCS codes established in the December 31, 2002 final rule.

GO262  Small intestinal imaging; intraluminal, from ligament of Treitz to the ileocecal valve, includes physician interpretation and report.

We created this code to describe a new diagnostic test for which we will make separate payment under the physician fee schedule. We assigned a work RVU of 2.12 to the code based on a comparison to the work of other diagnostic tests and procedures that require review of significant amounts of data.

Comment: Some commenters stated that that the time we used to establish the work RVU was greatly underestimated and may have been based on a misunderstanding of some of the time data contained in published literature. Based on limited survey data of physicians performing this procedure and comparison to the intensity of other services, commenters recommended a work RVU of 7.80.

Response: We are deleting HCPCS code G0262 since there is a new CPT code 91110, Gastrointestinal tract imaging, intraluminal (e.g., capsule endoscopy), esophagus through ileum, with physician interpretation and report, which will be used to report this service in 2004. We note that we accepted the RUC recommendation of 3.65 work RVUs for CPT 91110. If the commenters do not agree with the valuation of this service they may submit comments on this issue.

This code was created to allow payment to a physician who removes impacted cerumen on the same date as his or her employed audiologist performs audiologic function testing. We noted that routine removal of cerumen is not paid separately, because it is considered to be part of the procedure with which it is billed (for example, audiologic function testing). This code is to be used only in those unusual circumstances when an employed audiologist who bills under a physician uniform provider identifier number (UPIN) performs audiologic function testing on the same day as removal of impacted cerumen requiring physician expertise for removal. This code should not be used when the audiologist removes cerumen, because removal of cerumen is considered to be part of the diagnostic testing and is not paid separately.

Comment: Commenters stated that creation of this G code was problematic because there could be many other “incident to” services in which a physician performs a separate medically necessary procedure, that, if less extensive, would be considered to be included in a nonphysician provider service. The commenters suggested that a modifier could be used to describe this situation, avoiding the creation of a G code.

Response: We disagree and believe that this is a unique situation that is most appropriately handled through the use of a G code.

We created this code due to the inappropriate reporting of this service with codes for such procedures as “blood vessel repair” and “repair of arterial pseudoaneurysm”, and indicated that there would be no separate payment for this service as the work, practice expense, and malpractice risk is included in the main invasive procedure.

Comment: Commenters disagreed with the creation of this G code because it is intended to report a service that is a required component of another service and believed that the creation of this code may lead to the creation of many codes for reporting inclusive procedures separately. Some commenters suggested that the creation of parenthetical instructions in CPT to instruct that “referenced procedures ( i.e. , blood vessel repair, repair of arterial pseudoaneurysm) would not be appropriately reported in addition to the interventional vascular procedure” would address our concerns. Other commenters disagreed with our assertion that closure devices are included in the practice expense payment, as such devices are not typically used in every interventional or surgical case. Commenters suggested this code be a technical component service only and have RVUs commensurate with the cost of the device.

Response: As we indicated in the December 31, 2002, final rule, this code was created to address a specific concern about inappropriate reporting of this service using such procedures as “blood vessel repair” and “repair of arterial pseudoaneurysm.” Since this service is considered part of the main invasive procedure, to the extent this is typically part of the invasive procedure, it is accounted for under the practice expense methodology. We will continue to consider this code bundled for Medicare purposes, that is, no separate payment will be made under the physician fee schedule.

We indicated we were creating this code for use until an identical CPT code can become effective. We assigned this code a work RVU of 0.32.

Comment: Commenters disagreed with the 0.32 value assigned to this service and recommended that we replace the work RVUs with the RUC recommended work value for CPT code 43752.

Response: We are deleting HCPCS code G0272 and CPT code 43752, Naso-or oro-gastric tube placement, requiring physician's skill and fluoroscopic guidance (includes fluoroscopy, image documentation and report), will be used to report this service.

We created G0273 to describe radionuclide scanning to determine the biodistribution of Zevulin. We assigned 0.86 work RVUs to this code based on a comparison to CPT code 78802, Radiopharmaceutical localization of tumor; whole body. We established G0274 to allow appropriate reporting of this new service and assigned a work RVU of 2.07 to this code.

Comment: Commenters urged us to reevaluate the RVUs assigned to these codes and expressed concern that a lack of understanding about this service has led to its inappropriate valuation. Additionally, commenters requested that we present these codes to the AMA for consideration by the CPT Editorial Panel and RUC.

Response: We are deleting HCPCS codes G0273 and G0274. CPT codes 79403, Radiopharmaceutical therapy, radiolabeled monoclonal antibody by intravenous infusion, and 78802, Radiopharmaceutical localization of tumor or distribution of radiopharmaceutical agent(s); whole body single day imaging, will be used to report these services.

We created these add-on codes to assure proper reporting of and payment for renal and iliac angiography performed at the time of cardiac angiography. We determined the work value of 0.25 for these two add-on procedures by using the work values for CPT codes 75625, Aortography, abdominal, by serialography, radiological supervision and interpretation and 93544  Injection procedure during cardiac catherization; for aortography and adjusting for the procedure time.

Comment: Commenters suggested that, if the true intention for the creation of G0275 was to assure correct coding of selective renal angiography performed in conjunction with cardiac catheterization, the RVUs are too low and not commensurate with the work associated with selective unilateral and/or bilateral renal angiography. However, if CMS” intention for G0275 is non-selective renal angiography, then this should be stated clearly in the code descriptor. Commenters also considered the work RVUs assigned to G0278 to be too low. If G0278 is meant to be a selective procedure, then the work RVU should take into consideration the selective catheterization codes (CPT codes 36425 and 36425) and associated imaging codes (CPT codes 75710 and 75716).

Response: As announced in Program Memorandum, Transmittal AB-03-119, Change Request 2853) issued August 8, 2003, the descriptors for these two services specify that they apply to non-selective angiography and have been revised as follows:

In the December 31, 2002 final rule we incorrectly established RVUs for CPT code 0020T, Extracorporeal shock wave therapy; involving musculoskeletal system, which is an emerging technology code and also created two new HCPCS codes (G0279 and G0280) with payments based on our valuation of this CPT code. In the August 15, 2003 proposed rule we also requested additional information on these services.

Comment: Commenters on the December 2002 rule indicated that assignment of RVUs for CPT code 0020T is contrary to national policy established in the November 1, 2001 (66 FR 55269) final rule. They also indicated that the assumptions used to assign RVUs to these services were incorrect and undervalued these services.

Response: In a correction notice published May 30, 2003 (68 FR 32400) we indicated that we had incorrectly assigned RVUs to these services and they would be carrier priced.

Comment: Commenters on the December 2002 rule expressed concern that the G codes were not reflective of the changes in technology and FDA approval of ESWT. Commenters also disagreed with our categorization and portrayal of CPT 0020T as a procedure similar to other physical therapy modalities. Commenters urged us to correct and clarify that CPT 0020T is not physical therapy service but a physician procedure and thus should be removed from the list of codes identifying certain designated health services.

Response: We understand that this is a changing technology and believe the current descriptors accommodate these changes. We are removing CPT 0020T from the list of designated health services in Addendum F since we agree that, at this time, this service is predominantly performed by medical specialties such as orthopedists and podiatrists.

Comment: Commenters on the August 15, 2003 proposed rule urged us to continue to have these services priced by the carrier and expressed concern that our request for additional information indicated we would be establishing national payment amounts for these services. In addition, several physicians provided information on how this service is used in their offices, including cost information as well as a description of the procedure. Some commenters recommended that separate G codes be established to differentiate between the high and low energy levels that are currently used, as this impacts the treatment protocols as well as the resources used in these procedures.

Response: The purpose for soliciting information in the proposed rule was to gain a better understanding of the use of the various systems as well as the resources involved with this procedure. We appreciate the information the commenters provided and will continue to review this issue to determine if coding changes are warranted. We are retaining the current codes, G0279, G0280 and CPT code 0020T under the fee schedule and these will continue to be carrier priced. We believe this will enable the carriers to make appropriate payment for these services based on resources used. In addition, as previously discussed, we are removing CPT code 0020T from the list of designated health services in Addendum F.

GO288  Reconstruction, computed tomographic angiography of aorta for surgical planning for vascular surgery.

We created this code, which is a technical component code, to assure accurate reporting of this service by independent diagnostic testing facilities (IDTFs) that perform this service. This service includes receipt of a Computed Tomographic Angiogram (CTA), post-CTA processing using specialized software, and burning the 3D model onto a CD and returning it to the operating surgeon. This 3D only model is used to assist vascular surgeons in planning for, or monitoring the results of, endovascular aneurysm repair. The service is a technical service provided under the general supervision of a physician according to the supervision requirements for IDTFs.

Comment: Commenters requested clarification on whether this code could be used for the treatment planning both prior to surgery as well as for post-surgical monitoring. They also indicated that it should be expanded to include the use of enhanced computed tomography scans or magnetic resonance images and not just those generated by CTA. In addition, one commenter suggested that CMS ensure that this HCPCS code is used only for those technologies that meet the following criteria: (1) The ability to perform precise modeling of multiple clinically-relevant objects; (2) the ability to generate specific measurements essential for surgical planning and follow-up; (3) built-in quality control and self-validation capabilities; (4) FDA marketing clearance for use in surgical planning and follow-up treatment; and (5) conformance to standards adopted by the International Standards of Organization (ISO).

Commenters also suggested that the payment for this code be revised so that it is more in line with the payment for these services when administered in the outpatient setting.

Response: We agree that this service can be used for treatment planning prior to surgery as well as for post-surgical monitoring and have revised the code descriptor to clarify this point. The descriptor for this code is revised as follows:

G0288  Reconstruction, computed tomographic angiography of aorta for preoperative planning and evaluation post vascular surgery.

However, we are not expanding this service to include the use of enhanced computed tomography scans or magnetic resonance, as we have not been presented with information to support its use with these other data sources. We assume that physicians providing this service will abide by the FDA labeling requirements for the specific equipment used. Payment for services under the outpatient prospective payment system is based on a different methodology than services paid under the physician fee schedule. As required by section 1848 of the Act, payment under the physician fee schedule is based on national relative value units based on resources used in furnishing the service. We believe the RVUs established for this service are reflective of the resources used, and therefore do not believe this should be carrier priced.

GO289  Arthroscopy, knee, surgical, for removal of loose body, foreign body, debridement/shaving of articular cartilage (chrondroplasty) at the time of other surgical knee arthroscopy in a different compartment of the same knee.

We created this add-on code to permit appropriate reporting of arthroscopic procedures performed in different compartments of the same knee during the same operative session. We stated that this code should be reported only when the physician spends at least 15 minutes in the additional compartment performing the procedure. It should not be reported if the reason for performing the procedure is due to a problem caused by the arthroscopic procedure itself. We noted that this code is to be used when a procedure is performed in the lateral, medial, or patellar compartments in addition to the main procedure. We assigned a work RVU of 1.48 to this code RVUs based on a comparison to CPT codes 29874, 29877 and 29870, the base procedure for this family of codes.

Comment: Commenters appreciated our efforts to address the issue of reimbursement for this procedure. However, they expressed concern about the specific reference to a 15 minute time requirement. The commenters believed that this was inappropriate because using time in this manner rewards and encourages inefficient work and penalizes efficient physicians, which ultimately has an impact on the quality of care delivered to Medicare beneficiaries.

Response: We understand the concerns expressed by the commenters and regret any confusion that the time reference may have created. This reference to time was intended as a guideline to ensure that this add-on code is used only when the procedure performed is a substantive procedure needed to produce a significant improvement in the patient's condition. Documentation supporting this should be reflected in the operative note.

Establishment of Interim Work Relative Value Units for New and Revised Physician's Current Procedural Terminology (CPT) Codes and New Healthcare Common Procedure Coding System Codes (HCPCS) for 2004 (Includes Table titled American Medical Association Specialty Relative Value Update Committee and Health Care Professionals Advisory Committee Recommendations and CMS's Decisions for New and Revised 2004 CPT Codes)

One aspect of establishing RVUs for 2004 was related to the assignment of interim work RVUs for all new and revised CPT codes. As described in our November 25, 1992 notice on the 1993 physician fee schedule (57 FR 55983) and in section III.B. of the November 22, 1996 final rule (61 FR 59505 through 59506), we established a process, based on recommendations received from the AMA's RUC, for establishing interim work RVUs for new and revised codes.

This year we received work RVU recommendations for approximately 132 new and revised CPT codes from the RUC. Our staff and medical officers reviewed the RUC recommendations by comparing them to our reference set or to other comparable services for which work RVUs had previously been established, or to both of these criteria. We also considered the relationships among the new and revised codes for which we received RUC recommendations. We agreed with the majority of the relative relationships reflected in the RUC values. In some instances, when we agreed with the relationships, we nonetheless revised the work RVUs to achieve work neutrality within families of codes. That is, the work RVUs have been adjusted so that the sum of the new or revised work RVUs (weighted by projected frequency of use) for a family will be the same as the sum of the current work RVUs (weighted by projected frequency of use). We reviewed all the RUC recommendations. We accepted approximately 95 percent and we disagreed with approximately 5 percent of the RUC recommended values. In the majority of these instances, we agreed with the relativity established by the RUC, but needed to adjust work RVUs to retain budget neutrality.

We received 2 recommendations from the HCPAC. We agreed with both of the HCPAC recommendations.

Table 5, titled “AMA RUC and HCPAC Recommendations and CMS Decisions for New and Revised 2004 CPT Codes”, lists the new or revised CPT codes, and their associated work RVUs, that will be interim in 2004. This table includes the following information:

• A “#” identifies a new code for 2004.

• CPT code. This is the CPT code for a service.

• Modifier. A “26” in this column indicates that the work RVUs are for the professional component of the code.

• Description. This is an abbreviated version of the narrative description of the code.

• RUC recommendations. This column identifies the work RVUs recommended by the RUC.

• HCPAC recommendations. This column identifies the work RVUs recommended by the HCPAC.

• CMS decision. This column indicates whether we agreed with the RUC recommendation (“agree”) or we disagreed with the RUC recommendation (“disagree”). Codes for which we did not accept the RUC recommendation are discussed in greater detail following this table. An “(a)” indicates that no RUC recommendation was provided.

• 2004 Work RVUs. This column establishes the 2004 work RVUs for physician work. These work RVUs may differ from the work RVUs in Addenda B and C that reflect the adjustments to work RVUs to match the MEI weights.

Table 6, which is titled “AMA RUC ANESTHESIA RECOMMENDATIONS AND CMS DECISIONS FOR NEW AND REVISED 2004 CPT CODES”, lists the new or revised CPT codes for anesthesia and their base units that will be interim in 2004. This table includes the following information:

• CPT code. This is the CPT code for a service.

• Description. This is an abbreviated version of the narrative description of the code.

• RUC recommendations. This column identifies the base units recommended by the RUC.

• CMS decision. This column indicates whether we agreed with the RUC recommendation (“agree”) or we disagreed with the RUC recommendation (“disagree”). Codes for which we did not accept the RUC recommendation are discussed in greater detail following this table.

• 2004 Base Units. This column establishes the 2004 base units for these services.

The following is a summary of our rationale for not accepting particular RUC work RVU or base unit recommendations. It is arranged by type of service in CPT order. Additionally, we also discuss those CPT codes for which we received no RUC recommendations for physician work RVUs. This summary refers only to work RVUs or base units.

CPT code 43752  Naso- or oro-gastric tube placement, requiring physician's skill and fluoroscopic guidance (includes fluoroscopy, image documentation and report)

The RUC recommended a work RVU of 0.82 for this service based on a comparison of this procedure to CPT code 44500. While we agree that CPT code 43752 is similar in work intensity to CPT code 44500, we feel the intra-service time is more appropriately valued at the 25th percentile (15 minutes of intra-service time vs. 20 minutes of intra-service time). This reduces the total time associated with CPT code 43752 from 30 minutes to 25 minutes. We applied the ratio of the RUC recommended value of 0.82 work RVU over 30 minutes to the revised intra-service time of 25 minutes to assign 0.68 interim work RVUs for CPT code 43752.

CPT code 63103  Vertebral corpectomy (vertebral body resection), partial or complete, lateral extracavitary approach with decompression of spinal cord and/or nerve root(s) (e.g., for tumor or retropulsed bone fragments); thoracic or lumbar, each additional segment. (List separately in addition to code for primary procedure)

The RUC recommended a work RVU of 5.00 for this service based on a comparison of this procedure to CPT code 63088. It was unclear from the clinical vignettes supplied by the specialty society whether the additional corpectomy would more commonly involve the lumbar or the thoracic region of the spine. There is a significant difference in work intensity associated with the resection of an additional corpus in the thoracic region as opposed to the lumbar region. For this reason we applied the ratio of the reference service (CPT code 63088) to its primary service (CPT code 63087) to CPT code 63101 (primary service associated with CPT 63103) to assign 3.90 interim work RVUs for CPT code 63103.

CPT code 61863  Twist drill, burr hole, craniotomy, or craniectomy with stereotactic implantation of neurostimulator electrode array in subcortical site (e.g., thalamus, globus pallidus, subthalamic nucleus, periventricular, periaqueductal gray), without use of intraoperative microelectrode recording; first array and CPT code 61867 Twist drill, burr hole, craniotomy, or craniectomy with stereotactic implantation of neurostimulator electrode array in subcortical site (e.g., thalamus, globus pallidus, subthalamic nucleus, periventricular, periaqueductal gray), with use of intraoperative microelectrode recording; first array

The RUC recommended a work RVU of 19.00 for CPT code 61863 and 31.34 work RVUs for CPT code 61867. These two new CPT codes replace existing CPT code 61862 (work RVU=19.34). Although we agree with the relative relationship established by the RUC for these services, in order to retain budget neutrality, we adjusted the RUC recommended values. Thus, the recommended values were adjusted in order that the total relative values remain constant before and after the inclusion of the new CPT codes.

We assigned 13.92 work RVUs to CPT code 61863 and 22.96 work RVUs to CPT code 61867.

CPT code 38208  Transplant preparation of hematopoietic progenitor cells; thawing of previously frozen harvest, without washing and CPT code 38209 Transplant preparation of hematopoietic progenitor cells; thawing of previously frozen harvest, with washing

We continue to have the same concerns as outlined in the December 31, 2002 final rule (67 FR 80007) with respect to moving these codes from the laboratory fee schedule and thus establishing relative values under the physician fee schedule. We are maintaining a status indicator “I” for these services, making them not valid for payment under the Medicare Physician Fee Schedule.

CPT code 96111  Developmental testing extended (includes assessment of motor, language, social, adaptive, and/or cognitive functioning by standardized developmental instruments, eg Bayley Scales of Infant Development) with interpretation and report, per hour

Although we agree with the RUC recommended work RVU of 2.60 for CPT code 96111, we note that the tests under this code will no longer be paid on a per hour basis. That is, total payment for the services under CPT code 96111 is based on one hour of provision of the tests. It is our understanding that these tests can be completed typically in one hour. That is, some of the tests can be administered in less than one hour and some may require a little more than one hour, so that the average time for all of the tests works out to be one hour. Therefore, regardless of the total number of hours it takes to complete the services under CPT code 96111 or whether the services are split up and spread over a number of days, payment will be made for 96111 based on only one unit/hour at 2.6 RVUs.

We have developed a process for establishing interim practice expense RVUs for new and revised codes that is similar to that used for work RVUs. Under this process, the RUC recommends the practice expense direct inputs, that is, the staff time, supplies and equipment, associated with each new code. We then review the recommendations in a manner similar to our evaluation of the recommended work RVUs.

The RUC recommendations on the practice expense inputs for the new and revised 2004 codes were submitted to us as interim recommendations. We, therefore, consider that these recommendations are still subject to further refinement by the PEAC, or by us, if it is determined that such future review is needed. We may also revisit these inputs in light of future decisions of the PEAC regarding supply and equipment packages and standardized approaches to pre- and post-service clinical staff times.

We have accepted, in the interim, almost all of the practice expense recommendations submitted by the RUC for the codes listed in the following table titled “AMA RUC and HCPAC RVU Recommendations and CMS Decisions for New and Revised 2004 CPT Codes.”

We made the following minor changes to the inputs where relevant:

• We deleted the 3-minute phone calls in the post service period to conform to our established standard for all codes with 10 and 90-day global periods.

• We also deleted equipment when individual items did not meet the minimum $500 requirement.

• We deleted certain equipment items that represent indirect, rather than direct costs, including lead shielding, lead lined radioactive waste box and lead-lined sharps box.

• We deleted the L-Block table shield because it is included in the price and description of the dose calibrator, another CPEP equipment item.

• We made minor changes to clinical labor and supplies, for several central venous access (CVA) codes in order to bring uniformity to this new family of codes.

• We assigned, on an interim basis, the clinical labor RN designation for CPT code 95991, physician administered refilling and maintenance of spinal or brain implantable pump, until the PEAC has an opportunity to review the necessity for this clinical assignment.

On January 4, 2001 we published in the Federal Register a final rule with comment period, “Medicare and Medicaid Programs; Physicians Referrals to Health Care Entities With Which They Have Financial Relationships” (66 FR 856). That final rule incorporated into regulations the provisions in paragraphs (a), (b) and (h) of section 1877 of the Act. Section 1877 of the Act prohibits a physician from referring a Medicare beneficiary for certain “designated health services” to a health care entity with which the physician (or a member of the physician's immediate family) has a financial relationship, unless an exception applies. In the final rule, we published an attachment listing all of the CPT and HCPCS codes that defined the entire scope of the following designated health services for purposes of section 1877 of the Act: clinical laboratory services; physical therapy services (including speech-language pathology services); occupational therapy services; radiology and certain other imaging services; and radiation therapy services and supplies.

In the January 2001 final rule, we stated that we would update the list of codes used to define these designated health services (the “Code List”) in an addendum to the annual physician fee schedule final rule. The purpose of the update is to conform the Code List to the most recent publications of CPT and HCPCS codes. The last update of the Code List was included in the December 31, 2002 physician fee schedule final rule in Addendum E and was subsequently corrected in a notice that was published in the Federal Register (68 FR 32400) on May 30, 2003.

The updated all-inclusive Code List effective January 1, 2004 is presented in Addendum F in this final rule. We intend to publish annually the all-inclusive Code List in an addendum to the physician fee schedule final rule. The updated all-inclusive Code List will also be available on our Web site at http://www.cms.hhs.gov/medlearn/refphys.asp.

We received public comments on three issues relating to the most recent Code List. The comments and our responses are stated below.

Comment: One commenter noted that we added three new “Q” codes (Q3021, Q3022, and Q3023) for hepatitis B vaccines. Program Memorandum AB-02-185 issued on December 31, 2002 deleted these HCPCS codes. However, the Program Memorandum also reactivated the following CPT codes for hepatitis B vaccine: 90740, 90743, 90744, 90746 and 90747.

Response: The commenter is correct. We erred in adding the “Q” codes to the list of services that may qualify for an exception under 42 CFR 411.355(h) concerning exceptions for preventive screening tests, immunizations, and vaccines. This was corrected in the correction notice published on May 30, 2003 (68 FR 32400).

Comment: Some commenters objected to the addition of CPT code 0020T (Extracorporeal shock wave therapy; involving plantar fascia) to the list of physical therapy services for purposes of the physician self-referral prohibition. The commenters stated that CPT 0020T is currently a physician service involving anesthesia and therefore, should not be characterized as a physical therapy service.

Response: We agree with the commenters and have removed CPT code 0020T from the list of designated health services. Further discussion of this comment and response is included in section IV.C.2 of this preamble concerning the HCPCS codes G0279 and G0280 relating to extracorporeal shock wave therapy.

Comment: One commenter noted that the annual Code List update does not include codes for the following designated health services: Durable medical equipment and supplies; parenteral and enteral nutrients, equipment and supplies; prosthetics, orthotics and prothestic devices and supplies; home health services; outpatient prescription drugs; and inpatient and outpatient hospital services. The commenter recommended that we include the CPT and HCPCS codes for these designated health services in the annual update and in the quarterly updated Microsoft Excel spreadsheet of RVU values, global periods and supervision levels for Medicare covered-services posted on the CMS Web site. Alternatively, the commenter requested that we clarify that the Code List is not exhaustive and indicate where providers can obtain more information on the remaining categories.

Response: As explained in the January 4, 2001 final rule with comment (66 FR 923), we believe that the regulatory definitions of the designated health services at issue are sufficiently clear to permit entities and physicians to identify them readily. Moreover, some of these designated health services are not amenable to definition solely through codes. Regardless, to define these services through codes or to change the frequency of the Code List update would require a change in the text of the regulatory definitions for the various designated health services found in § 411.351. The purpose of this Code List is simply to make those ministerial changes necessary to conform the Code List to the current CPT and HCPCS code publications. Making substantive changes to the regulatory definitions is beyond the scope of this update and cannot be accomplished without first proposing the changes in a Notice of Proposed Rulemaking. Lastly, we cannot accept the commenter's suggestion that we explain that the Code List is not exhaustive because such a statement is false. The Code List is exhaustive with respect to the specific designated health services that it defines, and for the reasons noted above, we are not defining the remaining designated health services through codes.

Tables 7 and 8, below, identify the additions and deletions, respectively, to the comprehensive Code List last published in Addendum E of the December 2002 physician fee schedule final rule and subsequently corrected in the May 30, 2003 correction notice (68 FR 32400). Tables 7 and 8 also identify the additions and deletions to the lists of codes used to identify the items and services that may qualify for the exceptions in § 411.355(g) (regarding EPO and other dialysis-related outpatient prescription drugs furnished in or by an end-stage renal dialysis (ESRD) facility) and in § 411.355(h) (regarding preventive screening tests, immunizations and vaccines).

We will consider comments with respect to the codes listed in Tables 8 and 9 below, if we receive them by the date specified in the DATES section of this final rule.

The additions specified in Table 8 generally reflect new CPT and HCPCS codes that become effective January 1, 2004 or that became effective since our last update. It also reflects the addition of codes recently recognized by Medicare for payment purposes.

Additionally, we are adding two G-codes (G0173, “Stereo radiosurgery, complete” and G0251, “Linear acc based stero radio”) to the category of radiation therapy services and supplies. These codes became effective for Medicare payment purposes in August 2000 and July 2002, respectively and should have been reflected in previous Code Lists.

Table 8 also reflects the addition of 2 new HCPCS codes (Q4054 and Q4055) to the list of dialysis-related outpatient prescription drugs that may qualify for the exception described in § 411.355(g) regarding those items. The physician self-referral prohibition will not apply to these drugs if they meet the conditions set forth in § 411.355(g). Table X also reflects the addition of a screening mammography code (CPT 76083) and a flu vaccine code (CPT 90655) to the list that identifies preventive screening tests, immunizations and vaccines that may qualify for the exception described in § 411.355(h) for such items and services. The physician self-referral prohibition will not apply to these services if they meet the conditions set forth in § 411.355(h) concerning the exception for preventive screening tests, immunizations, and vaccines.

Table 8 reflects the deletions necessary to conform the Code List to the most recent publications of CPT and HCPCS codes, as well as additional deletions that we have determined are necessary as described below.

Under the category of physical therapy, occupational therapy and speech-language pathology services, we are removing CPT code 0020T, extracorporeal shock wave therapy for plantar fascia consistent with the response to the comment discussed in section IV.C.2 and VI.B of this preamble.

Under the category of radiology and certain other imaging services, we are deleting CPT code 76831 for an echo exam of the uterus. This code should never have appeared on the Code List. Our definition of “radiology and certain other imaging services” at § 411.351 specifically excludes any x-ray, fluoroscopy or ultrasonic procedure that requires “the insertion of a needle, catheter, tube, or probe”. The type of procedure described by CPT code 76831 involves infusion tubing and should be removed from the Code List.

Under the category of radiation therapy services and supplies, we are removing HCPCS code G0274 for radiopharmaceutical therapy for non-Hodgkin's lymphoma because it is a nuclear medicine service. Our definition of “radiation therapy services and supplies” at § 411.351 specifically excludes nuclear medicine procedures. Thus, HCPCS code G0274 should never have appeared on the Code List.

The physician fee schedule update is determined using a formula specified by statute. Under section 1848(d)(4) of the Act, the update is equal to the product of 1 plus the percentage increase in the Medicare Economic Index (MEI) (divided by 100) and 1 plus the update adjustment factor (UAF). For CY 2004, the MEI is equal to 2.9 percent (1.029). The UAF is −7.0 percent (0.930). Section 1848(d)(4)(F) of the Act requires an additional −0.2 percent (0.998) reduction to the update for 2004. Thus, the product of the MEI (1.029), the UAF (0.930), and the statutory adjustment factor (0.998) equals the CY 2004 update of −4.5 percent (0.9551).

The negative physician fee schedule update occurs under a mandatory statutory formula. The law gives us no alternative to reducing the physician fee schedule rates. Only Congress can change the law and avert a reduction in 2004 physician fee schedule rates. Without a congressional act to change the law, the Department is compelled to announce a physician fee schedule update for CY 2004 of −4.5 percent. The Department's calculations are explained below.

The Medicare Economic Index (MEI) is required by section 1842(b)(3) of the Act, which states that prevailing charge levels beginning after June 30, 1973 may not exceed the level from the previous year except to the extent that the Secretary finds, on the basis of appropriate economic index data, that a higher level is justified by year-to-year economic changes.

Beginning July 1, 1975, and continuing through today, the MEI has met this requirement by reflecting the weighted sum of the annual price changes of the inputs used to produce physicians' services. As such, the MEI attempts to be an equitable measure of price changes associated with physician time and operating expenses.

The current form of the MEI was detailed in the November 25, 1992 Federal Register (57 FR 55896) and was based in part on the recommendations of a Congressionally-mandated meeting of experts held in March 1987. Since that time, the structure of the MEI has remained essentially unchanged, with two exceptions. First, the MEI was rebased in 1998 (63 FR 58845), which moved the cost structure of the index from 1992 data to 1996 data. Second, the methodology for adjusting for productivity was revised in 2002 (67 FR 80019) to reflect the percentage change in the 10-year moving average of economy-wide multifactor productivity.

We are rebasing and revising the MEI for the 2004 physician fee schedule update. The terms “rebasing” and “revising”, while often used interchangeably, actually denote different activities. Rebasing means moving the base year for the structure of costs of an input price index, while revising means changing data sources, cost categories, or price proxies used in the input price index. As is always the case with a rebasing and revising exercise, we have attempted to use the most recently available, relevant, and appropriate information to develop the MEI cost category weights and price proxies. We detail below the updated cost weights for the MEI expense categories, our rationale for selecting the price proxies in the MEI, and the results of rebasing and revising the MEI.

The MEI was last rebased and revised in 1998 for the 1999 physician fee schedule update (63 FR 58845). The base year for that version of the MEI was 1996, which means that the cost weights in the index reflect physicians' expenses in 1996. However, we believe it is desirable to periodically rebase and revise the index so that the expense shares and price proxies reflect more current conditions. For this reason, we are rebasing the MEI to reflect physicians' expenses in 2000. In addition, we are revising the cost categories in the MEI and changing three of the proxies we currently use to ensure that the index is appropriately reflecting price changes. We will continue to adjust the MEI using economy-wide multifactor productivity.

The expense categories in the rebased and revised MEI were primarily derived from the 2003 AMA Physician Socioeconomic Characteristics publication (2003 Patient Care Physician Survey data), which measures physicians' earnings and overall practice expenses for 2000. The AMA data were used to determine expenditure weights for total expenses, physicians' earnings, and malpractice expenses, the only information detailed in this survey. To further disaggregate the weights into subcategories reflecting more detailed expenses, we used data from previous AMA surveys, the 1997 Bureau of Economic Analysis Benchmark Input-Output table (I/O), the 2003 Bureau of Labor Statistics (BLS) Employment Cost Index (ECI), and the 2002 Bureau of the Census Current Population Survey (CPS).

Developing a rebased and revised MEI requires selecting a base year and determining the number and composition of expense categories and their associated price proxies. We are rebasing the MEI to CY 2000. CY 2000 was chosen as the base year for two main reasons: (1) CY 2000 was the most recent year for which data were available from the AMA, and (2) we believed that the CY 2000 data were representative of the changing distribution of physicians' earnings and practice expenses over time.

Comment: One commenter suggested that we update the weights in the MEI to a more recent base year, possibly CY 2004. While the commenter agreed with us that there is a lack of data to do so, the commenter suggested using the price change in each of the proxies to estimate weights for 2004 as an alternative to 2000 data.

Response: We selected CY 2000 as the base year for two reasons: (1) CY 2000 data were the most recent data available from the AMA, and (2) we felt the CY 2000 data were representative of the changing distribution of physician earnings and practice expenses over time. We do not expect that the experience of the past 3 or 4 years would have a significant impact on the MEI for the CY 2004 update, particularly since changing the weights from 1996 to 2000 had such a minimal effect. In addition, the price proxies that we use capture the current price changes in each of the categories that make up the MEI.

While we agree that it would be optimal to develop MEI weights based on more recent data, we recognize the lack of data to do so. We also recognize that an alternative would be to use price changes in each of the proxies to update the weights to a more recent base year, similar to the methodology we used to develop the distribution of detailed practice expense categories in the current structure. In that case, we used price changes from 1998 to 2000 to develop weights for 2000.

However, as we indicated in the proposed rule, this method has a major drawback in that it assumes that the quantity of inputs would increase at the same rate as the price of those inputs. This may not be the case over longer time periods (for instance, 2000-2004) where there is likely to be substitution away from more costly inputs toward those which are less costly. Our experience with rebasing indexes has also shown that the weights for major categories do not change very much over time, even though the individual price changes for those categories can differ significantly. In addition, because the MEI is a Laspeyres-type index, the price changes between the base period and the current period are reflected in the relative importance of each category in determining the overall increase. Therefore, we feel that basing the index on CY 2000 data and reflecting current price changes likely represents a reasonable estimate of physicians' current experience.

We determined the number and composition of expense categories based on the criteria used to develop the current MEI and other CMS input price index expenditure weights. These criteria are timeliness, reliability, relevance, and public availability. For more information on these criteria, see the May 9, 2002 Federal Register (67 FR 31444) and the detail later in this preamble. Table 10 lists the set of mutually exclusive and exhaustive cost categories that make up the rebased and revised MEI.

To determine the expenditure weights for the rebased and revised MEI, we used currently available and statistically valid data sources on physician earnings and practice expenses. While we consulted numerous data sources, we used five data sources to determine the MEI expenditure weights: (1) The 2003 AMA Physician Socioeconomic Statistics (2000 survey data) for self-employed physicians, (2) the 2000-2002 AMA Physician Socioeconomic Statistics (1998 data) for self-employed physicians, (3) the March 2003 BLS Employment Cost Index, (4) the 2002 Bureau of the Census Current Population Survey, and (5) the Bureau of Economic Analysis (BEA) 1997 Benchmark Input-Output tables (I/O). No one data source provided all of the information needed to determine expenditure weights according to our criteria. The development of each of the cost categories using these sources is described in detail below.

The rebased and revised MEI uses AMA data on mean physician net income (physician earnings) for self-employed physicians to develop a weight for physician earnings. The weight for this expense category is based on AMA data for 2000 and is calculated as a percentage of total mean expenses (physician earnings and practice expenses, including malpractice). The physician earnings expenditure category also includes employee physician compensation. Currently, physician earnings and overhead expenses generated by employee physicians are included in the AMA practice expenses category. However, we believe it is appropriate, for our purposes, to place employee physician compensation in the MEI cost category of physician earnings. Including employee physician payroll in physician earnings in the MEI is consistent with the current payment methodologies in accordance with the physician fee schedule, where the work RVU is computed based on what service is provided and not on who provides the service. Since employee physicians perform the same services as self-employed physicians, employee physician time is reflected in the work RVU. By including the compensation of employee physicians in the physician earnings expense category, these expenses will be adjusted by the appropriate price proxies for time spent by a physician.

To obtain further detail for both wages/salaries and benefits, the ratio between these categories for 1996 (based on the 1996-based MEI) was updated to 2000 using the growth in the overall Employment Cost Index for private employees for wages/salaries and benefits. Alternative data for determining this split were not readily available from any other source. The main shortcoming of this method is that any changes in quantity and intensity (mix of physicians) are not reflected. However, faced with the lack of alternative data, we deemed this approach to be the most feasible, and the results appear to be consistent with anecdotal evidence on this ratio. Its application resulted in a wage-fringe benefit split of 81.4 and 18.6 percent, respectively, in the revised and rebased MEI compared with a wage-fringe benefit split of 81.2 and 18.8 percent, respectively, in the 1996-based MEI.

To determine the remaining individual practice expense weights other than malpractice expense, we updated AMA expense data from 1998 to 2000 using the relative price change in an appropriate price index. After the levels were updated to 2000 values, it was necessary to normalize these levels to equal the 2000 mean total expense data provided by the 2003 AMA survey. The detailed explanations for the derivation of the individual weights are listed below.

The cost share for nonphysician employee compensation was developed by updating the 1998 AMA Socioeconomic Survey data on nonphysician employee compensation costs for self-employed physicians to 2000, using the current proxy for this category, and dividing the resulting amount into total expenses (physician earnings plus practice expenses) for 2000 from the AMA survey. We further divided this cost share into wages/salaries and benefits using BLS Employment Cost Index data. The ECI survey contains data on the proportion of total compensation accounted for by wages/salaries and benefits (including paid leave) by private industry health services occupational category. These proportions can be used to distribute the total nonphysician employee compensation weight to wages/salaries and benefits for non-physician employees. We used 2000 data from the March 2003 publication. Although this survey does not contain data specifically for offices of physicians, data are available on wage/fringe shares for private industry health services, which include hospitals, nursing homes, offices of physicians, and offices of dentists. We believe the data for health services from the survey do provide a reasonable estimate of the split between wages and fringe benefits for employees in physicians' offices. Data for 2000 in the ECI survey for total health services indicate that wages and fringe benefits are 74.02 percent and 25.98 percent of compensation, respectively. As in the 1996-based MEI, we will use CPS data on earnings by occupation to develop cost shares for wages for nonphysician occupational groups shown in Table 6. To arrive at a distribution for these separate categories, we multiplied the overall share for nonphysician employee wages/salaries by each of the occupational proportions from the 2000 CPS. This distribution for the 1996-based and 2000-based MEI are presented in Table 10.

The weight for professional liability expense was derived from the 2003 AMA survey (2000 data) and was calculated as the mean professional liability expense expressed as a percentage of total expenses (physician earnings plus practice expenses). This calculation resulted in a 3.865 percent share of total costs in 2000 compared to a 3.152 percent share in the 1996-based index. The increase in weight for professional liability insurance represents the increases in both premiums and the amount of coverage purchased by physicians in 2000 compared to 1996. While the weight does not reflect the cost experience for 2001 and 2002, the proxy used in the rebased and revised index does reflect the price increases associated with the recent rise in malpractice costs.

Comment : Some commenters were concerned that the rebased and revised MEI does not appropriately reflect the recent increase in professional liability insurance (PLI) premiums that physicians are experiencing.

Response : As we indicated in the proposed rule, the weights in the rebased and revised MEI reflect the distribution of physicians' costs in CY 2000 and do not reflect the more recent experience of physicians, particularly as it pertains to PLI. While it would be optimal to base the weights on more recent data, there is not a more recent, comprehensive measure that would meet our criteria for determining weights in the MEI.

We also indicated that while the weights do not reflect the more recent experience, the proxy we use to measure the price change in this category does reflect more recent price changes in premiums and is the most current data available through the second quarter of 2003. This MEI PLI data, like that used in the development of the GPCIs, does not reflect total expenditures on PLI, which would be needed to develop more current weights for the MEI. In order to develop cost weights, expenditure data for all costs facing physicians are needed.

The 2003 AMA survey provides less detail for expenses with respect to prior years' publications. Therefore, we calculated the share of each of the above categories by updating the AMA data for 1998 to 2000 using an appropriate price proxy. The primary reason for using the price proxy was that we lacked other data to develop cost weights for each of these categories. As stated previously, the main deficiency of this method is that it does not directly account for changes in the quantity or intensity associated with these expenses. Our belief, however, was that it was important to continue using these detailed breakouts so that each would be proxied by an appropriate price index and that the quantity/intensity effects over a short period of time are not likely to be large. In fact, we have found that even over longer periods of time, the distribution of costs tends to be relatively similar.

Office expenses and medical equipment levels were moved to 2000 using the growth from 1998 to 2000 in their respective MEI price proxies. In the case of office expenses, we used the growth in the CPI-U Housing; for medical equipment expenses, we used the growth in the PPI for Medical Instruments and Equipment.

The share for pharmaceuticals (prescription drugs) and medical materials and supplies was calculated by separating out pharmaceuticals and other medical materials and supplies using 1997 BEA Benchmark Input-Output data. First, the sum of all the pharmaceuticals and medical supplies categories from the Benchmark Input-Output tables for 1997 was calculated. The share of pharmaceuticals and medical supplies was then calculated as a percentage of this total and applied to the 1997 AMA medical supplies data. These calculated levels were then aged to 2000 using the growth in an appropriate price proxy. We thought it was important and appropriate to account for each of these categories separately so that differences in relative price growth between pharmaceuticals (prescription drugs) and other medical materials and supplies would be more accurately represented. The resulting 2000 data for the two separate categories were then aggregated (summed) together to form the overall total for the share for the pharmaceuticals and medical materials and supplies category in the rebased and revised MEI. The pharmaceuticals category was aged using the Producer Price Index (PPI) for Pharmaceutical preparations and the medical materials and supplies category was updated using the PPI for surgical appliances and supplies.

Finally, the Other Expenses category was calculated as a residual (total expenses less the percentage of all categories currently accounted for). The additional detail for transportation expenses found in the 1996-based MEI was removed because the data were not readily available for measurement of a cost share for 2000. The effect on the MEI of removing the detail is negligible.

Comment: One commenter suggested for the purposes of future changes to the MEI, that CMS consider inputs that are vastly different than when the MEI was first developed, such as costs of complying with government regulatory requirements and interpreter services for patients.

Response: We thoroughly research many of the known data sources on a regular basis to determine the appropriate number of detailed categories that make up the MEI. If we determine that a different combination of inputs is needed we will revise the MEI to reflect a more current cost distribution. However, CMS does not have the detailed expenditure and price data for the types of expenditures the commenter indicated. CMS will continue to work with other outside entities in the future to ensure the MEI is as accurate and representative as possible. It should also be noted that these costs are already captured in the MEI, as all costs are captured in the index, just not separately broken out for the reasons previously stated.

After the 2000 cost weights for the rebased and revised MEI were developed, we reviewed the current set of price proxies to determine whether they were still the most appropriate for each expenditure category. As was the case in the development of the 1996-based MEI (57 FR 55901), most of the indicators we considered are based on BLS data and are grouped into one of the following five categories:

Producer price indices (PPIs) measure price changes for goods sold in other than retail markets. They are the preferred proxies for physician purchases at the wholesale level. These fixed-weight indices are a measure of price change at the producer or at the intermediate stage of production, a more likely mode of purchase for physicians.

Consumer price indices (CPIs) measure change in the prices of final goods and services purchased by consumers. Like the PPIs, they are fixed-weight. Since they may not represent the price changes faced by producers, CPIs were used if there were no appropriate PPI or if the expenditure category was similar to expenditure of retail consumers in general.

Average hourly earnings (AHEs) are available for production and nonsupervisory workers for specific industries as well as for the nonfarm business economy. They are calculated by dividing gross payrolls for wages/salaries by total hours. The series reflects shifts in employment mix and, thus, is representative of actual changes in hourly earnings for industries or for the nonfarm business economy.

These ECIs measure the rate of change in employee wage rates per hour worked. These fixed-weight indices are not affected by shifts in industry or occupation employment levels and measure only the pure rate of change in wages.

These ECIs measure the rate of change in employer costs of employee benefits, such as the employer's share of Social Security taxes, pension and other retirement plans, insurance benefits (life, health, disability, and accident), and paid leave. Like ECIs for wages/salaries, the ECIs for employee benefits are not affected by changes in industry output or occupational shifts.

When choosing wage and price proxies for each expense category, we evaluate the strengths and weaknesses of each proxy variable using four criteria. The first criterion is relevance. The price variable should appropriately represent price changes for specific goods or services within the expense category. Relevance may encompass judgments about relative efficiency of the market generating the price and wage increases.

The second criterion is reliability or low sampling variability. If the proxy wage-price variable has a high sampling variability or inexplicable erratic patterns over time, its value is greatly diminished, since it is unlikely to accurately reflect price changes in its associated expenditure category. Low sampling variability can conflict with relevance, since the more specifically a price variable is defined in terms of service, commodity, or geographic area, the higher the possibility of sampling variability.

The third criterion is timeliness of actual published data. For this reason, we prefer monthly and quarterly data to annual data. The length of time the time series data have been published is also important. A well-established time series is needed to assess the reasonableness of the series and to provide a solid base from which to forecast future price changes in the series. We need to forecast the MEI to make Federal budget and Trustees Report estimates.

The fourth criterion is public availability. We prefer to use data sources that are publicly available for our indices so that the public may track each of the individual components in the MEI.

The BLS price proxy categories previously described meet the criteria of relevance, reliability, timeliness, and public availability. Below we discuss the price-wage proxies for the rebased and revised MEI (shown in Table 5).

In the rebased and revised MEI, we are using the AHE for the private nonfarm economy as the proxy for the physician wages/salaries component; this is the same price measure used in the 1996-based MEI. In our judgment, this proxy still most closely comports with Congressional intent as expressed in the Senate Finance Committee's 1972 report (S. Rept. No. 92-1230 at 191 (1972)). It should be noted that AHEs change in accordance with changes in the type and mix of workers.

As we discussed extensively in the November 2, 1998 final rule (63 FR 58848) and again in the December 31, 2002 final rule (67 FR 80019), we believe that the current price proxy for physicians' earnings—AHE in the nonfarm business economy—is the most appropriate proxy to use in the MEI. The AHE for the nonfarm business economy reflects the impacts of supply, demand, and economy-wide productivity for the average worker in the economy. Using this measure as the proxy for physicians' earnings ensures parity in the rate of change in wages for the average worker and those for physicians. In addition, use of this proxy is consistent with the original legislative intent that the change in the physicians' earnings portion of the MEI parallel the change in general earnings for the economy. Since earnings are expressed per hour, a constant quantity of labor input per unit of time is reflected. The use of the AHE data is also consistent with our using the BLS economy-wide multifactor productivity measures since economy-wide wage increases reflect economy-wide productivity increases.

Using the ECI for professional and technical workers or other occupational-specific wage proxies has a major shortcoming; in many instances, occupations such as engineering, computer science, and nursing have unique characteristics that are not representative of the overall economy or the physician market. Specifically, wage changes for such occupations can be influenced by excess supply or demand for these types of workers. We believe it would not be appropriate to proxy the physician earnings portion of the MEI with a wage proxy that reflects these other occupation's unique characteristics. The 2000-based MEI will use the ECI for fringe benefits for total private industry as the price proxy for physician fringe benefits, the same proxy used for the 1996-based MEI. This means that both the wage and fringe benefit proxies for physician earnings are derived from the nonfarm private sector and are computed on a per-hour basis.

As in the 1996-based MEI, we used Current Population Survey data on earnings and employment by occupation to develop labor cost shares for the nonphysician occupational groups shown in Table 10. BLS maintains an ECI for each occupational group, and we use these ECIs as price proxies for nonphysician employee wages in the 2000-based MEI.

The skill mix shift in employees of physician offices in the last few years has been towards managerial occupations. While these skill mix shifts are captured in the expenditure weights, they are appropriately held constant in a Laspeyres price index such as the MEI. Skill mix shifts, which may reflect the changing intensity of services provided in physicians' offices, are accounted for in the payment system outside of the MEI. The 2000-based MEI will use the ECI for fringe benefits for white collar employees in the private sector as a proxy for nonphysician benefits since most nonphysician employees in physicians' offices are white-collar employees. This is the same proxy used for the 1996-based MEI.

Office expenses include rent or mortgage for office space, furnishings, insurance, utilities, and telephone. We continue to use the CPI-U for housing because it is a comprehensive measure of the cost of housing, including rent, owner's equivalent rent, and the types of goods and services associated with running an office. This proxy covers about 80 percent of the population.

This cost category includes drugs, outside laboratory work, x-ray films, and other related services. There is not one price proxy that includes this complete mix of materials and supplies. In the absence of one index, we separately accounted for pharmaceuticals and medical materials and supplies in the 2000-based MEI.

• Medical Materials and Supplies

We equally weighted two proxies together (the PPI Surgical Appliances and Supplies and the CPI-U for Medical Equipment and Supplies) since one proxy does not accurately measure the price change associated with these types of products used nor the mode of purchase used in physicians' offices. While both indexes include such items as bandages, dressings, catheters, I.V. equipment, syringes, and other general disposable medical supplies and nonprescription equipment, the indexes reflect significant differences in the mode of purchase. The PPI measures actual transaction prices at the wholesale level, the mode most likely used by physicians, while the CPI measures prices at the retail level or the final stage of production. The price movements in these two indexes can be different and we believe that it is appropriate to combine these indexes into one proxy since physicians likely use both purchasing methods when obtaining medical supplies.

• Pharmaceuticals

The PPI for pharmaceutical preparations is used to proxy pharmaceutical prices in other CMS market baskets and reflects the price change associated with the average mix of pharmaceuticals purchased economy-wide. We use the PPI for pharmaceutical preparations, rather than the CPI for prescription drugs, because physicians generally purchase drugs directly from a wholesaler. The PPIs we use measure price changes at the final stage of production and not intermediate production, however.

It is vital that the MEI accurately reflect the price changes associated with professional liability costs. Accordingly, we continue to incorporate into the MEI a price proxy that accomplishes this goal by making the maximum use of available data on professional liability premiums.

Each year, we solicit professional liability premium data for physicians from a small sample of commercial carriers. This information is not collected through a survey form but instead is requested, on a voluntary basis, from a few national commercial carriers via letter. Generally between 5 and 8 carriers volunteer this information. For the CY 2004 update we were able to obtain data from 7 carriers, all of which were in the top 15 companies in 2001 in terms of market share. While the sample size certainly does not cover the entire professional liability insurance market, we have attempted to maximize the market share in terms of both national coverage and coverage within States.

As we require for our other price proxies, the professional liability price proxy should reflect the pure price change associated with this particular cost category. Thus, it should not capture changes in the mix or level of liability coverage. To accomplish this result, we obtain premium information from commercial carriers for a fixed level of coverage, currently $1 million per occurrence and a $3 million annual limit. This information is collected for every State by physician specialty and risk class. Finally, the State-level, physician-specialty data are aggregated by effective premium date to compute a national total, using counts of physicians by State and specialty as provided in the AMA publication, Physician Characteristics and Distribution in the U.S.

The resulting data provide a quarterly time series, indexed to a base year consistent with the MEI and reflect the national trend in the average professional liability premium for a given level of coverage. From this series, quarterly and annual percent changes in professional liability insurance are estimated for inclusion in the MEI.

Our research has indicated that the most comprehensive data on professional liability costs are held by the State insurance commissioners but these data are available only with a substantial time lag and, therefore, the data currently incorporated into the MEI are much more timely. We believe that, given the limited data available on professional liability premiums, this methodology adequately reflects the price trends facing physicians.

Comment: Several commenters were concerned about the 6.6 percent increase in the PLI component of the MEI published in the proposed rule and felt that this did not represent the actual increase in premiums physicians are experiencing.

Response: We indicated in the proposed rule that the 6.6 percent increase in the PLI component of the index was based on a forecast. For this final rule we have incorporated actual data (through the second quarter of 2003) that indicates that the increase in the proxy for the PLI component of the MEI is 16.9 percent.

Medical equipment includes depreciation, leases, and rent on medical equipment. We will use the PPI for medical instruments and equipment as the price proxy for this category, consistent with the price proxy used in the 1996-based MEI and other CMS input price indexes.

This category includes the residual subcategory of other expenses such as accounting services, legal services, office management services, continuing education, professional association memberships, journals, professional car expenses, and other professional expenses. In the absence of one price proxy or even a group of price proxies that might reflect this heterogeneous mix of goods and services, we use the CPI-U for all items less food and energy, consistent with the price proxy used in the 1996-based MEI. We also condensed the structure compared to that used in the 1996-based MEI because we lack the data to develop a representative weight for transportation, as discussed above. This change resulted in only a negligible effect on the overall MEI over the past 8 years; the average annual increase differs by less than a tenth of a percentage point over that time.

In the December 2002 final rule, we indicated that we were changing the methodology for adjusting for productivity in the MEI. The MEI used for the 2003 physician payment update reflected changes in the 10-year moving average of private nonfarm business (economy-wide) multifactor productivity applied to the entire index; we had previously used economy-wide private nonfarm business labor productivity applied to the labor portions of the index. We will continue to use the new method, adjusting for multifactor productivity applied to the entire index, in the rebased and revised MEI.

As described in the December 31, 2002 (68 FR 9568) final rule, we use multifactor productivity because: (1) It is theoretically more appropriate to explicitly reflect the productivity gains associated with all inputs (both labor and nonlabor); (2) the recent growth rate in economy-wide multifactor productivity appears to be more consistent with the current market conditions facing physicians; and (3) the MEI still uses economy-wide wage changes as a proxy for physician wage changes. We also believe that using a 10-year moving average change in economy-wide multifactor productivity produces a stable and predictable adjustment and is consistent with the moving-average methodology used in the 1996-based MEI. The adjustment will be based on the latest available actual historical economy-wide multifactor productivity data, as measured by BLS. For the 2004 update, this means using the multifactor productivity data through 2001, the latest available information.

Because the rebased and revised MEI is similar in structure to the 1996-based MEI, updating the MEI from a 1996 base year to a 2000 base year resulted in small changes in expense category weights. Physicians' earnings dropped slightly, from 54.5 percent of the index in 1996 to 52.5 percent in 2000. The expense shares for non-physician employee compensation, office expenses, professional liability insurance, and medical equipment all rose slightly, while expense shares for medical materials and supplies and other expenses declined.

The update using the rebased and revised MEI for the 2004 Physician Fee Schedule is an increase of 2.9 percent. This incorporates historical data through the second quarter of 2003.

As is the case with this index rebasing, our experience in previous rebasing and revising indexes has been that there is usually a very small effect on the overall percent change. The difference is typically between zero and 0.3 percentage points per year on average. The rebased and revised MEI overall percent increase for the CY 2004 update is only 0.1 percentage point higher compared to the 1996-based MEI. This is also the case for this final rule. When the MEI was last rebased, there was no difference in the average annual percentage change from 1985 to 1998. When the PPS hospital indices were rebased, the average difference in the percentage change was less than one-tenth of a percentage point from 1995 to 2002.

The first reason for this small difference between the 1996-based and 2000-based MEI percent changes is that the weight of professional liability insurance increased, giving it a higher relative importance in the index in 2000. This category also increased at a faster pace than other index categories during 2002 and projected for 2003, resulting in an even greater relative importance for this index by 2004 and causing it to have a larger effect on the overall index compared to the 1996-based MEI.

In addition, the pharmaceuticals from the medical materials and supplies category grew faster than the overall medical materials and supplies in the 1996-based MEI. In addition, the faster growth in the aggregate medical materials and supplies category combined with a higher weight in the 2000-based index gave the category a higher relative importance. However, these increases were mostly offset by declines in weight of some of the other categories, most notably physician earnings.

As discussed in the August 15, 2003 proposed rule, section 1848(c)(2)(B)(ii)(II) of the Act requires that increases or decreases in RVUs may not cause the amount of expenditures for the year to differ by more than $20 million from what expenditures would have been in the absence of these changes. If this threshold is exceeded, we make across-the-board adjustments to preserve budget neutrality. Therefore, if we adjust the work, practice expense and malpractice RVUs to match the new MEI weights, we are required by statute to ensure that the adjustments do not increase or decrease Medicare expenditures by more than $20 million. To meet the requirements of the statute and ensure that aggregate pools of RVUs match the proposed new MEI weights, we considered two options. We considered either making no adjustments to the physician work RVUs and adjusting only the practice expense and malpractice RVUs or adjusting all 3 categories of RVUs. We proposed adjusting all 3 categories of RVUs rather than adjusting only the practice expense and malpractice RVUs, which would have resulted in a reduction to the physician fee schedule conversion factor in addition to the -4.2 percent reduction that was forecasted. Specifically, we proposed to reduce the physician work RVUs by an estimated 0.35 percent (0.9965) and the practice expense RVUs by an estimated 1.15 percent (0.9885) and to increase the malpractice RVUs by an estimated 21.7 percent (1.217) to match the rebased MEI weights.

Comment: We received comments from a number of physician organizations opposing any adjustment to the physician work RVUs. Several of the comments appreciated our reluctance to reduce the physician fee schedule conversion factor by an additional 0.3 percentage points when there will already be a large reduction in the physician fee schedule update. One commenter stated that any additional reduction to the physician fee schedule conversion factor would be inappropriate. However, these comments also stated that that the physician work RVUs should remain constant and stable. There were a number of comments that stated that across-the-board adjustments should never be applied to the work component of the Resource Based Relative Value System. One comment indicated that we should not make any adjustments to the work RVUs unless they are recommended by the RUC. Several of the comments stated that the proposed adjustments to the RVUs to match the MEI weights would not assist the physician community in addressing the professional liability crisis since any increase in physician fees for some services will be offset by reductions in other services. Additional payments by Medicare to cover increased professional liability costs, or congressional action, are necessary to alleviate this problem. Some of the comments indicated that CMS did not provide sufficient information to make a determination as to how the two proposals would affect individual codes because the adjustments were not applied to the RVUs shown in Addendum B of the proposed rule. Several of the comments stated that the stability of work RVUs is essential since they are used by private payors, physician compensation systems, and in productivity analysis. The RUC commented that they depend upon the stability in these values as they review new and revised codes, both in magnitude estimation and in any calculations regarding intra-work per unit of time (IWPUT). One comment suggested CMS create a separate adjustment factor to adjust payments without changing the conversion factor or the RVUs, as it did for the first five-year review of the Medicare physician fee schedule in 1995. We also received a comment urging us to review the Secretary's “ancillary policies” authority under section 1848(c)(4) of Act to determine whether CMS has statutory authority to increase PLI relative value units without reducing the work and practice expense relative value units.

We also received several comments that expressed support for maintaining stability in the practice expense RVUs. The comment stated “much like what is done with work relative values, any code-level refinements due to annual coding changes that result in a non-budget neutral impact should not result in a reduction of all practice expense relative values. The comment requested that CMS present an analysis of this issue in an upcoming proposed rule and recommended that adjustments related to the MEI rebasing not be applied to the practice expense relative values.

Response: We share the concern about establishing stability in the practice expense RVUs. As we indicated in the June 28, 2002 proposed rule (67 FR 43851), “once the refinement process is complete, we believe the physician community has a reasonable expectation that the practice expense RVUs will not change from year to year unless further refinement is undertaken.” We plan to analyze in an upcoming proposed rule whether there are any alternatives to our current practice of rescaling the practice expense RVUs to apply budget neutrality. However, we disagree with the comments that suggest we only increase the malpractice expense RVUs and not apply any adjustments to the work and practice expense RVUs to match the MEI weights. It is not possible to match the aggregate RVUs to the new MEI weights if we make no adjustments to both work and practice expense and adjust only the malpractice RVUs and the conversion factor. While it would be possible to maintain budget neutrality for the increase in malpractice RVUs by reducing the conversion factor, the aggregate number of RVUs for work and practice expense would not match the MEI weights unless we could adjust at least two of the three RVUs in combination with applying a compensating adjustment to the CF.

We have considered the comment suggesting that we use the Secretary's section 1848(c) “ancillary” policies authority to adjust the RVUs to match the MEI weights but not maintain budget neutrality. Section 1848(c) states that the Secretary may establish ancillary policies (with respect to the use of modifiers, local codes, and other matters) as may be necessary to implement this section.” We believe that this section of the statute must, nonetheless, be read consistently with the requirements of section 1848(c)(2)(B)(ii)(II) of the Act requiring that changes to RVUs cannot cause the amount of expenditures to increase or decrease by more than $20 million from the amount of expenditures that would have been made if such adjustments had not been made. We believe the statute is clear and any increase in the malpractice expense RVUs must be offset by decreases to the work and practice expense RVUs or the conversion factor.

We also do not believe that the work RVUs should be maintained and a separate “work adjustor” established. While such policy was adopted following the 5-year review of physician work in 1997, we used this procedure only because the effect of the work adjustor could be removed once resource-based practice expense RVUs were adopted in 1999. We did not find the work adjustor to be desirable. It added an extra element to the physician fee schedule payment calculation and created confusion and questions among the public who had difficulty using the RVUs determine a payment amount that matched the amount actually paid by Medicare.

We acknowledge the comments that indicate that the work RVUs are used for many purposes other than Medicare payment. While our proposal would slightly reduce the absolute value of the physician work RVUs, it would not change their relative values since there would be a uniform decrease to all of the RVUs. We believe the relative relationship among the values for the services makes them useful for analysis for purposes other than Medicare payment. Since the relative values will be left unchanged, we do not believe the work RVUs will lose their utility for these other uses.

We disagree that our proposed rule did not provide enough information upon which to determine the impact on payment for a given service. The proposed rule provided the specific level of the estimated adjustments. While we did not actually apply the adjustments to the RVUs shown in Addendum B, any interested party could determine the effect of our proposal on any given service with the information we provided. We further noted that the adjustments we provided were estimated and would change once we made final determinations of the work, practice expense and malpractice RVUs for 2004. For the final rule, we will reduce the work RVUs by 0.57 percent (0.9943), the practice expense by 0.77 (0.9923) percent and increase the malpractice RVUs by 19.86 percent (1.1986). We have also modeled the impact of our proposal by specialty in the impact section of this final rule.

With respect to the comments about our proposal and the large increases in professional liability premiums, we have not asserted that our policy to adjust the RVUs will resolve this issue. While the comments that our policy will increase payments for some service and decrease payments for payments for others are correct, we note that payments for services with high malpractice RVUs will increase the most in payment while there will be negligible impact on payment for most other services. Such a policy will improve our payment policies by giving more weight to the malpractice RVU in determining Medicare total payment consistent with the proportion that professional liability expenses represent of total physician expenses. As indicated in the impact section, services provided by cardiac and thoracic surgeons, neurosurgeons, orthopedic surgeons, vascular surgeons and emergency physicians are increasing in payment as a result of this proposal. There will be little impact of these adjustments on all other specialties.

Section 1848(d) of the Act provides that the physician fee schedule update is equal to the product of the MEI and an “update adjustment factor” or UAF. The UAF is applied to make actual and target expenditures (referred to in the law as “allowed expenditures”) equal. Allowed expenditures are equal to actual expenditures in a base period updated each year by the SGR. The SGR sets the annual rate of growth in allowed expenditures and is determined by a formula specified in section 1848(f) of the Act.

Under section 1848(d)(4)(A) of the Act, the physician fee schedule update for a year is equal to the product of—(1) 1 plus the Secretary's estimate of the percentage increase in the MEI for the year, divided by 100 and (2) 1 plus the Secretary's estimate of the UAF for the year. Under section 1848(d)(4)(B) of the Act, the UAF for a year beginning with 2001 is equal to the sum of the following—

• Prior Year Adjustment Component. An amount determined by—

• Computing the difference (which may be positive or negative) between the amount of the allowed expenditures for physicians' services for the prior year (the year prior to the year for which the update is being determined) and the amount of the actual expenditures for such services for that year;

• Dividing that difference by the amount of the actual expenditures for such services for that year; and

• Multiplying that quotient by 0.75.

• Cumulative Adjustment Component. An amount determined by—

• Computing the difference (which may be positive or negative) between the amount of the allowed expenditures for physicians' services from April 1, 1996, through the end of the prior year and the amount of the actual expenditures for such services during that period;

• Dividing that difference by actual expenditures for such services for the prior year as increased by the sustainable growth rate for the year for which the update adjustment factor is to be determined; and

• Multiplying that quotient by 0.33.

Section 1848(d)(4)(E) of the Act requires the Secretary to recalculate allowed expenditures consistent with section 1848(f)(3) of the Act. Section 1848(f)(3) specifies that the SGR (and, in turn, allowed expenditures) for the upcoming calendar year (2004 in this case), the current calendar year (2003) and the preceding calendar year (2002) are to be determined on the basis of the best data available as of September 1 of the current year. Allowed expenditures are initially estimated and subsequently revised twice. The second revision occurs after the calendar year has ended (that is, we are making the final revision to 2002 allowed expenditures in this final rule). Once the SGR and allowed expenditures for a year have been revised twice, they are final.

Table 13 shows annual and cumulative allowed expenditures for physicians' services from April 1, 1996 through the end of the current calendar year, including the transition period to a calendar year system that occurred in 1999.

Consistent with section 1848(d)(4)(E) of the Act, table 13 includes our final revision of allowed expenditures for 2002, a recalculation of allowed expenditures for 2003, and our initial estimate of allowed expenditures for 2004. To determine the update adjustment factor for 2004, the statute requires that we use allowed and actual expenditures from April 1, 1996 through December 31, 2003 and the 2004 SGR. Consistent with section 1848(d)(4)(E), we will be making further revisions to 2003 and 2004 SGRs and 2003 allowed expenditures. Because we have incomplete actual expenditure data for 2003, we are using an estimate for this period. Any difference between current estimates and final figures will be taken into account in determining the update adjustment factor for future years.

We are using figures from table 13 in the statutory formula illustrated below:

Section 1848(d)(4)(D) of the Act indicates that the UAF determined under section 1848(d)(4)(B) of the Act for a year may not be less than −0.070 or greater than 0.03. The calculated UAF of −0.090 is less than the statutory limit of −0.070. Therefore, the UAF for 2004 will be −0.70.

Section 1848(d)(4)(A)(ii) of the Act indicates that 1 should be added to the UAF determined under section 1848(d)(4)(B) of the Act. Thus, adding 1 to −0.070 makes the update adjustment factor equal to 0.930.

The SGR is an annual growth rate that applies to physicians' services paid for by Medicare. The use of the SGR is intended to control growth in aggregate Medicare expenditures for physicians' services. Payments for services are not withheld if the percentage increase in actual expenditures exceeds the SGR. Rather, the physician fee schedule update, as specified in section 1848(d)(4) of the Act, is adjusted based on a comparison of allowed expenditures (determined using the SGR) and actual expenditures. If actual expenditures exceed allowed expenditures, the update is reduced. If actual expenditures are less than allowed expenditures, the update is increased.

Section 1848(f)(2) of the Act specifies that the SGR for a year (beginning with 2001) is equal to the product of the following four factors:

(1) The estimated change in fees for physicians' services.

(2) The estimated change in the average number of Medicare fee-for-service beneficiaries.

(3) The estimated projected growth in real GDP per capita.

(4) The estimated change in expenditures due to changes in law or regulations.

In general, section 1848(f)(3) of the Act requires us to publish SGRs for 3 different time periods, no later than November 1 of each year, using the best data available as of September 1 of each year. Under section 1848(f)(3)(C)(i) of the Act, the SGR is estimated and subsequently revised twice (beginning with the FY and CY 2000 SGRs) based on later data. (The Consolidated Appropriations Reduction Resolution of 2003 (P.L. 108-7) contained a provision permitting revision of the FY 1998 and FY 1999 SGRs. See the February 28, 2003 Federal Register (68 FR 9567) for a discussion of these SGRs. Under section 1848(f)(3)(C)(ii) of the Act, there are no further revisions to the SGR once it has been estimated and subsequently revised in each of the 2 years following the preliminary estimate. In this final rule, we are making our preliminary estimate of the 2004 SGR, a revision to the 2003 SGR, and our final revision to the 2002 SGR.

Section 1848(f)(4)(A) of the Act defines the scope of physicians' services covered by the SGR. The statute indicates that the term “physicians' services” includes other items and services (such as clinical diagnostic laboratory tests and radiology services), specified by the Secretary, that are commonly performed or furnished by a physician or in a physician's office, but does not include services furnished to a Medicare+Choice plan enrollee. We published a definition of physicians' services for use in the SGR in the Federal Register (66 FR 55316) on November 1, 2001. We defined “physicians' services” to include many of the medical and other health services listed in section 1861(s) of the Act. For purposes of determining allowed expenditures, actual expenditures, and SGRs through December 31, 2002, we have specified that “physicians' services” include the following medical and other health services if bills for the items and services are processed and paid by Medicare carriers (and those items and services paid through intermediaries where specified):

• Physicians' services.

• Services and supplies furnished incident to physicians' services.

• Outpatient physical therapy services and outpatient occupational therapy services.

• Antigens prepared by or under the direct supervision of a physician.

• Services of physician assistants, certified registered nurse anesthetists, certified nurse midwives, clinical psychologists, clinical social workers, nurse practitioners, and clinical nurse specialists.

• Screening tests for prostate cancer, colorectal cancer, and glaucoma.

• Screening mammography, screening pap smears, and screening pelvic exams.

• Diabetes outpatient self-management training services.

• Medical nutrition therapy services.

• Diagnostic x-ray tests, diagnostic laboratory tests, and other diagnostic tests (including outpatient diagnostic laboratory tests paid through intermediaries).

• X-ray, radium, and radioactive isotope therapy.

• Surgical dressings, splints, casts, and other devices used for the reduction of fractures and dislocations.

• Bone mass measurements.

Section 211(b)(1) of the BBRA amended section 1848(f)(1) of the Act to require that three SGR estimates be published in the Federal Register not later than November 1 of every year. In this final rule, we are publishing our preliminary estimate of the SGR for 2004, a revised estimate of the SGR for 2003, and our final determination of the SGR for 2002. Consistent with section 1848(f)(3)(C) of the Act, we are using the best data available to us as of September 1, 2003 for all of the figures.

Our preliminary estimate of the 2004 SGR is 7.4 percent. We first estimated the 2004 SGR in March and made the estimate available to the Medicare Payment Advisory Commission and on our website. Table 13 shows our March estimates and our current estimates of the factors included in the SGR:

Our current estimate of the 2003 SGR is 6.7 percent. Table 14 shows our preliminary estimate of the 2003 SGR that was published in the Federal Register on December 1, 2002 (67 FR 80027) and our current estimate:

A more detailed explanation of each figure is provided below in section G.2.

The SGR for 2002 is 8.3 percent. Table 16 shows our preliminary estimate of the SGR published in the Federal Register on November 1, 2001 (66 FR 55317), our revised estimate published in the Federal Register on December 31, 2001 (67 FR 80028) and the final figures determined using the latest available data: