THE SOCIETY OF THORACIC SURGEONS 633 NORTH SAINT CLAIR STREET CHICAGO, ILLINOIS 60611-3658 Phone: 312.202.5800 Fax: 312.202.5801 E-mail: [email protected] Web: http://www.sts.org

July 25, 2011 Donald M. Berwick, MD, MPP Administrator Centers for Medicare and Medicaid Services Department of Health and Human Services Hubert H. Humphrey Building – Room 445G 200 Independence Avenue SW Washington, DC 20201

MICHAEL J. MACK, MD President The Heart Hospital Baylor Plano 1100 Allied Drive Plano, TX 75093 Phone: 469.814.4105 Fax: 469.814.4388 E-mail: [email protected]

Re: CMS-1582-PN Medicare Program; Five Year Review of Work Relative Value Units Under the Physician Fee Schedule Dear Dr. Berwick: On behalf of The Society of Thoracic Surgeons (STS), the largest organization representing cardiothoracic surgeons in the United States and the world, I am writing to provide comments on the Centers for Medicare and Medicaid Services Proposed Rule on the Five-Year Review of Work Relative Value Units Under the Physician Payment Schedule, published in the June 6, 2011 Federal Register. STS represents more than 6,200 surgeons, researchers, and allied health care professionals worldwide who are dedicated to ensuring the best possible outcomes for surgeries of the heart, lung, and esophagus, as well as other surgical procedures within the chest. We appreciate the opportunity to provide comments on this proposed rule. STS has carefully reviewed the proposed rule and offers a number of general comments as well as comments related to specific CMS proposals. In general, we are concerned about the apparent development of an unprecedented systematic bias in CMS‟s consideration of the American Medical Association/Specialty Society Relative Value Scale Update Committee (RUC)-recommended values. Below, Chart 1 depicts the RUC recommendations and CMS actions over the past five years for the 43 code proposals with respect to which STS was a participating specialty society:

Administrator Berwick CMS-3248-P July 25, 2011 Page 2

There was 100% acceptance of the RUC recommendations, which were arrayed in a distribution around the median survey physician work values, with a number of codes determined by STS, the RUC, and CMS to warrant values that were at or below the 25th percentile survey results. Chart 2 represents a similar display of the RUC recommendations and CMS proposal for the STS 2010 Five-Year Review:

Administrator Berwick CMS-3248-P July 25, 2011 Page 3 It is apparent that the RUC recommendations generally have been rejected (16/29) and proposed at the survey 25th percentile (11/12). Only one RUC recommended value was accepted based on the median survey result. Thus, despite the fact that the RUC Five-Year recommendation profile is similar or lower in relation to survey values compared to the 2005-2010 period, CMS has clearly established a systematic approach in which the RUC survey 25th percentile or less is a “new normal,” at least for our specialty, regardless of the RUC rationale, supporting crosswalks or attestation by surgeons who perform these services. STS also is concerned that the otherwise unsupported systematic adoption of the survey 25th percentile work value estimate, or even, on occasion, the survey low value, will have pernicious effects on the relative valuation system. The general knowledge that CMS policy apparently has shifted away from a considered evaluation of the survey results and informed integration into the relative values of the existing physician fee schedule, toward a fixed downward offset of surveyed work magnitude estimation, predictably will result in “work estimate inflation” by respondents. To this point, that effect has been mitigated by the expert opinions of specialty societies and multidisciplinary RUC members, as evident in Figure 1 above. We believe that a final rule published with current proposed values and current CMS rationales will cause irreparable damage to the random survey magnitude estimation process, which is the foundation of the Resource-Based Relative Value Scale (RBRVS). In addition, CMS has utilized methods to construct new proposed code values that are inconsistent with methodologies used in the past by both the RUC and CMS and that will create rank order anomalies and disturb the relativity of the existing physician fee schedule. The RUC currently is the accepted venue for specialty societies to provide input on code valuation. The RUC survey methodology has been the valuation standard used by CMS to assist in the establishment of codes values since the inception of the RBRVS. STS has utilized the approved venue and methodology to develop revised valuation recommendations for the fourth Five-Year Review. The RUC has evaluated those recommendations, making adjustments where necessary, and, until now, CMS has accepted those recommendations. One of the roles of the RUC is to ensure that relativity is maintained within and across specialties. STS has made a concerted effort to ensure that recommended values, as well as those values refined by the RUC, maintain relativity within and across specialties. The arbitrary and unsupported recommendations made by CMS in the proposed rule, even by a small amount, create rank order anomalies for many of the reviewed codes within the specialty. Until CMS establishes accepted methodologies other than the RUC to determine values, STS strongly encourages CMS to align its decisions with the concerted, informed, and evidence-based efforts of the RUC and the specialty societies representing physicians who perform the procedures. In summary, STS objects to the methods employed by CMS to propose alternate values for selected cardiothoracic surgery codes, and asks that CMS reconsider and accept the RUC recommendations for all of the cardiothoracic surgery procedures from the fourth Five-Year Review. STS would like to call CMS‟s attention to the fact that the cardiothoracic surgery procedures reviewed in the fourth Five-Year Review are low volume, high risk and highly specialized procedures. For the most part, they are performed in tertiary or quaternary hospital environments and at academic sites that are the training centers for our future thoracic and cardiac surgeons. In addition, we would point out that

Administrator Berwick CMS-3248-P July 25, 2011 Page 4 these procedures are proven to be life prolonging and to enhance the quality of life for Medicare beneficiaries. Our specialty has, at great expense, engaged in a national quality improvement exercise through the STS Adult Cardiac Surgery Database to enhance outcomes and ensure appropriate utilization. Below we discuss, in detail, each of the codes or families of codes for which alternate values have been proposed. It is important to note that each of the RUC-recommended values is amply supported by survey results, expert multidisciplinary opinion, and crosswalks or comparison to Multispecialty Points of Comparison (MPC) codes and comparable cardiothoracic surgery codes. Lung Transplant codes: 32854 (Double lung transplant with cardiopulmonary bypass) – RUC Recommendation: 95.00, CMS Recommendation: 90.00 We object to the methods employed by CMS to propose alternate values for the lung transplant codes, and ask that CMS reconsider and accept the RUC recommendations. It is clear that, in approaching this code family, CMS first determined to reject the RUC-recommended value for 32854, double lung transplant with cardiopulmonary bypass at the median survey RVW of 95.00. Instead, CMS proposes a work value of 90 (the 25th percentile), which was considered and rejected by the RUC workgroup and STS as not reflective of the physician work involved. The key reference service was 33945 (Heart Transplant), valued at the last Five-Year Review at 89.50. This was chosen by 70% of the survey respondents with a median performance rate of 5 annually. Survey respondents indicated that the median intra-service time was 400 minutes, 75 minutes more than the reference service code and that double lung transplant with cardiopulmonary bypass required substantially more physician work by every complexity and intensity rating as demonstrated in Table 1: Table 1 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean) The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making

(of those that selected Key reference code) 32854 33945 4.58

4.13

4.85

4.33

4.55

4.08

Technical skill required

4.83

4.25

Physical effort required

4.85

4.35

Technical Skill/Physical Effort (Mean)

Psychological Stress (Mean)

Administrator Berwick CMS-3248-P July 25, 2011 Page 5 The risk of significant complications, morbidity and/or 4.80 mortality Outcome depends on the skill and judgment of 4.73 physician Estimated risk of malpractice suit with poor outcome 3.73 Time Segments (Mean)

4.23

Pre-Service intensity/complexity

4.65

3.85

Intra-Service intensity/complexity

4.93

4.13

Post-Service intensity/complexity

4.75

4.10

4.18 3.65

We believe that the additional 5.50 work units provided by the RUC-recommended work value are well justified by the additional intra-service time, intensity, and complexity measures as compared to the key reference code. We submit that a 1.5 RVW difference resulting from the CMS proposal fails to maintain relativity between double lung transplant with cardiopulmonary bypass and heart transplant. We find the CMS rationale for this code confusing, as it maintains relativity to 32851, (Single lung transplant without cardiopulmonary bypass), which concerns a different patient population and substantially different procedure with a proposed RVW difference of more than 40 RVW. As will be discussed next, relativity to 32853, (Double lung transplant without cardiopulmonary bypass) is more reasonable to consider just as articulated above for heart transplant. 32853 (Double lung transplant without cardiopulmonary bypass) – RUC Recommendation: 90.00, CMS Recommendation: 84.48 For 32853, we would ask CMS to reconsider the RUC recommendation in light of the above valuation of 32854 and maintain relativity by changing the proposed value from 84.48 to 90. The latter value is well justified by the summary of recommendations submitted at the 5-Year Review and the RUC rationale, and will not be replicated here. Similar to 32854, the key reference service was 33945 with a RVW of 89.50. Here, survey respondents indicated an additional 50 minutes of intra-service time and increased intensity and complexity measures in all domains compared to 33945 as reflected in Table 2: Table 2 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean) The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed

(of those that selected Key Reference code) 32853 33945 4.68

4.14

4.86

4.32

Administrator Berwick CMS-3248-P July 25, 2011 Page 6 Urgency of medical decision making

4.55

4.09

Technical skill required

4.82

4.09

Physical effort required Psychological Stress (Mean)

4.86

4.18

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean)

4.86

4.27

4.86

4.32

4.18

3.95

Pre-Service intensity/complexity

4.59

4.05

Intra-Service intensity/complexity Post-Service intensity/complexity

4.82 4.55

4.23 4.18

Technical Skill/Physical Effort (Mean)

There is no rational justification for 32853 to be valued less than 33945. 32851 (Single lung transplant without cardiopulmonary bypass) – RUC Recommendation: 63.00, CMS Recommendation: 59.64 32852 (Single lung transplant with cardiopulmonary bypass) – RUC Recommendation: 74.37, CMS Recommendation: 65.50 For the single lung transplant codes, CMS proposes to employ a reverse building block method to discard the survey results for 32851, (Single lung transplant without cardiopulmonary bypass), thereby considering the RUC-recommended 25th percentile RVW as being too high. CMS calculates a proposed value of 59.64 as being 5.86 RVW less than 32852, (Single lung transplant with cardiopulmonary bypass) to account for the non-use of cardiopulmonary bypass. CMS chose the with/without cardiopulmonary bypass code pair of 33255/33256 to calculate this difference, and simultaneously rejected the RUC recommendation for 32852, proposing the 25th percentile survey result of 65.50 (which the RUC considered and rejected as insufficient). In proposing this methodology, CMS does not ground these recommendations in crosswalks to comparable codes or provide any rationale other than to “believe it to be appropriate.” In selecting the code pair 33255/33256, CMS ignores the other such code pairs RUC valued since 1993 and CMS accepted, which are 33925/33926 with a 13.43 RVW differential, 33470/33474 with a 17.86 RVW differential and 33305/33300 with a 31.96 RVW differential.

Administrator Berwick CMS-3248-P July 25, 2011 Page 7 We do not see evidence for a building block for the use of cardiopulmonary bypass, as its use is just one component of the physician work. The decision to use cardiopulmonary bypass is based on patient characteristics, and those characteristics determine the difference in physician work as much as the placement of cannulae, conduct of cardiopulmonary bypass, and differences in operative time and perioperative complications associated with cardiopulmonary bypass alone. As shown here, the difference ranges from 5.86 to 31.96, making the selection of any one value impossible to justify. Furthermore, the survey method of magnitude estimation of the total RVW accounts for both the use of cardiopulmonary bypass and the differences in patient characteristics, which contribute to any difference in physician work. Therefore it is not justified to employ this method, even if a reasonable and consistent work value for the use of cardiopulmonary bypass were available. In this instance, the use of cardiopulmonary bypass is urgent, determined upon failure of an attempt to perform single lung transplant without cardiopulmonary bypass, and is much more like the code pair 33305/33300 where a cardiac wound cannot be repaired without cardiopulmonary bypass. We do not argue for a differential of 31.96 with either code as the baseline. We ask that CMS accept the RUC recommendations, based on consideration of all aspects of these two codes, including the physician surveys, a review of the physician fee schedule for comparable values, differences in the procedures and patients, and consideration of the entire lung transplant family. The RUC-recommended value for 32581, 63.00 RVW (25th percentile) with an IWPUT of 0.109, 4 hours of intra-service time and a hospital length of stay of 11 days, is all perfectly reasonable to an objective observer. It is also consistent with the vast majority of CMS actions in the past decade. The proposed value of 65.50 for 32852 then becomes completely untenable in terms of relativity to 32581 at 63.00. This was the crux of the conundrum faced by the RUC workgroup and the specialty society in developing an appropriate work recommendation that necessarily must be between the 25th and 50th percentiles of the surveys. We believe that the actual RUC recommendation for 32852 is the correct value, and recognizes the additional 60 minutes of intra-service time, the added complexity of cardiopulmonary bypass, the typical urgency of converting a patient from off to on cardiopulmonary bypass during these procedures, and the additional 1.5 days of hospital stay. There are many examples of codes with this sort of crosswalk for additional time and visit pattern, such as 27216, 45160, 49402, 49492, 58260, 61154, 65756, 66982 and 67218. 33030 (Pericardiectomy, subtotal or complete; without cardiopulmonary bypass) – RUC Recommendation: 39.50, CMS Recommendation: 36.00 Although CMS is proposing to accept the RUC-recommended work value for this procedure when cardiopulmonary bypass is used (33031, RVW 45.00, 25th percentile), it proposes the 25th percentile survey value of 36.00 for 33030, which is less than the RUC-recommended median value of 39.5 (intraservice time 180 minutes). CMS offers no rationale whatsoever for this decision, simply declaring it “more appropriate for this service.” We disagree with this proposed value, and ask CMS to reconsider the RUC recommendation as more appropriate. If absolutely necessary, CMS could review its own logic for valuing 32851 with a differential value of 5.86 for the use of cardiopulmonary bypass -- which would, of course, support the RUC recommendation almost exactly for 33030. However, because we disagree with that methodology

Administrator Berwick CMS-3248-P July 25, 2011 Page 8 as discussed above, we would prefer that CMS accept the RUC recommendation based on the RUC rationale that has been forwarded. This would place the value of 33030 in an appropriate relative value for time and intensity of physician work in relation to 33031. Crosswalk codes with similar intra-service time and equal or greater RVW include 35082 (RVW: 42.09 intratime: 180), 35103 (RVW: 43.62 intratime: 108), 43313 (RVW: 48.45 intratime: 178), 43314 (RVW: 53.43 intratime: 178), 47361 (RVW: 52.60 intratime: 190), and 48105 (RVW: 49.26 intratime: 180. 33910 (Pulmonary Artery Embolectomy; with cardiopulmonary bypass) – RUC Recommendation: 52.33, CMS Recommendation: 48.21 STS requests that CMS reconsider its proposed work value of 48.21, and accept the RUC-recommended work value of 52.33, which is the median survey value. We strongly dispute that the CMS-proposed direct crosswalk to the value of 33542 is appropriate. Although some of the technical composition of the two codes (time and visits) is indeed similar, the intensity and complexity measures are completely different and easily account for the additional 4.12 work units that would result from utilizing the median survey work value. Pulmonary artery embolectomy is not simply an emergency procedure, as opposed to the crosswalk code which is typically (and virtually always) elective. Embolectomy always is performed in patients who are in shock, and almost always after initial treatment failure with thrombolytic therapy. The indication to proceed is „shock unresponsive to medical therapy,‟ and the alternative is imminent death. 33935 (Heart-lung transplant with recipient cardiectomy) – RUC Recommendation: 100.00, CMS Recommendation: 91.78 STS requests that CMS reconsider its proposed work value of 91.78 and accept the RUC-recommended survey median value of 100. In its rationale, CMS acknowledges the increased intensity, complexity and physician work compared to the key reference service 33945 Heart Transplant (89.5 RVW). This is borne out by the survey intensity and complexity measures which are substantially higher, by the intraservice time which is 55 minutes longer, and by the additional 2.5 days of hospital length of stay as demonstrated in Tables 3 and 4: Table 3 TIME ESTIMATES (Median)

CPT Code: 33935

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time

160.00 380.00 90.00 240.0 670.0

Key Reference CPT Code: 33945 272.00 325.00 85.00 240.00 535.00

Administrator Berwick CMS-3248-P July 25, 2011 Page 9 Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

55.0 118.0 0.0 1713.00

55.00 204.00 0.00 1,716.0

Table 4 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean) The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making

(of those that selected Key Reference code) 33935 33945 4.84 4.29

4.94

4.32

4.68

4.03

Technical skill required Physical effort required Psychological Stress (Mean)

4.94 4.77

4.19 4.23

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean)

4.65

4.03

4.65

4.16

3.94

3.65

Pre-Service intensity/complexity Intra-Service intensity/complexity

4.90 5.00

4.23 4.32

Post-Service intensity/complexity

4.81

4.32

Technical Skill/Physical Effort (Mean)

On further consideration, we believe that the differential of 2.28 work units proposed is insufficient to account for this additional physician work. Moreover, whereas the typical heart transplant patient has not had prior heart surgery, the typical heart-lung transplant patient has had prior cardiac surgery. In addition, these patients frequently will have intense vascular collaterals between chest wall and lungs due to congenital abnormalities in pulmonary blood flow that are the root cause of the need for heartlung transplantation.

Administrator Berwick CMS-3248-P July 25, 2011 Page 10 33875 (Descending thoracic aorta graft, with or without cardiopulmonary bypass) – RUC Recommendation: 56.83, CMS Recommendation: 50.72 STS requests that CMS reconsider its proposed work value of 50.72 and, instead, accept the RUCrecommended value of 56.83, which is the 25th percentile of the physician survey. The CMS rationale provided is simply a decision to directly crosswalk this value to 33465, (Replacement tricuspid valve with cardiopulmonary bypass). STS strongly disagrees with this methodology and the use of this, specific crosswalk. Although the technical components (intra-service time and visits) of the value of 33465 are similar to those of 33875, the patients and procedures are substantially different and in fact do not even involve operation on the same organ system. 33875 is much more technically demanding, and the risk of paraplegia dominates the psychological stress of this operation like only one other: the key reference service code 33877, (Repair of Thoracoabdominal aneurysm with graft; with or without cardiopulmonary bypass). Code 33877 is very similar to 33875, which is also an operation for aneurysm in the thoracic aorta but which does not extend in to the abdominal aorta. 33877 was valued at the last Five-Year Review with a physician work value of 69.03. We would ask CMS to review the comparison between the surveyed elements of 33875 and those in the RUC database for 33877 as presented in the Summary of Recommendations submitted for the Five-Year Review, reproduced in table 5:

Table 5

Median Pre-Service Time

CPT Code: 33875 100.00

Key Reference CPT Code: 33877 110.00

Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time

240.00 60.00 210.0 265.0 38.0 63.0

324.00 60.00 210.00 265.00 55.00 86.00

Prolonged Services Time

0.0

0.00

Median Total Time

976.00

1,110.0

TIME ESTIMATES (Median)

Source of Time RUC Time

The components are essentially identical except for the intra-service time, with 33875 having 84 fewer intra-service minutes. At the IWPUT of 33877, this intra-service time difference can be valued at 9.61 RVU (0.1144 X 84). Subtracting this from the physician work value of 33877 (69.03-9.61) results in a value for 33875 of 59.42. This compares favorably to the 25th percentile RUC recommendation of 56.83, and adds further support to the RUC rationale for this value.

Administrator Berwick CMS-3248-P July 25, 2011 Page 11 33120 (Excision of intracardiac tumor, resection with cardiopulmonary bypass) – RUC Recommendation: 42.88, CMS Recommendation: 38.45 CMS proposes a new value which is less than the 25th percentile RUC-recommended value of 42.88. CMS proposes a direct crosswalk to 33677, (Closure of multiple ventricular septal defects; with removal of pulmonary artery band, with or without gusset), with a physician work value of 38.75. STS would appreciate CMS reconsidering this proposal, and strongly disagrees with the direct crosswalk of these very dissimilar codes. 33677 is a congenital heart procedure involving the right side of the heart, and does not involve resection of tumor and cardiac tissue. We would ask CMS to review 33120 in relation to the key reference code selected by physicians who perform procedure, 33426, (Valvuloplasty, mitral valve, with cardiopulmonary bypass; with prosthetic ring). This procedure is an adult cardiac procedure and involves the left side of the heart and left atrium where the mitral valve resides. It is very similar to operating to remove the typical left atrial tumor, utilizing the same cardiac incision and the same cannulation strategy for cardiopulmonary bypass. 33426 is also an MPC list code and is performed frequently by adult cardiac surgeons who also perform 33120. Table 6 reflects the survey and reference code data that were submitted in the Summary of Recommendations form for 33120: Table 6 TIME ESTIMATES (Median)

Median Pre-Service Time

CPT Code: 33120 63.00

Key Reference CPT Code: 33426 95.00

Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

205.00 60.00 70.0 210.0 38.0 40.0 0.0 686.00

205.00 40.00 70.00 265.00 38.00 63.00 0.00 776.00

We submit that a careful review of these data would suggest that 33120 can reasonably be valued at the RUC-recommended 25th percentile physician work value of 42.88, which is 0.4 RVU less than the value of 33426 at 43.28 RVW. The intra-service times are identical. While there is somewhat less hospital time and office visit time as well as less pre-service time (although this is in part due to the introduction of pre-service packages), this is more than offset by the substantially increased measures of intensity and complexity as indicated in Table 7: Table 7 INTENSITY/COMPLEXITY MEASURES

(Of those that selected

Administrator Berwick CMS-3248-P July 25, 2011 Page 12 (Mean) Mental Effort and Judgment (Mean)

Key Reference code) 33120 33426

The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making

3.57

3.43

3.87

3.48

3.70

3.26

Technical skill required

4.17

3.74

Physical effort required Psychological Stress (Mean)

3.91

3.48

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean) Pre-Service intensity/complexity

3.96

3.35

4.22

3.78

3.70

3.26

3.78

3.48

Intra-Service intensity/complexity

3.87

3.52

Post-Service intensity/complexity

3.04

2.74

Technical Skill/Physical Effort (Mean)

We believe that the CMS proposal of 36.00, or 7.28 less than 33426, is unsupportable and departs from the norms utilized to maintain appropriate relativity within the physician fee schedule. 33980 (Removal of ventricular assist device, implantable intracorporeal) – RUC Recommendation: 40.00, CMS Recommendation: 33.50 CMS proposes the 25th percentile value for this code, 33.50, rather than the RUC-recommended median survey value of 40.00. Of note, this is the only one of the nine Ventricular Assist Device (VAD) codes valued here, and the only one where the 25th percentile survey value was not recommended by the RUC and the specialty society. CMS provides a rationale that there has been a significant reduction in the physician times and the number and level of post-operative visits that the RUC included in the value of CPT code 33980. This is inaccurate, as the 33980 was surveyed as an XXX code with no post-operative

Administrator Berwick CMS-3248-P July 25, 2011 Page 13 visits and cannot be compared in this way to any 90 day global without explanation. Nonetheless, it appears that CMS is not satisfied by the RUC rationale for this recommendation. STS urges CMS to reconsider, as this code is one of the most intense, complex, and demanding procedures that our specialty performs. This is consistent with our recommendation of the median value, something that is not done lightly nor without substantial justification. In the first place, this is an obligatory reoperation which is almost always performed during a one-six month time frame when the adhesions are new, tenacious, and very vascular. The reoperation code 33530 would not apply here, but its value (10.13 as a ZZZ code) should certainly be considered in this recommendation. Secondly, this procedure requires reconstruction of the large bore defect in the apex of the left ventricle, which is technically demanding, particularly in patients destined for survival with a fragile and compromised left ventricle that must now support the circulation without VAD support. These features certainly justify the RUC-recommended value, which results in an IWPUT of 0.121. This IWPUT is appropriate for this procedure, and cardiothoracic surgeons are the primary providers for 16 of the 58 existing codes with calculated IWPUT in the range of 0.115 to 0.125. The CMS-proposed value (which was considered and rejected by the RUC facilitation committee) would result in an IWPUT of 0.099, which would be the lowest IWPUT in this family when it should be the highest. As CMS seeks a better rationale for valuing this code, it should be noted that (and we believe this may be what CMS means in its published rationale) this code was originally surveyed as a 90-day global during the Five-Year Review. That 90-day global survey median value, as an XXX code without visits and with the same XXX inputs, would have been 37.25 with an identical intra-service time of 300 minutes and an IWPUT of 0.111 which is similar to the XXX survey supporting this recommendation. Alternatively, one could employ the existing IWPUT for the reoperation code (0.115) and build a value using the survey inputs as follows: 0.115*300+0.0224*(60+15+90)+0.008*20= 38.36 RVW We would appreciate if CMS would reconsider either the median value as recommended by the RUC (40 RVW) or one of the two mutually supporting values outlined above (RVW=37.25 or RVW=38.36), which would fall between the 25th and 50th percentiles of the XXX survey. Congenital Codes STS has particular concerns regarding several of the proposed valuations of the codes for congenital heart surgery. First, because these are codes for congenital heart surgery procedures, it is not surprising that there are extremely low numbers of cases using these codes in the Medicare database. Although the small numbers of cases result in an infinitesimal impact on the Medicare Fee Schedule, CMS must recognize that the Medicare Fee Schedule is the basis for the fee schedules of more than 70% of the private payers in the U.S. Therefore the values for the codes that are assigned by CMS for congenital heart surgery procedures have a much greater impact outside the agency‟s immediate fiscal responsibilities. Secondly, the pattern of uniform rejection of the RUC recommendations at the 50th percentile of the survey results by CMS and the recommendation for the 25th percentile of the survey for each of these codes suggests that there was no individual consideration of each code. In each case, the stated CMS

Administrator Berwick CMS-3248-P July 25, 2011 Page 14 rationale for proposing the 25th percentile consisted of a CMS judgment that the 25th percentile was “more appropriate.” STS devoted considerable physician time and resources to conducting the surveys that the RUC carefully considered during the Five-Year Review process. STS-proposed values for each code at the 50th percentile of the survey values, and both the multidisciplinary RUC workgroup and the full RUC approved the surveyed 50th percentile values. CMS representatives were present for the indepth review by the workgroup. Detailed explanations for each of the congenital heart surgery codes are provided on a code by code basis. STS respectfully requests that CMS reconsider its approach to the valuations of these congenital heart surgery codes for the reasons outlined above and additional rationale provided for each code below. 33412 (Replacement, aortic valve; with transventricular aortic annulus enlargement) – RUC Recommendation: 60.00, CMS Recommendation: 59.00 CMS did not agree with the RUC recommendation of 60.00 work RVUs for CPT code 33412, (the median survey work estimate). Instead, CMS proposes the survey 25th percentile value of 59.00 work RVUs. STS does not understand CMS‟s decision to reduce the value for this service by 1.7% without a meaningful rationale. The RUC workgroup closely reviewed this service and compared it to key reference service CPT code 33782 (Aortic root translocation with ventricular septal defect and pulmonary stenosis repair (i.e., Nikaidoh procedure); without coronary ostium reimplantation) (work RVU = 60.08 and intra-time = 300 minutes). STS and the RUC agreed that these two services require the same intensity and complexity, physician work and time to perform. This is evident from the survey results included in tables 8 and 9: Table 8 TIME ESTIMATES (Median)

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

Table 9

CPT Code: 33412 63.00 300.00 60.00 140.00 225.00 38.00 40.00 0.0 866.00

Key Reference CPT Code: 33872 63.00 300.00 60.00 140.00 225.00 38.00 40.00 0.00 866.00

Administrator Berwick CMS-3248-P July 25, 2011 Page 15 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean)

(of those that selected Key Reference code) 33412 33872

The number of possible diagnosis and/or the 4.72 number of management options that must be considered The amount and/or complexity of medical records, 4.61 diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making 4.06 Technical Skill/Physical Effort (Mean)

4.61

Technical skill required Physical effort required Psychological Stress (Mean)

4.89 4.83

4.83 4.78

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean) Pre-Service intensity/complexity

4.83

4.78

4.94

4.89

4.11

4.06

4.50

4.39

Intra-Service intensity/complexity Post-Service intensity/complexity

4.83 4.17

4.89 4.17

4.56

4.06

33468 (Tricuspid valve repositioning and plication for Ebstein anomaly) – RUC Recommendation: 50.00, CMS Recommendation: 45.13 CMS did not agree with the RUC-recommended work RVU of 50.00 for CPT code 33468. CMS proposes the survey 25th percentile work RVU of 45.13. STS does not understand the rationale provided by CMS for valuing this service at this work RVU. When the RUC reviewed this service it compared CPT code 33468 to key reference service CPT code 33427, (Valvuloplasty, mitral valve, with cardiopulmonary bypass; radical reconstruction, with or without ring) (work RVU = 44.83 and intratime = 221 minutes). The specialty society indicated and the RUC agreed that 33468 is more intense and complex, and requires more physician work and time to perform. This is evident from the survey results in tables 10 and 11:

Administrator Berwick CMS-3248-P July 25, 2011 Page 16 Table 10 TIME ESTIMATES (Median)

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

CPT Code: 33468 63.00 240.00 60.00 140.00 225.00 38.00 40.00 0.0 806.00

Key Reference CPT Code: 33427 90.00 230.00 60.00 70.00 115.00 38.00 23.00 0.00 626.00

Table 11 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean) The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making Technical Skill/Physical Effort (Mean)

(of those that selected Key Reference code) 33468 33427 4.44 3.88 4.50

3.88

3.69

3.31

Technical skill required

4.75

4.19

Physical effort required Psychological Stress (Mean)

4.19

3.88

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor

4.52

3.96

4.77

4.27

3.79

3.60

Administrator Berwick CMS-3248-P July 25, 2011 Page 17 outcome Time Segments (Mean) Pre-Service intensity/complexity Intra-Service intensity/complexity Post-Service intensity/complexity

4.38 4.63 4.13

3.94 4.19 3.65

The CMS-proposed work value is not relative to the key reference service, performed by cardiothoracic surgeons and is not appropriate given the strong relationship between the two services. The CMSproposed work value does not correct the rank order anomaly that existed relative to the reference service, and now would persist despite direct survey evidence and RUC agreement to the contrary. Therefore, STS requests that CMS accept the RUC recommendation of 50.00 work RVUs for CPT code 33468. 33645 (Direct or patch closure, sinus venosus atrial septal defect, with or without anomalous pulmonary venous drainage) – RUC Recommendation: 33.00, CMS Recommendation: 31.30 CMS did not agree with the RUC-recommended work RVU of 33.00 for CPT code 33645. CMS proposes the survey 25th percentile work RVU of 31.30. STS, again, does not understand CMS‟s rationale for valuing this service at this work RVU. When the RUC reviewed this service it compared 33645 to key reference service CPT codes 33641, (Repair atrial septal defect, secundum, with cardiopulmonary bypass, with or without patch) (work RVU = 29.58 and intra-time = 164 minutes) and 33681, (Closure of single ventricular septal defect, with or without patch) (work RVU = 32.34 and intratime = 150 minutes). The specialty society indicated and the RUC agreed that 33645, (Surveyed intraservice time = 175 minutes) requires more intensity and complexity to perform compared to these reference services as demonstrated in tables 12 and 13: Table 12 TIME ESTIMATES (Median)

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

CPT Code: 33645 63.00 175.00 45.00 70.00 115.0 38.0 40.00 0.0 546.00

Key Reference CPT Code: 33641 95.00 164.00 40.00 70.00 115.00 38.00 40.00 0.00 562.00

Administrator Berwick CMS-3248-P July 25, 2011 Page 18 Table 13 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean)

(of those that selected Key Reference code 33645 33641

The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making Technical Skill/Physical Effort (Mean)

2.50

1.96

2.54

1.96

2.04

1.67

Technical skill required

2.63

2.00

Physical effort required Psychological Stress (Mean)

2.42

2.00

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean) Pre-Service intensity/complexity Intra-Service intensity/complexity

2.38

1.75

3.00

2.25

3.46

3.29

2.04 2.25

1.71 1.75

Post-Service intensity/complexity

1.83

1.67

The RUC agreed with STS that the survey median work RVU of 33.00 for 33645 appropriately accounts for the work required to perform this service, as well as establishes the correct relativity among other similar services. STS also would note that there is an increase in total time for 33645 compared to the key reference code 33641 of 16 minutes if the same pre-service package were employed. Therefore, STS requests that CMS accept the RUC recommendation of 33.00 for CPT code 33645. 33647 (Repair of atrial septal defect and ventricular septal defect, with direct or patch closure) – RUC Recommendation: 35.00, CMS Recommendation: 33.00 CMS did not agree with the RUC-recommended work RVU of 35.00 for CPT code 33647. CMS proposes the survey 25th percentile work RVU of 33.00. Again, STS does not understand CMS‟s rationale for valuing this service at this work RVU. When the RUC reviewed this service it compared 33647 to key reference service CPT code 33681, (Closure of single ventricular septal defect, with or

Administrator Berwick CMS-3248-P July 25, 2011 Page 19 without patch) (work RVU = 32.34 and intra-time = 150 minutes). The specialty society indicated and the RUC agreed that 33647 is similarly intense and complex, and requires more physician work and time to perform as supported by the survey in tables 14 and 15: Table 14 TIME ESTIMATES (Median)

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

CPT Code: 33647 63.00 180.00 53.00 70.00 170.00 38.00 40.00 0.0 612.00

Key Reference CPT Code: 33681 52.50 150.00 60.00 0.00 120.00 38.00 86.00 0.00 506.50

Table 15 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean)

(of those that selected Key Reference code 33647 33681

The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making Technical Skill/Physical Effort (Mean)

2.67

2.64

2.58

2.61

2.55

2.55

Technical skill required

3.12

3.09

Physical effort required

2.85

2.85

2.85

2.70

3.18

3.21

3.42

3.39

Psychological Stress (Mean) The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor

Administrator Berwick CMS-3248-P July 25, 2011 Page 20 outcome Time Segments (Mean) Pre-Service intensity/complexity

2.73

2.70

Intra-Service intensity/complexity Post-Service intensity/complexity

2.91 2.52

2.85 2.52

Compared to the reference service code, 33647 includes all the services and in addition the closure of an atrial septal defect. Thus, there are 30 additional minutes of intra-service time as well as additional hospital visit time and total time; and indeed another cardiac defect repaired as compared to the reference code. CMS‟s proposed value of 33.00 provides an increment of only 0.66 RVU for this difference, which is completely inconsistent with the direct comparison observed above. For example, the additional intra-service time of 30 minutes at the same intensity as the reference code (by intensity survey) would result in an additional 4.11 work units alone. This, alone, would thus support a value of 36.45 for 33647, and thus easily confirms the RUC-recommended median value of 35.00. Therefore, STS requests that CMS accept the RUC recommendation of 35.00 for CPT code 33647. 33692 (Complete repair tetralogy of Fallot without pulmonary atresia) – RUC Recommendation: 38.75, CMS Recommendation: 36.15 STS appreciates that CMS has corrected the proposed PLI RVU for this code from 2.56 to 9.11 in other rule making, and has also proposed correcting this for the other procedures where STS identified incorrect identification of the dominant specialty provider. CMS did not agree with the RUC-recommended work RVU of 38.75 for CPT code 33692. CMS proposes the survey 25th percentile work RVU of 36.15. STS again does not understand CMS‟s rationale for value this service at this work RVU. When the RUC reviewed this service it compared the service to key reference service CPT code 33684, (Closure of single ventricular septal defect, with or without patch; with pulmonary valvotomy or infundibular resection (acyanotic)) (work RVU = 34.37 and intratime = 200 minutes). STS indicated and the RUC agreed that 33692 is more complex and requires more physician work and time to perform. This is supported by the survey results as indicated in tables 16 & 17: Table 16 TIME ESTIMATES (Median)

CPT Code: 33692

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time

63.00 218.00 60.00 70.00

Key Reference CPT Code: 33684 93.00 200.00 60.00 70.00

Administrator Berwick CMS-3248-P July 25, 2011 Page 21 Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

195.00 38.00 40.00 0.0 684.00

Table 17 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean)

115.00 38.00 40.00 0.00 616.00

(of those tat selected Key Reference code) 33692

33684

The number of possible diagnosis and/or the number of management options that must be considered

3.45

3.00

The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making

3.38

3.03

3.31

2.76

Technical skill required

3.83

3.38

Physical effort required

3.59

3.24

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean)

3.72

3.28

4.00

3.52

3.62

3.28

Pre-Service intensity/complexity

3.24

3.10

Intra-Service intensity/complexity

3.62

3.31

Post-Service intensity/complexity

3.28

3.10

Technical Skill/Physical Effort (Mean)

Psychological Stress (Mean)

In reviewing this code, compared to the reference service code, STS puts forward a building block method to support the magnitude estimation median RVW of 38.75. In calculating this value from the survey inputs, we assume that the IWPUT of the surveyed code is equal to the reference service code, rather than significantly greater as was found in the survey, as demonstrated in Table 18:

Administrator Berwick CMS-3248-P July 25, 2011 Page 22 Table 18 Building Block RVW for 33692 at same intensity/complexity as Key Reference Service 33684 33684 33692 Value 33684 33692 Pre eval 63 40 X 0.0224 1.4112 0.896 Positioning 15 3 X 0.0224 0.336 0.0672 SDW 15 20 X 0.0081 0.1215 0.162 SD Post 60 60 X 0.0224 1.344 1.344 Intraservice 200 218 X 0.0986 19.728 21.50352 99291 1 1 4.5 4.5 4.5 99233 1 1 2 2 2 99232 1 2 1.39 1.39 2.78 99231 1 3 0.76 0.76 2.28 99238 1 1 1.28 1.28 1.28 99214 1 1 1.5 1.5 1.5 Total RVW

34.37

38.31

This demonstrates that the additional value, compared to the reference code, as recommended by the RUC, is related primarily to the work in the three additional days of hospital length of stay (one 99232 and two 99231 and the additional 18 minutes of intra-service time). This results in a calculated value of 38.31 for 33692, which resolves to 38.75 if the IWPUT is increased to 0.100 as supported by the survey and well within the IWPUT range for congenital heart surgery. Put another way, STS objects to the CMS-proposed value which provides only 1.78 RVW for 3 additional hospital days and 18 additional minutes of more complex surgery, and instead requests that CMS accept the RUC recommendation of 38.75 work RVUs for CPT code 33692. 33710 (Repair sinus of Valsalva fistula, with cardiopulmonary bypass; with repair of ventricular septal defect) – RUC Recommendation: 43.00, CMS Recommendation: 37.50 CMS did not agree with the RUC-recommended work RVU of 43.00 for CPT code 33710. CMS proposes the survey 25th percentile work RVU of 37.50. STS again does not understand CMS‟s rationale for valuing this service at this work RVU. When the RUC reviewed this service it compared 33710 to key reference service CPT code 33405, (Replacement, aortic valve, with cardiopulmonary bypass; with prosthetic valve other than homograft or stentless valve) (work RVU = 41.32 and intra-time = 198 minutes). The STS survey indicated and the RUC agreed that code 33710 requires more intensity and complexity, physician work and time to perform than the reference code 33405 as follows:

Administrator Berwick CMS-3248-P July 25, 2011 Page 23 Table 19 TIME ESTIMATES (Median)

Median Pre-Service Time Median Intra-Service Time Median Immediate Post-service Time Median Critical Care Time Median Other Hospital Visit Time Median Discharge Day Management Time Median Office Visit Time Prolonged Services Time Median Total Time

Table 20 INTENSITY/COMPLEXITY MEASURES (Mean) Mental Effort and Judgment (Mean) The number of possible diagnosis and/or the number of management options that must be considered The amount and/or complexity of medical records, diagnostic tests, and/or other information that must be reviewed and analyzed Urgency of medical decision making

CPT Code: 33710 63.00 200.00 60.00 140.00 115.00 38.00 40.00 0.0 656.00

Key Reference CPT Code: 33405 95.00 198.00 40.00 70.00 265.00 38.00 63.00 0.00 769.00

(of those that selected Key Reference code) 33710 33405 3.40

3.20

3.20

2.90

3.10

2.80

Technical skill required Physical effort required Psychological Stress (Mean)

3.90 3.10

3.70 2.80

The risk of significant complications, morbidity and/or mortality Outcome depends on the skill and judgment of physician Estimated risk of malpractice suit with poor outcome Time Segments (Mean)

3.10

3.30

3.70

3.60

3.60

3.30

Pre-Service intensity/complexity

3.10

3.90

Technical Skill/Physical Effort (Mean)

Administrator Berwick CMS-3248-P July 25, 2011 Page 24 Intra-Service intensity/complexity

3.30

3.10

Post-Service intensity/complexity

2.50

2.60

Accordingly, the CMS proposal, which values 33710 below the key reference service, is not relative and does not appropriately align itself with other similar services performed by cardiothoracic surgeons. The RUC agreed with STS that the survey median work RVU of 43.00 for 33710 appropriately accounts for the work required to perform this service as well as establishes the correct relativity among other similar services. STS requests that CMS accept the RUC recommendation of 43.00 work RVUs for CPT code 33710. Finally, STS would appreciate clarification on the CMS acceptance of the RUC recommendation that the work value for 33411 remain at 62.07 without comment. This code was brought forward by CMS due to a change in site of service from virtually 100% in-patient to predominantly out-patient. Since this procedure involves a median sternotomy, surgical replacement of the aortic valve with enlargement of the aortic root, and obligatory cardiopulmonary bypass with elective cardiac arrest, it is clearly not an outpatient procedure. Despite requests to deal with this administratively as an obvious data error, STS and RUC staff were required to spend considerable time and effort to seek out an explanation for this error. That investigation revealed no systematic changes in billing for this procedure, for example by another specialty or in a geographic area for this low-volume code. The most likely cause of the error therefore resides within the CMS database and CMS methodology in determining outpatient billing status. Given that many of CMS‟s rejections of RUC recommendations are related to small changes in site-of-service leading to a shift of predominance to out-patient status, the credibility of these data are of considerable importance. We request that CMS determine the source of this error and reassure the RUC and the physician community of the accuracy of its site of service information. On behalf of STS, thank you for the opportunity to provide these comments on the Five-Year Review of Work Relative Value Units under the Physician Payment Schedule. If you have any questions, please contact Phil Bongiorno, STS Director of Government Relations, at (202) 787-1221 or [email protected]. Sincerely,

Michael J. Mack, MD