1 MEDICARE PAYMENT ADVISORY COMMISSION PUBLIC MEETING The Horizon Ballroom Ronald Reagan Building International Trade Center 1300 Pennsylvania Avenue, N.W. Washington, D.C. Thursday, January 12, 2012 9:33 a.m. COMMISSIONERS PRESENT: GLENN M. HACKBARTH, JD, Chair ROBERT BERENSON, MD, FACP, Vice Chair SCOTT ARMSTRONG, MBA KATHERINE BAICKER, PhD MITRA BEHROOZI, JD KAREN R. BORMAN, MD PETER W. BUTLER, MHSA RONALD D. CASTELLANOS, MD MICHAEL CHERNEW, PhD THOMAS M. DEAN, MD WILLIS D. GRADISON, MBA WILLIAM J. HALL, MD HERB B. KUHN GEORGE N. MILLER, JR., MHSA MARY NAYLOR, PhD, RN, FAAN BRUCE STUART, PhD CORI UCCELLO, FSA, MAAA, MPP 2 AGENDA Assessing payment adequacy: hospital inpatient and outpatient services - Jeff Stensland, Craig Lisk, Dan Zabinski Assessing payment adequacy: physician and other health professional services/ambulatory surgical center services - Ariel Winter, Dan Zabinski, Cristina Boccuti, Kevin Hayes PAGE 6 76 Assessing payment adequacy: outpatient dialysis services - Nancy Ray 101 Assessing payment adequacy: hospice services - Kim Neuman 113 Public Comment 128 Assessing payment adequacy: skilled nursing facility services - Carol Carter 135 Assessing payment adequacy: inpatient rehabilitation facility services - Christine Aguiar, Craig Lisk 161 Assessing payment adequacy: Long-term care hospital services - Dana Kelley 177 Encouraging the use of lower-cost medications by beneficiaries receiving Part D’s low-income subsidy - Shinobu Suzuki 186 The Medicare Advantage program: status report - Scott Harrison, Carlos Zarabozo 218 Public comment 270 3 1 P R O C E E D I N G S 2 MR. HACKBARTH: Okay. [9:33 a.m.] Good morning. Our agenda 3 today consists primarily of recommendations on update 4 factors for inclusion in our March report. 5 each of these at some length in our December meeting, and so 6 the discussions we will be having today will be a shortened 7 version. 8 information that was presented in December in the interest 9 of time because, in addition to our update recommendations, We discussed We are not going to review all of the same 10 we have some other issues that we want to get into the 11 agenda for the meeting today and tomorrow. 12 So for those of you in the audience who did not 13 attend the December session and are just entering the 14 conversation at this point, I want to emphasize that there 15 have been extensive discussions of these issues already in 16 the Commission. 17 I want to say a word about the legislative context 18 for the update recommendations this year. As everybody 19 knows, as part of the debt ceiling legislation, a committee, 20 a congressional special committee, was created to develop a 21 deficit reduction package to hit a particular target. 22 did not succeed in doing that, and so under the terms of the They 4 1 debt ceiling legislation, the alternative is a 2 percent 2 sequester in payment rates in Medicare to take effect in 3 February 2013. 4 not yet been formally triggered. 5 later point, but the common assumption is that the sequester 6 seems likely. 7 In formal legal terms, that sequester has That would happen at a Of course, there is the possibility that Congress 8 could pass new legislation between now and February that 9 would alter that. But, again, many observers are assuming 10 that the sequester will take effect in February 2013. 11 that raises the question of what are the implications of 12 that sequester for MedPAC's update recommendations that we 13 will be voting on today. 14 So We think of our update recommendations as 15 recommendations for a percentage increase in the existing 16 base rates in the various payment systems for physicians, 17 hospitals, nursing homes, home health agencies, and the 18 like. 19 There is a base rate, and then off of that base rate, 20 various adjustments are made for differences in wages, 21 differences in case mix, and the like to calculate the rate 22 for a particular service that is paid to a provider. So each of those payment systems has a schedule. 5 1 When we make update recommendations, they are 2 recommendations for a percentage change in that base rate. 3 And so just for the sake of illustration, if you think of 4 the base rate for a provider category as $100 and we 5 recommend a 1 percent update, what we're saying is we think 6 the base rate for that category of providers should go to 7 $101 for the fiscal year in question, and we are making 8 recommendations now for fiscal year 2013. 9 The sequester, if it were to occur, also would 10 affect those base rates. 11 the base rate below our $101 recommendation in the example, 12 that would be contrary, lower than what MedPAC recommended. 13 And so we're not going to make a sequester adjustment. 14 simply will recommend what we think the appropriate base 15 rates are. 16 recommendations, whether to allow the sequester to continue, 17 whether to make any modifications. 18 there, but you won't hear us referring to it in each of our 19 discussions. 20 existing base rates should be adjusted in percentage terms. 21 22 If the sequester were to reduce We Then Congress will make its judgment on our So the sequester is out We will make recommendations on how the With that preface, let's turn to our first session on hospital inpatient and outpatient services. 6 1 DR. STENSLAND: All right. Good morning. This 2 session will address issues regarding Medicare payments to 3 hospitals. 4 topics were covered in December, and the details are all in 5 your mailing materials. 6 As Glenn said, we're going to be concise. These First, we will recap how our payment adequacy 7 findings have led to the draft update recommendation for 8 inpatient and outpatient hospital payments, and then Dan 9 will discuss the rationale behind the draft recommendation 10 11 to adjust E&M payment rates. To evaluate payment adequacy, we use a common 12 framework across all sectors. 13 examine capacity, service volume, quality of care, access to 14 capital, as well as providers' costs and payments for 15 Medicare services. 16 When data is available, we This is the first set of payment adequacy 17 discussions you will hear today. 18 payment adequacy for the other sectors later today will use 19 this same set of indicators when the data is available. 20 The analysts discussing Recall that we have discussed how capacity was 21 increasing and how access to capital is adequate. We see 22 strong volume growth in outpatient services and a slight 7 1 decline in inpatient volume. 2 measures, particularly the growth in outpatient volumes, 3 point to adequate payments. 4 These capacity and volume All quality-of-care indicators are either 5 improving or stable. We monitor improvements in 30-day 6 mortality and see improvements in all the conditions we 7 monitor. 8 slightly. 9 Hospitals will have a stronger incentive to reduce We also see patient satisfaction has also improved However, readmission rates have not improved. 10 readmissions when a new readmission penalty starts in 2013. 11 As we discussed last month, margins improved from 12 2008 to 2010, and this was due to two factors: 13 First, with the introduction of the MS-DRGs, 14 documentation and coding changes resulted in increased 15 Medicare payments; 16 Second, with the downturn in the economy at the 17 end of 2008, hospitals responded by tightening expense 18 control. 19 2010. 20 The result was slower cost growth in 2009 and The combination of higher payment growth and 21 slower cost growth led to the better margins you see on this 22 slide. 8 1 We estimate that overall Medicare margins will go 2 from negative 4.5 percent in 2010 to negative 7 percent in 3 2012, which is about where margins were in 2008 prior to the 4 full effect of the documentation and coding on revenues. 5 The drop is primarily due to reductions in 6 inpatient updates that occurred in fiscal years 2011 and 7 2012 to partially offset changes in hospitals' documentation 8 and coding. 9 the update resulted in inpatient payments increasing by only 10 These documentation and coding adjustments to 1 percent from 2010 to 2012. 11 We also expect cost growth to increase from 12 roughly 2 percent to 3 percent per year. 13 prices and hospital costs to increase due to an improving 14 economy and a reduction in hospitals' pressure to constrain 15 costs after achieving high all-payer margins in 2010. 16 expect the net result will be slower payment growth and 17 faster cost growth, causing approximately a 2 or 3 percent 18 decline in margins over the two-year period from 2010 to 19 2012. 20 We expect input We It's important to note that the 2010 margin of 21 negative 4.5 percent is for the average hospital. There is 22 a subset of relatively efficient hospitals that on average 9 1 kept their 2010 costs down below Medicare payment rates. 2 As you recall from last month, we showed you that 3 this group of relatively efficient hospitals has been able 4 to keep its costs roughly 10 percent below average while 5 having better patient outcomes on average. 6 allowed them to generate a 4 percent positive Medicare 7 margin on hospital services. 8 9 Lower costs At the December meeting, Mike asked what was driving their 10 percent lower costs. He was wondering 10 whether it was fixed or variable costs, and I want to say 11 that each individual hospital has its own story and there is 12 a wide variation in costs. 13 this. 14 the more efficient group: 15 I wouldn't read too much into However, on average, there are two tendencies amongst The first tendency is higher occupancy. Due to 16 higher occupancy, the median efficient hospital has 10 17 percent more discharges for every bed than the average 18 hospital, and this could explain why they have roughly 10 19 percent lower building and 10 percent lower equipment costs 20 per discharge than the average hospital. 21 spreading their fixed costs over more cases. 22 They benefit from The second difference is in the distribution of 10 1 labor costs. Labor costs are also about 10 percent lower 2 for the efficient group, meaning the whole labor cost pie is 3 about 10 percent smaller. 4 hospitals tend to spend a little bit larger share of their 5 labor costs on direct patient care. 6 efficient hospitals tend to spend a little bit larger share 7 of their costs on indirect costs such as administrative and 8 general expenses. However, the relatively efficient In contrast, the less 9 Given the data presented on payment adequacy and 10 the need to recover past overpayments due to documentation 11 and coding, the draft recommendation now reads: 12 Congress should increase payment rates for the inpatient and 13 outpatient prospective payment systems in 2013 by 1 percent. 14 For inpatient services, the Congress should also require the 15 Secretary, beginning in 2013, to use the difference between 16 the increase under current law and MedPAC's recommended 17 update to gradually recover past overpayments due to 18 documentation and coding changes. 19 that The spending implication for 2013 is that it would 20 decrease spending by roughly between $750 million and $2 21 billion over one year -- probably close to $2 billion -- and 22 by over $10 billion over five years. These are our 11 1 preliminary estimates, and they are subject to change. 2 see no adverse impacts on beneficiaries from this 3 recommendation. 4 We Just to recap the rationale behind the update 5 recommendation, adjustments for documentation and coding are 6 needed to recover overpayments from 2010, 2011, and 2012. 7 CMS needs new authority from Congress to make these 8 recoveries. 9 recoveries to cause a financial shock to hospitals. However, we do not want the magnitude of the Given 10 the payment adequacy indicators, a 1 percent update is 11 sufficient to preserve payment adequacy for reasonably 12 efficient hospitals. 13 the 1 percent update could be applied to fully recover all 14 overpayments due to documentation and coding changes. 15 The difference between current law and The 1 percent increase on the outpatient side is 16 appropriate for two reasons: 17 volume growth of 4 percent. 18 projected update under current law of is probably too high. 19 However, a 0 percent update may be too low given the current 20 profitability of outpatient services. 21 percent update appears reasonable. 22 First, we see outpatient Given this volume growth, the Therefore, a 1 Part of the rapid growth in outpatient services is 12 1 due to a shift in the site of services from physician 2 offices for evaluation and management visits to hospital- 3 based offices. 4 Dan will now discuss that concern. DR. ZABINSKI: In the December meeting, we 5 discussed the issue of increased hospital employment of 6 physicians. 7 of services to shift from free-standing practices to OPDs 8 and that this shift has been accelerating. 9 Data suggests that this has caused the billing For example, for E&M office visits provided in 10 OPDs or in free-standing physician practices, the percentage 11 that is provided in OPDs has increased at an annual rate of 12 3.5 percent per year over 2004 through 2008, but by 9.9 13 percent in 2009 and by 12.9 percent in 2010. 14 incentives are present for this acceleration to continue. 15 A problem related to this shift is that payment Moreover, 16 rates for the same service are typically much higher in the 17 outpatient prospective payment system than in the Physician 18 Fee Schedule. 19 office visit is about 80 percent higher in the OPD than in a 20 free-standing practice. 21 services shifting from free-standing practices to OPDs is 22 increased program spending and beneficiary cost sharing even For example, Medicare payment for a mid-level Consequently, the result of 13 1 though the care received by the patient may not change at 2 all. 3 office visit is $15 if provided in the free-standing 4 physician practice but $25 if provided in an OPD. For example, beneficiary cost sharing for a 15-minute 5 So how should the Medicare program address this 6 issue of higher payment rates when the services are provided 7 in an OPD compared to a free-standing practice? 8 9 A simple approach would be to set the payment rates in the outpatient PPS so that payment rates for all 10 services are the same whether provided in an OPD or a free- 11 standing practice. 12 consider the following questions for services provided in 13 OPDs: However, for some services we need to 14 Do OPDs have more complex patients? 15 Do OPDs maintain standby capacity for that 16 17 18 19 service? And does the outpatient PPS have greater packaging of ancillary services than the Physician Fee Schedule? For E&M office visits indicated by CPT codes 99201 20 through 99215, we have found that patient complexity is 21 addressed through the CPT codes for these services; the 22 costs of standby capacity are generally allocated to other 14 1 areas of the hospital; and the cost of packaged ancillaries 2 is small when these services are provided in OPDs, about two 3 dollars of the total cost for these services. 4 believe it is reasonable to have equal payment rates across 5 these two sectors for E&M office visits. 6 Therefore, we If this policy is fully phased in, it would reduce 7 hospitals' overall Medicare revenue by 0.6 percent and 8 outpatient Medicare revenue by 2.8 percent. 9 I bring up the term "phase-in" because in December 10 there was general agreement on this policy in principle, but 11 there was concern that it could adversely affect some 12 hospitals that are longstanding institutions that provide 13 access to primary care for low-income patients in their 14 community. 15 To ease that transition, a three-year phase-in was 16 suggested. 17 reduce Medicare revenue by about 0.2 percent for each year 18 of the phase-in. 19 fully phased-in policy is relatively small, as 78 percent of 20 hospitals would have their overall Medicare revenue reduced 21 by less than 0.5 percent. 22 Under a phase-in, we found this policy would Also, for most hospitals, the effect of a However, the effect on hospital revenue differs by 15 1 hospital group. 2 decline by 1.1 percent among major teaching hospitals but by 3 0.4 percent among other teaching hospitals and non-teaching 4 hospitals. 5 We estimate that Medicare revenue would In addition, last month we were asked the effect 6 of this policy on efficient providers. 7 it would reduce their Medicare revenue by 0.7 percent and 8 drop their overall Medicare margin from just under 4 percent 9 to just over 3 percent. 10 The answer is that Finally, there is wide variation of the effect on 11 Medicare revenue of this policy as 10 percent of hospitals 12 would see no effect on their Medicare revenue and 10 percent 13 would have Medicare revenue decrease by at least 1.2 14 percent. 15 As I mentioned on the previous slide, there was 16 agreement among Commissioners about the general principle of 17 this policy, but some were concerned about the transition to 18 the policy for hospitals that are a critical source of 19 primary care to low-income patients. 20 transition, a three-year phase-in was suggested. 21 22 To ease the Features of the phase-in include that the impact of this policy would be limited to 2 percent of overall 16 1 Medicare revenue for hospitals with disproportionate share 2 percentages of 0.25 or higher, where 0.25 is the median DSH 3 percentage among all hospitals and the DSH percentage is the 4 sum of the percentage of Medicare inpatient days that are 5 for patients who are eligible for SSI plus the percentage of 6 total inpatient days that are for patients who have 7 Medicaid. 8 would affect about 4 percent of all hospitals or about 120 9 hospitals. 10 In the final year of the phase-in, this policy At the December meeting, Peter wanted to know the 11 profile of the hospitals that would be most affected by this 12 policy, so we analyzed the 4 percent of hospitals that would 13 have their losses limited during the phase-in. 14 We found that there is a broad mix of hospitals in 15 this group, but as the first column in this table indicates, 16 these hospitals do have some different characteristics from 17 other hospitals. 18 government-owned and have major teaching status; they have a 19 higher percentage of Medicaid patients; and they have a 20 lower all-payer margin. 21 22 In particular, they are more likely to be However, these hospitals also have a better Medicare margin than other hospitals, probably because of 17 1 the relatively high payments for their DSH and teaching 2 status. 3 So based on a goal of making payment rates for 4 office visits equal across free-standing practices and OPDs, 5 we have the following draft recommendation: 6 should direct the Secretary of Health and Human Services to 7 reduce payment rates for evaluation and management office 8 visits provided in outpatient departments so that total 9 payment rates for these visits are the same whether the The Congress 10 service is provided in an outpatient department or a 11 physician's office. 12 three years. 13 hospitals with a disproportionate share patient percentage 14 at or above the median should be limited to 2 percent of 15 overall payments. 16 17 These changes should be phased in over During the phase-in, payment reductions to That is long enough to deserve a copyright, I think. 18 [Laughter.] 19 DR. ZABINSKI: Okay. The spending implication is 20 that it is expected to decrease spending for 2013 and over 21 five years because of lower payment rates in the outpatient 22 PPS. For beneficiaries and providers, this policy may slow 18 1 or stop the shift of services from free-standing practices 2 to OPDs; it will reduce beneficiary cost sharing; and 3 because of the lower OPD payment rates, we may need to 4 monitor beneficiaries' access to these services. 5 In addition, it may be prudent to have a study 6 that evaluates how this policy affects access to ambulatory 7 physician services among vulnerable populations, so we have 8 this additional draft recommendation. 9 The Secretary of Health and Human Services should 10 conduct a study by January 2015 to examine whether access to 11 ambulatory physician services for low-income patients would 12 be impaired by setting outpatient evaluation and management 13 payment rates equal to those paid in physician offices. 14 access will be impaired, the Secretary should recommend 15 actions to protect access. 16 If This recommendation should have no effect on 17 program spending, and for beneficiaries and providers, it 18 may help identify problems beneficiaries are having in 19 regard to access to ambulatory physician services. 20 The rationale behind this draft recommendation is 21 to determine if there is a set of hospitals that serve as 22 the vital source of clinic-based services for low-income 19 1 2 patients in the community. The purpose of the study is identify any financial 3 difficulties these hospitals face due to this policy that 4 may affect low-income patients' access to primary care. 5 if low-income patients are found to be at risk, the 6 Secretary could recommend appropriate policy changes. 7 8 9 10 And Now I'll turn things over to the Commission for discussion and questions. MR. HACKBARTH: Thank you. What I'm going to propose is that we separate the 11 discussions on the update factors for inpatient and 12 outpatient services, recommendation 1, from the discussion 13 of the payment for E&M services in outpatient departments, 14 the later recommendations. 15 recommendation 1, the update recommendation, and what I 16 propose we do in this discussion and all day is just have 17 one round of comments from Commissioners inasmuch as we have 18 discussed each of these issues in December. 19 20 21 22 So let's just now begin with Scott, let me begin with you on recommendation 1. Any comments or questions? MR. ARMSTRONG: No questions. I would just tell you that I'm prepared to support this recommendation. 20 1 DR. BAICKER: 2 MR. BUTLER: I support the recommendation. I will support the recommendation, 3 but I would make one comment. We have said several places 4 in the chapter that the payments for outpatient are okay 5 because there has been growth in the last ten years and so 6 they must be decent. 7 fact is there have been many incentives for outpatient care. 8 We take who comes in the door for outpatient care, and I 9 don't think people celebrate the payment that you receive I'd take issue with that because the 10 for it. 11 that's the rationale, but I do support the recommendation. 12 That's the business we're in. DR. CHERNEW: So I'm not sure that I also support the recommendation 13 and just would like to note that this is the place for the 14 discussion about all the issues about hospitals not being 15 paid well enough or hospitals having financial problems or a 16 whole series of things because this is the part of the 17 discussion that's fundamentally related to the hospital 18 fiscal health overall as opposed to other places where one 19 might have that discussion. 20 DR. HALL: I support this. 21 DR. BERENSON: 22 DR. NAYLOR: I am supportive. I support the recommendation, but I 21 1 just have a question in terms of the relatively efficient 2 providers. 3 percent of Medicaid share hospitals, but do you have the 4 statistic on where they fall with respect to DSH? 5 the profile, the DSH profile of the relatively efficient 6 versus all other? 7 you could share it -- 8 9 10 I know you left out the highest -- the top 10 What's And if you don't have it now, maybe if DR. STENSLAND: I don't have it now, but I'll get back to you. MR. GEORGE MILLER: In light of Peter's comment 11 about outpatient payments across the system and my colleague 12 Michael's statement that this is the place where we should 13 deal with all the other issues, I do not support the 14 recommendation because I believe we should update the 15 outpatient piece here to deal with the other recommendations 16 you said not to talk about here, so I want -- the other two 17 recommendations, you want to separate them. 18 is correct. 19 payment on the outpatient side and this recommendation here, 20 and so I do not support the recommendation until we address 21 the outpatient payment adequacy. 22 for outpatient payments, and I agree with Michael, this is I think Michael We should deal with the inadequacy of the We've got negative margins 22 1 where we should fix those problems versus bringing it up in 2 addressing the third -- the second recommendation. 3 think the third recommendation should be done before the 4 second recommendation as well. 5 6 MR. HACKBARTH: We'll come to that. Comments on recommendation 1? 7 DR. STUART: 8 DR. CASTELLANOS: 9 And I do have a question. I support the recommendation. I support the recommendation. I On Slide 5, you mention that the 10 overall medical margin for 2011 was going to be minus 7 11 percent. 12 margin for 2011? 13 Do you have any guesstimate for the outpatient DR. STENSLAND: Usually when we make our 14 projections, we just make a projection of the aggregate 15 total of the two together. 16 DR. CASTELLANOS: I understand that, but that's 17 why I used the word "guesstimate." 18 DR. STENSLAND: 19 what, negative 9.6, Dan? 20 DR. ZABINSKI: 21 DR. STENSLAND: 22 Well, the outpatient margin now is Yeah. And basically the outpatient margin probably will not fall as much as the inpatient 23 1 margin. The outpatient margin, you know, may be somewhat -- 2 the guesstimate might be somewhat close to where it is, 3 because part of the reason we're seeing this decline in the 4 inpatient margin is they're doing some adjustments to the 5 updates to recover some of the documentation and coding 6 increases in payments, and those adjustments only affect the 7 inpatient side. 8 will be bigger -- was bigger than the inpatient update, so 9 we won't see as much of a decline in the outpatient margin. So the bottom line is the outpatient update 10 DR. CASTELLANOS: 11 MR. LISK: Thank you. And, in fact, you might actually see it 12 increase slightly because of volume growth and stuff like 13 that, you know, if you had the continued volume growth, 14 because we have seen the margin improving on the outpatient 15 side. 16 MR. GRADISON: I support this package. My 17 thinking is heavily influenced by the importance that I give 18 to trying to have the payment related to the service, 19 regardless of where the service is actually provided. 20 think this is kind of a classic example, but hardly the only 21 one. 22 concerns of the Commission over the next couple of years And I It would seem to me that one of the high-priority 24 1 ought to be looking to see whether there are other 2 situations in which similar questions could be raised and 3 should be examined. 4 outside to think about this, too. 5 the people who contact us -- or the losers, as they see it. 6 It's kind of interesting to me, at least my own experience 7 in the last couple of weeks is I haven't been hearing from 8 the ostensible winners, the ones who would get paid more. 9 don't know the significance of that, but it would seem to me And I also would encourage those on the We emphasize, or at least 10 that in looking at differential payment rates beyond this 11 one, other situations that might arise, that there may be 12 groups out there that monitor our activities, if you will, 13 the possible winners in future changes that ought to speak 14 up if they see distortions that they think we ought to 15 I examine. 16 17 MR. HACKBARTH: Other questions or comments on recommendation 1? 18 DR. BORMAN: I support the recommendation. 19 MR. HACKBARTH: Okay. Why don't we also go ahead 20 and do the official vote on 1 before we turn to 21 recommendations 2 and 3. 22 please raise your hand. All in favor of recommendation 1, 25 1 Opposed to recommendation 1? 2 Abstentions? 3 Okay. 4 So now we will turn to the recommendations related Thank you. 5 to payment for evaluation and management services in the 6 outpatient department, and I wanted to kick off that 7 discussion by saying a little bit about how I have thought 8 about this issue. 9 To me, the principle here is that, over time, 10 Medicare needs to move towards paying the same amount for 11 the same service regardless of the provider type. 12 current siloed payment systems, we have rates that differ 13 based on the type of provider even if they are providing the 14 same service to the same type of patient. 15 the existence of these payment silos in the Medicare program 16 and say that that makes sense neither from a financial nor a 17 clinical perspective. 18 In the We often lament There are a couple different approaches for 19 breaking through the existing silos. The one that we most 20 often have discussed in the past is developing new payment 21 methods that span the existing provider groupings, for 22 example, bundling around hospital admissions whereby there 26 1 would be a single payment that would cover the services 2 provided not just by the hospital, but also by physicians 3 and post-acute providers. 4 down the silos and moving forward. 5 course, is Accountable Care Organizations, where there is a 6 payment to an ACO for the full range of services and the 7 organization assumes both clinical and financial 8 responsibility for a defined population. 9 That is one approach to breaking Another example, of Over time, I strongly believe that these are the 10 preferred ways to deal with the silo problem. 11 think realistically, those changes in payment methods which 12 in turn require changes in the organization of the delivery 13 of care, will only come into Medicare gradually. 14 Significant reorganization of care is required under these 15 new payment methods. 16 not to mention new clinical relationships, are required. 17 And so this will be a gradual process of instituting payment 18 reform and care delivery reform. 19 However, I New legal and structural arrangements, As far as I can see into the future, we will 20 continue to have a traditional fee-for-service Medicare 21 program, so I believe it's also important to, within the 22 context of traditional Medicare, look at ways to start 27 1 bridging the existing siloed payment systems, and I think 2 one of our guiding principles should be to try to pay the 3 same amount for the same service regardless of provider 4 type. 5 I see this instance of evaluation and management 6 services provided by both hospital outpatient departments 7 and physician offices as but one example of that. 8 often, for example, in talking about post-acute care, noted 9 that we have very different payment rates based on type of 10 provider, even though in some instances they are providing 11 services to the same type of patients and we need to, in 12 post-acute care and other parts of the Medicare program, 13 begin bridging these silos, breaking down the payment silos, 14 and moving towards the goal of payment the same amount for 15 the same service regardless of provider type. 16 challenging to do this. 17 We have It will be This discussion of payment for hospital outpatient 18 services often comes up in the context of hospitals 19 acquiring existing physician practices or physicians 20 electing salaried or other financial arrangements under the 21 employment of hospitals. 22 both physicians and the hospital people, there is talk about Everywhere I go, in talking to 28 1 a shift in how care is organized, where more physicians are 2 seeking new relationships, often employment relationships, 3 with hospitals. 4 It is not clear to me, nor is it, I think, clear 5 to anybody how far and how fast that evolution will occur, 6 but we are seeing -- beginning to see signs of that in 7 Medicare data and other data that we see. 8 I want to be clear, though, that I don't see this 9 payment recommendation, recommendation two, of moving 10 towards equal payment, as motivated by trying to stop 11 acquisition of physician practices by hospitals. 12 that there are a lot of good reasons for both physicians and 13 hospitals to rethink their traditional relationships and 14 perhaps join together in new relationships. 15 not be to discourage that. 16 I think Our goal should Having said that, it's clear that, to the extent 17 that the shift occurs, it has important implications for 18 both the Medicare program and for Medicare beneficiaries. 19 If there is a significant shift of evaluation and management 20 services from physician offices paid under the Physician Fee 21 Schedule to hospital outpatient departments paid under the 22 hospital outpatient system, that will increase costs both 29 1 for Medicare and for Medicare beneficiaries. 2 for Medicare beneficiaries will come in the form of both 3 increased copayments at the time of services and increases 4 in their Part B premiums. 5 The increase So I think it's important to be aware of that 6 potential shift in the cost increases and adjust our payment 7 system so that there is not an adverse effect on either the 8 program or Medicare beneficiaries. 9 As I say, this is new ground, in a sense. What we 10 are saying is that no longer are we going to benchmark the 11 payment for OPD services within an OPD hospital-based 12 payment structure. 13 outside of the hospital system, and that raises a number of 14 complicated issues, sensitive issues, that have been called 15 to our attention. 16 last month or so. 17 We're going to look for benchmarks We've all heard a lot about this in the There are two issues that are of particular 18 concern to me. First is that we have often noted that 19 payments under the Physician Fee Schedule for evaluation and 20 management services are, we think, too low and have made 21 over the years a series of recommendations aimed at 22 increasing payment for E&M services, some of which have gone 30 1 into effect, and, in fact, in recent years, there has been a 2 significant increase in the relative values for E&M 3 services. 4 our October letter on SGR and the recommendations therein, 5 we don't think that that has run its course yet. 6 there is still further room for improving the accuracy of 7 pricing in the Medicare Fee Schedule, which we believe will, 8 among other things, increase the relative prices for E&M 9 services. 10 As we have made clear, however, as recently as We think One of the advantages of having the transition 11 that's described in draft recommendation two is that it does 12 allow some time for that work to continue and increase 13 payment for E&M within the Physician Fee Schedule. 14 The other issue that has struck me as particularly 15 important is the potential effect of this recommendation on 16 hospitals that are a critical source of care for 17 communities, where there may not be private practice 18 alternatives. 19 hospital outpatient department is an essential provider of 20 services to all patients, but in particular to low-income 21 patients. 22 those institutions, so we're proposing to take two steps in There are parts of the country where the And, of course, it is not our intent to damage 31 1 that regard. 2 The first is to have what we've referred to as a 3 "stop loss" for institutions that experience a greater-than- 4 2 percent loss in their total Medicare revenues and, in 5 addition to that, have at the median or a higher ratio of 6 disproportionate share patients, namely, low-income Medicare 7 or Medicaid patients, and that stop loss protection would be 8 in effect for the first three years. 9 Before the end of that period, we also recommend 10 in draft recommendation 3 that the Secretary do a study of 11 the potential impact of this change in payment for E&M 12 services on providers that are a critical source of 13 ambulatory physician services and, A, determine whether this 14 proposal would adversely affect access, and for those 15 institutions where it would adversely affect access in 16 communities, propose an approach for dealing with that issue 17 directly. 18 So, with that, I will pause and, Scott, we're 19 open. 20 comments or questions on recommendations 2 and 3. 21 22 Again, we're going to have just one round here for MR. ARMSTRONG: So just a brief question, and then I'll make a couple of comments. 32 1 With respect to this three-year transition period, 2 just to make sure I get this, this means that we would not 3 be applying these new rates to the hospital outpatient 4 services until the end of that three-year period? 5 that actually happen? 6 DR. ZABINSKI: I'll just start. How does It goes in a 7 three-step -- you know, a third of the way. 8 a -- you got your whole difference, and a third of the way 9 the first year -- 10 MR. ARMSTRONG: 11 DR. ZABINSKI: 12 MR. ARMSTRONG: 13 Okay, okay. I mean, there's I get it. Two-thirds the second year. So we do start with 30 percent of the impact of this in the first year and then -- 14 DR. ZABINSKI: Right. 15 MR. ARMSTRONG: 16 So, Glenn, the points you just made are points to Okay, got it. 17 a great degree that I agree with. 18 Okay, great. reiterate a couple of them. 19 I just would want to One, we have affirmed, I think actually quite a 20 few times, that we believe the principles that you just laid 21 out, and whether they are the principles about paying 22 comparable rates for comparable services or Medicare's 33 1 responsibility not necessarily to subsidize for losses 2 through Medicaid or through other programs. 3 more time we should be affirming that those principles are 4 our responsibility to uphold as Commissioners on MedPAC. 5 And I think one Second, I really like the point that you made that 6 we've discussed at length that one more time we are dealing 7 with a symptom of a payment structure that's highly 8 fragmented and that it's a lot of what we do, and it 9 sometimes simply doesn't make sense, and there's real 10 compromises that we make. 11 we are trying to deal with what is indeed a symptom of a 12 payment structure that pays in a highly fragmented way. 13 But this is not unusual in that I'd also just have to say that I think most people 14 agree -- at least people I speak with, even people who run 15 hospital systems dependent upon this payment structure agree 16 -- this payment structure doesn't necessarily make sense. 17 The issue we're dealing with is that they have come to rely 18 on the revenues that have been supporting their financial 19 structure, and I think that just reinforces the fact that it 20 is time for us to deal with the symptom and apply these 21 principles and come up with some other way of dealing with 22 the consequences on the financial performance. 34 1 And then, finally, I know there has been a concern 2 about whether we have studied this enough to move forward 3 with our recommendations today. 4 this more than enough. 5 of organizations that want to study this more is problematic 6 in many ways and keeps our industry from moving forward with 7 changes that are long overdue, and I think this is, again, 8 one of those cases where we know more than what we need to 9 know to move forward with this recommendation. 10 11 I think we have studied I think that, in fact, the influence And I will support it. DR. BAICKER: I support the recommendation and, 12 again, agree with the principle that we need to harmonize 13 payments across silos, and this seems like a great first 14 step in that direction because of the well-defined nature of 15 the service and the well-defined nature of the patients. 16 I think it's important that we're being cognizant 17 of the effects on access for particular populations that may 18 primarily access these services through these venues, but I 19 think the transition period is a great way to address that 20 and that in the long run we don't want to preserve access 21 for populations to critical services through the mechanism 22 of introducing a wedge in payments between services 35 1 delivered in one venue versus services delivered in another. 2 It's a very inefficient way to promote access to care to 3 preserve the wedge for that purpose. 4 broader look at how can we preserve access in geographic 5 locations where it may be limited is a good endpoint for 6 that transition, but I'm glad that we're not keeping the 7 wedge as a permanent mechanism for guaranteeing that access. 8 9 MS. UCCELLO: So I think moving to a I agree with all the comments made thus far and strongly support the recommendations. Our goal 10 is to get a good value for Medicare expenditures, and we're 11 trying to balance payment adequacy, beneficiary access, 12 along with affordability to taxpayers and beneficiaries. 13 And I think the way that the E&M payments are currently 14 structured doesn't provide good value. 15 sense to pay a lot more for pretty much the same service in 16 the hospital setting than the free-standing physician 17 setting. 18 It just doesn't make And agreeing with Scott on this, I think we need 19 to continue to explore whether there are similar 20 recommendations we can make for other types of services that 21 are provided in different settings by different providers. 22 There is one thing that I am somewhat bothered by, 36 1 and that is, because of this stop loss, the reductions in 2 beneficiary co-pays for those hospitals that are subject to 3 the stop loss are actually going to be higher than the co- 4 pays in the hospitals that are not subject to the stop loss. 5 So some of those -- or even many Medicare beneficiaries in 6 those stop loss settings, they're still going to pay less 7 than they would under current law, and many of them are 8 already going to have their cost sharing paid through other 9 sources, in particular Medicaid. However, for those people 10 who are low-income yet don't qualify for Medicaid, they're 11 still going to have to pay more, and so there's an equity 12 issue here. 13 program, I think that it's fine, and so we'll eventually get 14 to where there's equity in cost sharing across people. 15 16 17 But because this is a temporary stop loss So, again, I strongly support these recommendations. MR. BUTLER: Okay. I think this is one of those 18 occasions that what we say and capture on the transcript is 19 as important as our votes, because I think there are a lot 20 of messages that need to be carried forward for sure. 21 22 If we could turn to Slide 13, I just wanted to make a couple of points and questions on this. I've been 37 1 concerned all along that we presented these numbers in 2 overall Medicare revenue versus outpatient and also the way 3 we display them by percentile. 4 question and try to make a point. So let me just both ask a 5 So if we look at the 95th percentile, for example, 6 and we say that 2.6 percent impact on those organizations in 7 overall revenue, first of all, you've already identified 8 that that top 5 percentile tends to be large public teaching 9 hospitals. So it may be 5 percent of the hospitals. If 10 they're four times the size of other institutions, it could 11 be 20 percent of the dollars. 12 DR. ZABINSKI: 13 MR. BUTLER: Right? Potentially, yeah, sure. Well, not potentially. 14 is certainly skewed that way. 15 It probably is. 16 I don't know what the number Secondly, if you were to say, and arbitrarily but 17 not totally randomly, that 25 percent might be outpatient 18 and 75 percent inpatient, that 2.6 percent is more like a 10 19 percent reduction in outpatient payments. 20 -- in other sectors where we've displayed things, at least 21 we've shown the impact -- and one other point along those 22 lines. And so we kind of We've highlighted 9.6 percent loss on outpatient 38 1 overall. 2 outpatient, that would, all other things being equal, be a 3 20 percent loss for those institutions in that last 5th 4 percentile. 5 So if you were to apply a 10 percent reduction to The point I'm trying to make is the impact on the 6 outpatient side, when you isolate that, is really, really 7 significant. 8 go forward that Congress, they look at, well, 0.6 or 2.6, 9 this isn't a big deal. And the way we -- I fear that as these numbers But when you isolate those that are 10 impacted and you isolate the outpatient, it is a very 11 significant percentage of the outpatient money. 12 MR. HACKBARTH: I follow your logic and agree with 13 that, with the proviso that this doesn't include the stop 14 loss effect. 15 MR. BUTLER: Right. 16 MR. HACKBARTH: 17 MR. BUTLER: Okay. This is just today's world. Yes. This is today's world. And, by the 18 way, I very much appreciate the staff's effort to try to 19 identify where the impact is and where it is skewed. 20 So that's my comments on just how we portray this. 21 encourage us -- or more than encourage us, would like to see 22 something in the chapter that shows the percentage impact on Okay. I would 39 1 the outpatient per se versus just overall revenue. 2 understand we're trying to look at the overall financial 3 health of the institution, and that's why it's displayed 4 this way. 5 business a little different than the inpatient businesses, 6 and you need to understand the impact on the outpatient 7 revenue. 8 9 I But I think we need to -- we run our outpatient Okay. So the more I have learned about these groups of institutions -- and I still don't have a great appreciation 10 for them -- my support for this initially was in no small 11 part due to the fact that I saw a lot of sites off of 12 academic medical centers in the suburban communities that 13 were converting and making money off of this -- or at least 14 making the conversion easy, with no apparent value added to 15 the Medicare program and at greater expense. 16 and strongly believe that those things should not be 17 permitted to occur. 18 And I still do There's a second group that are kind of the 19 multispecialty group practices -- some are in urban areas, 20 some are in rural areas -- that have been in this business 21 for a long time. 22 And then the third is not just teaching hospitals 40 1 but principally the ones that are the primary affiliates of 2 a medical school. 3 the ones I'm most familiar with, not just in my experience 4 at Rush in Chicago but Henry Ford Health System in Detroit, 5 so I'm going to speak to those because I don't pretend or I 6 don't want to suggest my expertise is all that thorough in 7 the other areas. 8 9 They sit in urban areas. And those are So what do these buildings -- and they're not just buildings, they're services -- look like? Because I think 10 sometimes a qualitative description is as useful as a 11 quantitative description. 12 the requirements, the regulatory requirements, or conditions 13 of participation or Joint Commission or standby capacity. 14 think it's a little bit more or different than that. 15 We have previously talked about I For those of us that have been in these settings, 16 they're often one or more large buildings that house a large 17 faculty that is employed and has been employed for a long 18 time. 19 in these buildings. In our case, we probably have half a million visits 20 So it's a big operation. They also tend to be, I think, disproportionately 21 supported by electronic health records compared to other 22 settings. I don't have all the data on that, but I have 41 1 read literature that these settings are also more advanced 2 in terms of having electronic support. 3 It's also where a lot of the ambulatory training 4 occurs that we want to encourage and support as we try to 5 move to a new system. 6 We don't want to discourage that. But from a patient standpoint, what you see come 7 in the door of these places is an unusual population, often. 8 They're often in wheelchairs. 9 services. They often need social They often need interpreter services. None of 10 those things are regulatory issues, but these are often 11 complicated patients coming to specialty clinics, and these 12 things aren't captured in E&M codes as much as E&M codes 13 capture complexity. 14 know what? 15 if they're insured, they often do not really want these 16 patients because they're a heck of a lot of work. 17 But it's a support system that, you The free-standing practices, they don't -- even What kinds of diseases are treated? Well, you 18 look to academic medical centers, and sometimes you think 19 that these are -- they are things like comprehensive cancer 20 clinics that provide supportive care and team care and are 21 doing wonderful things, and they do -- beyond provide care, 22 we, for example, look at a lot of disparities in breast 42 1 cancer, and there's research associated with these. 2 things like movement disorders clinics that treat 3 Parkinson's. 4 we even have NCQA-designated medical homes that are in 5 geriatric care. 6 gives you a little sense of the kind of environment that 7 it's in, and I think we need to protect it, I think 8 particularly those that are serving the poor. 9 There's Alzheimer's clinics. They're And in our case So it's not just specialty care. So it We need to think, too, about an ACO world, however 10 you want to bundle payments and bring them together. These 11 kind of enterprises aren't going to go away, and we need to 12 find ways to partner and make sure that, you know, further 13 integration is accelerated, not, you know, picked away at 14 one step at a time. 15 physician, you know, needs to be partners in the reform of 16 the system. 17 do, too. 18 kind of get these environments flipped where needed, have 19 teaching in them, have them do the kinds of models and 20 convert these kinds of clinics into ways that are really 21 managing care in a cost-effective, highly service-oriented 22 way, and can continue to serve the poor. And I think Ron articulated the I think that these big academic medical centers As archaic as they can be, we need to find ways to 43 1 I know I'm making a lot of statements on these, 2 but I think this environment is a very important one, and we 3 can criticize each part of the system, but I think all of us 4 need to be partners and find ways to kind of make this occur 5 as expeditiously as we can. 6 sure. 7 Culture are hard to change for So when all of that is said and done, the problem 8 is these provider-based clinics are bad policy. 9 it's a bad technical policy. It really is. I mean, And as much as 10 I believe that the academic medical centers and others 11 desperately need this support, this mechanism is probably 12 not the way to do it. 13 think we need to stop these conversions where they're not 14 appropriate, where they're occurring remotely, is time 15 sensitive and should be addressed. 16 And coupled with the fact that I So in the end, with great angst for sure, you 17 know, while I would prefer a moratorium, while I would 18 prefer something to be done, you know, that is more off 19 campus than on campus from what I know about particularly 20 academic medical centers, and while I would prefer actually 21 the -- and I understand the problems with using DSH and 22 inpatient versus outpatient, I think we should convey that 44 1 MedPAC is not in the business of cooking up formulas, and it 2 is better to use ones that exist. 3 serving the poor, and I do think that the access is in the 4 end the key issue along with the support of, as I said, some 5 of these environments that are contributing to the future 6 reform of the system. 7 It's a proxy, though, for So when all is said and done, not with great 8 enthusiasm, I can support this, particularly with the 9 knowledge that the Secretary will closely monitor this and 10 11 that it has a three-year phase-in. DR. CHERNEW: So I support the recommendation and 12 would just like to say that the challenge here is, I think, 13 to get the structure of payment right and then worry about 14 the level one way or the other. 15 criticism of various things, and I think Peter's comments 16 were eloquent, and I understand about some of the unique and 17 different things. 18 lot of the arguments that we've received in letters and 19 other people who have spoken with me is, despite what I 20 think is the truth behind a lot of those arguments, I just 21 find the magnitudes that were presented before so large that 22 you simply can't argue that they're taken into account by And there has been a lot of The concern that I have sort of about a 45 1 any of some of these other arguments. And so I think it is 2 certainly a fruitful exercise to figure out which of those 3 particular things are concerning and what the right 4 magnitudes should be, but I think to a first order we would 5 never, if we thought there were problems in the hospital and 6 the payment rates were equal, say we should solve those 7 problems by raising the E&M fees. 8 some other way. 9 the basic principle is an important first step towards I think we'd deal with it And so I think that this basic approach and 10 moving us forward, and we can try and get the actual levels 11 that we're moving payments to right and make adjustments 12 where we can. 13 DR. DEAN: I certainly -- I do support the 14 recommendation with some of the same hesitations that Peter 15 mentioned. 16 deeply flawed payment structure that is poorly targeted, 17 it's filled with perverse incentives, and it clearly needs 18 to be changed. 19 difficult things and creates some pain that all of us would 20 prefer to avoid. 21 22 I think as we have said many times, we have a But change clearly brings with it some very At the same time, if we're going to move toward a system that more rationally uses the resources that we have, 46 1 I think these changes are the direction that we need to go. 2 Having said that, you know, I'm concerned about a 3 lot of the questions that have been raised about this 4 proposal and the impact it will have. 5 well, to support it, I'm concerned right now about the 6 differential in beneficiary co-payments, which I think is a 7 significant concern, a number of sort of not very well tied 8 together concerns. 9 I'm concerned -- We've been told a number of times that outpatient 10 departments are the most poorly supported, most poorly 11 reimbursed section of hospital services, but I have asked 12 several times: 13 I've never really gotten a straight answer to that, and I 14 tend to think it's not because of Medicare payments. 15 much more -- I believe it's most likely because of 16 inadequate Medicaid reimbursement plus uninsured folks and 17 all that. 18 be sure that we don't too seriously limit the facilities 19 that are willing to try to tackle these challenges and still 20 try to provide the care. Is that because of Medicare payments? And I'm And that's a real problem, and we want to try to 21 At the same time, I think these kind of movements 22 hopefully will refocus some attention on those problems and 47 1 hopefully help to move the overall system in a more logical 2 way, and that Medicare really is not the solution to the 3 overall payment system for the rest of the population. 4 There are lots of concerns, but I guess rather 5 than ramble on, I think that this is a movement in the right 6 direction, even though it clearly has some impacts that are 7 going to be a real concern. 8 9 DR. HALL: I'm speaking in favor of the recommendation as well, and I'd like to just say a few words 10 particularly concentrating on the beneficiaries and access 11 to care for present beneficiaries and for the burgeoning 12 number that are coming down the chronological pipe. 13 I was very much impressed by the letters and calls 14 from a variety of interested parties, mostly health systems, 15 that we've received, and they were respectful and they were 16 passionate. 17 two themes that I saw in all of these institutions, it was: 18 How do we preserve care to the most disadvantaged in the 19 population? 20 many of these centers, which is the education of the next 21 generation of care providers? 22 of addressing those concerns? And I think if I had to break it down to the And how do we preserve the basic function of So what does this do in terms 48 1 Well, I guess I'm persuaded to believe that this 2 is just one additional step to start to rectify, as many 3 people have said here, some of the difficult discrepancies 4 in payment systems that we deal with, not only in Medicare 5 but indirectly through all the other payment systems. 6 think it has to be looked at in the context if it's not -- 7 it is one decision, but it's a decision that fits into a 8 much larger plan of really aligning our interests, which as 9 a new member of the Commission, I've come to really respect, And I 10 and that is that we really do care about access to care for 11 beneficiaries and we do really care about the educational 12 function of some of our premier institutions. 13 one of a number of steps, and I think we could see some 14 positive things flowing from that. 15 So this is I'm also reminded that our primary responsibility 16 is to the beneficiaries, and as others have said, it's not 17 to continue to perpetuate and extend the life of the system 18 that is taking money from one pot and putting it into 19 another, and inevitably these kinds of conflicts arise. 20 I believe that from the educational standpoint, 21 although we're not directly talking about medical education, 22 it makes little sense to continue to foster education, 49 1 particularly in teaching people about evaluative and 2 management services, in a broken system. 3 for us to do is to fix the system and then the education 4 will follow from that. 5 forward into a continued improvement of our educational 6 system. 7 The better thing So this I think will also move us And I think as we'll see in recommendation 3, we 8 put in some safeguards here that I think are going to allow 9 us to really think through this problem a little more deeply 10 as it becomes implemented over 36 months. 11 Thank you. 12 DR. BERENSON: Yeah, I also thought the letters 13 and phone calls were important, laid out a number of 14 concerns, and did raise the issue of why are we moving so 15 far on a complicated issue. 16 lot about that and generally came up with the counterview 17 that, in fact, the provider-based payment policy has been 18 festering for many years. 19 it creates in the discussions around changing oncology 20 payment -- Herb was living through that time -- in which one 21 of the arguments made for not moving from the previous 22 system, which provided substantial profits to oncologists And I guess I had to think a I became aware of the distortions 50 1 based on getting paid AWP, the argument was, well, we'll 2 just refer the patients to the hospital and Medicare is 3 going to wind up paying a lot more money, so why would you 4 do this? 5 More dramatically, 18 to 24 months ago, when the 6 fee schedule reduced overpayments, substantial overpayments 7 for some cardiac studies, nuclear studies and all, the 8 response was, similarly, you're penny-wise and pound- 9 foolish, we're just going to refer these patients to the 10 hospital outpatient department and you're going to wind up 11 spending a lot more money. 12 It actually got worse than that. The 13 cardiologists went to the hospital, not just their patients. 14 And, indeed, I've seen marketing materials suggesting to 15 cardiologists how valuable they were because the hospital 16 was going to be able to use provider-based payments to 17 generate substantial revenues. 18 of an academic health center buying a cardiology practice 19 and raising the prices for echocardiograms by 400 percent. 20 Medicare won't pay all of that, but Medicare policy ripples 21 through the entire health care system. 22 a broken payment system here, and to take more time to study I heard just the other day So we actually have 51 1 it I don't think is appropriate. 2 move towards a system, as the Chairman suggested, of pay 3 equivalence or differentials that can be empirically 4 justified. 5 there is a lot of pretty good data there suggesting that the 6 case mix is really very different between an ASC and an 7 outpatient department, and we would not want to recommend 8 going to payment equivalence there. 9 And I think we do want to We will later on in the day talk about ASCs, and I think on E&M it is a simpler case because, as 10 staff have made the case -- and I think it's largely 11 correct, not 100 percent correct -- that the levels of E&M 12 services are in a sense a case mix adjuster. 13 the hospitals appropriately code, I think they'll have a 14 greater interest right now since the profits will be much 15 less to actually get the coding right. 16 Whether or not I do have a regret that we are starting with E&M 17 services. The examples I used are not largely around E&M 18 services, and I've been, I guess, one of the leading 19 protagonists in arguing that we have distortions in the 20 Physician Fee Schedule with overpayment for tests and 21 imaging largely and underpayment for E&M services. 22 has been some correction of that in recent years. There It's 52 1 especially a problem because there's a growing recognition - 2 - and this was reflected in the physician payment rule this 3 year -- that we actually have to do fundamental rethinking 4 of even the code definitions of E&M services. 5 is deferring to an ASPE study which is looking at how we can 6 better capture the work of E&M services, especially in 7 primary care, because it is not well captured in the current 8 code definition. 9 CMS basically So I'm not happy that we've started with E&M. 10 There are reasons why we started with E&M. 11 want to move very quickly and send a clear signal that for 12 other services where there really is no empirically 13 justified differential, we would adopt the same 14 recommendations for those such as echocardiograms, where I 15 have trouble imagining that there is a patient severity of 16 illness difference in how we would pay. 17 So I'm in favor of the policy. I do think we I guess what I 18 would want to iterate or reiterate is that we have a lot of 19 work to do in adopting this policy, we and CMS and ideally 20 the RUC, and one of the perhaps unintended consequences of 21 this policy would be to bring a new party to the table of 22 trying to work through the correct coding for E&M services 53 1 and the correct payment for E&M services, that those would 2 be hospitals, and I would actually welcome some chief 3 medical officers of hospitals to the kinds of meetings that 4 are being held, need to be held, to try to rationalize the 5 Physician Fee Schedule coding and payment. 6 So with that, I'm in favor of the policy, although 7 I did want to point to one concern that was raised in some 8 of the letters which I was not sympathetic with. 9 sympathetic to some of those that were raised about time and I am 10 impact on the kinds of hospitals we're concerned about. The 11 argument that basically said MedPAC is in favor of ACOs, of 12 more integration, and now you're reducing payment for E&M 13 services to hospitals that's going to make it difficult for 14 us to integrate. 15 says we need lots of extra money so we can entice docs to 16 come to our place so that we can become more efficient. 17 doesn't work for me. 18 hospitals owning doctors and being integrated systems. 19 There are certainly other models around multispecialty group 20 practices and IPAs, and I don't think we need to -- in a 21 sense, the implication was we need to sort of subsidize the 22 hospitals for a while so they can develop their integrated I'm not sympathetic with the argument that It And that's one model of integration, 54 1 systems with the promise that they are then going to be much 2 more efficient. That one didn't convince me. 3 So I'm in favor of the policy. 4 MR. KUHN: Thank you, Glenn, and I want to join 5 Bill and Bob in expressing my appreciation to all the groups 6 that sent in letters and comments through the system that 7 Glenn shares with everybody at every meeting. 8 worked very well, and people have supplied us with some good 9 information to help us work through this issue. 10 I thought it Let me state at the outset that I'm going to 11 oppose this recommendation, and I want to make three general 12 comments about that. 13 I agree with, as Glenn laid out at the beginning, 14 and others, the same price for the same service, and I think 15 that's an important principle to be on. 16 little bit concerned that are we comparing the same service 17 in the two different settings. 18 although I don't think they're as great as some folks would 19 think they are, but there is a bit of a differential there. 20 We have standby capacity issues. 21 issues. 22 case mix issues. But I'm just a We've got packaging issues, We have life safety code And I think as some have expressed, we have some I think we recognize that the acuity level 55 1 of some of the patients presenting themselves to some of 2 these hospital-based clinics are far greater than others. 3 And so I just don't know if it's a true apples-to-apples 4 comparison in my mind. 5 The second issue -- and both Peter and Glenn 6 talked a lot about this -- was the clinic system that exists 7 out there now, and I think Glenn did a nice job of saying, 8 well, what if there is no private practice alternative? 9 I do want to compliment Glenn and all the other And 10 Commissioners who raised these issues because I think the 11 changes that are in this draft recommendation 2 recognize 12 that issue and do a lot to try to get us to that place. 13 With the three-year phase-in, the stop loss, the retargeting 14 of dollars back to those organizations that are actually 15 doing the work, not supporting others who are doing the 16 things that we've heard about that are probably 17 inappropriate behaviors that are out there, and 18 recommendation 3 for the Secretary to come back and look at 19 these issues, I think are all good improvements, and I 20 appreciate those being in there. 21 little concerned that when we look at 2014 and we know all 22 the new individuals that are going to have coverage as a But for me, I'm still a 56 1 result of the expansions of PPACA, are we going to continue 2 to have that robust clinic system for those that can't get 3 care elsewhere? 4 in the back of my mind for that as we go forward. 5 And I just have a bit of a nagging concern And then the final issue -- and it's one I raised 6 at the last meeting, among others -- is the issue of the 7 changes that this will bring about in the APC payment 8 system. 9 with the cost-to-charge ratio, and if we move down this It is a charge-based system that's put in place 10 direction of reducing charges for E&M codes, that means that 11 other codes will change dramatically, particularly device- 12 dependent codes, and will we see a skewing of the system in 13 the future with more payment being driven to device- 14 dependent codes, does that create even more access issues 15 for us for E&Ms in the future for these clinic systems, and 16 particularly in rural areas, is a concern of mine. 17 So, again, Glenn, I want to thank you for bringing 18 this issue forward. I think it's an important one for us to 19 look at. 20 is sound. 21 still have some nagging concerns here, so I will oppose 22 recommendation 2, but I believe I'll be able to support I think the principle of what you're trying to do I think the changes you've made are good. But I 57 1 2 recommendation 3. DR. NAYLOR: So I would like to acknowledge my 3 support of recommendations 2 and 3. I think that we have -- 4 and everyone here has talked about the critical principles 5 that we've outlined that should guide everything that we do, 6 our decisionmaking, and the notion of accelerating 7 accessible, affordable, equitable services for those in the 8 Medicare program is who we are and what we are supposed to 9 do, and the opportunities to align our payment across sites 10 and services to make sure that everyone gets that access and 11 we all benefit from the affordability of the program is 12 critically important. 13 related to each of these principles. 14 So let me just highlight a couple Under access, the higher co-pays experienced by 15 Medicare beneficiaries who are seeking and receiving 16 services in OPDs in many ways can serve as disincentives for 17 them to access those services, and I think that this is 18 really, in addition to the fact that I don't know how you 19 can justify a $10 differential for a mid-level or a mid- 20 range visit, 15 minutes in one setting versus another, the 21 very fact that that exists can serve and other evidence 22 suggests might be a disincentive for them accessing it. And 58 1 I certainly agree with all the transition plan, with the 2 attention that needs to be paid on assuring that hospital 3 OPDs both continue the critical services that Peter 4 described and particularly where they are in communities 5 that they are the critical access point. 6 I also think that we need to be really 7 accelerating and supporting other types of provisions of 8 these services in other contexts: 9 patient-centered medical homes, nurse-managed clinics. independence at home, We 10 have numbers of provisions already in play that are 11 providing and adding bonuses to providers in health 12 profession shortage areas. 13 ongoing to promote access, and I think that we want to have 14 the kind of policies that support and align and accelerate 15 those efficient providers. So we have a lot of change 16 On the issue of equity I keep coming back to, it's 17 so difficult to justify paying differences and justifying an 18 issue of affordability and equity for the system, Medicare's 19 use of increasingly finite resources for a growing 20 population of people and not using them as wisely as we can. 21 So really taking into consideration all that Peter and 22 others have said about the critical role that outpatient 59 1 departments play and will continue to play in the future, we 2 also need to recognize that we've got big challenges out 3 there, we've got to consistently apply these principles. 4 I don't know if we -- two minor points. On the 5 studies that will go forward, I think it's really important 6 to look at access to primary care or E&M services broadly. 7 In other work, we know that many Medicare beneficiaries are 8 not just accessing their care in physicians' offices or by 9 physicians -- 11 percent by NPs and PAs, and 33 percent 10 getting care elsewhere. 11 look at access, I hope we'll include those individuals as 12 well. 13 MS. BEHROOZI: So as we think about studies to Can I just ask a question first on 14 Slide 13? The margin in the 95th percentile, the per 15 transition year hit to the margin is 0.9 percent in the 16 lower right corner, right? 17 years. 18 between that and the 2 percent, the protection with respect 19 to the 2 percent. 20 percent of hospitals would end up being protected, I guess, 21 in the final year. 22 years? So that's for each of the three And so I'm trying to figure out the relationship So then on Slide 14 you say that about 4 Does that differ in the first and second 60 1 DR. ZABINSKI: Yes. 2 MS. BEHROOZI: Okay. 3 DR. ZABINSKI: It's a lot fewer in the first year. Do you have those figures? 4 It's like 17 the first year, 44 the second year, 120 in the 5 third. 6 MS. BEHROOZI: Thanks. Okay. So I'm going to try 7 to go back to something you asked us to do in prior times, 8 just to say what you like first before you say what you 9 don't like. 10 11 If you don't mind going back to the recommendation, please, Jeff? Thank you. I like the principle, I absolutely do like the 12 principle of paying the same for the same service in 13 different settings, and that's why I was one of those people 14 whom Dan referred to, I guess, when we discussed this 15 previously, the appeal of the principle is undeniable. 16 where it doesn't make a difference as applied -- you know, 17 the principle is absolutely correct, and when as applied it 18 turns out the way the principle should make it turn out, 19 then it's fine, like when I take my kid to the pediatrician 20 to one office on Tuesday and pay one rate and then take him 21 to a different office on Thursday for the same service and 22 pay a different rate because, you know, it's affiliated with And 61 1 a hospital, it doesn't make a difference. 2 him at the first place, at the physician office, and my 3 access is not impaired. 4 absolutely then should be applied. 5 I can keep seeing And, you know, the principle But where it does make a difference, I think, is 6 where I feel a responsibility to say that I don't think that 7 I can support the recommendation as written, and where it 8 makes a difference is in low-income communities. 9 all others, I heard a lot of, you know, some rhetorical and And like 10 some very, very persuasive and important information 11 presented by hospitals in response to this potential 12 recommendation. 13 But of all of that stuff that was deliberately 14 presented to all of us, I don't think any of it had quite as 15 much impact on me as listening to a hospital CEO who runs an 16 independent safety net hospital in Brooklyn talking about 17 when another hospital in Brooklyn in a different 18 neighborhood had closed -- part of the unfortunate trend in 19 New York City of the Catholic Church getting out of the 20 business of running hospitals at a loss -- and that hospital 21 closed, and that community would then be unserved by not 22 only the hospital but the clinics that provide primary and 62 1 other care. And so her hospital took over those clinics and 2 sought to run them, but then ultimately -- and it's not just 3 the fault of Medicare. 4 the opposite, I think, of what we're considering here. 5 York State has moved via the Medicaid program more money 6 into the outpatient side of hospitals. 7 talking about outpatient departments versus physicians, but, 8 you know, I think it would be helpful to put it more within 9 the context of inpatient versus outpatient and the This was before New York State did New By the way, we're 10 incentives to shift site of service within the hospital. 11 But, anyway, that's a whole other discussion. 12 about trying to run these outpatient clinics and that they 13 were just losing too much money and threatened to drag the 14 rest of the hospital down, so she ended up having to close 15 them. 16 about the fact that the hospital would close, but those 17 neighborhoods then lost their access to the services, the 18 primary services provided by those clinics. 19 So she talked So it's not about, you know, the threat -- it's not So it's the fact that they're already losing 20 money, can the hospitals sustain the additional hit, and 21 what choices will they make based on that additional hit. 22 And in the slide after the recommendation where it talks 63 1 about what the potential impacts are, I think it doesn't 2 sufficiently acknowledge that there is a potential for 3 outpatient clinics to close. 4 say there will be an impairment of access. 5 That's what it means when we So I like that in the second part of the 6 recommendation there is a recognition of the potential for 7 that access impairment, but I really don't think it's 8 sufficient. 9 first year from, you know, a hit and to take 2 percent out I mean, I think to protect 17 hospitals in the 10 of their overall Medicare revenues based on the fact that 11 they're running these outpatient clinics could very well 12 mean that they're not going to run those outpatient clinics. 13 And I think that it's just not protective enough. 14 Another concern that I have with the 15 recommendation as written is that it doesn't sufficiently -- 16 it's not sufficiently tied, it's not sort of sufficiently 17 conditional on getting E&M codes right, which isn't just 18 about E&M as a whole area, you know, a big lump, but I think 19 kind of goes to Herb's point as to being able to 20 differentiate better within the scope of E&M services and 21 the patients' accessing those services, and then it kind of 22 gets into is there a difference between at least some of the 64 1 services, the E&M services provided in hospitals and those 2 provided in doctors' offices based in large part on the 3 different characteristics of the beneficiaries. 4 percent of those accessing outpatient services in hospitals 5 in New York City are Medicaid and uninsured, and I kind of 6 doubt it's that high a percentage in doctors' offices. 7 the extent to which the patient mix is different, are there 8 at least some differences in the types of E&M services 9 provided in hospitals? 10 I mean, 68 So So I love the third recommendation, and I think 11 that it should be more tied to the second recommendation, or 12 more conditional. 13 like overall the principle and the protection of those who 14 could be harmed, I don't think the principle is sufficiently 15 tied to other principles we hold dear, and I don't think the 16 protection is strong enough. 17 MR. GEORGE MILLER: So because of those reservations, while I I, too, do not support the 18 recommendation, and much of what I will say has been said 19 before, so I'll try to be brief. 20 again, like Mitra, I'll start with the things that I like. 21 I certainly support and understand the issues of the 22 structure of the business of paying the same price for the But my concern -- and, 65 1 same service and certainly support that in theory and hope - 2 - and know this Commission will move to that point. 3 certainly understand the desire to move to reform the very 4 broken payment system, and we're addressing that. 5 And I But I guess the problem that I have is that we're 6 picking one silo to fix that versus doing a comprehensive 7 program to fix everything at one time. 8 me angst because, again, I'll make the point -- I think 9 Peter made it very eloquently earlier -- that the outpatient And that then causes 10 department is a different structure, particularly for those 11 vulnerable populations in the areas where private physicians 12 would not go and provide those services. 13 areas, I think it's the same thing. 14 provide these services because those patients would not have 15 primary care services if they did not go to the hospital- 16 based physicians. 17 And in some rural The rural hospitals I realize we want to try to get the structure of 18 payments right, and, again, I support that and that's the 19 right thing to do, as Mitra just illuminated. 20 I'll go back to my example, what Dr. Ron Anderson at 21 Parkland has put together. 22 has been built over a long period of time, they've made an But, again, That network of clinics, which 66 1 investment -- and, again, because of certificates of 2 participation, like safety code issues, EMTALA, the Joint 3 Commission, that clinic structure has to be vast and has EMR 4 and has to be put together right. 5 this recommendation -- at least in my opinion, this 6 recommendation will dismantle that infrastructure, again, 7 which is an access issue for the very poor, for minority 8 populations, and I would say part of the disparity problem 9 in America can be tied to this issue as well. And we will dismantle by 10 philosophically, I support the method. 11 So, again, recommendation is the way to go. 12 I don't think this And like Mitra, I can support recommendation 13 number 3, but I think it should be tied closer to 14 recommendation 2 so it has less of an impact. 15 all agree that E&M services are too low, but the 16 recommendation doesn't address that issue. 17 is, I think, harm safety net hospitals in America. 18 I mentioned earlier, and I'll say it again, Grady would take 19 currently as proposed a $20 million hit with this proposal, 20 which they can ill afford to take, a safety net hospital in 21 Atlanta, Georgia. 22 DR. STUART: I think we But what it does I think I support both recommendations, and I 67 1 support the order of the recommendations. 2 do not believe that we should have a moratorium on this 3 until we study it further. 4 In other words, I I also believe that we should make a very strong 5 case for the philosophic underpinning of equal pay for equal 6 service, regardless of the provider setting. 7 I happen to think that focusing on E&M first is 8 appropriate, particularly given the volume of services that 9 are provided, much more volume than you're going to have 10 with specific tests or specialty procedures. 11 convinced by the need to do this to prevent even more 12 erosion in the value of these services being provided in 13 hospitals that are taking up physician practices in order to 14 increase their bottom line. 15 that. 16 And I'm So I thoroughly agree with I think, following up on Bob's point, that we 17 should take this opportunity to perhaps in the first 18 chapter, the introductory chapter, to say, look, this is the 19 first time that the Commission has really forcefully moved 20 in this direction, and we recognize that this is just the 21 first of other areas in which we really need to look at 22 this, and so to put us on record as saying this is a first 68 1 step in an ongoing process that needs to be given some -- 2 needs to be continually addressed by the Commission. 3 I do have one technical question regarding Slide 4 13, and it gets to the eloquent point that Peter was raising 5 about the impact on specific facilities, and that is, do 6 these numbers, both by transition year and when fully phased 7 in, take account of the stop loss? 8 DR. ZABINSKI: 9 They don't. it's not -- I have something. Let's see. You know, It's buried in here 10 somewhere. 11 effect would drop from 0.6 percent to 0.5 percent. 12 have a small effect on each number. 13 You know, stop loss, for example, the overall DR. STUART: It would Well, it strikes me that it could 14 very well have a very large effect on a specific 15 institution. 16 percent, then for the facilities that Peter was referring 17 to, presumably it would be a lot less. 18 that's important, particularly in light of recommendation 3, 19 because the way I read recommendation 3, if somehow we've 20 really screwed up and there is going to be some particular 21 harm done, this will provide a mechanism by which we can 22 address that. In other words, if there is a cap of 2 And so I think And so I don't necessarily see the stop loss 69 1 disappearing after the third year. I mean, it might be 2 there, but we just don't know. 3 that I support having the study following the implementation 4 of this so that we have some experience with it, so we see 5 where it's headed. So I think that's the reason 6 DR. CASTELLANOS: I'll be as brief as I can. 7 First of all, I support both recommendations very 8 strongly. In a previous discussion, a discussion with Peter 9 and his comments, he made a very astute comment. If we were 10 going to pay the E&M charges for physicians in their office 11 higher or more appropriately, we wouldn't be having this 12 discussion. 13 But because we're on MedPAC and we're the prudent spenders 14 of the taxpayers' money, I think it's very appropriate that 15 we're making the decisions and the comments that we're 16 making. 17 We would not be having this discussion today. I strongly recommend going forth both with the 18 payment and delivery reforms as discussed. 19 would like to -- two things. 20 One, can you go back to Slide 16? The only thing I We had a lot of 21 discussion earlier about some of the definitions, and I'm 22 not sure what I'm voting for unless we clearly state the 70 1 outpatient department and the physician's office, a 2 clarification on those issues. 3 that before we vote. 4 So perhaps we could clarify And the second thing is, within the text, I would 5 like it to be stated where this dollars and savings is going 6 to be applied to, not the general revenue but where we 7 intend to have this savings applied to. 8 discussion, there was strong recommendations for applying to 9 the SGR and that issue. 10 MR. GRADISON: In our previous I've already expressed my support 11 for all three recommendations and am delighted that the 12 transition, which I and others were talking about in our 13 previous meetings a month or so ago, is included here. 14 I want to hit head on a concern I have about the 15 argument that basically says leave things alone. I'm not 16 talking about the things that have been expressed around 17 this table so much as some of the things that we've heard 18 from outside, because I think the argument in favor of the 19 status quo is a pretty hard argument to square with the 20 notion that the payment should be equivalent for 21 substantially equivalent services. 22 notion that they should be equivalent for substantially And if you buy the 71 1 equivalent services, then there are basically two ways to do 2 it: 3 outpatient departments to the physicians that provide the 4 services outside the hospital structure; or what is here, 5 which is the other way around. increase the payment to the level of the hospital 6 I frankly would be a lot more comfortable with the 7 recommendations of the critics if their real feeling is 8 let's increase -- do it the other way around, although I 9 would prefer this recommendation, than to say, well, let's 10 just stick with the status quo. 11 that way, but that's really what it comes down to, in my 12 opinion. 13 DR. BORMAN: It isn't usually expressed I support these two recommendations. 14 I've had some internal conflicts in thinking through this, 15 having practiced in hospital-based clinics as well as in 16 other settings and recognizing the large amount of good that 17 can be done through these practices and the populations that 18 in the best of all possible worlds that they target. 19 At the end of the day, we are obligated to advise 20 the Congress about the best investment of dollars into the 21 program and to make sure that the beneficiaries are being 22 best served and that their precious dollars are going to 72 1 best effect. 2 move down this road. 3 And because of that, I think that we need to I would make a couple comments about the study. 4 We usually talk about cost, access, and quality. We have 5 not talked a lot about quality in this conversation. 6 like to hope that in the study that we make sure that we 7 include the appropriate quality evaluations because that 8 may, in fact, also help us understand where better to target 9 these dollars in the future. I'd And I think that needs to -- 10 building on one of Mary's point, it needs to include all the 11 things that are going on in terms of the EMR use, the 12 quality bonus program already set up and so forth that, 13 surprisingly, populations of hospitals may or may not be 14 eligible for and will figure into this conversation about 15 where the dollars are best spent. 16 And the last thing I would mention is that -- it's 17 not something that we can deal with here today, but in one 18 of those rather peculiar quirks of the program, the specific 19 coding of the visit by the physician is not required to 20 match the coding by the facility. 21 history there, and I would submit that perhaps part of the 22 study going forward would be to also include looking at the And there's a lot of 73 1 underpinnings of that so that this becomes more transparent 2 and more appropriate and more easily valuable, because I 3 think that probably confounds a little bit some of the 4 analysis that underpins our recommendation today. 5 overall I support the two recommendations. 6 DR. MARK MILLER: But So the language point that Ron 7 raised for everybody is what we're trying to do here is 8 equalize the rates for an E&M visit that occurs in the OPD 9 and make it comparable to what happens in the physician's 10 office under the Physician Fee Schedule. 11 - one thing before we just discuss language, I just want to 12 remind the Commissioners and the rest of the world, in the 13 report that we've put in front of the Commission and 14 ultimately would be published, this is laid out in a fair 15 amount of detail as to how these rates are actually arrived 16 at. 17 in there. 18 This is laid out - In your current drafts, it's around pages 50, 51, 53, So there's a fair amount of detail that explains 19 what we actually mean when we say this, but the actual words 20 in question is in the sixth line down in that 21 recommendation. 22 department or physician's office." You see the phrase that says "outpatient And the question is 74 1 whether the "physician's office" is enough clarity. 2 So one thing we could do is stick with this and 3 make sure in the text we say this is what we mean by it. 4 Alternatively, there is another construction that -- you 5 know, because we knew this was an issue that a couple 6 Commissioners had -- that we could go to, Jeff -- and you'll 7 see also in the sixth line "non-facility setting paid under 8 the Physician Fee Schedule." 9 complicated. 10 It's a little bit more Either way the text would explain what we meant by these words. 11 I could go either way, but I'm not voting. So the 12 attempt is just to make sure that we're linking the payment 13 rate to the right level. 14 some detail how we got, you know, where we got. 15 question is which of these kind of captures the concept more 16 clearly. 17 MR. HACKBARTH: It's laid out in the chapter in And the We are substantially behind 18 schedule here, and so I propose what we do is go with the 19 original language and amplify in the text. 20 language is a little awkward. 21 out, "Explain me." 22 more commonplace English, and then address the issue in the I think this It almost sort of screams And so I'd rather go with the simpler, 75 1 text, if that's okay with you, Ron. 2 3 DR. CASTELLANOS: [off microphone] Okay. And Mary. 4 MR. HACKBARTH: 5 DR. NAYLOR: 6 MR. HACKBARTH: 7 recommendation number 2. 8 please raise your hand. 9 Okay. 12 Okay. All in favor of recommendation 2, Abstentions? 14 It's time to vote on Opposed? 13 Okay. Abstentions? 11 It's totally okay. All opposed? 10 And Mary. On recommendation number 3, all in favor? Thank you. I'd like to thank the staff in 15 particular this time. 16 in a relatively short time frame, and so they made heroic 17 efforts. 18 19 20 This is a complicated issue and done Thank you very much. Okay. It's much appreciated. We are now moving ahead to physician and other health professional payment adequacy and update. In fact, as our staff get in place here, let me 21 just say a word about this for the audience. Most of you 22 are familiar with the fact that in October, we sent a letter 76 1 to the Congress about the sustainable growth rate system in 2 the Physician Payment System and included in that letter 3 several recommendations, including a recommendation to 4 repeal SGR, and then we presented in the letter options for 5 how that might be financed if Congress were to decide that 6 SGR repeal must be financed fully out of the Medicare 7 program. 8 policy on the Physician Payment System, so we will not be 9 having separate update recommendations on physician payment. That letter still stands as MedPAC's statement of 10 Cristina, are you -- 11 MS. BOCCUTI: 12 13 Actually, we're going to start with the ASC group. DR. ZABINSKI: Okay. Today, Ariel and I will 14 present our payment adequacy analysis for ASCs and evaluate 15 the purchasing program for ASCs and Cristina will discuss 16 the payment adequacy analysis for the Physician Fee Schedule 17 that she and Kevin Hayes have done. 18 Before getting into the presentation itself, I 19 want to address the specific question that Mary asked in 20 December about how cost data could be collected for ASCs. 21 And as we discussed in the paper that the Commissioners 22 have, we see two possible mechanisms. First, CMS could use 77 1 annual surveys of a random sample of ASCs each year with 2 mandatory response to them. 3 require all ASCs to submit cost reports that are more 4 streamlined than what you, say, have with hospital cost 5 reports. 6 still sufficient to assess the payment adequacy while 7 minimizing the burden on CMS and ASCs. 8 9 Alternatively, CMS could All in all, the intent is to obtain data that is Okay. On to the payment adequacy. Important facts about ASCs in 2010 include that Medicare payments to 10 ASCs were about $3.4 billion. 11 beneficiaries served in ASCs was 3.3 million. 12 number of Medicare certified ASCs was 5,316. 13 The number of fee-for-service And the In addition, about 90 percent of ASCs have some 14 degree of physician ownership. 15 status, physician owners may furnish more surgical services 16 in ASCs than they would if they had to furnish their 17 ambulatory surgeries in hospital outpatient departments, 18 which is the sector with the greatest overlap of surgical 19 services with ASCs. 20 21 22 Because of this ownership And finally, ASC payment rates will receive an update of 1.6 percent in 2012. At the December meeting, we discussed measures of 78 1 payment adequacy in detail and we also provide detailed 2 discussion in the Commissioners' papers. 3 interest of time, today, we will cover measures of payment 4 adequacy more briefly. 5 payment adequacy for ASCs were all positive in 2010. 6 is, access to and supply of ASC services was adequate as the 7 number of beneficiaries served, the volume of services per 8 fee-for-service beneficiary, and the number of ASCs all 9 increased in 2010. 10 11 But in the In particular, our measures of That Also, the increase in the number of ASCs indicates that access to capital has been at least adequate. Finally, Medicare payments per fee-for-service 12 beneficiary increased in 2010. 13 use margins or other cost-dependent measures because ASCs do 14 not submit cost data to CMS, even though the Commission has 15 recommended submitting cost data in 2009, 2010, and 2011. 16 In addition, we cannot assess quality of care because ASCs 17 do not yet submit quality data, but they are slated to begin 18 doing so in October of this year. 19 However, we are unable to So for the Commission's consideration, we have the 20 following draft recommendation. The Congress should update 21 the payment rates for ambulatory surgical centers by 0.5 22 percent for calendar year for 2013. The Congress should 79 1 also require ambulatory surgical centers to submit cost 2 data. 3 In regard to the first part of this 4 recommendation, given our findings of payment adequacy and 5 our stated goals, a moderate update is warranted. 6 this is a lower update than the one percent that we just 7 recommended for outpatient departments. 8 provide motivation to satisfy the second part of the 9 recommendation, submitting cost data, which could be used to However, The purpose is to 10 better evaluate payment adequacy and help develop a market 11 basket for ASCs. 12 Spending implications of this recommendation are 13 that ASCs are poised to receive an update in 2013 of 1.2 14 percent. 15 small budget savings. 16 found growth in the number of ASCs and the number of 17 beneficiaries treated in ASCs as well as providers being 18 willing and able to furnish services under the ASC payment 19 system. 20 no impact on beneficiaries' access to ASC services or 21 providers' willingness or ability to furnish them. 22 ASCs would incur some administrative costs to submit the Therefore, this recommendation would produce a For beneficiaries and providers, we Therefore, we anticipate this recommendation having However, 80 1 cost data. 2 3 Now let me turn things to Ariel, who will discuss a value-based purchasing program for ASCs. 4 MR. WINTER: So, as Dan was saying, CMS has 5 adopted a quality reporting program for ASCs for 2012 and 6 ASCs will begin reporting five claims-based measures in 7 October. 8 will receive a lower payment update in 2014 and thereafter. 9 However, payments to ASCs will not be affected by how well ASCs that do not report data on these measures 10 they perform on these measures. In fact, CMS does not 11 currently have statutory authority to establish a value- 12 based purchasing program for ASCs that would rewards high- 13 performing facilities and penalize low-performing 14 facilities. 15 The Commission has outlined general criteria for 16 performance measures that should apply to any VBP program. 17 In the interest of time, I won't mention them here, but they 18 do appear on the slide and they're discussed in more detail 19 in your draft chapter. 20 Based on these criteria, the VBP program for ASCs 21 should include a small set of measures to reduce the burden 22 on ASCs and on CMS. We discuss several potential measures 81 1 in more detail in the draft chapter. 2 indicators are focused on outcomes, including patient safety 3 measures, such as patient fall or patient burn, hospital 4 transfer after an ASC procedure, and surgical site 5 infection. 6 process, structural, and patient experience measures. 7 Several measures are already part of the ASC quality 8 reporting program, but others would need to be developed. 9 Most of these The measure set should also include some I also want to mention some other key design 10 principles. 11 attain certain thresholds of quality as well as lower 12 performing providers who improve their quality over time. 13 14 First, it is important to reward providers who And second, funding for the pool of VBP payments should come from existing ASC spending. 15 So the second draft recommendation reads, the 16 Congress should direct the Secretary to implement a value- 17 based purchasing program for ASC services no later than 18 2016. 19 gain experience with reporting them, we think that 2016 is a 20 reasonable time frame for starting this program. 21 22 Given the need to develop additional measures and With regards to spending implications, because funding for VBP payments should come from existing ASC 82 1 spending, this recommendation will not increase Medicare 2 spending. 3 program to create small savings. 4 However, the Congress or CMS could design the With regards to beneficiary-provider impacts, this 5 should increase the quality of care provided to 6 beneficiaries. 7 submit quality data, and high-performing or consistently 8 improving ASCs would receive higher payments than under 9 current law, while low-performing ASCs would receive lower 10 ASCs will incur some administrative costs to payments. 11 And now I will hand things over to Cristina. 12 MS. BOCCUTI: So this final slide provides just a 13 brief summary of the issues and results that we discussed at 14 last month's meeting and also in your draft chapters, 15 particularly on the SGR. 16 chapter draft and I'll start here at the top. 17 And, of course, you have the Drawing on our patient survey and other national 18 studies, we found that most Medicare beneficiaries are able 19 to get timely appointments with physicians and can find a 20 new one when needed. 21 patients look for a new physician in a year, we continue to 22 find that among those seeking a new physician, finding one Although only a small share of 83 1 in primary care is more difficult than finding other ones in 2 other specialties. 3 patients reported receiving some or all of their primary 4 care from nurse practitioners or physician assistants. 5 We also found that about a third of Also in our analysis, we found that growth in the 6 volume of services provided, that is on a per beneficiary 7 basis, slowed in 2010, which is after about a decade of 8 rapid increases. 9 On claims-based measures of ambulatory quality 10 designed specifically for the elderly, we saw that most 11 indicators either improved or did not change significantly. 12 Looking at payments, we found that the ratio of 13 Medicare payments to private PPO rates remained steady, but 14 some payment factors are not included, like performance 15 bonuses, for example, that Bob and Peter have raised in the 16 last meeting. 17 As in previous years, the vast majority of 18 services are paid on assignment, meaning that physicians 19 accept the Medicare Fee Schedule rate as payment in full. 20 And the final bullet here on the slide calls 21 attention to the letter that MedPAC submitted to the 22 Congress this past October. That letter included four 84 1 recommendations, one of which was to repeal the SGR and 2 replace it with a ten-year path of specified updates. 3 this recommendation, updates for primary care are higher 4 than those for other fee schedule services in the first 5 three years. 6 7 8 9 Okay. In I think we can open it up for questions now. MR. HACKBARTH: Thank you. Let me kick off the discussion by saying a little bit more about physician 10 payment and our October recommendations. We took up again 11 the issue of SGR and recommended repeal in October. 12 course, MedPAC has a long history with the SGR issue, having 13 first recommended repeal of SGR in 2001. 14 discussing the SGR issue again last spring, it was prompted 15 by a concern that continuing the SGR was posing an 16 increasing threat to the Medicare program, a threat of 17 destabilizing the program by undermining both physician and 18 beneficiary confidence in Medicare. 19 has for me, at least, and other Commissioners can speak for 20 themselves, has only grown since the spring, prompted us to 21 delve into what is a very difficult area, and it's difficult 22 because of the fiscal situation and the concern that is Of As we began So that concern, which 85 1 widespread, if not universal, in Congress that we cannot 2 repeal SGR without having a way to pay for it. 3 So as I talked to people on the Hill back in the 4 spring about our taking up this issue, really, there was 5 bipartisan support for MedPAC delving into the SGR issue 6 again, but also a caution that if we are to do that and 7 we're to be effective, we need to also address the cost of 8 SGR repeal. 9 any notion of how to pay for it, would not be seen as Just another recommendation, repeal SGR without 10 particularly helpful to the Congress. 11 they've got. 12 get rid of it. 13 for it. 14 That's the barrier There's widespread agreement that they want to The problem is, they don't know how to pay In addition to that, at least some key people on 15 the Hill made it clear that we are the Medicare Payment 16 Advisory Commission and, therefore, our ideas about how to 17 pay for SGR ought to be from within the Medicare program. 18 To be blunt about it, our views on tax policy or on how big 19 the deficit should be or whether we ought to be funding 20 wars, not a war in Afghanistan, our views on those topics 21 are not of much interest to the Congress. 22 talk about repeal and how to finance it, we ought to focus So if we want to 86 1 within our domain, our area of expertise, the Medicare 2 program. 3 So the question that we faced as we addressed this 4 issue in September and October was did we think we wanted to 5 recommend repeal of SGR even if it had to be financed out of 6 Medicare, and we made the recommendation for repeal again 7 and suggested options within Medicare for how repeal might 8 be financed if Congress determined that it had to be fully 9 financed out of the Medicare program. 10 I want to be clear about that. We were not 11 recommending that SGR repeal should be fully financed out of 12 Medicare. 13 purview. 14 even if it had to be financed out of Medicare, and if so, 15 what were the options for financing that, and that's what 16 was in our October letter. 17 There are other options. Those are beyond our The question for us is did we recommend repeal At the end of our discussion in October, I made a 18 point that I will reiterate here. There are two, actually, 19 very important messages in the fact that we went through 20 what we did in October to recommend repeal. 21 one is that if Congress elects to fund SGR repeal solely out 22 of Medicare, it's going to be difficult and painful and it's Message number 87 1 going to have an effect on a lot of different parties -- 2 physicians, hospitals, Medicare Advantage plans, drug 3 companies, home health agencies, skilled nursing facilities, 4 really across the board. 5 Congress elects, it will be a difficult one and painful one 6 for some people. So if that's the path that 7 The other message in the fact that we went ahead 8 and recommended repeal anyhow is how urgently we feel that 9 it's time to get rid of the SGR system and that the threat 10 that it poses to the Medicare program and the beneficiaries 11 it serves is growing and just kicking the can down the road 12 is increasingly problematic. 13 So with those comments, I will open up the general 14 discussion, beginning with Karen. 15 both the ASC -- well, really, we only have the ASC 16 recommendation to vote on, but feel free to address either 17 ASCs or the physician issues. 18 DR. BORMAN: Yes. In this case, we have I support the ASC 19 recommendations and I think we've been over the issue 20 sufficiently while I don't have much to add. 21 Relative to the SGR, I think I imagine we have all 22 been hugely disappointed at the current two-month short-term 88 1 temporary stopgap and that how all the things come into play 2 about how demoralizing, how destabilizing, how unfortunate, 3 inappropriate, any adjective or adverb will apply about 4 right now. 5 options for change and a pathway to do that. 6 disagree with differential payment based upon specialty, but 7 still endorse the notion of repeal of the SGR. I think that we certainly outlined a lot of 8 MR. GRADISON: 9 DR. CASTELLANOS: 10 I continue to I support the recommendation. I support the recommendations on the ASCs. 11 I just have one question on the ASCs and it's just 12 a clarification question. 13 the ASCs, has there ever been any consideration to have a 14 single market basket for both the hospital outpatient and 15 the ASCs? 16 MR. WINTER: Once we get a cost report from Yes. In prior reports, we talked 17 about that. Our concerns with the current market basket use 18 for ASCs, which is the CPIU, and we actually used some older 19 cost data collected by GAO to look at whether the hospital 20 market basket or the Physician Medicare Economic Index would 21 be a better proxy of ASC input costs and found that ASCs -- 22 their cost structure in some ways resembled hospitals. In 89 1 other ways, they resembled physicians' offices. 2 left that decision about whether one of those market baskets 3 should be used or a completely new market basket should be 4 developed to the Secretary, once they collect the cost data 5 and analyze it. 6 And so we So that's certainly a possibility. DR. CASTELLANOS: Talking about physician pay, I 7 don't have to recapture the whole argument. Both, as you 8 remember, Karen and myself were very opposed to the October 9 recommendations for some very good reasons. Needless to 10 say, it was voted on as positive. 11 extremely concerned over these short-term fixes. 12 extremely disruptive to a physician who's trying to run a 13 practice, to CMS, and to the Medicare beneficiary. 14 present Congress, I don't see that Congress is very likely 15 to make a good recommendation and I think we're going to 16 continue to have these temporary patches with bigger budget 17 deficits and costs. 18 Like Karen, I am It is In the Within the medical community, there's a strong 19 ongoing feeling that we hope maybe Congress will enact the 20 27.4 percent cut. 21 make the Medicare beneficiary and the physician community 22 and the hospital community really seriously thinking about What that will do, if it is put in, it'll 90 1 this issue, and we really believe -- a lot of the doctors 2 believe that Congress will finally do something. 3 I've been on this Commission now five-and-a-half 4 years and we bring up the same issue. Glenn, you and I have 5 talked about it. 6 were talking about our recommendations, I said to you, 7 Glenn, what are we going to do in January when we have 8 another fix? 9 have this. In fact, in the September meeting when we You know, we both hope we haven't. We need an answer. We can't And I don't know how to 10 stress this more passionately, to say there needs to be a 11 message from MedPAC that this is just totally unacceptable. 12 We really need to consider going ahead and making -- getting 13 this -- so I can go ahead. 14 I'm trying to be able to provide care for them not just 15 today but in the future. 16 other business, and I'm sure the hospitals, if they were in 17 my situation and running a practice, they would feel exactly 18 like I do. 19 I'm not abandoning my patients. We just need some -- there's no It's just totally, totally unacceptable. I don't 20 know how to get that message across any better. I don't 21 think we -- you know, I don't like the idea of letting the 22 cut go into effect, but certainly if that happens, I think 91 1 we'd get an answer. 2 I just am a little concerned. You know, we're 3 talking about payment reforms and we're talking about equal 4 payments across different sectors. 5 continue that argument and continue that philosophy, 6 especially with the issues of some payment cuts, I think 7 we're going to see some significant differences in payments, 8 especially in the ASCs and the hospital HOPDs for the 9 physicians. 10 If we go ahead and I would like to only suggest that we seriously 11 consider some discussion on the same philosophy we've had 12 about appropriately paying for the same service across all 13 areas. 14 15 16 DR. STUART: I agree with Ron on that. I'd like to see that recommended, as well. I support the recommendation. the only thing that 17 I thought about when I looked at this in terms of one of the 18 rationales for only offering 0.5 is because we don't have 19 any information, and one might take this as to say, well, we 20 know that asking ASCs to provide cost information is going 21 to be cost increasing. 22 be to say, well, maybe we should give them a one percent So one way to think about this would 92 1 increase if they provide cost information and a zero percent 2 increase if they don't. 3 the language on that because that's going to take more time 4 than we have today, but something to that effect, that to 5 the extent that the costs rise because of this, that should 6 be considered as part of the activity that MedPAC goes 7 through in terms of recommending payment updates. 8 9 I'm not suggesting that we change MR. GEORGE MILLER: Yes, I support the recommendation and I kind of like what Bruce just said to 10 get the cost data. But I am struck in reading the chapter 11 previously concerning the demographic information that ASCs, 12 and I don't think it's improved over the year as far as 13 minorities and low-income and Medicaid patients. 14 proportion is very small compared to, like, the hospital 15 outpatient departments. 16 DR. ZABINSKI: 17 MR. GEORGE MILLER: Their Is that still correct? Yes, that's correct. Okay. All right. So those 18 patients probably go to the hospital outpatient departments 19 to get their care, but they don't get them at the ASCs. 20 DR. ZABINSKI: That is right. 21 MR. GEORGE MILLER: 22 MS. BEHROOZI: Okay. Thank you. I support the recommendations, and 93 1 just a comment. 2 recommendations, perhaps it's just something to note for the 3 text, you note in the principles for the ASC reporting, or 4 for any reporting -- I'm sorry, quality-based program that 5 the Commission would support that it should be aligned 6 across similar settings, in this case, including OPDs and 7 physician offices, but then the rest of the text just talks 8 about an ASC quality program and it might be worth kind of 9 reiterating that that should be aligned and where, in 10 11 Rather than messing with the words of particular, the obvious alignments are. DR. NAYLOR: I support the recommendations. I 12 think recommendation two, in particular, highlights the 13 critical importance the Commission is placing on consistent 14 sets of expectations regarding reporting quality and cost 15 data, but then goes to the next step of a value-based 16 purchasing recommendation that rewards high performers. 17 I think that this is very consistent with what we do. 18 MR. KUHN: 19 DR. BERENSON: So I support both recommendations. Yes. I just want to ask, since it 20 was an important part of the last conversation, you've added 21 some text to our chapter which talks about the severity of 22 illness, more complexity of patients in the OPD rather than 94 1 the ASC. 2 What's that based on? 3 complexity -- I mean, the burden of that complexity manifest 4 in longer OR times? 5 demonstrate that, in fact, that there are increased costs 6 that support such a differential? 7 Could you give me a little more about that? And is the sort of the measure of the Do we have anything else to be able to MR. WINTER: Okay. So the language about how ASCs 8 -- how OPDs serve patients who are more medically complex 9 than ASCs is based on primarily two sources of information. 10 One was a study that the Commission staff did that was 11 published in the 2003 Report to Congress that looked at 12 differences in average HCC risk scores for ASC patients 13 versus OPD patients and found that they were consistently 14 higher for OPD patients, recognizing that HCC risk scores 15 are an imperfect proxy for patient complexity and higher 16 cost. 17 In addition to that, the Commission funded a study 18 by RAND published in 2006 where they looked at specific 19 comorbidities -- presence of specific comorbidities for ASC 20 patients and OPD patients for two common outpatient 21 procedures, colonoscopy and cataract surgery, found that the 22 OPD patients were more likely to have these comorbidities 95 1 than ASC patients. But that report was not able to quantify 2 the impact of those differences on cost or on surgical time. 3 Separately from the RAND study we funded, RAND did 4 a study for ASPE which was published, I think, last year, 5 where they looked at differences in surgical time for ASCs 6 and OPDs using data from a CDC NCHS survey and found that 7 ASC procedures were about 40 percent less time than OPD 8 procedures. 9 treating sicker patients or they're less efficient or have The question is, is that because OPDs are 10 greater demands because they provide emergency care? 11 don't know, and the study didn't get into that and I don't 12 think the data are in the NCHS survey to be able to 13 disentangle those factors. 14 just can't quantify the impacts. 15 16 DR. BERENSON: We So it is -- unfortunately, we Do we know if there's any more coming, any more in this area? 17 DR. ZABINSKI: I don't know of any -- 18 DR. BERENSON: Okay. 19 DR. ZABINSKI: -- further research, but we can 20 take another look at what we can do with the data we have. 21 DR. HALL: I'm in support of the recommendations. 22 DR. DEAN: I support the recommendations. I would 96 1 just, for the sake of time, say I absolutely support in as 2 strong a possible way as we can the concerns that Ron raised 3 about the SGR. 4 Congress refuses to deal with this, and the fact that the 5 whole system suffers because they refuse to deal with it. 6 We've stated it about as strong as we can, but we just need 7 to keep doing it because it is a huge failure on the part of 8 our political system to deal with a major problem. 9 It just is so immensely frustrating that So -- and I also support the concerns that George 10 raised about the apparent cherry picking that goes on in 11 terms of selecting patients. 12 to look at and we need to respond to. 13 of what we can do right now. 14 15 16 DR. CHERNEW: That also is something we need It's beyond the scope I support the recommendations and the concern about the SGR. MR. BUTLER: I'll repeat the SGR, too. It doesn't 17 hurt, particularly in light of the conversation we just had, 18 where we offer up yet another reduction in hospital 19 outpatient which is out of the same pot, in effect. 20 imagine another 27 percent on top of that for some of these 21 institutions? 22 Can you It just makes no sense. MS. UCCELLO: I support the recommendations and 97 1 the comments made, and I just want to clarify something. 2 The issue with respect to minority patients in ASCs, is it - 3 - not that this necessarily makes it right, but is it 4 because of the physical location of the ASC, or is there -- 5 I think that's what it is, as opposed to something else -- 6 MR. GEORGE MILLER: 7 [Laughter.] 8 DR. ZABINSKI: 9 Cherry picking. There's probably some of that, but I look at this a little bit -- I haven't had a lot of time 10 to do it. 11 just looked at, you know, for example, in the Washington, 12 D.C. area where the ASCs are and it's probably some location 13 issue. 14 and this probably gets into a little bit of the cherry 15 picking. 16 they're more likely to lack supplemental coverage for their 17 Medicare, things like that. 18 part in it, as well. 19 more digging to make a real definitive answer to that. 20 21 22 Some of it's probably due to the location. I It also might be some factors of type of insurance, The minorities that were likely to be on Medicaid, DR. DEAN: And so that probably plays a But I've definitely got to do a little I would say that the location is a version of cherry picking. DR. BAICKER: I support the recommendations. 98 1 MR. ARMSTRONG: Yes, same. 2 MR. HACKBARTH: Just a few concluding comments, 3 one on ASCs and one on physician payment. I just want to be 4 clear that for my part, I see ASCs as serving an important 5 role in the system. 6 surgery both in the hospital outpatient department and in 7 ASCs, I know from working with our surgeons that they often 8 preferred working in the ASC. 9 productive, give a better experience for their patients. Having run a large group that did It allowed them to be more 10 Through no fault of the hospital, often working in the 11 outpatient department, ambulatory surgery suites, you're 12 subject to disruption because of the emergency department 13 and the management of the scheduling and the case flow is 14 just different. 15 And so in our group, we split our cases between 16 ASCs not owned by our group but by others and the hospital 17 outpatient department, the Brigham, in recognition of -- and 18 then we distributed the patients between those largely based 19 on the severity of risk. 20 experience complications and need the back-up that was in 21 the hospital and so they were directed to the hospital 22 outpatient department and, frankly, the easier cases, less Some patients were more likely to 99 1 risky cases, were directed to the ASCs. 2 was clinically the appropriate thing to do, and through our 3 payment, we recognized that selection of patients by paying 4 lower rates, significantly lower rates at the ASC than we 5 paid at Brigham for the hospital outpatient department. 6 And I think that I think ASCs are an important part of the care 7 delivery system. 8 just as we need to do so for hospital outpatient 9 departments. 10 That said, we need to pay them accurately On the physician issue, my concluding thought 11 about the urgency of SGR repeal is this. 12 October that there were three reasons that we thought it was 13 important to deal with this now as opposed to later. 14 that repeal of SGR will only get more expensive. 15 numbers will go up, not down. 16 We noted in One is Those Second is that given the fiscal challenges that 17 the Congress is grappling with, the likelihood that they're 18 just going to forgive all this seems to me, and this is just 19 my opinion, the likelihood that they're going to forgive it 20 and write off the SGR debt is going down, not up. 21 And then third is that Medicare savings that could 22 be used to finance SGR repeal are being snapped up for other 100 1 purposes, whether it be deficit reduction or for funding the 2 Affordable Care Act. 3 let me be clear. 4 without it being offset and then take the Medicare savings 5 for other purposes and leave this destabilizing element at 6 the heart of the Medicare program is problematic. 7 Those are both legitimate purposes, But to say that we can't repeal SGR And it was for those three reasons that I felt a 8 real sense of urgency about making the recommendation to 9 repeal SGR. 10 Thank you. 11 We will now move on to outpatient dialysis. 12 DR. MARK MILLER: 13 Wait a second. We've got to take the vote. 14 15 Good work. MR. HACKBARTH: Oh, I forgot the votes. A little detail. 16 [Laughter.] 17 MR. HACKBARTH: 18 recommendations on ASCs. Okay. MS. BOCCUTI: 20 MR. HACKBARTH: 21 Okay. Would you put those up? 19 22 raise your hand. So we have two I've got to find it. I led Cristina astray, too. On recommendation one, all in favor, please 101 1 Opposed? 2 Abstentions? 3 Okay. 4 Opposed? 5 Abstentions? 6 Now I think we're done. 7 [Pause.] 8 Nancy, you can start whenever you are ready. 9 MS. RAY: And number two, all in favor? Good morning. Thank you. During today's 10 presentation, I am going to summarize information about the 11 adequacy of Medicare's payments for outpatient dialysis 12 services. 13 was presented in detail at the December meeting. 14 presentation, I will be addressing specific questions that 15 Commissioners raised. 16 for you to consider about updating the payment rate for 17 calendar year 2013. 18 the March report. 19 Nearly all of the payment adequacy information During my I will present a draft recommendation This is the last presentation before So here is a snapshot of the dialysis sector in 20 terms of beneficiaries, facilities, and total Medicare 21 spending. 22 briefing paper. You saw this last month and it is also in your 102 1 Overall, the dialysis payment adequacy indicators 2 are positive. 3 capacity to meet demand. 4 the number of facilities and dialysis stations. 5 Dialysis facilities appear to have the There has been a net increase in Looking at the volumes of services furnished by 6 dialysis facilities, growth in the number of dialysis 7 treatments continues to match beneficiary growth. 8 use of erythropoietin, the leading drug in the ESA drug 9 class, declined in 2010. However, Looking at patients who received 10 erythropoietin in January and December, we see a seven 11 percent decline in mean units furnished per month. 12 Mitra, you asked for us to look at this by 13 ownership type. 14 decline was six percent. 15 facilities, the decline was nine percent. 16 For the two large dialysis chains, the For all other freestanding There was also a question about changes in the use 17 of other dialysis drugs. 18 volume of iron agents increased by one percent. 19 2009 and 2010, the total volume of Vitamin D analogs 20 decreased by two percent. 21 22 Between 2009 and 2010, the total And between Beneficiary access appears to be generally good. There were few facility closures in 2010, and it did not 103 1 disproportionately affect any beneficiary group, including 2 African Americans, the elderly, and duals. 3 Quality is mixed. Some measures are high or 4 improving and others still need improvement. One of the 5 improving measures is the use of AV fistulas. 6 reference to your question that not all -- well, in 7 reference to your comment that not all patients may be 8 candidates for AV fistulas, in the draft chapter, we have 9 referred to CMS's quality initiative goal that 66 percent of Karen, in 10 hemodialysis patients have an AV fistula and noted some of 11 the reasons why patients may not have one. 12 Glenn, in response to your question about 13 international AV fistula use, in 2010, seven out of ten 14 Western countries had higher rates of AV fistulas. 15 Mike, in response to your question about the 16 effect of the decline in ESA use, between 2009 and 2010, the 17 proportion of patients with hemoglobin levels less than ten 18 that might be suggestive of undertreatment of anemia 19 increased slightly, from 6.2 to 6.6 percent. 20 said, the FDA has not specified a lower bound for the 21 hemoglobin level. 22 Medicaid outcomes, including changes in the use of blood That being This spring, we will begin looking at 104 1 transfusions, and we will baseline blood transfusion use in 2 2009 and 2010, and once 2011 data become available, we will 3 make the comparison. 4 Bill Gradison, the draft chapter includes rates of 5 hospitalization and mortality by age, the decrease in rates 6 over time that we see overall and by race, we see the 7 similar changes by age. 8 9 Herb, you referred to the PPACA 3014 provision that requires the Secretary to establish a Federal pre- 10 rulemaking process for the selection of quality and 11 efficiency measures. 12 vascular access infection, adequacy, hypercalcemia, which 13 has to do with the bone and mineral drugs, dialysis drugs, 14 and bloodstream infections. 15 anemia. 16 There are five ESRD measures, one on None of the measures relate to George, among the PPACA 3014 measures that I 17 thought you would be interested is the one in the Physician 18 Quality Reporting System that would measure the percentage 19 of patients age 18 years and older on dialysis for 90 days 20 or longer who are referred to a transplant center for a 21 kidney transplant evaluation within a 12-month period. 22 Access to capital appears to be good. Merger and 105 1 acquisition data suggests that access is available for the 2 two large dialysis chains and other freestanding chains. 3 4 Glenn, we are still in the process of following up with the FTC and we will get back to you about that. 5 Scott, you asked about the cost growth by provider 6 type. 7 four percent lower than other freestanding facilities. 8 Between 2005 and 2010, cost growth was 2.6 percent per year 9 for the two large dialysis chains and two percent for all 10 In 2010, the cost per treatment for the two LDOs was other freestanding facilities. 11 The 2010 margin here includes composite rates -- 12 the 2010 margin for composite rate services and drugs is 2.3 13 percent. 14 across the different provider types. 15 16 17 As in previous years, the Medicare margin varies Mitra, in response to your inquiry, I have also included in this table the 25th and 75th percentile. We are concerned about the direction of margins 18 for rural facilities. Tomorrow morning, Jeff and Adaeze 19 will address the dialysis low volume adjustor that began in 20 2011 under the new payment method and that this low volume 21 adjustor is anticipated to disproportionately benefit rural 22 facilities. Recall that 2010, the year that the margin is 106 1 for, is the last year of the old payment system for most 2 facilities. 3 So the projected Medicare margin for 2010 is 2.6 4 percent. 5 noted on the slide. 6 behavioral offset to account for efficiencies in drug 7 delivery expected under the new payment method. 8 9 This includes all of the 2011 and 2012 policies It also includes a conservative So this leads us to the draft recommendation, which attempts to balance being cost conscious and ensuring 10 that providers can handle cost growth, and it reads, the 11 Congress should update the outpatient dialysis payment rate 12 by one percent for calendar year 2013. 13 would decrease spending relative to current law. 14 CMS's ESRD marketbasket projects providers' costs would 15 increase by 2.8 percent in 2013, and under current law the 16 update is marketbasket minus a productivity factor. 17 adverse impact on beneficiaries is expected. 18 recommendation would decrease beneficiary copayment relative 19 to current law. 20 financial pressure on some providers, but overall, a minimal 21 effect on providers' willingness and ability to care for 22 beneficiaries is expected. This recommendation Currently, No The draft The draft recommendation might increase 107 1 And I look forward to your questions. 2 MR. HACKBARTH: 3 Okay. Thank you, Nancy. Well done. 4 For anybody in the audience who just joined us, 5 the presentations today on most of the update 6 recommendations are short, concise, as well, exemplified by 7 Nancy's presentation. 8 we could allow some more time in our schedule for some of 9 the more complex issues that we are wrestling with. I asked the staff to do that so that I want 10 people to rest assured that we discussed these issues in 11 detail at the December meeting and this is a summary 12 focused, as Nancy did, on questions that came up in the 13 December meeting. 14 15 So, let's see. lead off. 16 17 20 Again, one round, comments or questions. MR. ARMSTRONG: Nancy has addressed all of my questions and I'm inclined to support this recommendation. 18 19 Scott, I think it's your turn to DR. BAICKER: I support the recommendation, as MS. UCCELLO: I support the recommendation, but I well. 21 want to revisit an issue I raised in the December meeting 22 where I asked -- the chapter made note that facilities were 108 1 having difficulty documenting certain comorbidities and one 2 of the public comments was that they were unaware of the 3 existence of certain comorbidities. 4 those comorbidities are not related to the costs of 5 providing care for those patients. 6 something that we can look into more moving forward and 7 perhaps provide insights on whether the current comorbidity 8 factors are appropriate. 9 MS. RAY: So that suggests that So I hope that this is Just two follow-up points. Once we 10 begin to get 2011 data, we will, of course, examine the 11 extent to which facilities -- you know, how facilities are 12 billing in terms of the case mix adjustments and so forth. 13 In terms of where these comorbidity adjustments 14 came from, they were based -- CMS's contractor conducted an 15 analysis of claims data and cost reports, and it was through 16 the use of claims data that these predictors in these 17 conditions were found to have some sort of effect on 18 providers' costs and payments. 19 text. 20 MR. BUTLER: 21 DR. CHERNEW: 22 DR. DEAN: But I can add that into the I support. I support. I made a mess. 109 1 [Laughter.] 2 DR. DEAN: I support the recommendation. I would 3 just say, I appreciated the concise report and I especially 4 appreciate the addition of the variation in margins that 5 exists, because one of the frustrations I think I've 6 expressed before is that, too often, we look at facilities 7 that really may differ immensely in their location, their 8 challenges, their settings, all those things, and we sort of 9 lump them all together, and I think looking at the variation 10 in margins is a first step to doing a better job of 11 targeting the resources we have. 12 DR. HALL: I am in favor of the recommendations. 13 DR. BERENSON: 14 MR. KUHN: 15 DR. NAYLOR: 16 MS. BEHROOZI: 17 MR. GEORGE MILLER: As am I. I support the recommendation. As do I. Same, and thanks for the answers. Yes, I support the 18 recommendations. However, at some point in the future, I'd 19 like to make one quality standard that we would support 20 dealing with the kidney transfer issue, as I've raised 21 before. 22 issue, but I appreciate the information. I'm concerned about the disparities about that 110 1 DR. STUART: I support the recommendation. 2 DR. CASTELLANOS: 3 MR. GRADISON: I support the recommendation. I do, too. I have a quick 4 question, looking forward. I've heard some concerns that 5 some folks are initiating treatment through dialysis where 6 the treatment is rather marginal in terms of whether it's 7 actually needed at that point in time, but are doing so 8 arguably in order to get some priority in terms of kidney 9 transplants. I'm not asking for -- this is not related to 10 this recommendation, it's a chance to ask you to take a look 11 at it if you haven't already done so. 12 DR. BORMAN: Thank you. I appreciate the answer to my 13 question and I support the recommendation. 14 however, leave a couple of thoughts for the next time we 15 address this particular sector. 16 I would, One would be that, as noted in the chapter, that 17 the biggest chunk of increasing costs was general and 18 administrative costs. 19 for efficiency, I think that that worries me a bit and that 20 we should continue that line of analysis and perhaps not be 21 so generous going forward if that persists. 22 In a time when we're trying to press A second piece is that I'm troubled a bit by some 111 1 of what appear to be behavioral adjustments in prescribing 2 patterns that have been documented through this, and 3 obviously we need to keep an eye on that. 4 counterweighted by -- you know, it becomes very difficult. 5 We start setting some of these targets -- it starts to 6 wander dangerously into individual medical practice. 7 so, for example, while there's not a guideline about how low 8 can you go, it's a very individual judgment based on a 9 particular patient about what they will tolerate in terms of It's And 10 anemia. 11 not a lower bound. 12 important issue, but it just means that we need to be a bit 13 cautious about thinking we can have these specific things 14 for so many different kinds of things. 15 the things that are relevant, that we can measure, and that 16 are less subject to that individual variation. 17 And so it may, in fact, be appropriate that there's That doesn't mean that anemia is not an We need to stick to The flip side of that is I would encourage us to 18 push for things that assess bone disease management. I see 19 this in my -- I have seen this in my own personal practice 20 related to the fact that I operate on hyperparathyroidism 21 and secondary hyperparathyroidism is a result of end stage 22 renal disease. Management of these patients, particularly 112 1 the -- well, not even particularly in the young ones -- 2 relative to their bone disease can make a dramatic 3 difference in their functionality. 4 patients for their hyperparathyroidism who have had ruptured 5 Achilles and patellar tendons, so at their knee and their 6 ankle, tremendous morbidity because of relatively poor 7 management and sometimes non-compliance with that 8 management, which is key, about that. 9 disability and so forth is something that I think bears I've operated on And that kind of 10 looking at, because we have already seen some variation in 11 the prescribing patterns of those drugs already, and I would 12 worry that we put people at risk for a lot of bone disease 13 that will lead to additional disability that in a very crass 14 kind of way will cost the program more, much less the human 15 cost. 16 So I would think we should look at that. And then, finally, as I mentioned last time, I 17 don't perceive that we've subjected the relationships among 18 the elements of this system to the same scrutiny that 19 perhaps we have, for example, with ambulatory surgical 20 centers, the use of radiologic imaging, high-end imaging and 21 things that people own and refer to and so forth. 22 I think we nicely summarize it actually in the appendix -- I And while 113 1 thought that was a wonderful outlay of the current state of 2 the market -- are there data that we can come to relative to 3 appropriateness of some of those relationships in ways that 4 they may have unintended consequences. 5 data continue to mount that more aggressive times per week 6 hemodialysis leads to a better outcome, we want to make 7 absolutely sure that we're getting exactly these 8 relationships in the services appropriately targeted to our 9 best ability. 10 11 MR. HACKBARTH: Okay. Particularly if the All in favor of the recommendation, please raise your hand. 12 Opposed. 13 Abstentions. 14 Okay. 15 Our next session is on hospice. 16 MS. NEUMAN: Thank you, Nancy. Good morning. Today I'm going to 17 review our indicators of payment adequacy for hospice. 18 discussed these data in detail at the December meeting, and 19 your paper has more detail as well. 20 though, I'll give a brief overview of hospice and respond to 21 questions from the December meeting. 22 We Before I do that, This slide provides a snapshot of Medicare hospice 114 1 services in 2010. 2 services from over 3,500 providers, and Medicare spent about 3 $13 billion. 4 Over 1.1 million beneficiaries received At the December meeting there were several 5 questions. First, on the hospice aggregate cap, George, you 6 asked about the process for repaying cap overpayments. 7 Generally, how it works is CMS and the hospice work out a 8 repayment plan. 9 overpayments, and interest is charged. Hospices have at most 5 years to repay the Due to the court 10 challenges, the collection of cap overpayments was put on 11 hold in 2010 and for some hospices in earlier years. 12 that CMS has issued a regulation establishing an alternate 13 cap methodology, we expect overpayment collections will 14 restart soon, and we'll have to wait and see how that goes. 15 Now Bruce, you asked about the characteristics of 16 above-cap hospices. 17 than 90 percent are free-standing, and they have smaller 18 caseloads than below-cap hospices. 19 About 87 percent are for-profit, more Herb, you asked about the impact of the original 20 cap methodology versus the new methodology. We've added to 21 the mailing materials the results of some modeling of the 22 two approaches and pointed out that the impact depends in 115 1 part on how the formulas are implemented operationally. 2 Mitra, you asked about the trend in live discharge 3 rates. 4 rate was were stable. 5 Between 2008 and 2009, overall the live discharge Kate, you asked about how much of the increase in 6 length of stay that we've seen over the last decade is due 7 to the changing diagnosis profile of the hospice population 8 versus growth in length of stay within diagnosis categories, 9 so we have a slide on that. 10 11 This shows the components of length-of-stay growth. In the chart on the left, we see the changing 12 diagnosis profile of the hospice population. For example, 13 in 2010, cancer patients account for a smaller share of the 14 hospice population than they did in 2000, whereas patients 15 with neurological conditions and debility account for a 16 larger share. 17 in length of stay within diagnosis categories. 18 decompose the overall growth in length of stay into these 19 two components, it's the growth in length of stay within 20 diagnosis categories rather than the changing profile of the 21 hospice population that accounts for most of the growth in 22 overall length of stay. In the chart on the right, we see the growth And if you 116 1 So now to review our payment adequacy indicators. 2 First, our indicators of access to care are positive. The 3 supply of hospices continues to grow, increasing more than 4 50 percent since 2000 and almost 3 percent in the most 5 recent year. For-profits account for most of that growth. 6 The percent of Medicare decedents using hospice 7 continues to grow, and hospice use grew in 2010 among all 8 beneficiary groups examined. 9 decedents also increased in 2010. 10 Average length of stay among In terms of quality, we do not have any data on 11 which to evaluate trends in hospice quality. 12 2013 per PPACA, hospices will begin reporting quality data. 13 Those that do not report that data will face a 2 percent 14 reduction in the update beginning in 2014. 15 However, in As we discussed in September, MedPAC convened a 16 technical panel on hospice quality in November. 17 has been added to your mailing materials on that, and I'd be 18 happy to discuss on question. 19 Information As far as access to capital, hospice is less 20 capital intensive than some other provider types. For free- 21 standing hospices, we continue to see entry of for-profit 22 providers and modest entry by nonprofits. Publicly traded 117 1 chains have generally reported positive financial results 2 and adequate access to capital. 3 have access through their parent institutions, so overall 4 access to capital appears adequate. 5 Provider-based hospices Now, in terms of margins, the aggregate margin is 6 7.1 percent in 2009, up from 5.1 percent in 2008. 7 recall this estimate does not count cap overpayments as 8 revenues and excludes nonreimbursable bereavement and 9 volunteer costs. 10 You'll Looking at margins by type of hospice, 11 freestanding hospices have more favorable margins than 12 provider-based; for-profits have higher margins than 13 nonprofits; and below-cap hospices have margins that are 14 slightly above the industry-wide average, while above-cap 15 hospices have high margin before the return of overpayments 16 but low margins assuming full return of overpayments. 17 As we've noted before, margins are higher for 18 providers with longer stays and for providers with more 19 patients in nursing and assisted living facilities. 20 In terms of urban and rural, margins are higher 21 for hospices serving urban areas. 22 appears to be driven by volume. Some of the difference Higher-volume hospices have 118 1 higher margins. 2 counterintuitive, looking at hospices that serve rural 3 counties, margins aren't lower for hospices serving more 4 remote counties as defined by our typology. 5 And although it might seem So this brings us to our 2012 margin projection, 6 and our assumptions are shown on the slide. In 2012, we 7 project a margin of 5.1 percent, 8 note for 2013 is that there will be the continued phase-out 9 of the wage index budget neutrality adjustment, which will And then one policy to 10 be an additional 0.6 percent reduction in payments that 11 year. 12 So this brings us to the draft recommendation. 13 It reads: 14 hospice for fiscal year 2013 by 0.5 percent. 15 preliminary estimates of the spending implications are that 16 it would decrease spending by between $50 million and $250 17 million over one year and less than $1 billion over five 18 years. 19 beneficiaries or providers' willingness to serve them. 20 The Congress should update the payment rates for And And we do not expect any adverse impact on We also, in addition to an update recommendation, 21 plan to reprint two of the Commission's prior 22 recommendations on hospice in the March report. We are 119 1 reprinting these recommendations because action has not been 2 taken yet in these areas. 3 The first is the payment reform recommendation. 4 This is the U-shaped curve. 5 payments at the beginning of the episode and at the end of 6 the episode and lower them in the middle, and this would 7 better align payments with the service intensity of care and 8 has the potential to make the payment system neutral toward 9 length of stay rather than favoring long stays as it 10 11 It would increase the per diem currently does. The second recommendation is for focused medical 12 review of claims exceeding 180 days for hospices with 13 unusually high percentages of long-stay patients. 14 recommendation was in response to concerns we heard from the 15 hospice community about the need to target regulatory 16 scrutiny toward those providers where it is most warranted. 17 So that concludes the presentation. 18 This forward to your discussion and any questions. 19 MR. HACKBARTH: 20 Prepare yourselves for this. 21 start with you. 22 this is a big change. Okay. I look Thank you, Kim. Herb, we're going to We're going to change the order here. No, 120 1 2 3 4 DR. MARK MILLER: [off microphone] Wait, wait. We didn't -MR. KUHN: You are making sure everybody is staying awake today, aren't you? 5 I support the recommendation. 6 DR. BERENSON: 7 DR. HALL: As do I. 8 DR. DEAN: I support the recommendation. 9 DR. CHERNEW: I support the recommendation. I support it. As do I. 10 MR. HACKBARTH: 11 MR. BUTLER: 12 [Laughter.] 13 MR. BUTLER: People are speechless. Not so easy. I have an observation on Slide 4 that 14 I would -- I support the recommendation, but it did start 15 piquing my interest that this shows obviously the dramatic 16 decrease in the percent that are cancer, and then in the 17 text it shows there are about a little more than double the 18 number of users over the same time frame, which suggests 19 that cancer has increased but not much, which is kind of 20 surprising to me. 21 22 But the other thing that it triggered is that I wonder in all these per diem payments, we're almost starting 121 1 to get into a case mix thing. 2 provided for these -- I'm just speculating that some of 3 these areas require less care than others on a per diem 4 basis in terms of the challenge of the disease. 5 MS. NEUMAN: Is the palliative care So in our look at sort of service 6 intensity, what we've seen -- and this is all based on the 7 number of visits. 8 length of stay that is the biggest determinant of sort of 9 the average number of visits you get per week over your What we've seen is that it's really 10 episode rather than diagnosis, but diagnosis is highly 11 correlated with length of stay. 12 We do see secondary effects of different types of 13 service mix. 14 as a share and less aide visits, and so there's potentially 15 those kinds of differences, also duration differences which 16 we will look at. 17 but our sort of first look at this has found that length of 18 stay is much more of a driver than diagnosis. 19 Cancer patients tend to have more nurse visits So we're continuing to investigate that, MR. BUTLER: Even the nature of a nurse's visit 20 for a cancer patient with certain kinds of pain could be 21 very different from a nursing visit for something -- for one 22 of the other neurological diseases, for example, right? 122 1 MS. NEUMAN: Yeah, that's true. 2 MR. BUTLER: Okay. 3 MS. UCCELLO: I support the recommendation. 4 DR. BAICKER: I think that extra detail on 5 decomposition of length of stay by type is really helpful, 6 and I support the recommendation. 7 MR. ARMSTRONG: So do I. You've heard me say 8 before I'm glad to see that the utilization of hospice 9 services is going up, but I think it's really smart the way 10 that we've pulled together a recommendation to adjust 11 payments to neutralize the financial benefits of long- 12 length-of-stay patients. 13 different recommendations here is a very nice one, so I will 14 support this, too. So I think the balance between our 15 DR. BORMAN: I support the recommendation. 16 MR. GRADISON: 17 DR. CASTELLANOS: As do I. I have a couple of 18 clarifications. I've been traveling. I've been out of the 19 country, actually, but I saw something about the OIG just 20 completed a report or a study against a few of the hospices, 21 and I'm just curious what that was about and whether that 22 explains any of the data we've seen in the past. Are you 123 1 familiar with that? 2 3 MS. NEUMAN: Are you talking about the OIG study of hospices that heavily focus on nursing homes? 4 DR. CASTELLANOS: 5 MS. NEUMAN: I think so, yeah. Yeah, so there's some discussion in 6 the chapter about that, and they found some results 7 consistent with what we found, that patients who are in 8 nursing homes tend to have longer stays, diagnoses that tend 9 to require a mix of care, more aide services, less nurse 10 services, those kinds of things, and they recommended that 11 CMS monitor this and also to lower the payment rates for 12 hospice and nursing homes. 13 DR. CASTELLANOS: Just another clarification. On 14 Slide 2 there's about $1.1 million and we spend over $13 15 billion. 16 does that fit into other terminal care costs per patient? 17 Like Scott, I'm a big supporter of hospices. MR. HACKBARTH: How Are you asking, Ron, whether costs 18 for terminal patients are higher in hospice or lower in 19 hospice than -- 20 21 22 DR. CASTELLANOS: stand. I'm just asking. I just don't know where we I don't know. DR. MARK MILLER: Yeah, we've gone through this. 124 1 I'm just going to trigger you to start talking about it. 2 think the way I would answer this question is the studies 3 that say is hospice more or less expensive than staying in 4 the acute-care system is highly dependent on when the 5 patient hits the system and how long they stay. 6 pick it up here if you wanted to. 7 MS. NEUMAN: I You could Yeah, and also diagnosis is a factor. 8 Most of the research shows that it saves for cancer, 9 generally; it's more mixed for the other diagnoses. 10 11 Some studies say yes, other studies say no. DR. MARK MILLER: Ron, the reason we kind of went 12 through a lot of this, and also to Kate's question about the 13 increase in the length of stay, we were seeing all these 14 increases in lengths of stay, and some people were 15 asserting, well, this is just a shift in the diagnoses. 16 we were also documenting financial incentives to increase 17 the length of stay that I think was just mentioned over 18 here, and that kind of got us into some of the questions 19 that you were asking about, and it's highly dependent on 20 whether -- when people say hospice saves money, it's highly 21 dependent on sort of where that experience occurs. 22 people bring them in very early and keep them for long If But 125 1 periods of time, then it doesn't save money. 2 DR. CASTELLANOS: I guess you're digging down to 3 each individual or individual category. 4 it in bulk and wondering if -- how do those numbers match up 5 to other terminal-care care? 6 don't, you know, maybe you can get back or something. 7 MR. HACKBARTH: I'm just looking at Do we have any idea? If we You need to dig down to particular 8 situations as opposed to think about it in the aggregate. 9 And I think we talked about this some in December, I 10 believe, and I think Mike reminded us that whether it costs 11 less or more, that's not the only issue at stake here. 12 it's the care that the patients need at the end of life, 13 it's a good thing to do, and it shouldn't be solely 14 evaluated on whether it saves money or not. 15 that, there has been research trying to assess whether the 16 hospices save money, and Kim gave you the brief summary of 17 that. 18 DR. CASTELLANOS: 19 totally agree. 20 If But having said I appreciate your answer, and I I'm a supporter of hospice. looking at those numbers. 21 MR. HACKBARTH: 22 DR. STUART: I was just Okay. I support the recommendations, Kim. 126 1 Thanks for answering my question. 2 putting in the previous recommendations. 3 MR. GEORGE MILLER: And, Glenn, thank you for Support, and thanks for the 4 answer to the questions and the previous recommendations as 5 well. 6 MS. BEHROOZI: Yes, I support the recommendations, 7 but, you know, especially with the reiteration of the prior 8 recommendations, and I'm just going to use my last 9 opportunity to emphasize the point on volunteers. You know, 10 it just struck me. We were talking about for-profits having 11 almost twice the margins of not-for-profits, and, you know, 12 we're worrying about spending money unwisely in the program. 13 It just seems like if for-profits want to enter the field 14 that was when it was established an all-voluntary field, an 15 all not-for-profit field, they should have to enter it like 16 any other market where they have to hire people and provide 17 all the services that their patients will want in order to 18 feel satisfied. 19 the services for which they use volunteers have higher 20 patient satisfaction, well, that's what, you know, providers 21 spend money on in other contexts. 22 the things that will make their patients satisfied. The fact that places that use volunteers or They spend money on doing And so 127 1 certainly on the for-profit side, they should have to do 2 that in hospice, and maybe we'll -- and so if the 3 requirement to have volunteers as part of the mix is 4 eliminated, volunteers certainly can still participate. 5 They will still find their way to the not-for-profits. 6 Maybe the for-profits will find they still have to keep 7 providing those services, but they'll have to pay to provide 8 them in order to retain patients, but their margins will 9 come down, and we'll know better what the money is being 10 11 spent on and not just going into profit. DR. NAYLOR: I support the recommendations. I 12 also hope that as we go forward -- because I think the 13 Commission has really encouraged the shared responsibility 14 and accountability for reporting of quality measures, and I 15 think the expert panel that you brought and the 16 recommendations that came from that both pointed to 17 challenges in the existing measures, especially measurement 18 of pain among the growing population of neurological -- 19 those who have neurological conditions. 20 can continue the conversation about advancing expectations 21 around quality reporting measures, building on the expert 22 panel's idea. So I hope that we 128 1 MR. HACKBARTH: 2 recommendation. 3 Okay. It's time to vote on the All in favor of the recommendation, please raise your hand. 4 Opposed to the recommendation? 5 Abstentions? 6 Okay. 7 MR. HACKBARTH: Thank you, Kim. Next up is public comment. And I 8 would ask people to keep their comments to no more than two 9 minutes. Begin by identifying yourself and your 10 organization. 11 the end of your time. 12 When the red light comes on, that signifies I would also remind people that this isn't your 13 only or even your best opportunity to provide input into the 14 Commission's work. 15 this morning, Commissioners are eager to get input from 16 various sources and various mechanisms and pay close 17 attention to it, so please avail yourself of all of the 18 alternatives, most importantly working with the MedPAC 19 staff. As people could see from the discussion 20 Sharon. 21 MS. McILRATH: 22 Sharon McIlrath, AMA. looking for a clarification. I just was If the Congress were to enact 129 1 your proposed changes on physician payment, i.e. a freeze 2 for the primary care and 5.9 percent a year cuts for the 3 other physicians who are providing those services, does that 4 then transfer over to the visits that are done in the 5 hospital clinics or not? 6 answer is problematic. 7 consistent? 8 problem that they have with the reductions that they're 9 getting. 10 11 And either way, it seems like the If the answer is no, then how is it But if it does, then it even expands the MR. HACKBARTH: question, Sharon. Consider that a rhetorical You -- 12 MS. McILRATH: 13 Well, honestly -- MR. HACKBARTH: You know the ground rules here. 14 We don't engage in dialogue. 15 time, we can talk about that, but for now, we'll go on to 16 the next speaker. 17 MS. BAER: Okay. If you want to talk at another My name's Ivy Baer. I'm with 18 the Association of American Medical Colleges and I just want 19 to go on the record as saying that we're very disappointed 20 that the Commission voted to reduce payments in outpatient 21 departments. 22 adversely affect access, as a number of the Commissioners We have very serious concerns that this will 130 1 noted. 2 it may be too late, that a lot of these clinics may have 3 either closed or already reduced the services that they're 4 providing and we don't really think that it will be helpful 5 to not look at this until 2015. 6 And although there is a study, we feel that by 2015, And finally, we certainly regret that we didn't 7 have information about the stop losses. We were unable to 8 model that and did not include anything in the letter that 9 we've already submitted to you about that, but we will be 10 returning back to our offices and taking a look at that to 11 see what the actual impact would be on our members. 12 you. 13 MS. UPCHURCH: Hi. Thank My name is Linda Upchurch and 14 I work with NxStage Medical. 15 December meeting to hear a great deal of discussion around 16 more frequent dialysis and the benefits of more frequent 17 dialysis, and so I was discouraged to not hear that 18 continued in conversation, particularly in this meeting. 19 I was very encouraged in the I want to put a challenge out for you to continue 20 to look at the mountain of clinical evidence showing the 21 benefits of more frequent dialysis in survival, in 22 morbidity, in transplantation. The list goes on and on -- 131 1 in bone disease. 2 It's important that you continue to look at it. 3 asked for additional follow-up from Nancy, and Nancy does a 4 great job in providing that follow-up. 5 and hope that that discussion will continue. 6 It's important that you focus on that. I know you So I can only assume But to point out again that the disparities in 7 access to more frequent dialysis are evident with Medicare 8 beneficiaries routinely being discriminated against if you 9 look at the population on more frequent dialysis versus 10 traditional thrice-weekly. 11 these patients has improved. 12 are improved. 13 status are improved. 14 attention in the March report and in meetings subsequent to 15 this to focus on the benefits of more frequent dialysis. 16 Thank you. 17 Again, the transplant status of Survival and cardiac outcomes Health-related quality of life and employment MS. HUANG: It's certainly worthy of time and Hi. My name is Xiaoyi Huang with the 18 National Association of Public Hospitals and Health Systems, 19 and our members are two percent of hospitals but provide 20 20 percent of all hospital uncompensated care. 21 22 The impact of reducing payment for E&M services would disproportionately impact these hospitals. They're 132 1 estimated to absorb some 15 percent of MedPAC's annual 2 estimated impact of $1 billion. 3 that the recommendation number two doesn't adequately 4 protect these hospitals and that the third recommendation 5 isn't strong enough and may come a little too late. 6 safety net hospitals serve critical roles for vulnerable 7 populations. 8 homes for these patients. 9 multiple chronic illnesses and mental and behavioral health We are also disappointed These They provide primary care and serve as medical And these patients also have 10 issues and require more comprehensive services in the 11 outpatient setting that's not available in the freestanding 12 physician office. 13 MR. MAY: Thank you. Thank you. I'm Don May with the 14 American Hospital Association. 15 my colleagues at the AAMC and NAPH have said. 16 I'll try not to repeat what We are very disappointed with this recommendation 17 and we think this will lead to access concerns. 18 the transition is helpful and an improvement over the last 19 recommendation that was here in December, it's simply not 20 enough. 21 22 And while We also think it's misleading that the impact is being shown overall and not at the outpatient PPS system. 133 1 We would encourage you to show that in the chapter. 2 look at this, we're looking at taking assistance from 90 3 cents on the dollar down to 87 cents on the dollar when the 4 transition is over, and then don't forget the two percent 5 sequester. 6 dollar of costs. 7 When we So we're really talking about 85 cents on the Our objections are really not about keeping the 8 status quo. Hospitals are doing a lot to coordinate care, 9 improve quality, and lower costs. But simply making this 10 cut is not a first step, it's a misstep, and that makes a 11 broken system even more broken. 12 Finally, one of the most concerning parts of this 13 recommendation is ignoring the costs of the very system you 14 are trying to pay, the outpatient system, and taking away 15 that link to the hospital-level costs. 16 going to move away from what is adequate for a payment 17 system, MedPAC really needs to look at how are stand-by 18 capacity going to be covered. 19 stand-by costs and we're going to pay site neutral, how do 20 we then ensure that hospitals have the funding to keep an 21 emergency room open 24/7, to be able to be in communities 22 where private physicians aren't practicing, and to be able If the system is If we're not going to look at 134 1 2 3 4 5 to have those services available that we can improve care. We want to move forward. backwards isn't the right way. MR. HACKBARTH: Okay. Taking us two steps Thank you. We're adjourned for lunch and we'll reconvene at 1:30. 6 [Whereupon, at 12:36 p.m., the meeting was 7 recessed, to reconvene at 1:30 p.m., this same day.] 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 135 1 AFTERNOON SESSION 2 MR. HACKBARTH: Okay. [1:33 p.m.] It is time to begin our 3 afternoon session with skilled nursing facility payment. 4 Carol? 5 DR. CARTER: Okay. Before I get started, I wanted 6 to follow up on a couple of questions from the December 7 meeting. 8 Cori, you asked about how much variation there is 9 in Medicare and Medicaid days across SNFs, and I added that 10 11 information to the chapter. Bill Hall, you asked about hospice use in nursing 12 facilities, and the association reports that about 6 percent 13 of nursing home residents receive hospice services. 14 I'll start with a thumbnail sketch of the 15 industry, and on this slide you can see the numbers of 16 providers and users and spending and the Medicare shares. 17 We went over a lot of this, the next set of information, at 18 the December meeting, and all the detail is in the chapter, 19 and so I'm going to go through the update very quickly. 20 MR. HACKBARTH: 21 DR. CARTER: 22 MR. HACKBARTH: Carol, could I just interrupt? Yes, sure. It occurs to me that for purposes 136 1 of people who weren't here this morning, I ought to say that 2 I have asked the staff to keep the presentations today as 3 concise as possible so they will not be reviewing in detail 4 material that was discussed at the December meeting when we 5 talked about the draft recommendations for updates. 6 I've asked Carol and the others to focus on what questions 7 came up in December, and as I say, keep it to the point. 8 So, Carol, thank you. 9 DR. CARTER: And So, in summary, access appears stable 10 for most beneficiaries. 11 year. There was no change in bed days or in occupancy 12 rates. Covered days and admissions decreased slightly 13 between 2009 and 2010, and that reflects the lower hospital 14 use which is required for Medicare coverage. 15 Supply is about the same as last In terms of quality, risk-adjusted rates of 16 community discharge and rehospitalization showed almost no 17 change between 2008 and 2009. 18 capital, capital was adequate this year, and lending and 19 borrowing is expected to be slow in 2012, reflecting 20 uncertainties about state and Medicare policies. 21 22 In terms of access to Comparing payments and costs, the aggregate Medicare margin for free-standing SNFs was 18.5 percent in 137 1 2010. 2 this slide, and in part this reflects cost differences 3 across facilities. 4 costs per day that are 7 percent higher than other 5 facilities, after adjusting for differences in wages and 6 case mix. 7 that raise payments for high therapy care above these 8 patients' costs. 9 There is quite a bit of variation, as you can see on For example, nonprofit facilities have The variation also reflects the biases in the PPS We use consistent performance over three years to 10 define a group of SNFs that are relatively efficient, and 11 relatively efficient SNFs had costs that were 10 percent 12 lower, community discharge rates that were 38 percent 13 higher, and rehospitalization rates that were 17 percent 14 lower compared to other SNFs. 15 To project margins for 2012, we model revenues and 16 costs using the assumptions on this slide. 17 any behavioral responses to current policies, and we project 18 the average margin in 2012 to be 14.6 percent. 19 We did not model The projected margin for 2012 continues the trend 20 of double digit margins in this sector since 2000, 21 indicating that the PPS has exerted too little fiscal 22 pressure on providers. Several other facts support the need 138 1 to rebase payments: 2 3 The variation in Medicare margins is not explained by differences in patient mix; 4 5 Cost differences are not explained by differences in wage levels, case mix, or beneficiary demographics; 6 Some SNFs have both low costs and high quality; 7 And some MA payments are considerably lower than 8 fee-for-service payments. 9 Three key concerns have been raised about rebasing 10 Medicare payments. 11 reduced by 11 percent in 2012. 12 reduction was made to correct unintended overpayments in 13 2011 that we estimate resulted in margins in the 24 percent 14 range. 15 be over 14 percent in 2012. 16 First, Medicare's payments were already But remember that this Even after the reductions, we estimate margins will A second concern is that some argue that 17 facilities need high payments from Medicare to finance low 18 payments from Medicaid. 19 subsidize Medicaid payments is poor policy for a number of 20 reasons that we went through in December and are summarized 21 on the slide. 22 with high Medicaid mix, then a separately financed, targeted However, using Medicare payments to If Congress wishes to help nursing facilities 139 1 program should be established to do this. 2 A third concern is that the variation in Medicare 3 margins could mean that some SNFs would fare poorly with 4 rebased payments. 5 performance under Medicare partly reflects the shortcomings 6 of the PPS. 7 PPS would correct these known shortcomings and would shift 8 payments from SNFs with high shares of patients receiving 9 intensive therapy to those with high shares of medically We appreciate that the financial Your recommendations back in 2008 to revise the 10 complex patients. 11 across different types of patients, the differences in 12 Medicare margins would be narrower. 13 for hospital-based facilities, nonprofit facilities, and 14 SNFs that treat high shares of medically complex patients, 15 and these are facilities that have, on average, lower 16 Medicare margins than other SNFs. 17 By improving the equity of payments Payments would increase While the empirical basis and policy rationale for 18 rebasing are sound, we recognize the need to proceed 19 cautiously but deliberately to help ensure that there are no 20 unintended disruptions caused by rebasing. 21 recommendation includes several elements that reflect this 22 caution. The draft It states that the PPS should be revised first so 140 1 that payments are redistributed before reductions occur. 2 Rebasing would not occur until the following year, in 2014. 3 It also acknowledges that a transition is important. 4 reductions would be taken in steps, with the reduction in 5 2014 seen as the first step in aligning payments and costs. 6 And we will be monitoring industry performance each year as 7 rebasing is implemented. 8 9 The draft recommendation reads: The The Congress should eliminate the market basket and direct the Secretary 10 to revise the prospective payment system for skilled nursing 11 facilities for 2013. 12 initial reduction of 4 percent and subsequent reductions 13 over an appropriate transition until Medicare payments are 14 better aligned with provider costs. 15 Rebasing should begin in 2014, with an Last month, we talked about how both revising and 16 rebasing are necessary to reform the PPS. 17 corrects the distortions in the payment system, and the 18 second brings payments closer to costs. 19 The first This recommendation would decrease program 20 spending relative to current law. For 2013, the preliminary 21 projection is that spending would be lowered by $250 to 750 22 million lower, and over 5 years it would be between $5 to 10 141 1 billion lower. 2 all types of care may result in better access for 3 beneficiaries. 4 differences in Medicare margins across facilities will be 5 smaller. 6 of their mix of patients and their current practice 7 patterns. 8 differences in Medicare margins due to the large variation 9 in costs. 10 For beneficiaries, fairer payments across Provider payments will be lower, but the Impacts on individual providers will be a function The recommendation would not eliminate all of the Last month, we also talked about a policy to 11 discourage unnecessary rehospitalizations. 12 such a policy are to improve the care beneficiaries receive 13 in SNFs, improve the transition care, and to lower program 14 spending on rehospitalizations that could have been averted. 15 Because not all factors that influence rehospitalization are 16 within a provider's control, it is important that a policy 17 accommodate the variation across patients and their changing 18 circumstances, but still encourage SNFs to improve the care 19 that they furnish. 20 The goals of This slide shows the wide variation in 21 rehospitalization rates for five potentially avoidable 22 conditions across facilities and suggests that there is 142 1 ample opportunity for some SNFs to lower their 2 rehospitalization rates. 3 Last month, you discussed the key parameters of a 4 rehospitalization policy: 5 the time period covered by the policy, and the importance of 6 using a rate to gauge performance to avoid evaluating how 7 individual cases were handled. 8 multiple years ensures that providers are not being 9 penalized for having one "bad" year. 10 the definition of the measure, Looking at rates over Mary, you asked about the alignment of the 11 hospital and SNF incentives if a SNF rehospitalization 12 policy was implemented. 13 are at risk under a rehospitalization policy that would 14 include a 30-day window after discharge from the SNF. 15 hospital stay is shown in the first bar and the SNF stay is 16 in the second bar. 17 30 days of discharge from the hospital, both the hospital 18 and the SNF would be at risk, helping to ensure good 19 transitions between these providers. 20 red bars. 21 22 This chart shows which providers The If a rehospitalization occurred within That's the overlapping If the rehospitalization occurred on day 31, however, only the SNF would be at risk. Some asymmetry in 143 1 risk may make sense since at some point the hospital is less 2 accountable for care provided considerably after a 3 beneficiary is discharged from it. 4 after the beneficiary is discharged from the SNF, the SNF 5 would be at risk for rehospitalizations that occurred within 6 30 days, helping to ensure adequate transitions after the 7 SNF stay. 8 9 Looking at the risks And with that summary of the policy, here is the draft recommendation: The Congress should direct the 10 Secretary to reduce payments to skilled nursing facilities 11 with relatively high risk-adjusted rehospitalization rates 12 for their Medicare-covered stays. 13 The recommendation language is purposefully 14 general to indicate flexibility about the measure and time 15 periods covered by a policy. 16 the recommendation outlines a phased approach that would 17 allow CMS to move forward with a policy with a risk-adjusted 18 measure that is readily available. 19 could be expanded to include a time period after discharge 20 from the SNF and could include all causes for the 21 rehospitalization once risk-adjusted measures become 22 available. In the chapter, the text below Over time, the measure The key point is to put a policy in place so 144 1 that providers have an incentive to lower their 2 rehospitalizations. 3 In terms of implications, it would lower spending 4 relative to current law due to fewer hospital stays and 5 payments would be lower for those SNFs with high rates. 6 Fewer beneficiaries would be hospitalized unnecessarily, and 7 more beneficiaries would receive better transition care. 8 Payments to providers with patterns of high rates would be 9 lower. 10 As required by PPACA, we include information on 11 Medicaid trends in spending, utilization, and financial 12 performance for nursing homes. 13 chapter, but here are the highlights. 14 between 2009 and 2010, and while spending decreased 15 slightly, since 2002 spending has increased 16 percent. 16 Average daily payments vary twofold across states. 17 non-Medicare and total margins improved between 2009 and 18 2010. 19 while total margins were positive. The details are in the Service use increased Both In 2010, non-Medicare margins were slightly negative 20 And with that, I look forward to your discussion. 21 MR. HACKBARTH: 22 Since these recommendations sort of fit together Okay. Thank you, Carol. 145 1 as a group, I think we should discuss them all in one round, 2 and Bill is going to lead off. 3 DR. HALL: I'll speak in favor of the 4 recommendation. One thing I have been kind of puzzling 5 about are some of the discrepancies between the for-profit 6 and nonprofit nursing homes and whether there are some 7 unintended consequences of our recommendations. 8 Specifically, right now in many parts of the country 9 Medicare patients are considered really excellent patients 10 for SNFs. 11 the alternative payment, which is almost invariably 12 Medicaid. 13 The payment is good -- it flows -- relative to The other way that costs can vary is the sort of 14 subtle quality of the experience that people are given. 15 There aren't very many ways to cut overhead in an SNF 16 facility. 17 There are fairly stringent requirements that tabulate kind 18 of the daily activities of staffing. 19 human part of it that gets this into trouble, and at least 20 some people feel that the nonprofit homes are the ones that 21 have some kind of an institutional affiliation, seem to do a 22 better job of that. Usually there are state laws about staffing. But it's sort of the 146 1 So I guess I'm saying I think that we should do 2 this, but I think we also have to watch very carefully as to 3 whether we exacerbate these kind of discrepancies between 4 the for-profit and nonprofit facilities. 5 DR. CARTER: There are two factors that are going 6 to influence how facilities fare under the rebasing and the 7 revised PPS, and those are: 8 And the second is: 9 much therapy are they currently furnishing? 10 11 What is their cost structure? What's their mix of patients and how Because the revised PPS really does move money away from therapy. So if you're a nursing home that has a high cost 12 structure and is really in the high therapy business, you're 13 going to be the most affected by this. 14 DR. HALL: 15 DR. CARTER: Right. But there's a lot of institutions 16 that really didn't get into the therapy game, and so the 17 impact on them is much less. 18 factors that affect kind of the distribution of the impacts. 19 DR. HALL: Yeah. And so it's really those two Well, the data you presented on 20 the margins really gives me confidence that we should 21 proceed with this. 22 DR. DEAN: I support the recommendations. I guess 147 1 I don't have any other questions. 2 DR. CHERNEW: 3 MR. BUTLER: Me, too. Just a thought. I support the 4 recommendations, and I think it's a very important step to 5 get the SNFs involved in the rehospitalization efforts. 6 They're not equipped in general with the data or with kind 7 of the infrastructure and the case management typically that 8 the hospitals have in this. 9 suggest in terms of anything in the chapter, but I think And I don't know what to 10 just as we had kind of the QIO discussion and things like 11 that, I'm wondering whether we need to think ahead and how 12 to support kind of the effort that it will take to do this 13 correctly and not just put the recommendation on the table, 14 because I think it will be an important part of the 15 partnership. 16 17 18 MR. HACKBARTH: Would you say a little bit more, Peter, about what kind of data you're referring to? MR. BUTLER: Well, let me contrast the hospital 19 side, which there is much electronic information available, 20 there's discharge planners, there's a whole army of people 21 that has analytical tools that already know where patients 22 are being discharged to and with which diagnosis and where 148 1 to zero in on the focus. 2 their side are sitting with paper records, are sitting with 3 a gut sense of where they're sending people, under what 4 conditions, but they don't have kind of the same information 5 to kind of understand where they are falling short now and 6 where the opportunities are for improvement. 7 MR. HACKBARTH: I think that the nursing homes on So, Carol, would you put up your 8 bar graphs where you showed the responsibility? Within the 9 hospital readmission window, where there's a shared 10 responsibility, that's not the part you're worrying about. 11 You're more focused on the discharge from the SNF issue 12 where they may be discharging into the community and not 13 know as much as hospitals know about their patients. 14 15 MR. BUTLER: Well, and also sometimes the discharge is back to the hospital as well, and -- 16 MR. HACKBARTH: 17 MR. BUTLER: Well, yeah. Wherever that discharge is going I 18 think is -- you're right, it's -- I don't think they have 19 the same information, and, yeah, you're right, they get out 20 into the home, and they don't -- if they don't have the 21 information, you can imagine what sits out in the home, 22 almost nothing sometimes once they're discharged. 149 1 MR. HACKBARTH: 2 MS. UCCELLO: Okay. I support the recommendations. I 3 think taken together they're a really good, important set of 4 recommendations. 5 too high, they're distributed poorly, and there's not enough 6 that's targeting quality. 7 more toward value. 8 9 It's clear that the levels of payments are So, taken together, this moves us And thank you in particular for adding in the distribution of shares. I think it helps more explicitly 10 show that Medicare would not -- using Medicare payments to 11 help offset low Medicaid rates is not well targeted. 12 DR. BAICKER: I support the recommendations, and I 13 think it makes a lot of sense to be flexible in 14 recommendation 2 about the manner in which 15 rehospitalizations would be incorporated into payments, 16 especially because we think we're going to be getting better 17 information over time. 18 MR. ARMSTRONG: Yeah, I do, too. I think just to 19 build on a couple points already made, I like how we are 20 thinking about within a section of our payment structure 21 we're organizing payments that actually extend beyond it as 22 a mechanism for sort of slowly working our way toward 150 1 something that just kind of holds together a little bit 2 better. So it is, I think, just good to point that out. 3 4 Was this recommendation in the packet of our SGR proposals? 5 MR. HACKBARTH: The rebasing? 6 MR. ARMSTRONG: Yeah. 7 MR. HACKBARTH: Yes. 8 MR. ARMSTRONG: Was it? 9 DR. MARK MILLER: 10 MR. ARMSTRONG: Okay. And the rehospitalization. I just couldn't remember. And 11 then another point, and it's not specific to Carol or this 12 but more broad. 13 recommendations that in the next couple of years make some 14 payment structure change, and we're saying at the same time 15 let's monitor the impact of this and, if we need to, let's 16 make adjustments to that. 17 actually felt like we over-rebased and had to make an 18 adjustment? 19 It seems that we're approving more MR. HACKBARTH: My question is: Have we ever Rebasing, the first provider group 20 for which we recommended rebasing was home health, and that 21 is -- remind me, Mark, the schedule for the home health 22 rebasing. 151 1 DR. MARK MILLER: 2 MR. HACKBARTH: 2013 Yeah, so it begins in 2013. So 3 there isn't a case where it's actually happened and we've 4 had a chance to evaluate the impact. 5 MR. ARMSTRONG: Maybe my question shouldn't be so 6 specific to just rebasing, but it seems like when we're 7 making what seems like fairly significant adjustments to our 8 payment structure or the payment rates themselves, we often 9 talk about, well, and let's monitor this over the next 10 couple of years and make adjustments if we feel like it 11 didn't really turn out the way we thought it might. 12 just wondering whether actually it ever turned out 13 differently than we thought it might, or whether we're just 14 inherently just very conservative about this and that -- I'm 15 just interested. 16 MR. HACKBARTH: 17 DR. MARK MILLER: And I'm Yes, do you have some thoughts? The only thing I would say, 18 actually I've never thought about that question, so I 19 actually -- and I wouldn't want to answer it on the fly. 20 is an interesting question. 21 22 It What I would say, though, is more of a looking -an evolution in thinking looking forward where people -- you 152 1 know, and thinking about my experience, which has been 2 several years now, where people are like these payment rates 3 and these margins, for example, just don't make sense, and 4 people would talk about payment rates. 5 unsatisfactory, and we started driving down into the 6 underlying payment systems and finding things like, look, 7 this is therapy and medically complex. 8 not a very scientific statement, but you see what I'm 9 saying. But then that was You know, this type of thing. I realize this is And also beginning 10 to build around that things like -- and, you know, hospice 11 is a good example of this, looking at patterns that aren't 12 fraud but appear to be abuse. 13 And so the evolution of thinking is you can't just 14 push down on this rate. 15 the underlying guts and the world that it's operating, and 16 I've seen that kind of evolution as opposed to what you've 17 said where we start something and then we go, Wait a second, 18 stop it. 19 getting a little more detailed about it. 20 21 22 You've got to start thinking about It's more as we've been going forward we've been But I'm going to think about your question directly. MR. ARMSTRONG: The reason why I ask it is that my 153 1 sense is that we are inherently so conservative and that 2 we're cautious and we make recommendations and then we sort 3 of mitigate the potential impact by saying we're going to 4 monitor the impact. 5 little bit, it might free us to be a little less 6 conservative about some of these things. 7 know that for sure. 8 9 I just think if we reflected on that a MR. HACKBARTH: worthy of more thought. But I don't really It's a good question, Scott, and My initial reaction to it was, you 10 know, our update recommendations, other than the rebasing, 11 again, they're always incremental, and so you wouldn't 12 expect huge effects. 13 The area that occurs to me where there was a more 14 significant change was in the case mix change in the 15 hospital where we've moved around a fair amount of money 16 based on recommendations for changing how the weights are 17 derived. 18 for example, the effect of that on the development of 19 physician-owned specialty hospitals. 20 And, you know, I don't know if we've looked at, Now, there are some confounding factors like the 21 moratorium and the like. In fact, this was an issue that 22 George and I talked about on the phone the other day. So, 154 1 you know, that would be an opportunity to look at a 2 significant policy change that we made and what effect did 3 it have in the real world. 4 that we anticipated or a different one? 5 So a good question. 6 DR. BORMAN: 7 MR. GRADISON: 8 DR. CASTELLANOS: 9 DR. STUART: Did it have the sort of effect More thought on that. I support the recommendations. I do as well. So do I. I support the recommendations. I do 10 have a question about the rehospitalization, and that is, 11 whether there has been an effort to determine whether this 12 might have unintended consequences for certain patients in 13 the sense that the SNF would have an incentive not to accept 14 patients with a higher risk of rehospitalization independent 15 of what they might be able to do about it. 16 something that you've looked at? 17 DR. CARTER: Is this No, I haven't, but I understand where 18 your question is coming from. If a patient has something 19 like CHF where those patients typically, if they're not well 20 managed, do get rehospitalized, but, I mean, part of this 21 recommendation -- and at least starting with potentially 22 avoidable, those are in theory conditions that not for every 155 1 patient but in general one can manage the care to try to 2 lower rehospitalization. 3 4 DR. STUART: hospitalizations than those that might not be avoidable. 5 6 I'm less worried about the avoidable DR. CARTER: Well, some of those are not predictable. 7 DR. STUART: Then the question is whether this 8 policy would make a distinction between rehospitalizations 9 for CHF or COPD or others that might be managed well in a 10 well-run SNF from others over which the facility really has 11 no control 12 DR. CARTER: Well, and I think that we've left 13 this recommendation general. 14 how you define the measure and whether -- I mean, we've 15 talked about potentially avoidable because those are risk 16 adjusted. 17 cause, but they want to throw out or not include things that 18 are scheduled, planned readmissions, like in chemotherapy or 19 something. 20 measure could maybe address some of your concerns. 21 22 There's a lot of debate about Some of the industries think that they like all And so that I think in how you define the DR. MARK MILLER: I would say it's actually, I think, somewhat striking to both Carol and me that we 156 1 thought, in part from some of our previous conversations a 2 few years ago, that the industry's general posture would be 3 potentially avoidable. 4 down to places where there are cases that can be avoided. 5 Remember, this is a rate-based thing, not specific cases. 6 But we've been having conversations with the SNFs on this, 7 and they've come in and had a lot of interesting ideas about 8 how to approach this, and they're not taking a position yet 9 or anything, but some discussion about wanting, some of Let's make sure that we define this 10 them, all-cause measures. 11 the effect and the organization is more broad based instead 12 of focused on specific either activities or patients and 13 some argument that that's a more effective way of going at 14 things. 15 And just because it means that And so I think what Carol and you are trying to do 16 in this conversation right now is leave some flexibility to 17 approach this from a couple of different direction. 18 at least somewhat surprising for us that they said those 19 things. 20 MR. GEORGE MILLER: 21 MS. BEHROOZI: 22 It was I support the recommendations. I support the recommendations, particularly because of the -- I think you used the word 157 1 "caution," Carol, with which it proceeds. 2 about protecting those that are providing services that we 3 want them to keep providing them for the patients that do, 4 but, in fact, rewarding them, you know, shifting the balance 5 to make it easier for them to do that and not lose money. 6 And so it's an improvement of access policy, recommendation 7 1, so I think that's great. 8 9 DR. NAYLOR: I support the recommendation, as I chop myself and need a rehospitalization. 10 DR. NAYLOR: I was -- [Laughter.] 11 It's not just But I do want to highlight how 12 building on Peter's comment, you know, the industry, a lot 13 of attention has been paid in the last decade to 14 rehospitalizations focused on hospitals and technical 15 assistance, starting with the QIOs and building through CMS 16 efforts, including CMMI's Partnership for Patients, 17 community-based care transitions, a lot of attention on 18 getting better data, data flow, getting better competencies 19 in the workforce to be able to really manage people in those 20 24 hours after hospital discharge. 21 stress whatever opportunities might exist to help in this 22 plan to build the technical assistance and support that's So I think if we could 158 1 2 needed. The second thing, I would totally like the 3 rehospitalization, Slide 18, the policy, and I would also 4 stress the flexibility. 5 example, why I think your data show about 12 percent of the 6 SNF beneficiaries have a length of stay 0 to 5 days. 7 don't know how many of them are going back to the hospital, 8 but there might be a reason why something that happens 9 within 0 to 3 days is the hospital's accountability, not the I mean, there may be a reason, for Now, I 10 SNF's, or something in 24 hours. 11 flexibility in this policy as we monitor what is actually 12 the experience of people and who should be accountable might 13 be important to include. 14 MR. KUHN: So the notion of I support the recommendations, and like 15 many others, I'm particularly appreciative of the work on 16 the rehospitalization. 17 particularly in the post-acute care provider setting, is 18 going to be very critical in order to be successful across 19 the board here. I think better alignment, 20 In particular, I know in the draft chapter and I 21 think in a chart that was distributed at the last meeting, 22 where you looked at state by state in terms of the great 159 1 variation out there, there is remarkable variation around 2 the country; and I think anything we can do to get some 3 better alignment and move that forward will be great. 4 Then speaking of the chart that you have up here 5 on page 18 slide, I think that's a very good chart to show 6 about the shared risk. 7 horizontal bar on there that talks about home health, so 8 that if a person is discharged from the SNF into home 9 health, and, you know, maybe for future conversation we may But in the chapter you have another 10 want to think about rehospitalization policies for home 11 health in the future as we think about that. 12 that's a nice set-up in this chapter. 13 design, but I think this would be the next part in terms of 14 making sure all post-acute care settings are pulling in the 15 same direction as we look at this policy. 16 DR. MARK MILLER: And I think Maybe that was by And that is the thinking, and we 17 just simplified the chart to make the SNF-hospital point 18 here. 19 we'll be back to you eventually to talk about. 20 But the more complex one is there, and that is what DR. BERENSON: I support the recommendations. I 21 would reiterate that I am a big supporter of recommendation 22 2 to try to figure out how to provide incentives for SNFs to 160 1 address -- to reduce rehospitalizations not only for the SNF 2 patients but, as we talked about at the last meeting, as 3 developing some skills that would apply more broadly to 4 their larger nursing home population. 5 To Bruce's issue, it's not clear that -- depending 6 on how this is constructed, the nursing home might still be 7 better off accepting a penalty on a high readmission and not 8 forgo the patient. 9 constructed and all need to be flexible. 10 11 And so the details of how this is We can't be overspecific here. I like very much the idea that we will align the 12 incentives of the nursing home and the hospital while 13 separately there will be experiments in bundling the 14 payment, that actually bundling the payment is much more 15 complicated and involves other design features than simply 16 aligning incentives. 17 out -- whether we can be successful with this kind of 18 parallel strategy as opposed to the bundled payment, which 19 also deserves good testing. So it will be nice to know which turns 20 So I'm all in favor. 21 I think we're in a good place. 22 MR. HACKBARTH: This has been very good, and Thank you. Okay, Carol, would you put up the 161 1 first recommendation? 2 please raise your hand. 3 Okay. 4 Abstentions? 5 Okay. All in favor of recommendation 1, 6 All opposed to recommendation 1? Recommendation 2, please. All in favor of number 2? 7 Opposed? 8 Abstentions? 9 Okay. Number 3. 10 [Comment off microphone.] 11 MR. HACKBARTH: 12 DR. NAYLOR: 13 MR. HACKBARTH: 14 Oh, that's right. That's right. [off microphone] You're on a roll. Yeah, right. I'm trying to make up for the fact that I didn't ask for a vote -- 15 [Laughter.] 16 DR. MARK MILLER: 17 everybody? 18 19 MR. HACKBARTH: 22 Well, I'll leave others to judge whether I'm okay or not, but thank you, Carol. 20 21 [off microphone] You okay, Okay. isn't it? So where are we here? Good job. This is MedPAC, Inpatient rehab facility services. MS. AGUIAR: Thank you. Our analysis was 162 1 presented in detail in December and the Commissioners have a 2 paper with the details of the analysis in their mailing 3 materials. 4 the analysis and the draft recommendations. 5 During today's session, I will briefly review First, to address Commissioner questions from 6 December. 7 of IRFs and that is included in the mailing materials. 8 9 George, you asked for a map with the distribution Herb asked for more information on studies comparing outcomes for hip and knee patients across post- 10 acute care settings. 11 conclusively identify one setting as having better outcomes 12 for rehabilitation patients. 13 In general, research studies do not Peter asked for more information on the 14 differences in case mix between hospital-based and 15 freestanding IRFs and tables showing the top ten cases and 16 distribution of cases by tier are included in the mailing 17 materials. 18 have relatively similar patient populations. 19 In general, hospital-based and freestanding IRFs As a reminder, this slide presents some key 20 characteristics of IRFs. 21 characteristics are included in your mailing materials. 22 The details of these As a quick reminder, we use the same framework for 163 1 payment adequacy as the other sectors. 2 We will begin with access to care measures and 3 supply. 4 least one IRF located in every State. 5 2010 vary by IRF category, but the overall picture suggests 6 that supply is adequate. 7 In 2010, there were close to 1,180 IRFs with at Changes in supply in The number of rehabilitation beds and occupancy 8 rates are measures of IRF capacity. As you can see on the 9 top half of this slide, in 2010, there was a decrease of 10 hospital-based IRF beds and an increase in freestanding IRF 11 beds. 12 occupancy rates in 2010 were higher for freestanding IRFs 13 than for hospital-based IRFs. 14 occupancy rates remains above 62 percent. 15 number of beds and occupancy rates indicate the IRF capacity 16 is adequate to handle current demand. 17 As you can see on the bottom half of the slide, However, total industry Overall, both the This chart presents our measures of volume and 18 fee-for-service spending. As you can see, IRF volume 19 remained relatively stable in 2010. 20 spending increased by close to five percent, and payment per 21 case increased by three percent. 22 likely due to a 2.25 percent update to the base rates in Fee-for-service This spending increase is 164 1 2010, an increase in outlier payments, and an increase in 2 patient severity. 3 Turning now to quality of care, we worked with 4 researchers at RAND to develop preliminary risk adjustment 5 models. 6 between 2004 and 2009 across five quality indicators. 7 However, as you can see on the last row of the final two 8 columns, more than nine percent of IRF patients that were 9 initially discharged home were subsequently readmitted to As you can see, there was incremental improvement 10 the hospital, and almost three percent of IRF patients that 11 were initially discharged home were admitted to a SNF. 12 These represent areas for improvement in IRF quality of 13 care. 14 Turning to access of capital, hospital-based units 15 have access to capital through their parent institution and 16 their access to capital appears adequate. 17 although the cost of accessing the debt and equity markets 18 increased in 2011 for one major national chain of 19 freestanding IRFs, this chain is still able to access the 20 capital markets because of positive revenue growth. 21 22 In addition, In terms of Medicare margins, as you can see, margins increased from 8.4 percent in 2009 to 8.8 percent in 165 1 2010. 2 margins, with margins increasing as bed size increases. 3 Margins vary substantially between freestanding and 4 hospital-based IRFs. 5 almost 42 percent of total IRF spending, had over 21 percent 6 margins in 2010. 7 account for 58 percent of total IRF spending, had lower 8 margins of negative 0.2 percent. 9 There is also a relationship between volume and Freestanding IRFs, which account for In comparison, hospital-based IRFs, which It is likely the hospital-based IRFs have lower 10 margins because they tend to have lower occupancy rates, as 11 we saw on Slide 4, and have lower volume and higher costs. 12 More than half of hospital-based IRFs have less than 21 13 beds, and as we saw on the previous slide, margins for IRFs 14 with fewer than 21 beds are negative. 15 also have higher direct, indirect, and standardized costs 16 per case than freestanding IRFs. 17 hospital-based IRFs have negative Medicare margins, on 18 average, they have margins of 34 percent on their direct 19 costs. 20 IRF unit are 1.6 percentage points higher than acute 21 hospitals without an IRF unit. 22 units are able to cover their direct costs and financially Hospital-based IRFs However, even though We also see that margins for acute hospitals with an Therefore, we see that IRF 166 1 contribute to their parent hospital. 2 As we discussed, aggregate Medicare margins for 3 IRFs in 2010 were 8.8 percent. 4 Medicare margin for 2012, we modeled the policy changes 5 indicated on the slide for 2011 and 2012. 6 Medicare margins for 2012 will be eight percent. 7 To project the aggregate We estimate that In summary, our indicators of Medicare payment 8 adequacy for IRFs are positive. Supply and capacity are 9 stable and adequate to meet demand, and volume is relatively 10 stable. 11 indicate that quality incrementally improved since 2004 and 12 that there are still areas for quality improvement. 13 addition, access to credit appears adequate for both 14 hospital-based and freestanding IRFs. 15 that 2012 aggregate Medicare margins will be about eight 16 percent. 17 Preliminary risk-adjusted quality of care estimates In Finally, we project The draft recommendation for your review is, the 18 Congress should eliminate the update to the Medicare payment 19 rates for inpatient rehabilitation facilities in fiscal year 20 2013. 21 could absorb cost increases and continue to provide care 22 with no update to the payment in 2013. On the basis of our analysis, we believe that IRFs We estimate that 167 1 this recommendation would decrease Federal program spending 2 relative to current law by between $50 million and $250 3 million in 2013 and by less than $1 billion over five years. 4 We do not expect this recommendation to have adverse impacts 5 on Medicare beneficiaries. 6 the financial pressure on some providers, but overall, a 7 minimal effect on providers' willingness and ability to care 8 for Medicare beneficiaries is expected. 9 10 This recommendation may increase This concludes the presentation and Craig and I welcome your questions. 11 MR. HACKBARTH: Thanks, Christine. 12 George, why don't you lead off this time. 13 MR. GEORGE MILLER: [Off microphone.] 14 all, I support the recommendation. 15 MR. HACKBARTH: 16 MR. GEORGE MILLER: First of Microphone. I support the recommendation. 17 But in the chapter, could you spend a little bit of time, 18 and you did an excellent job in the chapter differentiating 19 between freestanding and hospital-based, but I was struck by 20 where there's a difference in pattern of different reasons 21 for patients being in the hospital. 22 conclusions, or could you expound on that analysis? Could you draw any One 168 1 was, for example, in the chapter, stroke patients were more 2 likely to be in a hospital basis much more than in a 3 freestanding, and the neurological patients were more apt to 4 be in a freestanding than in a hospital. 5 why that difference. 6 7 MS. AGUIAR: Right. I'm just curious So I think you're referring to the Table 7 -- 8 MR. GEORGE MILLER: 9 MS. AGUIAR: Yes. -- that has the distribution -- 10 MR. GEORGE MILLER: 11 MS. AGUIAR: Yes. You know, we have to still look more 12 into that. 13 the differential. 14 sources. 15 that we've been thinking to look more into, to try to 16 understand -- and again, that study did not include that 17 table on the slide, so for those who are not familiar with 18 it -- 19 20 21 22 Unfortunately, we don't know exactly why it is I mean, it could have to do with referral But, again, we're not sure, and that is something MR. GEORGE MILLER: Right. Yes. It's in the chapter. MS. AGUIAR: As that said, the sort of main take- away from that was that the top ten diagnoses are the same 169 1 for both hospital-based and freestanding IRFs. 2 MR. GEORGE MILLER: 3 MS. AGUIAR: 4 Right. Five of those ten are not conditions that count towards the threshold -- 5 MR. GEORGE MILLER: 6 MS. AGUIAR: Right. -- the compliance threshold. But 7 again, that there were some differences, and like you said, 8 stroke was one of them -- 9 MR. GEORGE MILLER: 10 MS. AGUIAR: Yes. -- with hospital-based having about 11 22 percent of their patients with stroke and freestanding 12 it's about 16 percent. 13 MR. GEORGE MILLER: 14 MS. AGUIAR: 15 Yes. So, I mean, in short, that is something that we need to find out more about. 16 MR. GEORGE MILLER: 17 DR. STUART: Thank you. Actually, I have a similar question 18 regarding the differences between freestanding and hospital- 19 based. 20 these facilities that had been freestanding. 21 now hospital-based or is it continuing to be freestanding? 22 In other words, does hospital-based mean a financial I'm familiar with one system that purchased one of Now, is that 170 1 relationship with a hospital or does it mean a physical 2 relationship with a hospital? 3 MR. LISK: I don't know what the specific facility 4 you're talking about would be, so it might depend upon how 5 the financial arrangement is made. 6 facilities are part of the hospital, but there could be a 7 financial relationship where it's not exactly attached to 8 the facility, too, where it could be considered hospital- 9 based in some circumstances. I mean, usually, these So it depends upon kind of 10 both things. 11 transaction there that you're talking about. 12 I'm not sure what actually happened in the DR. STUART: Yes. Well, I guess this actually 13 cuts across not just this particular segment, but other 14 segments, as well, and I guess I just leave that as a 15 question without asking for an answer now to the extent that 16 it might have an implication for other providers. 17 DR. CASTELLANOS: First of all, I support the 18 recommendations. Could you show Chart 8, please, or page 19 eight? 20 among the margins, from minus-two percent and minus-zero- 21 point-two percent to 21.4 percent. 22 mentioned something about size of hospital beds, they're a Again, you explained it, but that's such a big swing That's a -- you 171 1 little smaller than units, but that really seems high. 2 MS. AGUIAR: Sure. So I'll try to go into a 3 little bit more detail, and again, we did have more in the 4 December presentation. 5 that here. 6 I apologize for not having included The reason that we think may account for some of 7 the differences is really one of the issues is on economies 8 of scale. 9 them have 22 or more beds. Freestanding IRFs, I believe about 50 percent of As you can see when you look at 10 bed size, those 22-plus tend to have positive margins, 11 whereas hospital-based IRFs, I believe it's about -- I think 12 it's half of them -- and so, yes, more than half of 13 hospital-based IRFs have less than 21 beds, and as you can 14 see, they tend to have negative margins. 15 economies of scale at play, as well. 16 Slide 4, hospital-based IRFs also tend to have lower 17 occupancy rates, as well. 18 So there is an We think, also, as on So we subsequently did an analysis which is 19 included in the paper, but again, I apologize, not in the 20 presentation, trying to look at the differences in direct 21 cost per case and direct cost per case. 22 some numbers out for you, hospital-based IRFs have about 30 And just to throw 172 1 percent higher direct cost per case than freestanding IRFs, 2 almost 11 percent higher indirect costs per case than 3 freestanding IRFs. 4 costs per case are about 24 percent lower. 5 a mix of economies of scale due to low volume, higher costs, 6 direct and indirect costs and costs per case, standardized 7 costs per case, but then also lower occupancy rates. 8 that's sort of our hypothesis about why that is. 9 And freestanding, their standardized MR. LISK: So we think it's So There's also one other factor on the 10 freestanding, too, is there is a kind of, you might say, a 11 corporate efficiency, as half of the freestanding are in one 12 chain. 13 broad scale efficiency that's gathered there because that 14 group has a higher margin than the other freestanding -- all 15 of the freestandings have higher margins, but there's that 16 that may come into play, too. 17 18 19 20 So there may be a corporate efficiency in terms of DR. CASTELLANOS: It's just a big swing, that's all. MR. GRADISON: I'm in support of the recommendation. 21 DR. BORMAN: I support the recommendation. 22 MR. ARMSTRONG: Yes, same. My questions have been 173 1 answered. 2 DR. BAICKER: I support the recommendation. 3 MS. UCCELLO: Me, too. 4 MR. BUTLER: So one observation is we're midway 5 through our three-facility post-acute facility provider -- 6 that I'd support the recommendation, but it's a reminder of 7 in truly an ACO world, all this stuff would be rationalized. 8 These more expensive IRFs which we think are doing a pretty 9 good job and we've got this 60 percent threshold and all 10 these, it would take care of itself if we were managing in 11 an ACO world and putting them all in the right place at the 12 right time. 13 DR. CHERNEW: I also support the recommendation, 14 and in the spirit of what Peter said, thinking about these 15 in terms of cohorts of patients, because some of them are 16 getting cared for in different places is much more 17 productive than cutting them up the way we cut them up here, 18 but that said, since we have to do it, I support the 19 recommendation. 20 DR. DEAN: I support the recommendations, and I 21 would agree entirely with what Peter just said, that if we 22 had a more sensitive way to get people to the services they 174 1 really need, I suspect we would do a much better job. 2 DR. HALL: 3 DR. BERENSON: 4 MR. HACKBARTH: 5 information here. 6 I support the recommendation. And I support the recommendation. I'm waiting for a piece of we wait for that. 7 8 9 Actually, we can go ahead and vote while DR. CASTELLANOS: [Off microphone.] You've got these people down here -MR. HACKBARTH: Oh, right. 10 [Laughter.] 11 [Discussion off microphone.] 12 MR. HACKBARTH: 13 Yes, right. Right. If I had a gavel, I'd be giving it to Bob at this point. 14 [Laughter.] 15 MR. HACKBARTH: 16 MR. KUHN: 17 MR. HACKBARTH: 18 DR. NAYLOR: Herb. I support the recommendation. Mary. One question. I support the 19 recommendation. Have we as a Commission, because I know you 20 mentioned in the report the emphasis we want to place on 21 pay-for-performance versus pay-for-reporting, and have we 22 ever made a recommendation directly related to moving from 175 1 2 P4R to P4P? MS. AGUIAR: To my knowledge, no, and I'm just 3 looking because there were things that happened before I 4 started working here, so no -- 5 6 DR. MARK MILLER: Do you mean for this sector or more broadly? 7 DR. NAYLOR: 8 DR. MARK MILLER: 9 For this sector. For this sector, we have not, but we have pursued it in other sectors, and I think what's 10 going on here is we're trying to develop the quality 11 measures -- 12 MS. AGUIAR: Exactly. 13 DR. MARK MILLER: -- so that we can begin to move 14 in that direction, and I think that's some of the work that 15 you're doing with the RAND folks. 16 MS. AGUIAR: Yes, exactly. 17 DR. NAYLOR: Thank you. 18 MS. BEHROOZI: 19 I support it. I also support with concerns raised by my colleagues over there where I used to sit. 20 MR. HACKBARTH: 21 [Laughter.] 22 MR. HACKBARTH: Okay. So one of the issues that came up 176 1 at the December meeting, I think it was Bruce that raised 2 it, and that is wouldn't it be good to have an introduction 3 to the post-acute providers that talks about -- puts the 4 whole post-acute sector in context and the fact that often 5 they're treating the same patients in different facilities 6 at different rates and we need to move towards a more 7 integrated approach to post-acute care. 8 planning to include something along those lines in the 9 report. And so we are And we did that, I think it was -- was it two years 10 ago we had that kind of a prelude to the post-acute? 11 might have been three, but we'll do something similar again 12 this year. 13 Okay. It Now, I think we're ready to vote, right? 14 So on the recommendation that's on the screen, all in favor, 15 please raise your hand. 16 Opposed? 17 Abstentions? 18 19 20 Are you stretching, Bill, or are you -DR. HALL: [Off microphone.] No, no. Aggressively voting. 21 MR. HACKBARTH: 22 And abstentions? Aggressively voting yes? No hands? Okay. Okay. 177 1 2 Thank you very much. I appreciate it, Christine and Craig. 3 So next, Dana is going to present on long-term 4 care hospital services, and this will be the last of the 5 update recommendations. 6 MS. KELLEY: Last month, we discussed in detail 7 our update analysis and the Chairman's draft recommendation 8 for long-term care hospitals, and you have the chapter and 9 the recommendation in your mailing materials. So today, 10 I'll review our findings on payment adequacy for LTCHs and 11 then you'll vote on the draft recommendation. 12 You'll recall, of course, that LTCHs furnish care 13 to patients with clinically complex problems who need 14 hospital-level care for extended periods. 15 118,000 Medicare beneficiaries had almost 135,000 LTCH 16 stays. 17 LTCHs filed Medicare cost reports in 2010, and Medicare's 18 payments to LTCHs are made on a per discharge basis based on 19 the MS-LTC-DRGs, which are the same groups that are used in 20 the acute care hospital PPS but with weights that are 21 specific to LTCHs. 22 In 2010, about Medicare spent $5.2 billion on this care. About 412 Before I go on with the update summary, I wanted 178 1 to answer a few questions that were raised last month. 2 First, Mary, you asked about the use of hospice care 3 following discharge from an LTCH. 4 LTCH discharges were directly to hospice care, and about 60 5 percent of those discharges were to hospice facilities as 6 opposed to hospice within a home setting. 7 know is how many LTCH patients who are discharged to other 8 post-acute care facilities eventually are discharged to 9 hospice and that's something we might be able to answer in 10 the future with some linked -- analysis of claims that are 11 linked together to see an entire episode of care. 12 In 2010, 3.3 percent of What we don't Herb, you asked whether the LTCH quality measures 13 to be collected in the forthcoming pay-for-reporting program 14 will be available online on Hospital Compare. 15 the Secretary is required to establish procedures for making 16 the measures available to the public and CMS is currently 17 developing procedures that will allow LTCHs an opportunity 18 to review the information before it is posted online. Under PPACA, 19 Glenn, who's not here, asked about the effect, if 20 any, of LTCHs on SNF margins in an area, and in her work on 21 SNF margins, Carol found little relationship between the 22 presence of an LTCH in a county and a SNF's margins. 179 1 Bill Hall and Mary, again, you asked what we know 2 about LTCH outcomes and how they compare with those for 3 medically complex patients in other settings. 4 a text box in the chapter that summarizes recent studies on 5 the issue, and generally speaking, researchers have been 6 unable to clearly distinguish LTCH patients from the 7 medically complex patients that are served in other 8 settings, acute-care hospitals and some SNFs. 9 discussed last time, I think lack of assessment data does I've included As we 10 limit what can be said about outcomes, but in several 11 analyses performed for CMS, RTI has shown that outcomes that 12 can be measured for medically complex beneficiaries who 13 receive care in LTCHs are comparable to those for similar 14 patients who don't have an LTCH stay. 15 And finally, Tom, you asked about the distribution 16 of length of stay in LTCHs, and you can see here that there 17 is a long tail of the distribution showing that some cases 18 really are quite lengthy. 19 this is for all MS-LTC-DRGs combined, as you can see up top 20 there. 21 24. 22 The mean length of stay over -- The mean length of stay is 27 days and the median is So turning now to our update framework, our first 180 1 consideration, as you know, is access to care. 2 direct indicators of beneficiaries' access to LTCH services, 3 so we focus on changes in capacity and use. 4 important to keep in mind with this service, in particular, 5 that the product here is not very well defined. 6 no established criteria for an admission to an LTCH, so it's 7 not clear when patients treated there require that level of 8 care. 9 live in areas without LTCHs and so receive similar services 10 11 We have no But it's There are And remember, too, that many Medicare beneficiaries in other facilities. To gauge access to services, we first looked at 12 available capacity. 13 nationwide. 14 in facilities. There was a net increase of one LTCH between 15 2009 and 2010. The line representing the number of LTCH 16 beds looks very similar to this one, so we're seeing the 17 same pattern there. 18 beds remained steady, as well. 19 This slide shows the number of LTCHs A moratorium on new LTCHs has slowed the growth Between 2009 and 2010, the number of And looking at growth in the number of cases per 20 10,000 fee-for-service beneficiaries, we see an increase of 21 3.5 percent between 2009 and 2010. 22 trends suggest to us that access to care has been maintained So taken together, these 181 1 2 during the period. Turning now to quality, as you know, LTCHs don't 3 submit quality data to CMS, so we rely on claims data to 4 examine trends and in-facility mortality, mortality within 5 30 days of discharge, and readmission to acute care, and we 6 use these to assess more high-level changes in quality of 7 care in LTCHs. 8 declining for most of the top ten diagnoses, or most of the 9 top diagnoses, rather. 10 In 2010, these rates were stable or And, of course, we've long been concerned about 11 this lack of quality data in LTCHs, and you'll recall that 12 last year, we convened a panel, an expert panel, to elicit 13 information on how best to measure LTCH quality. 14 mentioned, CMS is implementing an LTCH pay-for-reporting 15 program beginning October 2013 with data collection 16 beginning October 2012. 17 As I We also look at facilities' access to capital. 18 The moratorium, of course, limited opportunities for 19 expansion, but in 2010, the two largest LTCH chains, which 20 together own slightly more than half of all LTCHs, acquired 21 other LTCHs and other post-acute care providers. 22 to the chains' filings with the SEC, they have access to According 182 1 revolving credit that they've tapped to finance these 2 acquisitions. 3 diversified both horizontally and vertically, which may 4 improve their ability to control costs and better position 5 the companies for payment policy changes. 6 chains and nonprofits probably do not have the same level of 7 access to capital. 8 9 These two LTCH companies are increasingly But smaller As you can see in the top row here, the aggregate Medicare margin for 2010 was 6.4 percent. There's wide 10 spread in the margin, similar to what you've seen in other 11 settings, with a quarter of LTCHs having 2010 margins of 12 minus-2.9 percent or less and another quarter having margins 13 that are 14.6 percent or more. 14 For purposes of projecting 2012 margins, we 15 modeled a number of policy changes. We included updates in 16 2011 and 2012 and also, of course, the PPACA mandated 17 reductions. 18 percent in 2011 and an increase of 1.8 percent in 2012. 19 Altogether, we estimate that the effects will This resulted in an update of minus-half-a- 20 result in somewhat slower growth in provider payments 21 relative to costs over the next year, so we've projected a 22 margin of 4.8 percent in 2012, assuming providers' costs go 183 1 2 up at projected market basket levels. Before I put up the draft recommendation, I'll 3 just remind you of some policy changes that are expected in 4 fiscal year 2013 that will have an impact on both LTCH 5 payments and provider behavior. 6 percent rule and the short-stay outlier policy and the 7 moratorium on new LTCHs will end. 8 full in your paper, but I can take any questions that you 9 might have about that. We'll see changes in the 25 These are discussed in 10 We make our recommendation to the Secretary 11 because there's no legislated update to the LTCH PPS. 12 draft recommendation reads, the Secretary should eliminate 13 the update to payment rates for long-term care hospitals for 14 fiscal year 2013. 15 basket as a starting point for establishing updates to LTCH 16 payments, so eliminating the update for 2013 will produce 17 savings relative to a market basket, even assuming the PPACA 18 mandated reductions. 19 on beneficiaries or on providers' willingness to care for 20 beneficiaries. The CMS historically has used the market We don't anticipate any adverse impact 21 So with that, I will turn it over to you. 22 MR. HACKBARTH: So when the Secretary has 184 1 discretion on the update, what is the baseline that we use 2 for determining savings? 3 DR. MARK MILLER: 4 MS. KELLEY: 5 MR. HACKBARTH: 6 Market basket. Market basket. Okay. Cori, you're starting this time, please. 7 MS. UCCELLO: I support the recommendation. 8 MR. HACKBARTH: 9 DR. BAICKER: Kate. As do I. 10 DR. BORMAN: I'm fine. 11 MR. GRADISON: 12 DR. CASTELLANOS: 13 [Members around table shaking heads.] 14 MR. HACKBARTH: 15 MR. KUHN: [Off microphone.] As I. Same here. Herb. I support the recommendation, but I'd 16 like to come back to Glenn's question because I had the same 17 thought in my mind as you were putting the information up 18 there. 19 update, so in the CBO OMB baseline, they use just regular 20 market basket to move each and every year accordingly. 21 Okay. 22 So the Secretary has complete discretion on the Thank you. DR. BERENSON: I support the recommendation. 185 1 DR. HALL: I support the recommendation. 2 DR. DEAN: I support the recommendation. I think 3 we're still struggling to figure out where does this model 4 fit within the overall scope of things, and I think -- I 5 would hope that gradually some of these changes, we will 6 begin to get the information that will help us to answer 7 that question. 8 MR. HACKBARTH: 9 MR. BUTLER: 10 Mike. Peter. Support, but this is our last update, right? 11 MR. HACKBARTH: 12 MR. BUTLER: Yes. So collectively, how much have we 13 contributed off the baseline here? 14 question to -- we never kind of revisit that -- 15 MR. HACKBARTH: 16 MR. BUTLER: 17 18 It's also an interesting Yes. -- in terms of what the law says versus what damage we've done. DR. MARK MILLER: And we could do -- I don't 19 happen to have that accumulated across them. And remember, 20 as an institution or a Commission, we don't say that. 21 this whole bucket approach because CBO in the end does kind 22 of the final statement to the Hill as to how much all this We do 186 1 stuff saves. 2 ranges because we aren't in the estimation business. 3 could put together a back-of-the-envelope and let you guys 4 know in general what happens across these updates, but I 5 don't happen to have it on me. 6 So we go through this process of estimating MR. HACKBARTH: Okay. We The draft recommendation is 7 on the screen,. 8 the recommendation, please raise your hand. 9 It's time for our vote. All in favor of Opposed to the recommendation? 10 Abstentions? 11 Okay. 12 MR. HACKBARTH: Thank you, Dana. So our next discussion is on 13 encouraging the use of lower-cost medications by 14 beneficiaries in the LIS program. 15 MS. SUZUKI: Good afternoon. Last year, we spent 16 several meetings discussing the issues related to the 17 different patterns of drug use we've observed for 18 beneficiaries receiving the low-income subsidy compared to 19 those who don't. 20 of the key points from the previous presentations and 21 present a draft recommendation to increase the use of 22 generics by LIS beneficiaries. In this session, I'll briefly review some Additional detail on this 187 1 2 issue was included with your mailing material. As you recall from the previous sessions, there 3 are some notable differences in drug use patterns for Part D 4 enrollees who receive the LIS compared to those who don't. 5 Specifically, we found that: 6 more prescriptions, and the cost of each prescription was 7 higher, on average, compared to non-LIS enrollees; and we 8 also found that the use of more brand name medications by 9 LIS enrollees was contributing to the higher per 10 11 LIS enrollees tend to fill prescription cost for this population. We also discussed how we might change the LIS 12 cost-sharing structure to encourage the use of lower-cost 13 medications, when clinically appropriate, through generic 14 substitution and therapeutic substitution, which involves 15 switching to a different drug in the same therapeutic class. 16 Generic drugs cost significantly less than their 17 brand counterparts, so a policy that encourages generic 18 substitution has the potential to reduce Part D's program 19 spending without limiting access to medications. 20 to lower-cost drugs can also reduce what beneficiaries pay 21 out-of-pocket. 22 A switch In the example we gave in December, which I'll 188 1 come back to in a minute, the generic co-pay was lowered to 2 widen the spread between brand and generic drugs. 3 that scenario, in addition to those who switch from brand to 4 generic drugs, beneficiaries who are already on generic 5 medications would also see a drop in their out-of-pocket 6 costs. 7 adherence to their medication therapy. Studies have shown that lower costs can improve 8 9 Under Here are the two key features of the policy we discussed in December: 10 First, the policy would modify co-pays for LIS 11 beneficiaries to encourage the use of lower-cost drugs when 12 available in a given therapeutic class. 13 already use tiered cost sharing for their non-LIS enrollees 14 to encourage the use of lower-cost drugs. 15 enrollees, plans have limited ability to do this because the 16 co-pays are set by law, and the amounts in the statute 17 provides weaker financial incentives to choose the lower- 18 cost drugs compared to the incentives faced by non-LIS 19 enrollees. 20 Many plan sponsors But for their LIS So the policy would change the co-pays for LIS 21 enrollees to increase the spread between generic and brand 22 name drugs so that generic drugs become relatively more 189 1 attractive. 2 increase generic substitution or, in some cases, increase 3 both generic and therapeutic substitution. 4 Depending on the drug class, this policy could The policy would not apply to classes with no 5 generic substitutes, meaning that the cost-sharing amounts 6 for brand name drugs in classes with no generic substitutes 7 would remain unchanged. 8 that beneficiaries continue to have access to medications 9 they need. This is because we wanted to ensure 10 The policy would only apply to those LIS 11 beneficiaries currently subject to co-pays, so it would not 12 apply to dual eligibles residing in institutions, where in 13 many cases beneficiaries are not the ones making the 14 decisions about the choice of brand versus generic drugs. 15 The second key feature of the policy is that the 16 Secretary would review the therapeutic classes to determine 17 the appropriate classifications for the policy. 18 Secretary could define a drug class broadly or narrowly 19 depending on the appropriateness of therapeutic 20 substitution, and the Secretary may exclude certain classes 21 for clinical reasons. 22 The There's been some concerns raised about this 190 1 policy, so I wanted to explicitly review protections that 2 would be in place under the policy. 3 First, the draft recommendation directs the 4 Congress to provide broad authority and flexibility to the 5 Secretary to determine appropriate therapeutic 6 classifications for the policy. 7 previous slide, this would allow the Secretary to: 8 class narrowly or broadly, depending on the clinical 9 appropriateness of therapeutic substitution, and exclude 10 11 As I mentioned on the define a certain classes of drugs from the policy. Our intent here is to encourage substitution only 12 when it is clinically appropriate. 13 classes, therapeutic or generic substitution may not be 14 clinically appropriate for all individuals. 15 However, for some We would maintain the current exceptions and 16 appeals process under the policy to ensure those individuals 17 have access to the brand name drugs when clinical reasons 18 prevent them from substituting to a lower-cost medication in 19 the same therapeutic class. 20 monitor this process to ensure that beneficiaries continue 21 to have access to medications they need. 22 In the future, we plan to Finally, the same out-of-pocket limit will 191 1 continue to apply. 2 limit the out-of-pocket spending for beneficiaries who incur 3 very high drug spending. 4 means that once they reach the catastrophic phase of the 5 benefit, they no longer pay cost sharing for their 6 medications. 7 As you recall, Part D is structured to For these LIS beneficiaries, that Here is an example that should look familiar to 8 all of you. This illustrates how one might change the co- 9 pays to make generic drugs relatively more attractive. The 10 example shows co-pays under current law and under the policy 11 for LIS enrollees with incomes at or below 100 percent of 12 poverty. 13 for generics and $3.30 for all brand name drugs. 14 Currently these LIS enrollees pay a little over $1 Instead of this $1/$3, we eliminate the cost 15 sharing for generic drugs and increase the cost-sharing 16 amounts to $6 for brand name drugs when there are generics. 17 This is shown in red. 18 For brand name drugs in classes with no generic 19 substitutes, cost-sharing amounts would not change so that 20 beneficiaries would have the same access to those drugs as 21 under current law. 22 There are many ways to do this. For example, many 192 1 plans have separate non-preferred tiers for generic and/or 2 brand name drugs. 3 to allow some flexibility in setting the co-pays for LIS 4 enrollees to encourage the use of drugs that are placed on 5 preferred tiers. For those plans, the Secretary may want 6 Here's the draft recommendation. It reads: The 7 Congress should modify the LIS Part D co-payments for 8 Medicare beneficiaries with incomes at or below 135 percent 9 of poverty to encourage the use of generic drugs when 10 available in selected therapeutic classes. 11 should direct the Secretary to develop a co-pay structure 12 giving special consideration to eliminating the cost sharing 13 for generic drugs. 14 Secretary to determine appropriate therapeutic 15 classifications for the purposes of implementing this policy 16 and review the therapeutic classes at least every three 17 years. 18 The Congress The Congress should also direct the We expect this policy to decrease Medicare 19 spending relative to current law. For beneficiaries, on the 20 one hand, a lower generic co-pay could lower beneficiaries' 21 out-of-pocket costs and increase access to medications. 22 the other hand, a higher brand co-pay could negatively On 193 1 affect access if beneficiaries aren't able to effectively 2 use the exceptions and appeals process to obtain the 3 medications they need. 4 Some plan sponsors may experience a decrease in 5 the costs of providing the benefit, which would tend to 6 lower the premiums for all beneficiaries and the subsidy 7 payments Medicare makes to Part D plans. 8 9 Finally, some pharmacies may experience an increase in profits from dispensing more generic drugs. 10 So, with that, I'll turn it over to you. 11 MR. HACKBARTH: 12 DR. STUART: Thank you, Shinobu. Thank you. I had some additional 13 suggested language on this draft recommendation, and part of 14 it comes about because the chapter actually does not include 15 the graphic that you had a couple pages earlier about the 16 changes that might include both not just a zero co-pay for 17 generics but also higher co-pays for preferred and non- 18 preferred drugs. 19 language is not reflected in the recommendation, and so I 20 thought just to be honest about this, we really should say 21 that we're giving the Secretary the authority to recommend 22 or to increase co-pays for these selected therapeutic And that's not reflected in the -- that 194 1 categories for brand drugs. 2 My own thinking on this is that if the only effect 3 of this was to reduce the generic co-pay to zero, that could 4 very well increase total cost rather than decrease total 5 cost. 6 cost for the generic side as well as the potential stick of 7 higher costs on the brand side, recognizing, as we all do, 8 that you have to be very careful in terms of targeting this. 9 That would be the other part, is to make sure that that's I really think that you need both the carrot of lower 10 really ironclad, not just in terms of the recommendation but 11 in the language in the chapter as well. 12 MR. HACKBARTH: So, Bruce, your comments suggest 13 two possible paths. One is to amend the language of the 14 recommendation. 15 incorporate the table, whatever slide that is. 16 could you put up the slide that has the table with the 17 example of the different -- yes. 18 into the text and explain, leave the recommendation language 19 the same. A second would be to change the text to 20 DR. STUART: 21 MR. HACKBARTH: 22 MR. GEORGE MILLER: Shinobu, So we could put this back That would be my recommendation. Okay. I support the recommendation. 195 1 MS. BEHROOZI: I share Bruce's concern, so I 2 support -- I actually support the recommendation as written, 3 but, Shinobu, would you mind turning to Slide 7 where you 4 discuss the implications? 5 up a little more boldly than the rest of the text. 6 like the second sentence in the first bullet, "A higher 7 brand co-pay could negatively" -- I feel like that's a 8 little too cautious because we're already giving -- we want 9 the Secretary to have the authority to decide in which Because I guess that maybe shows I feel 10 classes any of these rules should apply and then to say, "A 11 higher brand co-pay could negatively affect" if they can't 12 "effectively use the exceptions," well, what does that mean, 13 if they can't get an exception anytime they wanted? 14 know, I think by saying that the Secretary should decide the 15 appropriate drug classes first, that we're being quite 16 cautious. 17 process should stay in place. 18 this is one of those things we know a lot about already, as 19 we were talking about earlier, and all the evidence that you 20 cite of what private plans do, that there's still suspicion 21 and whatever distrust of generics over brands, but we really 22 do know a lot clinically -- not me, but, you know, You We're saying that the exceptions and appeals But I think there's really -- 196 1 clinicians know a lot about the fact that you don't really 2 need the brand when the generic is exactly the same. 3 feel like the way that second sentence reads, it kind of 4 undermines too much, you know, what Bruce said, that the 5 Secretary really ought to have the ability to set a co-pay 6 structure that maximally encourages generic utilization, and 7 we wanted to emphasize the zero, but I think we shouldn't 8 sort of de-emphasize or take away the ability to add 9 additional cost if she finds it appropriate. 10 11 DR. MARK MILLER: right? So I Well, you're looking at me, Yeah, I knew it. 12 I made this more cautious because, you know, there 13 were people who commented after the last session saying that 14 we weren't being enough -- and I see your point, and I 15 agree, and I think there are -- we've designed the policy to 16 have safeguards. 17 little bit, and so we'll go back to the other way. But I asked them to amp up the concern a 18 MR. HACKBARTH: 19 DR. MARK MILLER: 20 Yeah. So I just want you to know I was responsible for that notching up. 21 [Laughter.] 22 MR. HACKBARTH: Mitra, since this isn't in the 197 1 formal recommendation, if you want to suggest language, that 2 would be welcome. 3 DR. NAYLOR: I support the recommendation. 4 MR. KUHN: 5 DR. BERENSON: 6 DR. HALL: Support. 7 DR. DEAN: A question. I support the recommendation. I support the recommendation. The statement that there 8 should be no change in cost-sharing amounts for drugs in a 9 class with no generic substitutes, does that mean that for a 10 preferred brand drug it would stay at $3.30? 11 it would be? 12 MS. SUZUKI: 13 DR. DEAN: Is that what Yes. Okay. And a broader comment that 14 really isn't addressed here, I think this is partly a 15 beneficiary issue, but it's also a prescriber issue. 16 think some of this, the use brand name drugs really relates 17 to what the prescribers choose and what they tell their 18 patients. 19 terms of the amount of leverage we have in terms of how much 20 we can adjust these co-pays given the population that we're 21 dealing with. 22 and I don't exactly know how to merge it in with this, but And I And it would seem that we're very limited in And I guess the thing that occurs to me -- 198 1 some kind of requirement for, say, a step care approach or, 2 for instance, in the statin family, you know, there's a 3 couple of very widely known drugs that everybody knows about 4 and there are some lesser-known ones that are every bit as 5 effective and have just as much science behind them and sell 6 for, you know, a quarter or a fifth as much. 7 those selected situations to somehow develop policies that 8 would encourage that use. 9 issue, not a beneficiary issue. 10 MS. SUZUKI: We need in And that probably is a prescriber So I guess we're trying to balance a 11 plan's -- plans have the ability to use step therapy and 12 other prior auth. or other tools to manage the use. 13 apply that to both LIS and non-LIS equally. 14 we saw the difference was in maybe the financial incentives 15 that they face, and part of it is that plans create their 16 formulary structure individually. 17 Secretary to determine versus plans making decisions about 18 how to structure and how to encourage use of lower-cost 19 drugs -- we were trying to add a little bit more leverage by 20 increasing the spread between generics and drugs, and maybe 21 that would prompt beneficiaries themselves to ask about 22 maybe cheaper alternatives, which we've seen happen for the They One place where So how much you want the 199 1 non-LIS population, especially when they hit the doughnut 2 hole. 3 DR. DEAN: I support the recommendation and the 4 direction we've gone. I'm just saying that -- and maybe 5 what you've just said answers my concern, that maybe the 6 incentive is there for the plans to put these things in 7 place rather than Medicare. 8 we don't need to worry about it. 9 DR. CHERNEW: If that's the case, then maybe Yeah, I'm sorry for being a little 10 confused in that the recommendation is relatively vague with 11 broad authority the way the recommendation is written, but 12 some of the discussion and some of the implications, we're 13 assuming something that seemed a little bit different. 14 the recommendation really was just the Secretary having the 15 authority to develop a co-pay structure with consideration 16 for eliminating the cost sharing for generic drugs, but it 17 doesn't say a lot about the increasing part -- although they 18 would have the authority there, but the implications seem to 19 flow a little bit from that. 20 the comment that -- 21 MR. HACKBARTH: 22 DR. CHERNEW: Like And so it's hard -- which is [off microphone] That was Bruce's. Right. And I think that it's 200 1 important to think through what that is. My general sense 2 is, although I'm supportive of the recommendation as 3 written, one of the things that's not in this whole process 4 is something that was analogous to the discussion we were 5 having when we were talking about the outpatient department 6 E&M codes, which is some process to measure the impact. 7 depending on how the co-pay structure -- if you're just 8 lowering generics, that's -- you know, I don't have a 9 problem with that one. So When you start raising branded, 10 which I think you can make a rationale to do, but a lot of 11 complicated things happen when you do that because a lot of 12 times people don't substitute the way they want. 13 we're talking about people switching drugs most of the time 14 in a therapeutic class, but still switching drugs. 15 people may drop. 16 for example, using the appeals process for a whole bunch of 17 reasons. 18 to adherence in some of these areas, particularly the 19 chronic care areas and other areas, I think would be very 20 important if you were going to start having a more 21 aggressive interpretation of our recommendation than the 22 text of the recommendation actually implies. Remember, A lot of I'm a little skeptical of this population, So I think some monitoring about what's happening 201 1 DR. MARK MILLER: So I think, you know, just to 2 draw a couple of points together, some of the comments and 3 concerns that you see that we built in there derive from 4 this kind of perspective. 5 carefully on the fact that the implications that we've put 6 in here sort of presume the policy more in the chart that we 7 put up than the language. 8 DR. CHERNEW: 9 DR. MARK MILLER: You obviously have tracked very [off microphone] Right. But I just want you to 10 understand why the language is a little bit more open-ended, 11 is because we don't know whether it's, you know, $0 and $6 12 or $0 and $5.50. 13 numbers in. 14 around, and we've talked to plans and gotten some sense. 15 You know, we didn't want to start writing And in this area, there is still some feeling So my question to you, although this is really a 16 Glenn question, would be: 17 the statement that you made in the text? 18 for something to change in the recommendation? 19 tracked through the thought process correctly. 20 DR. CHERNEW: Would you be satisfied to make Or are you calling But you have So the recommendation right now 21 relatively vaguely says the Secretary shall have the 22 authority to think about a better co-pay structure for drugs 202 1 for low-income subsidies, right? 2 3 DR. MARK MILLER: [off microphone] And then the table, not -- 4 DR. CHERNEW: Right, but the table is -- 5 DR. MARK MILLER: 6 DR. CHERNEW: [off microphone] Illustrative. So in terms of the recommendation, 7 I'm a supporter of that. I think there's a lot of 8 improvements one could do to the co-pay structures just in 9 general. That said, as you just pointed out, you don't know 10 what the actual changes are going to be, so it's hard to 11 figure out -- it really has a lot more to do with your trust 12 in the Secretary than it has to do with particular changes 13 that are going on. 14 I'm fine with the discussion. 15 are things the Secretary could conceivably do that I would 16 have concerns over. 17 The recommendation is vague than that. 18 So I'm fine with the recommendation, and I think some concern -- there We aren't recommending she do them. So I think the easy answer to your question is I 19 am fine with the recommendation. I think the text is 20 reasonable. 21 of the stuff was inferring something a little bit more than 22 just what's in the recommendation, and I would be a little I did have a concern about the implication part 203 1 happier one way or another if we had some lip service to the 2 text is fine to having to monitor whatever the changes the 3 Secretary does do, which we don't know what they will be. 4 We should monitor that for utilization changes. 5 at our -- we should caution about assuming that all the 6 substitution patterns work the way that you would think in a 7 perfectly rationale world -- I think it's pretty well known 8 -- Bruce can comment -- that if you take people on a branded 9 medication and raise their co-pays, you might think they And I think 10 would just substitute to the generic equivalent. 11 them just stop taking the medications. 12 depending on the classes that were done, adherence to the 13 medications turns out to be very important. 14 Many of And I think I would just add, because I like to say these 15 things, the merit of whatever the Secretary does shouldn't 16 be judged based on the cost. 17 out the cost with the quality of care. 18 that there was some change like the reduction in generics 19 increased to zero, if that's all they did, and that 20 increased spending, we should not view that as a bad thing. 21 That could be a good thing. 22 medications in many of these classes is really our goal if The whole point is to balance So if it turned out Getting people to keep their 204 1 we can do it in a cost-effective way, and in many cases I 2 think these drugs -- the ones at least I have in the back of 3 my mind -- are very cost-effective ways of treating illness, 4 not necessarily cost savings but at least very cost- 5 effective ways of treating the illnesses, and people should 6 be encouraged to take medications for chronic diseases. 7 MR. HACKBARTH: I certainly don't disagree with 8 anything that you've said, and I think based on previous 9 discussions, I think your statement reflects a consensus 10 among the Commissioners. 11 include language along those lines in the text discussing 12 the recommendation as opposed to amending the recommendation 13 itself. 14 think it's important for the Secretary to monitor the impact 15 of these changes. 16 What I would suggest is that we And that language in the text can also say that we I don't want to set the precedent that every time 17 we recommend a change there's a formal recommendation to 18 monitor the effects. 19 little old. 20 they ought to be included in the text. 21 22 I think after awhile that gets a But your points are well taken, and I think MR. BUTLER: A question. We dumb this down in our example so that we have this $6 co-pay example, and in 205 1 reading the text, it says the high-cost beneficiaries have 2 an average of 111 prescriptions per year and that 83 percent 3 of the LIS people fit into that category, which means to me 4 there may be nine or ten prescriptions -- these are 30-day 5 prescriptions, so they're holding basically nine or ten 6 prescriptions, most of these. 7 month -- if they went all to preferred, in the example it's 8 like -- I'm trying to get a sense of what the real financial 9 burden is here for the co-pays, not just on an individual So it's really like a $60-a- 10 drug but on, you know, the total out-of-pocket. 11 follow me? 12 13 14 DR. MARK MILLER: I think so. Do you Shinobu, if you [off microphone]. MS. SUZUKI: So one thing that we were trying to 15 make clear is that Part D limits the out-of-pocket spending 16 by using true out-of-pocket calculation. 17 much you've spent in gross spending, not just your out-of- 18 pocket spending but gross spending, and your actual out-of- 19 pocket spending and see how much it is, and once you hit 20 that limit, then you're in this catastrophic phase where, 21 for these LIS beneficiaries, their cost sharing would be 22 waived once they hit that limit. So plans track how 206 1 For LIS beneficiaries, their cost sharing is right 2 now $103 if you're under 100 percent of poverty, but the LIS 3 subsidy also counts towards these troops. 4 for example, taking the $6 drug but that $6 drug turns out 5 that the cost sharing on plan formulary is $80 because it's 6 a non-preferred brand, then the whole $80 plus the $6 counts 7 toward the troop limit. 8 quicker in some ways, and so relative to someone who takes a 9 lower-cost medication, it's possible that they get to that 10 So if they are, So it gets you to that limit much limit much quicker. 11 And so I guess you were trying to compare current 12 law versus under the policy if their out-of-pocket amount 13 would be much greater, and our sense is that it's still 14 going to be pretty limited. 15 version, that just gets you much quicker to the catastrophic 16 phase. 17 Even if they take the brand Does that make sense to you? MR. BUTLER: I understand how you get credit for 18 the $80, and that's a lot towards getting towards your 19 limit, even if you're only paying 6 bucks. 20 math to figure out, you know, what is the true -- what are 21 they really going to pay out-of-pocket. 22 MR. HACKBARTH: I can't do the That's what I -- So Peter gave an example that he 207 1 derived from the averages and how many prescriptions these 2 people are taking, or at least a sub-group of the population 3 are taking. 4 example? 5 drugs, how quickly are they going to hit the out-of-pocket 6 limit, as you describe it? 7 8 9 10 Can we answer his question in terms of that That person who's taking nine or ten brand name MS. SUZUKI: I guess that would depend on the mix of drugs. DR. MARK MILLER: MS. SUZUKI: Right. It's a case-by-case sort of 11 calculation, so it's hard to say whether someone who's 12 taking 111 drugs under current law with a different mix of 13 brand and generics gets to the catastrophic limit under the 14 policy more quickly or not, it just depends on how they 15 change their behavior. 16 example where it's less under the policy, maybe it's more 17 under the policy in some cases where they change behavior, 18 if that's helpful. I mean, we can try to work up an But I think it just depends. 19 MR. HACKBARTH: 20 DR. STUART: Bruce, do you have -- I guess I'm a little confused. The 21 LIS are not subject to the benefit thresholds, so the out- 22 of-pocket limit wouldn't come into play, would it? 208 1 MS. SUZUKI: It applies because after their gross 2 spending gets to a certain limit -- so assuming they have no 3 supplemental coverage, they count out-of-pocket, they count 4 the low-income subsidy, and they tally that up to figure out 5 whether they reach the true out-of-pocket limit that's 6 applied to non-LIS people. 7 limit. 8 for LIS it's the out-of-pocket co-pays plus the subsidy, and 9 you get to that limit. 10 So say $3,600 per year is the From non-LIS that would be a true out-of-pocket, but Above that limit you don't pay any cost sharing. 11 MS. UCCELLO: I'm sorry, but I think the issue 12 here is that after you reach that cap, Medicare is paying 13 for the reinsurance portion of things. 14 MS. SUZUKI: 15 MS. UCCELLO: 16 case. Mm-hmm. And that's why it matters in this It seems like -- you know -- 17 MS. SUZUKI: 18 MS. UCCELLO: That and -It's not necessarily out-of-pocket 19 for the LIS enrollee, but it has financial implications for 20 the program. 21 22 MS. SUZUKI: implications. It definitely does have financial It also changes what -- so the law defines 209 1 those co-pays, but those co-pays only apply until they hit 2 that catastrophic limit, and that's what I was trying to 3 bring up. 4 MR. HACKBARTH: I agree that it has implications. 5 The way I understood Peter's question, though, he was 6 focused on the financial impact for these low-income people, 7 and if the brand name co-pay goes up to, say $6 and you're 8 taking nine or ten prescriptions, he's trying to assess the 9 implications of that for their out-of-pocket costs. 10 DR. MARK MILLER: 11 MR. BUTLER: And I think -- Exactly. And when you just look at, 12 oh, one drug, 6 bucks, what's the big deal, anybody can come 13 up with 72 bucks in a year. 14 these, is it -- and that's just an example, anyway. 15 not even the recommendation. It says the discretion, the 16 Secretary is going to do it. But it's a little bit like 17 this morning. 18 multiply them together, and it might be -- I just want to 19 make sure it's not as big a financial burden as we're -- our 20 portrayal of it is an understatement of what the real 21 financial burden might be. 22 But if you're taking ten of It's You get in these numbers, and suddenly you MS. BEHROOZI: But, you know, it's about driving 210 1 behavior. 2 effectiveness drug -- and it's only for those cases where 3 it's the same effectiveness -- for free. 4 be paying $33 a month for the ten brand name drugs, but then 5 they could be going to zero, getting the same effectiveness, 6 because they would be helped to make that choice. 7 clinical effectiveness for that patient, I want to [off 8 microphone]. 9 The idea is to get people to take the same DR. CHERNEW: So now they could Only same But I think the thing is it's 10 actually not written for only the same clinical 11 effectiveness, because it's all done within a therapeutic 12 class. 13 some of these, like some of the mental health drugs, the 14 effectiveness of drugs in a class could be very different. 15 So you could see an implementation of this that it isn't 16 substituting branded Drug X for the generic version of X. 17 It's substituting branded Drug Y for a generic version of 18 Drug X in that class, and that in many cases could be a 19 bigger deal, and that's the way the recommendation -- 20 And within a therapeutic class, particularly for MS. BEHROOZI: But that's what the appeals and 21 exceptions would be for, and that's the cases they would be 22 applied to. That's the way I understand -- 211 1 DR. MARK MILLER: 2 DR. CHERNEW: 3 MS. BEHROOZI: [off microphone]. [off microphone]. As well as how the Secretary -- 4 exactly. So there are all those protections to prevent 5 that. 6 situation where people would have to pay more for the drug 7 they need, that they clinically need based on them as an 8 individual and the drug -- the spread of the effectiveness 9 of the drug within the class. As I understand this and what I'm supporting is not a That's not the intention. 10 It's only where it for them is the clinically effective 11 equivalent. 12 DR. CHERNEW: And I think the issue is how much 13 you think this appeals process and all those safeguards that 14 were just discussed, how effective would they be? 15 think there's reasons in certain areas to worry about that, 16 but as a general principle, given the Secretary the 17 authority to do a better job on co-pays is probably a good 18 thing, but they have to be careful of [off microphone]. 19 DR. STUART: And I [off microphone] -- it's not the 20 appeals process here that we're going to. It's the 21 selection of the appropriate therapeutic classes so that you 22 don't say that you're going to have high co-pays and expect 212 1 to have therapeutic substitution in classes where there is 2 evidence that, in fact, they're not substitutable. 3 MR. HACKBARTH: 4 DR. MARK MILLER: 5 So -[off microphone] didn't answer your -- 6 MR. HACKBARTH: Yeah. Before we leave Peter's 7 question -- and I don't know how feasible this is to do, 8 Shinobu, but if we -- I don't recall this being addressed in 9 the text. If we could add something to the text that 10 addresses Peter's core question of how much of a financial 11 burden would this represent for LIS beneficiaries, I think 12 that would be helpful. 13 DR. MARK MILLER: I know you want to move on. The 14 only thing I want to say is we may have to end up doing a 15 few illustrative examples, because the clutch of drugs that 16 the person has could say this one's going to stay under 17 current law, this one's going to move to zero, this one's 18 going to move to 6, assuming 6 is the number. 19 that we have to show you hypothetically how ten 20 prescriptions in a month could be affected by this up or 21 down, is my sense, unless you're sitting on some data and 22 can move through it faster than I'm thinking. So it may be 213 1 MS. UCCELLO: I support the recommendation, and I 2 especially support the way it is worded now. Last month, it 3 was more vague in terms of giving the Secretary authority to 4 change the co-pays to encourage more generics, and Mitra and 5 I and maybe some others with some confirmation from staff 6 thought that, you know, one of the -- the bigger driver was 7 moving to zero versus raising the branded co-pay. 8 like the way that that has been incorporated into this 9 recommendation, and the text itself, especially if that And so I 10 table is included, with examples or not, can show the 11 potential to also increase the brand co-pay without 12 necessarily having to be explicit in this. 13 strike the right balance here. 14 We're trying to In a total topic change, in the text it kind of 15 highlights the need for to have this be a competitive market 16 where people are choosing plans in part based on premiums, 17 they have to be willing to switch using the price signals, 18 but there seems to be a low frequency of switching in 19 general, and in particular for those low-income subsidy 20 folks who may now have to actually pay a premium if they 21 don't switch, but they still choose to stay. 22 So I'm just wondering what kind of information is 214 1 provided to the LIS beneficiary kind of every year notifying 2 them of choices, and even more generally to all 3 beneficiaries, you know, making it clear that they can 4 change plans if they want. 5 MS. SUZUKI: I may have to get back to you on the 6 details, but for LIS beneficiaries, they do provide a lot of 7 information about, well, your plan is no longer qualifying 8 for the zero dollar premium. 9 conversation about maybe providing information about, you I vaguely remember some 10 know, switching to other plans in your area, but I can 11 definitely look into that and get back to you. 12 DR. BAICKER: I think this is moving in a very 13 good direction of promoting higher-value use as we do across 14 all things, and we have a reasonably well-ground idea that 15 people are underusing generic drugs and underadhering in 16 general. 17 things that people are worried about, the methods that we're 18 using to address seem good but are never going to be 19 perfect, and I think we just have to acknowledge that, that 20 there isn't a world in which for some patients there are 21 perfect substitutes and for others there aren't. 22 continuum where, you know, it's going to work almost as And so this promotes that kind of use. All of the It's a 215 1 well, for some people it will work exactly as well, but for 2 a lot of people there are going to be these small 3 differences, and we have to acknowledge that by introducing 4 a price wedge, some people are going to have to pay a little 5 bit more for some drugs that work a little bit better, and 6 we're not going to be able to perfectly separate and no 7 appeals process is going to be able to perfectly separate 8 that because it's a continuum. 9 By the same token, I think we're never going to be 10 able to say that there are certain -- I'll stop there, with 11 the overall idea that it's worthwhile to move people towards 12 more use of generics and that some people will inevitably 13 have to pay more than they would have otherwise to achieve 14 the same therapeutic outcome. 15 MR. ARMSTRONG: I support the recommendation. 16 Actually, I think building in a way on Kate's points, first, 17 I think it's so admirable the accountability that we feel as 18 Commissioners to the beneficiaries of our program and the 19 kinds of concerns we've been expressing. 20 part of what makes me proud to be a Commissioner is the 21 degree to which we're looking for ways of safeguarding any 22 potential negative implications of some of these policy I also am just -- 216 1 changes. 2 But I just feel like this is an area that is so 3 five years ago, I mean, these kinds of -- there are experts 4 and this kind of benefit design has been going on now for a 5 long time. 6 the recommendation. So, anyway, I think we should move forward with 7 DR. BORMAN: 8 MR. GRADISON: 9 question. I support the recommendation. I do, too. I have a quick My recollection is that in the new health act, 10 133 percent is the cut-off point for mandatory Medicaid 11 coverage. This is 135. Is that statutory? 12 MS. SUZUKI: Yes 13 MR. GRADISON: So there's a 2 percent difference. 14 In other words a dual eligible is up to 133. 15 correct? 16 want to make sure I understand that because -- it's just a 17 factual question, and I'm confused. 18 But LIS is at 135? DR. CASTELLANOS: Is that I'm just -- later. I support this. I just Could you go to 19 Slide 4 for a second? I just want to bring out a real-world 20 problem. 21 process that should remain in effect. 22 excellent presentation and in the chapter that you give us, You talk about the current exception and appeals Then during your 217 1 you said the Commission had planned to monitor the exception 2 and appeals process for its effectiveness. 3 - and I'm sure Tom and Bill and Karen will say this, too -- 4 if you go through this process, it is a difficult process to 5 go through, not just for the Medicare beneficiary but for 6 the physician. 7 quite sure how you're going to monitor that, and I'm really 8 not quite sure what you plan to do with that. 9 would help us significantly trying to ease the burden of As a physician - And it's such a mishmash of things. I'm not But that 10 especially the primary care doctor who's doing this work and 11 it's uncompensated. 12 So I really would like you to -- I know I'm 13 talking about something that's probably not the biggest 14 thing on the agenda, but it's something that in the real 15 world it would help. 16 MR. HACKBARTH: Okay. Would you put up the draft 17 recommendation, Shinobu? 18 recommendation, please raise your hand. All in favor of the draft 19 Opposed? 20 Abstentions? 21 Okay. 22 And our final session is on Medicare Advantage. Thank you. 218 1 Each year in the March report, we provide a status report on 2 Medicare Advantage even though there is not an update, 3 payment update involved. 4 DR. HARRISON: Scott. Good afternoon. We are here to 5 report the current status of the Medicare Advantage, or MA, 6 program in terms of enrollment in the program, the 7 availability of plans for 2012, projected Medicare payments, 8 and quality indicators for those plans. 9 Our March chapter is a view of the landscape of 10 the program and contains no formal recommendations. 11 is more detail in the chapter than we are presenting here 12 today just to keep things short, but we invite your 13 questions and comments on all the material in the draft 14 chapter. 15 There Let me begin by describing the MA program and 16 payment system. The Medicare Advantage program allows 17 Medicare beneficiaries to receive their Medicare Parts A and 18 B benefits through a private plan rather than through the 19 traditional fee-for-service Medicare program. 20 who enrolls in a plan continues to pay the Part B premium 21 and any additional premium that the MA plan charges. 22 Medicare program plays the MA plan a monthly capitated A beneficiary The 219 1 amount that is adjusted for the health risk of the 2 individual beneficiary. 3 the Medicare A and B benefits and usually provides coverage 4 for additional benefits. 5 Medicare beneficiaries were enrolled in MA plans. 6 The plan then provides coverage for In 2011, about 25 percent of In some of the analyses, I will differentiate by 7 plan types and other plan characteristics and I just want to 8 define some of them for you here. Coordinated Care Plans, 9 or CCPs, are either HMOs or PPOs. CCPs have provider 10 networks and various tools to coordinate or manage care. 11 Under the MA program, there are local PPOs and regional 12 PPOs. 13 individual counties while regional PPOs are required to 14 serve entire regions, which are made up of one or more 15 complete States. 16 The difference is that the local PPOs can serve The MA program also includes private fee-for- 17 service plans which historically had no provider networks 18 and paid providers' Medicare fee-for-service rates. 19 legislation changed the plan requirements, and as of last 20 year, these plans must either have networks or can only be 21 offered in areas where there are fewer than two network 22 plans. Recent 220 1 We sometimes make other distinctions. Special 2 Needs Plans, or SNPs, limit their enrollment to either 3 Medicare-Medicaid dual eligibles or to those beneficiaries 4 who have either certain chronic or disabling conditions or 5 who require institutionalization. 6 available to individual beneficiaries but only to employer 7 or union groups. 8 do not include these so-called employer group plans or SNPs 9 because these plans are not available to all beneficiaries. And there are plans not Our availability numbers that you will see 10 But our enrollment and payment numbers generally include 11 them. 12 Plans submit bids each year for the amount they 13 think it will cost them to provide Parts A and B benefits. 14 There is a separate bid for Part D drugs, but the MA plans 15 just get paid for Part D as if they were stand-alone Part D 16 plans. 17 CMS actuaries review the bids to make sure they 18 are reasonable. Each plan's bid is compared to a benchmark, 19 which is a dollar amount set for each county. 20 are administratively set based on historical payment rates 21 for each county. 22 benchmarks of the counties it serves, and beginning in 2012, Benchmarks A plan's benchmark is based on the 221 1 on the plan's quality rating. 2 Carlos will discuss the plan quality ratings shortly. 3 If a plan bids above the benchmark, Medicare pays 4 the benchmark and beneficiaries make up the difference with 5 a premium. 6 pays the bid plus a rebate calculated as a percentage of the 7 difference between the bid and the benchmark. 8 must be used by the plan to provide extra benefits to the 9 beneficiaries. If a plan bids below the benchmark, Medicare The rebate These extra benefits can take the form of 10 reduced cost sharing for A-B services, additional non- 11 Medicare benefits, such as dental, vision, hearing, or gym 12 memberships. 13 D benefits, including lower Part D premiums. 14 They could also take the form of improved Part There has been growth in Medicare Advantage 15 enrollment each year since 2003. From November 2010 to 16 November 2011, enrollment grew by about six percent, or 17 700,000 enrollees, up to 12.1 million beneficiaries. 18 Among plan types, HMOs, at eight million, 19 continued to enroll the most beneficiaries, and as a result, 20 16 percent of all Medicare beneficiaries were in HMOs in 21 2011. 22 1.7 million to about 600,000, with enrollment shifting to Private fee-for-service enrollment shrank from about 222 1 network plans. 2 local PPO increasing about 65 percent and regional PPO 3 enrollment increasing about 34 percent. 4 PPOs exhibited rapid enrollment growth, with Going forward, plan bids project overall 5 enrollment growth in the seven to eight percent range for 6 2012, but almost all of the projected growth is in HMOs. 7 Enrollment patterns differ in urban and rural 8 areas. About 26 percent of urban Medicare beneficiaries are 9 enrolled in Medicare Advantage, compared with about 14 10 percent of beneficiaries residing in rural counties. What's 11 not shown here is that urban MA enrollees are much more 12 likely to be in HMOs and rural enrollees are much more 13 likely to be in non-HMO plans. 14 So let's look at plan availability. 15 beneficiaries have a large number of plans from which to 16 choose. 17 Zero-point-three percent of beneficiaries do not have a plan 18 available. 19 Medicare MA plans are available to almost all beneficiaries. Looking at the top line here, in 2012, 93 percent 20 of Medicare beneficiaries have an HMO or local PPO plan 21 operating in their county, up from 92 percent in 2011 and 67 22 percent back in 2005. And if you combine the local CCPs 223 1 with the regional PPOs, you would find that 99 percent of 2 beneficiaries have a Coordinated Care Plan available in 3 2012, but don't look too hard at the chart because that 4 number is not on there. 5 percent of beneficiaries in 2012, down from 86 percent in 6 2011 due to the withdrawal of the only California regional 7 PPO for 2012. 8 decreased from 63 percent to 60 percent of beneficiaries. 9 Regional PPOs are available to 76 Access to private fee-for-service plans In most counties, a large number of MA plans are 10 available to beneficiaries, although the number varies by 11 county. 12 York City can choose from more than 50 plans in 2012. 13 average, 12 plans, including eight Coordinated Care Plans, 14 are offered in each county in 2012, and that's the same as 15 was in 2011. 16 At the high end, beneficiaries in Miami and New On And in 2012, 80 percent of beneficiaries have 17 access to at least one MA plan that includes Part D drug 18 coverage and charges no premium beyond the Medicare Part B 19 premium, and that's compared with 90 percent in 2011. 20 So we use the plan bid projections to compare 21 projected MA spending with projected fee-for-service 22 spending on a like set of fee-for-service beneficiaries. 224 1 Because we are comparing fee-for-service expenditures with 2 plan bids and the resulting MA payments, we are using a 3 growth factor for 2012 similar to what the plans use to 4 develop their bids. 5 modestly and discount the likelihood that physician services 6 would be cut by the SGR. 7 we have done for the previous two years. 8 assume no SGR cut during 2012. 9 Plans generally assume costs will grow We also used this assumption, as The numbers here So looking at the top row, we estimate that, on 10 average, 2012 MA benchmarks, bids, and payments will be 112 11 percent, 98 percent, and 107 percent of fee-for-service 12 spending, respectively. 13 average approximately three percent less than the benchmarks 14 for 2011 before you took into account any of the quality 15 bonuses that are starting this year. 16 2012 plan enrollment is projected to be in plans that will 17 receive add-ons to their benchmarks for quality. 18 quality demonstration in effect offset most of the PPACA 19 benchmark reductions that were scheduled for 2012. 20 For 2012, the county benchmarks However, almost all The The pre-quality benchmark reductions, however, may 21 have encouraged plans to tighten costs and lower their bids 22 for 2012. The average bid is now 98 percent of the 225 1 projected fee-for-service spending for similar 2 beneficiaries, and the HMO bids average 95 percent of fee- 3 for-service. 4 generally above fee-for-service, but they are closer to fee- 5 for-service than they were last year. 6 Now, the bids of the other plan types are As a result, we project that plan payments in 2012 7 will move closer to fee-for-service spending. 8 exception of employer group plans, the payments for all plan 9 types are projected to be closer to fee-for-service levels 10 And with the in 2012 than they were in 2011. 11 Before we go on to quality, let me make a few 12 summary observations about the direction of some areas of 13 MA. 14 managed care plans, namely HMOs, which we think have the 15 potential to increase the quality and efficiency of Medicare 16 services. 17 has grown by seven percent and plans project similar growth 18 for 2012. Enrollment continues to grow in the more tightly Over each of the past four years, HMO enrollment 19 Led by the HMOs, the plans have responded to the 20 tighter benchmarks by lowering their bids relative to fee- 21 for-service spending. 22 fee-for-service spending. The average bid is now lower than 226 1 Payments, even after including the quality 2 bonuses, have moved closer to fee-for-service spending 3 levels, and at the same time, the benefit packages offered 4 by MA plans have not eroded in terms of the dollar values of 5 the extra benefits offered. 6 7 Now, as promised, Carlos will present findings on quality and the quality bonuses. 8 9 MR. ZARABOZO: I'll give you a quick overview of the most recent quality indicators in the Medicare Advantage 10 program, which are discussed in detail -- in fact, in 11 excruciating detail -- in your mailing material. 12 Looking at the clinical process and intermediate 13 outcome measures that plans report, we find that between 14 last year and this year, there's been improvement in those 15 measures. 16 that we track showed improvement between 2010 and 2011 among 17 plans reporting in both years. 18 measures improved. 19 private fee-for-service and regional PPOs, their results are 20 generally not as good for HMOs and local PPOs. 21 see major improvement between this year and last to the 22 extent that we can evaluate results for the small number of Among HMOs, for example, 14 of the 45 measures For local PPOs, nine For the other two types of plans, We did not 227 1 2 these plans that report quality measures. For some measures, we're able to compare the 3 performance of Medicare Advantage plans with the performance 4 of fee-for-service Medicare. 5 survey administered in both sectors tracks vaccination rates 6 for influenza and pneumonia. 7 period, the surveys show that vaccination rates are about 8 the same in each sector and have improved in each sector 9 compared to last year's rates. 10 For example, a beneficiary For the most recent time As we found in the past, newer plans tend to have 11 lower levels of performance on quality measures than more 12 established plans. 13 Special Needs Plans in particular, sometimes using a proxy 14 method to identify those plans in the broader data. 15 find, in general, that the Special Needs Plans are lower 16 performers than other types of plans, albeit with major 17 exceptions. 18 And this year, we attempted to look at We Beginning this current year, 2012, Medicare 19 Advantage plans will receive bonus payments based on their 20 performance on quality indicators. 21 rating, between one and five stars, based on their 22 performance on clinical measures, patient experience Plans are awarded a star 228 1 measures, and contract performance measures. 2 bonuses are based on the year 2011 star ratings that were 3 announced at the end of 2010. 4 The 2012 Although the statute established the manner in 5 which the bonus system was to be implemented, CMS is using a 6 program-wide demonstration project to determine bonus 7 payment amounts. 8 rating of four stars or higher would have been eligible for 9 bonuses, in the demonstration, bonuses begin at the three- 10 star level, meaning that based on the bids that we saw and 11 the projected enrollment, 93 percent of enrollees are 12 expected to be in plans that will receive quality bonuses in 13 2012 versus 25 percent of enrollees that would have received 14 bonuses under the statute because they're in four- to five- 15 star plans. 16 higher payment amounts to the bonuses themselves, the 17 demonstration is expected to result in an additional cost of 18 $2.8 billion in 2012. 19 While under the statute only plans with a As a result of this and because CMS has set The Commission sent a letter to CMS expressing its 20 concern over the use of demonstration authority for this 21 purpose, that is, to do a program-wide demonstration, and 22 over the cost of the demonstration, the resulting cost. 229 1 Another concern that we raised regarding the star 2 system is the actual method for determining the stars of 3 plans. 4 stars placed too great an emphasis on contract performance 5 measures, that is, things like how responsive the call 6 centers were and whether or not they had foreign language 7 interpreters versus clinical measures such as outcome 8 measures, intermediate outcome measures, and patient 9 experience measures that are from the CAHPS regarding 10 The concern that we raised last year was that the patients' perceptions of their care. 11 CMS did address this concern and they are now, for 12 the bonuses that will be effective in 2013, they are using a 13 weighting system to assign different weights to different 14 types of measures, resulting in a system where contract 15 performance measures are not as heavily weighted. 16 weights are now attached to outcome measures and patient 17 experience measures with less weight given to process 18 measures and contract performance measures. 19 new system will be the basis of bonus payments in 2013. 20 This concludes our presentation. Higher And again, the We look forward 21 to any comments you may have on the draft chapter, any 22 additional information you would like to see included in the 230 1 chapter, and any questions you may have on the draft chapter 2 or on the material that we've just presented. Okay. Thank you. 3 MR. HACKBARTH: Thank you. 4 Scott, could you put up the slide that has the 5 comparison of the bids and payments, Slide 7. 6 give just a brief reminder of what the recent trend has been 7 on the bids to fee-for-service ratio? 8 that over the last several years, that has been trending 9 down. 10 11 DR. HARRISON: MR. HACKBARTH: 13 DR. HARRISON: 14 MR. HACKBARTH: The payments -- yes, it seems to me Yes. Last year, the payments were 110. Yes, and I'm focused in particular on the bids to fee-for-service ratio. 16 17 My recollection is they were a high of about 114, maybe, a few years ago. 12 15 Could you DR. HARRISON: The bids last year were 100 percent -- 18 MR. HACKBARTH: 19 DR. HARRISON: Yes. -- with the HMOs being 97 and 20 everybody else above. And before that, they had been -- 21 they'd sort of hovered there a little bit, but they may have 22 been a little bit over fee-for-service for a few years. 231 1 MR. HACKBARTH: Yes. Yes. That's my 2 recollection, also. 3 trend, especially this year, is consistent with what we've 4 seen in other sectors, so when there's pressure on rates, 5 the providers, in this case plans, respond to that by 6 reducing costs. 7 they can have money to provide added benefits, which is 8 their mechanism of attracting customers. 9 strikes me as good news and consistent with past MedPAC 10 So one hypothesis is that this downward And here, they want to reduce costs so that Now, so this thinking about the Medicare Advantage program. 11 What I wanted to ask about, though, is there any 12 way that this could be confounded by the quality bonus 13 program? 14 got this source of money that, as I understand it, is 15 outside this table? Are they somehow bidding less because, oh, we've 16 DR. HARRISON: 17 MR. HACKBARTH: 18 DR. HARRISON: 19 MR. HACKBARTH: 20 DR. HARRISON: 21 22 No, it's inside the table. Oh, it is inside. Right, so the -Okay. I misunderstood -- The benchmarks have been puffed up for the quality bonus. MR. HACKBARTH: Okay. I misunderstood that point. 232 1 DR. HARRISON: 2 MR. HACKBARTH: 3 Yes. Okay. Scott, why don't you lead this. 4 DR. MARK MILLER: Can I say one other thing about 5 that? So that also means, because the bonuses are in there, 6 that the 107 on payment is probably closer to fee-for- 7 service than 107. 8 9 DR. HARRISON: Yes, I would bet maybe like 104 or 105, probably. 10 DR. MARK MILLER: So that's the other thing. Just 11 to your point, because it is, in fact, ground into this 12 table, it also means the payments over fee-for-service are 13 closer if you had those bonus payments out. 14 DR. HARRISON: 15 DR. MARK MILLER: 16 -- everybody with me? 17 Mm-hmm. If you see what I'm saying. Okay. MR. HACKBARTH: So I got one nod, so I'm done. Why don't we still just plan on 18 doing one round with clarifying questions and comments in 19 one round. 20 Scott? MR. ARMSTRONG: First, I probably should know 21 this, but the discretionary bonuses that take the three 22 stars, is there a time limit on that? I mean, how will that 233 1 2 3 4 5 6 7 play out over the next couple of years? MR. ZARABOZO: [Off microphone.] is the demonstration. MR. ARMSTRONG: And at that point, the payments are limited -MR. ZARABOZO: And it goes back to the statutory four-star and above. 8 MR. ARMSTRONG: 9 MR. ZARABOZO: 10 11 MR. ZARABOZO: You said 2014? Yes -- 2015. MR. ARMSTRONG: 12 13 Twenty-fourteen Twenty-fifteen? [Off microphone.] The demonstration is 2012, 2013, and 2014. MR. ARMSTRONG: Thank you. Thanks. No, I don't 14 think I have any comments. Glenn, I just was -- I had 15 reacted to this information in a way that's similar to 16 yours, and that is that it seems that there is good news in 17 some of this data in that I don't know what MedPAC's role is 18 in drawing conclusions like that, but the relative cost to 19 the Medicare program of MA plans compared with fee-for- 20 service is getting closer to where we want it to be. 21 Second, you see a real distinction in the 22 progress, and I suppose this has been historical between the 234 1 different plan types, and I know we describe the 2 differences, but I don't know that we've ever weighed in on 3 those things that we may do through our role in encouraging 4 faster growth of those plan types that are giving us the 5 better results and the better return on our investments, and 6 it just might be something to think about. 7 And then, finally, given that we are going to be 8 teeing up a conversation on down the road around premium 9 support, it really does seem to me that the closer these MA 10 plans get to fee-for-service 100 percent benchmarks, the -- 11 I mean, that starts becoming really an interesting part of 12 the conversation that we would have there. 13 14 But in terms of the chapter itself, I don't really have anything more to say about it. 15 DR. BAICKER: I think it's a really useful point 16 to make. Even though on some level we're all aware of it, I 17 feel like it hasn't quite percolated through into the public 18 discussion that a big reason we pay MA plans more is not 19 because of their bids, but because of the benchmark. 20 thought this laid that out really clearly. 21 wasn't as clear on in thinking about the star system is our 22 goal is to pay plans -- reward plans for providing high- So I And what I 235 1 quality care in an efficient way, and I wonder how the star 2 system does or is intended to evolve over time in two ways. 3 One, if fee-for-service gets higher quality, 4 should the stars get ratcheted up? 5 be meeting or exceeding fee-for-service performance or an 6 absolute performance metric is one dimension. 7 Should we expect MA to And the second dimension I wondered about is it 8 seemed from the chapter that people, if I'm remembering 9 correctly, disproportionately enroll in higher star 10 programs, plans, which you would hope. 11 moving towards higher quality if we're giving them enough 12 information about it. 13 enrollment patterns in the sense that will every plan that a 14 person is actually enrolled in be a five-star plan 15 eventually, or will that move up as people move into higher- 16 performing plans? 17 air in a pilot, but how is it supposed to function? 18 People should be Eventually, do the stars track actual And then I know all of it is up in the MR. ZARABOZO: Well, one thing about the five-star 19 plans is that they can enroll year-round, so that's an 20 advantage to being a five-star plan, in addition to the 21 higher bonus payments, especially in the absence of the 22 demo, you are going to get those bonus payments. Part of 236 1 the -- you know, there was a table that showed distribution 2 by type of plan. 3 example, regional PPOs are all three-star, pretty much. 4 that distribution happens to be the current situation, but 5 it does reflect that HMOs tend to have better quality than 6 the other plans, at least in the current measure. There are no five-star plans -- for 7 Now, in terms of the benchmarks and where the 8 standard would be, CMS is working on how to incorporate 9 improvement into the star system. So Right now, it's relative 10 within MAs to a certain extent. 11 levels for, like, the four-star performance is at a certain 12 threshold level and that is typically unchanged from one 13 year to another. 14 MA. 15 compared to fee-for-service, and hopefully at some point 16 there will be a comparison to fee-for-service. 17 18 They do have threshold But otherwise, it is in relation within So the point about we have always said you need to be MR. HACKBARTH: Kate, we did a report, was it two years ago now, Carlos? 19 MR. ZARABOZO: 20 MR. HACKBARTH: March 2010, it was released. Yes. It was a report that 21 Congress specifically requested on how to go about making 22 comparisons of MA plans to fee-for-service. There are some 237 1 challenges there, suffice it to say. 2 Cori. 3 MS. UCCELLO: 4 suggestions on the quality part. 5 information there, a lot of good and important information, 6 but my eyes did start glazing over at all of the detail. 7 I tried to think about, well, what are the questions I want 8 this information to answer, and I think, actually, in your 9 presentation, you kind of laid things out this way. Okay. I have some comments and There's a lot of So But one 10 thing is are the metrics themselves meaningful and do they 11 produce meaningful differences between the plans? 12 are the overall trends in quality? 13 are there specific areas that need more attention? 14 how does the quality vary by plan type and between MA and 15 fee-for-service, and then also with the dual eligibles and 16 SNPs plans and those things. 17 are big differences between these plan types, are there 18 lessons that can be learned from MA and transferred to fee- 19 for-service or vice-versa? 20 the big issues that I'd hope that this kind of quality 21 information can answer. 22 MR. HACKBARTH: And what Is it improving? And And then And then in areas where there I mean, those are just kind of Peter. 238 1 MR. BUTLER: So, Scott, as you were referencing 2 our incrementalism, the flip side of this is that these are 3 a little bit so five years ago, too. 4 moved, but not much. 5 Medicare -- another way to look at this is as the enrollment 6 has grown in Medicare Advantage, it has been more expensive. 7 So everybody that's enrolled in there is more than the fee- 8 for-service, as much as we as a Commission have valued the 9 incentives associated with them. 10 These numbers have And as the enrollment has grown in Now, what I wanted to ask you to clarify in my own 11 mind, we annually report on the status. 12 legislative activity, or is that -- did we just decide to do 13 an annual status report? 14 DR. MARK MILLER: Is that part of our I'm pretty sure -- I would 15 really like to just put my eyes back on the legislative 16 language again. 17 statements to the Congress on fee-for-service and MA that 18 goes into the March report. I think it says if we're making payment 19 MR. BUTLER: But we're not -- 20 DR. MARK MILLER: 21 MR. BUTLER: 22 DR. MARK MILLER: Well, but we have -- Right. -- so it turns out this time, we 239 1 haven't, but some of this movement that you see in these 2 numbers that are so five years ago, in your words -- or 3 Scott's -- came from recommendations that we did make. 4 it turns out this time in the chapter, no, we don't 5 necessarily have recommendations, but we have in the past. 6 MR. HACKBARTH: So Peter, was your question 7 specifically about payment recommendations, whether we're 8 required to make payment recommendations, or whether we're 9 required to -- 10 MR. BUTLER: I know we don't make payment 11 recommendations unless we have a change in the system. 12 There are no updates to provide. 13 DR. MARK MILLER: 14 MR. BUTLER: 15 DR. MARK MILLER: 16 MR. BUTLER: Correct. Exactly. I understand that. Yes. I understand that, and we choose 17 nevertheless to report in the March chapter how the program 18 is doing -- 19 MR. HACKBARTH: 20 MR. BUTLER: Yes, and my recollection -- -- and we have had selected studies 21 that we are required to do by Congress, and so I'm just 22 trying to clarify. But beyond that, we don't in our 240 1 legislative -- I mean, as we were constituted, we do not 2 have to report beyond that level, right? 3 MR. HACKBARTH: Beyond that level. So my 4 recollection, Mark, and help me out, is that there was a 5 period when we didn't have Medicare Advantage in the March 6 report and one of the Congressional committees said, you 7 know, the way we read MedPAC's statutory requirements for 8 the March report, we think you need to report each year on 9 Medicare Advantage, not make payment recommendations since 10 that's the way the MA payment system works, but at least 11 report on the status. 12 that. 13 that's what you want, we will do it, and we've done it each 14 of the last, I don't know, three or four years. 15 And we said, of course, we will do The language was ambiguous to us, but we said, if MR. BUTLER: Yes. So, I mean, I don't need to 16 tell you, Glenn, you repeat it fairly often, the original 95 17 percent of AAPCC is still a long way away from what we are 18 seeing from the plans. 19 kind of interesting that we have 35 pages out of probably 20 hundreds that we report, and I just kind of wonder as we 21 move more to likely, whether it's ACOs or more Medicare 22 Advantage, whether we should be a little more proactive in So we have 35 pages here. It's just 241 1 thinking about how to shine a little brighter light and a 2 little bit more of our attention on the issue, because this 3 is complicated. 4 haven't been involved with Medicare Advantage, this thing 5 takes a long time to go through and understand if you 6 haven't been a part of it and I think it's really important. 7 And I'm sure the new Commissioners who MR. HACKBARTH: For the benefit of the new 8 Commissioners, when I first joined the Commission, Medicare 9 Advantage payment rates were a very hot issue and we made a 10 number of recommendations on those that went basically 11 unheeded for a number of years. 12 moved in the general direction that we have been 13 recommending. 14 had recommended and it has some bells and whistles, like the 15 quality bonuses that we didn't necessarily contemplate, or 16 at least not in the magnitude. 17 in a way we have been describing as far back as 2001, 2002, 18 2003. 19 next policy debate around the role of private plans in 20 Medicare, as Scott indicated, is in the premium support 21 context, and so we as a group will need to decide how we 22 wish to engage in that debate. And then in PPACA, Congress The PPACA approach isn't precisely what we But directionally, it moved PPACA passed a couple of years ago. Now, I think the 242 1 Mike. 2 DR. CHERNEW: So I always like these reports, so 3 thank you, and I just have one comment. 4 have about the number of plans, I think, going from 21 on 5 average to 12, or -- on one of the slides. 6 Between 2010 and 2011, the average number of choices -- I 7 assume that means the average number of plans -- dropped 8 from 21 to 12, and I assume a lot of that was consolidation 9 of private fee-for-service plans that needed to have a 10 There's a slide you But -- this one. network and dropped out? 11 DR. HARRISON: 12 DR. CHERNEW: That's correct. Right. And then there were also 13 some efforts to encourage plans to consolidate in general, 14 and you don't see that generally picked up in the 15 enrollment. 16 In fact, the enrollment was rising. So my comment is that the last time we had really 17 big MA payment changes -- well, actually not -- the last 18 time we had really big MA payment reductions, we saw a ton 19 of plans existing, and although you see plans consolidating 20 here, it's not clear you see a lot of exiting, all of which 21 I think is a good thing, although I just want to say, we've 22 been doing some work looking at this in more detail and you 243 1 do see different patterns, and I can show you the results 2 across the quartile. 3 the generosity, if you will, varies by quartiles, and 4 there's a lot of noise going on for all the reasons you said 5 -- privacy, plans leaving, and plans being asked to 6 consolidate within the same insurer and stuff like that. 7 But there does seem to be some effect within quartiles. 8 9 So now they've set up a system where The other thing that's going on which I think is really important to understand and why I think you see 10 growth in MA amongst other reasons is there's a bunch of 11 reasons why I think the non-MA system, the traditional 12 system, is going to become less appealing to individuals. 13 Employers are cutting back. 14 MA remains a place where you can get access to some of the 15 benefits that you might want in a way that, you know, still 16 might be cost effective for individuals. 17 seeing the demand for individuals for MA plans really rising 18 and so you see a lot of enrollment, despite the fact that 19 the payment rates are becoming -- are scheduled to become 20 less generous. 21 it. 22 consolidation, although we didn't see it in enrollment. A bunch of things are going on. So I think you're And we saw our results when they just froze Before they even implemented them, we saw some level of So 244 1 2 the enrollees are finding ways to go somewhere else. So again, I think that this is really good. It 3 will provide a great baseline. You are going to see a lot, 4 even more as we go forward, as these cuts come in, because 5 there's a big change across certain counties when you move 6 from the benchmark system the way it had been set up to the 7 benchmark system the way it's going to be set up. 8 some places that actually -- we had great presentations 9 before about how overly generous some places were because of There's 10 some of the craziness of the payment system and some of 11 that's getting unraveled by the Affordable Care Act and we 12 can trace what happens, and this is the start. 13 My only point is, as you go forward, thinking 14 about that in the sort of quartile world is important. 15 I'll share with you the work that we've done. 16 DR. DEAN: These are encouraging numbers. And I had 17 some of the same questions that Cori did about the 18 measurement of the quality and also the impact that it has, 19 and it sounds like those things, too, are moving in a way 20 where the measurements are more sophisticated and hopefully 21 more accurate. 22 How much difference is there for the plans to move 245 1 from, say, three to four stars, or four to five stars? 2 assume that there's still incentives to keep pushing on 3 those measures, is that correct? 4 MR. ZARABOZO: Right. I There are different levels 5 of bonus payments, depending on the star level. 6 after the demonstration, you have to be at four, four-and-a- 7 half or five stars to get a bonus, so -- 8 DR. DEAN: 9 MR. ZARABOZO: And again, For each step up -- 10 DR. DEAN: 11 MR. ZARABOZO: 12 DR. DEAN: You have -- -- there will be an additional bonus --- different levels -- It's enough of a step to really provide 13 an incentive, as far as you know? 14 MR. ZARABOZO: 15 DR. DEAN: 16 17 Well, it is more money, so -- But is it enough to make up for the difference of the investment that they'd likely make to -MR. ZARABOZO: Well, the other thing is, again, 18 the five-stars get to enroll year-round, so that's sort of a 19 little boost there, also. 20 21 22 DR. DEAN: Are there any other non-monetary rewards besides the year-round enrollment? MR. ZARABOZO: The marketing. At the Medicare.gov 246 1 website, the five-stars plans are indicated as being 2 exceptional plans, and the plans that are at the other end, 3 the 2.5-star plans, are labeled as this plan has not been 4 good for a couple of years or something, so -- 5 6 DR. DEAN: And the five-star plans are the only ones that are labeled that way? 7 MR. ZARABOZO: 8 DR. DEAN: 9 MR. HACKBARTH: 10 Right. Okay. So how many five-star plans are there? 11 12 Right. MR. ZARABOZO: nine. Thank you. I think we have -- well, this -- My consultant here says nine. 13 [Laughter.] 14 DR. HALL: Just a couple of comments. One is, 15 this was really a good chapter. 16 of MA descriptions in a variety of different places and this 17 was really very, very informative and I really thank you for 18 that. 19 This was -- I've seen lots In terms of the star system, certainly, that's 20 been a good idea, but it's a little bit like what do we do 21 when everybody's five-star? 22 things, or is it sort of like Lake Wobegon, that when all Do we then stop worrying about 247 1 the children become above average, you don't really know 2 what to do with them anymore? 3 guess I'm -- so I think there's a -- we have to keep a close 4 eye on whether we really have reached nirvana or perfection 5 when we use these star systems. 6 So one of the things, and I The other thing that I think is correct, and 7 there's sort of some other data that I've looked at 8 recently, is that for any of the major national providers, 9 underwriters for MA services, the rubric or the metric they 10 often use is that in 80 percent of our localities, we are 11 four-star, something like that. 12 out the regional data, there will be pockets of the country 13 where they're not really very good, even though on a 14 national basis they're really quite good. 15 variation still exists and this is very, very difficult for 16 consumers to understand. 17 look this stuff up, and I'll tell you, it is a real 18 challenge when you go do that. 19 But then when you dissect So the regional They can go to Medicare.gov and So not that I have any really good ideas on this, 20 but as we follow this, I think at some point, when everybody 21 gets to be five-star, we have to push the industry to do 22 what it really should be doing and that is to become even 248 1 more innovative than five stars, and I don't know what that 2 means. 3 MR. ZARABOZO: Well, I'm not sure that's 4 necessarily the case, that everybody will become five-star, 5 again, because of the relatives, that is, within -- I mean, 6 it will be the best plans that are the five-star plans. 7 Other plans will be below that. 8 DR. HALL: But they're trending that way. 9 MR. ZARABOZO: 10 DR. HALL: 11 MR. HACKBARTH: 12 They're -- Yes, I know, that was an overstatement. The measures in the system are not fixed, either -- 13 DR. HALL: Right. 14 MR. HACKBARTH: -- and so the measures presumably 15 will evolve -- hopefully, they'll evolve to be more 16 effective measures of quality, and so this isn't a static 17 system where people are moving at a fixed target. 18 Bob. 19 DR. BERENSON: Yes. I have two data issues and 20 then one comment. Scott and I talked earlier about two data 21 things which I think it's useful to bring up publicly as a 22 question. One is that Marsha Gold's analysis of Medicare 249 1 Advantage that she produces at Mathematica for Medicare -- 2 I'm sorry, for Kaiser Family Foundation has a different 3 number of Medicare Advantage enrollees and it's -- Scott and 4 I, I think, have figured out that it's because she includes 5 the costs of HMO enrollees. 6 explain who they are and whether it makes more sense to 7 include them in Medicare Advantage? 8 percent. 9 Medicare Advantage, and I just want to understand that. 10 And so, just quickly, could you The difference is two In her data, we're now up to 27 percent of DR. HARRISON: Yes. I think one percent is 11 probably due to the numerator and one percent is due to the 12 denominator. 13 numbers of beneficiaries, so you could end up picking ones 14 that are slightly different from one another. 15 thinking that's where part of it is. 16 CMS puts out all kinds of different total So I'm But the part that I do know for sure is that so 17 there are these HMOs called cost HMOs and what they are is 18 you can sign up for a closed panel -- you as a beneficiary 19 could sign up to get your care from this closed panel, but 20 you don't have to get all of it from the closed panel. 21 You're allowed to go outside of that network and whenever 22 you do, you pay normal Medicare cost sharing. When you're 250 1 in the network, you're going to pay nothing or something 2 lower. 3 Colorado is covered by a cost plan completely, I believe. 4 Scott and White has lots of parts of rural Texas. 5 actually even Kaiser in a couple locations is considered a 6 cost plan. 7 not really paid the same way. 8 worth looking at in the next year or two. 9 10 So there are some big pockets of the country -- And those plans do not submit bids, so they're MR. HACKBARTH: But they may be something Scott, is it possible for a plan to get a new cost contract -- 11 DR. HARRISON: 12 MR. HACKBARTH: 13 And No. -- or these are people grandfathered in under old contracts? 14 DR. HARRISON: Right. 15 DR. BERENSON: So these are clearly not risk -- I 16 mean, they're not Medicare Advantage per se, but they're 17 also not pure fee-for-service, so they're in some middle 18 ground, but okay. 19 DR. HARRISON: Right. 20 DR. BERENSON: The more important issue, if you 21 could go to Slide 7, Peter said that the new Commissioners 22 may have some difficulties. I've written a half-a-dozen 251 1 articles in Health Affairs and don't think I understood 2 this, because Scott and I were talking. 3 that because MA plans get to decide where they want to go 4 and that they would typically disproportionately settle in 5 places with high benchmarks, high fee-for-service spending, 6 that the fact that the ratio wasn't adjusted for equal 7 distribution -- I mean, I basically assume that if they were 8 equally distributed across the country, that the number 9 might be somewhat different rather than what you're showing I have long assumed 10 here, which I assume showed a bias towards going to high- 11 cost fee-for-service areas. 12 you're making an adjustment. 13 you're doing? 14 DR. HARRISON: You suggested earlier that Could you explain that, what Right. So let's just say we have 15 one plan. 16 enrollees are and I know the risk scores of their enrollees, 17 and so I assume that those are -- and so I, in a sense, 18 create a different plan with the same enrollment patterns 19 and the same risk patterns and see what those people would 20 cost under fee-for-service, okay. 21 a demographically and geographically similar population. 22 So I would take the plan, and I know where their DR. MARK MILLER: And so we're comparing to [Off microphone.] It's MA to 252 1 fee-for-service -- 2 DR. BERENSON: Right. 3 DR. MARK MILLER: Right. -- not across the country, and 4 your comment almost sounded like it was an across the 5 country kind of -- 6 DR. BERENSON: Well, now I'm not sure what -- I 7 mean, basically, if my hypothesis is that there's a lot of 8 MA plans in Miami and New York, which you've told me there 9 are, and those are very high fee-for-service spending areas, 10 that if I -- and those are all the plans we had, we would 11 show a pretty low ratio of bid to fee-for-service which 12 might give a misleading picture of how efficient MA plans 13 were if they were actually equally spread across the 14 country. 15 kind of an adjustment so that your -- do you see what I'm 16 asking? 17 18 19 And so I guess the question is, have you made that DR. HARRISON: Yes. These numbers are all based on where the plans draw their enrollment. DR. BERENSON: All right. So that adjustment 20 hasn't been made. The one that I'm suggesting needs to be 21 made to be able to say anything about our MA plan is lower 22 cost than fee-for-service, which is a question that policy 253 1 makers like to ask. 2 DR. HARRISON: Now -- 3 DR. BERENSON: We don't really know that answer. 4 DR. HARRISON: Now, the problem is that you may 5 not trust or like the fee-for-service numbers in Miami, but 6 that's what they are, and so -- 7 DR. BERENSON: No, I understand that -- 8 DR. HARRISON: Yes -- 9 DR. BERENSON: -- but if one were wanting to 10 project what a policy might be, one would want to know what 11 the limitations are of the data and one of them might be, 12 well, these plans have gone disproportionately to high cost 13 fee-for-service areas. If they were actually -- 14 DR. HARRISON: No -- 15 DR. BERENSON: -- if we did away with traditional 16 Medicare and all we had were plans, what might we expect? 17 We don't really know that without doing that kind of 18 geographic adjustment -- 19 DR. HARRISON: 20 DR. BAICKER: No -Doesn't Scott's number give us a 21 better sense of what that number would be than what you're 22 proposing in that the counterfactual is what fee-for-service 254 1 is spending, you know. And so if we're observing 2 beneficiaries in MA disproportionately in areas where fee- 3 for-service spends a lot and they're spending less than fee- 4 for-service there, we want to know that they're located in 5 areas where fee-for-service is spending a lot and not say, 6 oh, wow, the enrollees in Medicare Advantage in Miami are 7 spending a lot more than the fee-for-service enrollees in 8 Minneapolis, that that's not the informative comparison. 9 The more informative comparison is what the MA enrollees in 10 Miami are spending relative to the fee-for-service enrollees 11 in Miami and what the MA enrollees in Minneapolis are 12 spending relative to the fee-for-service enrollees in 13 Minneapolis. 14 DR. BERENSON: That would be better than -- 15 comparing Miami to Minnesota wouldn't be helpful, but also I 16 think it would be helpful to do a geographic adjustment to 17 estimate what the spend would be if the plans were sort of 18 naturally distributed or equitably distributed. 19 DR. BAICKER: I don't -- But then you'd have to do the same 20 thing to fee-for-service, right, because that looks 21 different in different parts of the country, too. 22 DR. MARK MILLER: [Off microphone.] And the other 255 1 thing that occurs to me -- the other thing that occurs to me 2 in this, that one of the issues that would get raised in 3 this is you can't necessarily assume that a managed care 4 plan would go to every part of the country, and I get 5 worried that if you make an assumption that says, okay, 6 let's take this and assume it all the way across the 7 country, if you really paid in such a way relative to fee- 8 for-service, there are certain areas of the country where 9 managed care plans wouldn't go. 10 MR. ZARABOZO: We can't do what Bob would like us 11 to do, I don't think, because you would have to say, when 12 this company goes to Iowa, let's say, that had never been 13 there before, we think their bid will be such, and that's X 14 percent of fee-for-service. 15 what that bid would be of an HMO going to Iowa. 16 is what is out there. 17 MR. HACKBARTH: We can't do -- we don't know All we know I think it is reasonable to 18 believe that private plans will be more successful in some 19 parts of the country than others relative to fee-for-service 20 because -- not just because the level of fee-for-service 21 cost varies across the country, but also because the care 22 delivery systems vary. 256 1 So in a market where you have few providers, a 2 sparsely populated rural area, and those providers have few 3 competitors and therefore substantial market power when 4 dealing with a private insurer, that's a tough market to 5 make a private health plan work and beat Medicare's 6 administered price system, especially if you're talking 7 about the swaths of the country where the utilization rates 8 are also low. 9 utilization is low. 10 So the unit prices are low and the In other parts of the country, you have the 11 reverse situation, where you've got many providers, 12 opportunities for plans to negotiate, steer their 13 beneficiaries towards particular providers, and play one 14 provider off against the other in the negotiating process. 15 And in addition to that, they have high utilization rates. 16 So Bob's observation is exactly the right one. 17 Plans historically have tended to cluster in the areas that 18 have high fee-for-service costs and many providers. 19 the rich environment for them and they've avoided the low 20 utilization sparse provider areas of the country until 21 Congress said, well, we want private plans everywhere and 22 adopted the policy of paying way above fee-for-service rates That's 257 1 in those sparsely populated areas with few providers in 2 order to attract private plans in. 3 And that's how we got into this situation that 4 somebody alluded to earlier that we had payments way above 5 fee-for-service rates in many parts of the country, ratios 6 like 140 percent, as I recall, Scott, in some parts of the 7 country. 8 And what's happening now with PPACA is we're moving away 9 from that and moving closer to having MA payments broadly MA rates are 140 percent of fee-for-service cost. 10 linked to underlying fee-for-service costs with very 11 important exceptions to that. 12 plans cluster in the high-cost areas for understandable 13 reasons. But I do think it's true that 14 And as I understand these numbers, Scott -- now, 15 to get back to Bob's specific question -- these numbers in 16 this table -- whoops, that table -- the fact that HMOs are 17 95 percent of fee-for-service, that reflects the markets 18 that they choose to be in, the markets that HMOs are 19 actually in. 20 21 22 DR. HARRISON: Right. Now, I would like to point out, though, that they are going into more and more markets. MR. HACKBARTH: Yes. 258 1 DR. HARRISON: The HMOs cover, and I was looking 2 for the number and I don't see it, but I think something 3 like 83, 84 percent of beneficiaries now have access to an 4 HMO, whereas that's way higher than it was before. 5 DR. BERENSON: But the penetration is obviously 6 very different, with some counties being 60, 70 percent and 7 others being -- in any case, I don't think we can continue 8 this now. 9 and editorial writers like the New York Times a few weeks I would just point to the fact that policy makers 10 ago are going to want to argue as to, in a context of 11 premium support discussions, which is more efficient, 12 traditional Medicare or Medicare Advantage, and to the 13 extent that we're able to help ground that debate in facts - 14 - 15 MR. HACKBARTH: 16 DR. BERENSON: 17 Yes --- I mean, that editorial writer had it wrong -- 18 MR. HACKBARTH: 19 DR. BERENSON: Right. -- it would be helpful. And so I 20 actually did, then, understand what you had done and hadn't 21 been wrong for all those Health Affairs articles, so -- 22 MR. HACKBARTH: That's good. 259 1 2 3 DR. BERENSON: -- so it's helpful. Let me finish my comments and then I'll move it along. The comment I wanted to make was on the quality 4 side, which I found quite disturbing, the SNP findings. 5 fact, for regular MA, we do have a choice construct and 6 people can look at the data and deal with their own 7 preferences and decide whether to join an MA plan or not. 8 9 In There's an increasing discussion now about for the dual population about mandatory enrollment in managed care, 10 and so I find the one area that at least some people are 11 contemplating that seems to be the area which was supposed 12 to be sort of the sweet spot of what private plans could do 13 is working on patients with chronic conditions and multiple 14 comorbidities, et cetera, and in that area, at least, for 15 the quality measures we've got, they're actually doing worse 16 than sort of unmanaged fee-for-service. 17 there's anything more you can amplify on that, but I find 18 that one is not everybody is getting to five-plus in that 19 area and that's the area that may, in fact, be the most 20 immediate one of policy relevance for us. 21 22 MR. ZARABOZO: So I'm not sure if Well, I think the SNP plans would say that some of these measures are not appropriate for 260 1 them, and the famous one is the colorectal cancer screening 2 measure, which they say -- which is percent up to age 85 3 that have the screening, and they say for many of their 4 people, it is inappropriate to get that screening even 5 though they will suffer in the HEDIS measures by not doing 6 it. 7 DR. BERENSON: No, that -- I mean, clearly, that's 8 a problem across the whole program, where you should stop 9 that at 75, I believe, seems to -- no. Okay. So I'd like 10 to know a little more about what we think about those 11 measures. 12 I mean, the ones -- the flu vaccination -MR. ZARABOZO: Right. That's where -- I mean, 13 it's clear there it's lower and I don't see an explanation 14 why it should be lower -- 15 DR. BERENSON: Yes. Okay. I think that one 16 deserves some -- if we're sort of setting out priorities for 17 future work, I think understanding what the SNPs are doing 18 in quality, to me, would be a high priority. 19 MR. ZARABOZO: Yes. And again, it's difficult to 20 identify them because they're often under a larger contract, 21 which is the reporting entity. 22 out what exactly is happening in the Special Needs Plans. So we sometimes can't figure 261 1 MR. KUHN: A couple quick questions. One, Scott, 2 take me out to, if we could, the 2015, when this current 3 demo ends on the quality that's going on, or maybe Carlos, 4 whoever is working on that one. 5 this demo and see the numbers here where the bids are 6 starting to come down. 7 the demo's done that we'll see bids start to creep up, maybe 8 be above the benchmark, the extra benefits that people get 9 as coming below the benchmark start to disappear, and we So we go out to the end of Is there a likely scenario that when 10 start to see a migration back to fee-for-service and out of 11 MA? 12 this demo ends? 13 Is that a plausible scenario that we could see when DR. HARRISON: It's kind of hard to know what 14 would happen there. 15 continuing quality program, and as time goes on, that will 16 get a little bit stronger because it's being phased in. 17 won't be fully phased in until 2017. 18 going to be quality money. 19 be sort of for the average plans. 20 four- and five-star plans. 21 22 MR. KUHN: I mean, there's going to be a It So there is still It's just that it's not going to It's going to be for the And those four- and five-star plans -- or, I'm sorry, the three-star plans are the ones that are 262 1 getting this money as the result of a demo. 2 of where Bob was talking about, where plans are kind of 3 clustered, are they clustered in parts of the country that 4 are at risk of losing access in the future, or do we know 5 where they are? 6 MR. ZARABOZO: You know, kind The -- I did look at the geographic 7 distribution to the extent possible because of the way the 8 contracts are set up and the lower-star plans tend to be -- 9 well, Puerto Rico is one situation, but in the South. The 10 five-star plans are spread across the country, so you have 11 Massachusetts, Maine, Washington, California. 12 MR. KUHN: Okay. 13 MR. ZARABOZO: I do have a correction. In the 14 four-and-a-half and five-star plans, the increase in the 15 benchmark is the same percentage, so that in 2014, it will 16 be a five percent increase in the benchmark for all those 17 four-and-a-half and five-star plans. 18 MR. KUHN: I guess the reason I'm just curious is 19 that it appears to me that we may be looking at a cliff in 20 2015 and I'm just curious, order of magnitude, what that 21 cliff may or may not be and could it hit certain geographic 22 areas of the country as we go forward. 263 1 My second quick question has to do with coding 2 intensity. You know, this morning we made recommendations 3 on payment updates for hospitals because of coding and 4 others out there. 5 adjustments that are part of the MA plan experience now, 6 started, I think, in 2010 and continues to go forward. 7 we -- based on the analysis that you're all doing, do you 8 think the coding intensity that's in place now and moving 9 forward is sufficient to capture the coding that's going on 10 in the MA plan world or do we think there's opportunity for 11 further recommendations or review in the future in this 12 area? PPACA had some coding intensity 13 DR. HARRISON: 14 how well the risk system is doing. 15 actuary types think that there's even more coding that's 16 being corrected for. 17 up starting next year. 18 maybe even 2019. Do 19 There's a variety of opinions about I think that some The coding adjustment factor will go It goes up through, I think, 2017 or It keeps going up for a while. But we don't have a really good way to figure out 20 what we think the coding adjustment should be. You know, 21 first of all, CMS changes the model every couple of years 22 and there's also coding changes going on in fee-for-service. 264 1 So we're really not comfortable saying what we think the 2 true adjustment should be at this point. 3 DR. MARK MILLER: But we do have on our agenda to 4 look at some of the risk adjustment issues here as they 5 relate, how well HCC does, thinking about how it would apply 6 in a dual eligible type of world. 7 topic, while what you've said is correct, we don't have an, 8 okay, we know the answer to it type of posture, we are kind 9 of poking around a little bit and looking at this. And at least on this So it's 10 not just completely stagnant. 11 be cautious as to whether we're going to have something to 12 bring to you, but we are kind of messing around a little bit 13 here. 14 DR. NAYLOR: I think Scott is correct to Just let me echo Bill's comment. 15 think this was an excellent report. 16 I questions. 17 MS. BEHROOZI: I have no further Yes, it's really thorough, and 18 after six years, I'm starting to understand it. And I would 19 just echo the concern that you raise with respect to the 20 star measures applying to broad contracts rather than to 21 units. 22 should know what we're measuring, and when somebody sees a It kind of goes to Cori's point about, like, we 265 1 star rating, they should know what it's measuring as opposed 2 to some disparate geography, like is McDonald's better than 3 Burger King? 4 corner, that kind of thing. 5 What about your local McDonald's on the MR. GEORGE MILLER: I generally support the 6 recommendation. I guess I would have a question on the 7 total cost as compared to fee-for-service if the MA plans 8 weren't there, if we had the total impact of the entire 9 system for the MA plans and then compared that if there was 10 no MA plans and fee-for-service just to show the magnitude 11 of impact to the system. 12 13 DR. HARRISON: -- 14 15 MR. GEORGE MILLER: DR. HARRISON: We have put that number in the report in the past. 18 MR. GEORGE MILLER: 19 DR. HARRISON: 20 MR. GEORGE MILLER: 21 22 In billions of dollars, yes. Soon, we're talking about real money. 16 17 You mean in billions of dollars or Yes. I guess we could do that. Just curious. It's billions, right? MR. HACKBARTH: -- all part of magnitude in the 266 1 past. 2 MR. GEORGE MILLER: 3 DR. HARRISON: Twenty billion? Yes. 4 I'll bet it's maybe ten. 5 MR. HACKBARTH: I bet it's lower now, though. 6 7 Yes. I was thinking ten, 11, something in that neighborhood. MR. GEORGE MILLER: And then the follow-up 8 question, do we get value for that. 9 patient satisfaction, it seems that we do, but I'm just -- 10 DR. HARRISON: Certainly, measured by There are extra benefits that the 11 beneficiaries get and now you could argue that there's also 12 you're getting quality. 13 MR. GEORGE MILLER: 14 DR. STUART: Okay. Thank you. I agree with Bill. I think this is a 15 fascinating chapter. I've enjoyed reading it every year. 16 It's a dynamic area and so seeing what's happening within MA 17 is really interesting. 18 that it would be useful to think about whether we could do 19 this with the Part D MA-PD plans. 20 chapter, the prior chapter, there are some very useful 21 comparisons between MA-PD and PDP. 22 raise is are there differences in such things as average And having said that, it strikes me In other words, in the But the question I would 267 1 premium, premium-free plans, proportion of plans that offer 2 gap coverage among the different MA-PD types. 3 DR. CASTELLANOS: 4 that you put into this. 5 I really appreciate the effort forward to it each year. 6 I thought it was great and I look George, I'm a little concerned that you're 7 supporting a recommendation and Glenn may ask for a vote, so 8 maybe I've been asleep and I don't know. 9 [Laughter.] 10 MR. HACKBARTH: 11 [Laughter.] 12 MR. GRADISON: I'm one vote behind, Ron, so -- Thank you for your work. The jury 13 is still out on these programs, but this is very helpful to 14 have an annual review at this level of depth. 15 DR. BORMAN: Good job. 16 MR. HACKBARTH: Thank you. No questions. So, let me just go back to Herb's 17 question about what happens at the end of the demonstration, 18 the quality demonstration, and there's a significant 19 reduction in the total payments going out to plans. 20 a critical and interesting policy question. 21 22 This is My belief has been that while we could, as a result of that, have somewhat fewer plans and somewhat lower 268 1 levels of enrollment, on the other hand, there may be some 2 benefits in terms of the type of plan that we get. 3 to continue to offer attractive benefits at lower levels of 4 Medicare payments, we may -- my hypothesis would be that we 5 would see plans start to become more organized and more 6 effective and doing what we want private plans to do, which 7 is effectively manage care for Medicare beneficiaries. 8 To put it another way, when we had very high 9 In order levels of payment relative to fee-for-service Medicare, 10 there was little reason for plans to aggressively manage or 11 try to identify efficient providers and steer beneficiaries 12 towards them. 13 generous benefits to attract new enrollees while not doing 14 much to actually manage the care, and that's sort of the 15 worst case scenario for the program. 16 enrollment at a high cost and private plans are not engaged 17 in doing the things that we need private plans to do. 18 have them in the program because they have the potential to 19 do things that traditional Medicare finds difficult. 20 They could, with the excess payments, offer You get a high We And so I am hopeful that as PPACA unfolds and 21 there's more pressure on the plans, that we will see the 22 private sector innovate and restructure itself, redesign how 269 1 they work so that they can compete effectively at lower 2 rates. 3 there's lots of room for them to do very well in Medicare 4 even at lower rates. I believe in private plans in the market. 5 Scott. 6 MR. ARMSTRONG: I think Just one more point to build on 7 what you just said. First, just to disclose, so I do work 8 for an organization that has a Medicare Advantage plan that 9 is one of the nine five-star plans in our country. But I 10 will tell you that it's very difficult, and I would just 11 tell you, we do not presume we will be able to maintain our 12 five-star status, even with all the advantages that our 13 organization has. 14 structure, as we've been talking about it, there are very 15 strong incentives for our organization to figure out how we 16 maintain that. 17 I would just say I think it's hard to imagine that the whole 18 industry clustering around five-star, I just don't think 19 that that's going to happen. 20 And so in today's bonus payment And I think to the degree that -- so, first, To the degree there continues to be strong 21 incentives through this bonus structure for advancing 22 exactly the kind of results that we're trying to advance in 270 1 so many other ways, we ought to be paying very close 2 attention to how this informs us. 3 you know, we're about to talk tomorrow morning about benefit 4 design. 5 quality program, but the Medicare Advantage plans 6 themselves, I think, very naturally flow into some of the 7 goals that we'll have in that chapter, too. And I was just thinking, There are a lot of features of this five-star 8 MR. HACKBARTH: Thank you, Scott and Carlos. 9 We will now have our public comment period. 10 Before you begin, let me just quickly review the rules. 11 more than two minutes. 12 signifies the end of the two minutes, and please begin by 13 identifying yourself and your organization. 14 No MR. SPERLING: When the light comes on, that Thank you. My name is Andrew 15 Sperling. 16 NAMI is the largest organization representing Medicare 17 beneficiaries living with severe and persistent mental 18 illness in this country. 19 I'm with the National Alliance on Mental Illness. This certainly probably goes in the category of 20 futility, but I will make the comment nonetheless. NAMI 21 would again like to express extreme disappointment in the 22 decision made by this Commission relative to cost sharing 271 1 for low-income subsidy and dual eligible beneficiaries in 2 the Medicare program. 3 lead to severe disruptions in adherence and treatment for 4 dual eligible beneficiaries, particularly vulnerable 5 beneficiaries, not just with a single condition but with 6 multiple chronic conditions who take many, many medications. 7 This is a profound mistake and could We're disappointed that this proposal, quite 8 frankly, appears to be divorced from the real prescribing 9 patterns that go on for these low-income beneficiaries that 10 have multiple chronic conditions, and the idea that, 11 somehow, their brand prescribing is driven by them insisting 12 on brand medications, in fact, they have very little control 13 over the prescribing decisions that are made for them on 14 behalf of their physicians. 15 impairments and other disabilities that, quite frankly, make 16 adherence difficult, and we're extremely concerned that this 17 proposal is going to disrupt their ongoing treatment and, in 18 fact, cost the Medicare program more in the long run. 19 Many of them have mental We recognize this Commission tried to make some 20 concessions by deferring to the Secretary to make the 21 decision about which therapeutic classes will be subject to 22 higher cost sharing. That gives us little confidence, quite 272 1 frankly, and a decision over time where the Secretary is 2 making more and more decisions that are driven by cost 3 control rather than ensuring strong, good clinical outcomes 4 for vulnerable beneficiaries that, in fact, cost will 5 overcome these decisions. 6 therapeutic classes subject to higher cost sharing. 7 We'll see more and more We're disappointed that in terms of the lowering 8 of the cost sharing to zero, that appears to be only for 9 generics in classes that are designated by the Secretary. 10 So, in fact, many beneficiaries would only get that if their 11 class were designated for the higher cost sharing. 12 I want to be clear. For someone to stay adherent 13 to a brand name medication that's prescribed by their 14 doctor, specifically by their physician, they could be 15 subject under this proposal to cost sharing increases as 16 high as 200 and 300 percent. 17 beneficiary to stay adherent to a particular compound 18 prescribed by their prescription, regardless of specialized 19 circumstances in terms of polypharmacy, other drugs, 20 contraindications, the fact that other compounds from that 21 class may have been tried and failed for that individual. 22 That would be the cost to that So, again, we understand the decision has been 273 1 made. It will now be transmitted to Congress. 2 to engage our grassroots advocates and all 1,100 affiliates 3 that are in all 50 States and all 435 Congressional 4 districts to educate their members about what the potential 5 of this proposal is for ongoing strong clinical care for 6 dual eligibles and LIS beneficiaries in the program. 7 you. Thank 8 MR. HACKBARTH: 9 We will reconvene at 8:00 a.m. tomorrow morning. 10 Okay. We will have Thank you. [Whereupon, at 4:32 p.m., the proceedings were 11 adjourned, to reconvene at 8:00 a.m. on Friday, January 13, 12 2012.] 13 14 15 16 17 18 19 20 21 22 1 MEDICARE PAYMENT ADVISORY COMMISSION PUBLIC MEETING The Horizon Ballroom Ronald Reagan Building International Trade Center 1300 Pennsylvania Avenue, N.W. Washington, D.C. Friday, January 13, 2012 8:05 a.m. COMMISSIONERS PRESENT: GLENN M. HACKBARTH, JD, Chair ROBERT BERENSON, MD, FACP, Vice Chair SCOTT ARMSTRONG, MBA KATHERINE BAICKER, PhD MITRA BEHROOZI, JD KAREN R. BORMAN, MD PETER W. BUTLER, MHSA RONALD D. CASTELLANOS, MD MICHAEL CHERNEW, PhD THOMAS M. DEAN, MD WILLIS D. GRADISON, MBA WILLIAM J. HALL, MD HERB B. KUHN GEORGE N. MILLER, JR., MHSA MARY NAYLOR, PhD, RN, FAAN BRUCE STUART, PhD CORI UCCELLO, FSA, MAAA, MPP 2 AGENDA Reforming Medicare’s benefit design - Julie Lee, Scott Harrison, Joan Sokolovsky PAGE 3 Mandated rural study: rural payment adequacy and plan for the final report - Jeff Stensland, Adaeze Akamigbo 110 Public comment 173 3 1 P R O C E E D I N G S 2 MR. HACKBARTH: Okay. Good morning. [8:03 a.m.] We have two 3 sessions today, one on reforming the Medicare benefit 4 package and the other on our mandated report on rural 5 payment adequacy. 6 design? Julie? 7 So who is leading off on the benefit Okay. DR. LEE: Good morning. In today's presentation, 8 we continue our discussion of potential changes in 9 Medicare's benefit design. 10 Recall that we began this discussion last October and November. 11 We have the following goals for today's 12 presentation: 13 in spring. 14 characteristics of the new benefit design you wish to 15 include and address policy questions relative to 16 implementing new benefits, such as an excise tax on 17 supplemental insurance. 18 We are working toward draft recommendations To make that happen, we need to define the key The Commission has been considering ways to reform 19 the traditional benefit package for several years to give 20 beneficiaries better protection against high out-of-pocket 21 spending and to create the incentives for beneficiaries to 22 make informed decisions about their use of care. 4 1 The Commission has been also particularly 2 concerned about the potential impact of such changes on low- 3 income beneficiaries and those in poor health. 4 So let's begin with a brief recap of the two 5 previous meetings on the topic. 6 we discussed the current fee-for-service benefit design and 7 presented three alternative benefit packages for you to 8 consider. 9 that they had an out-of-pocket maximum and a combined 10 11 In October's presentation, All three shared a common benefit structure in deductible for Part A and Part B services. In addition, the first package, presented as a 12 reference point, had 20 percent coinsurance on all services 13 between the deductible and the out-of-pocket maximum. 14 second and third packages had a set of co-payments instead 15 of co-insurance. 16 for co-payments because they are more predictable for 17 beneficiaries. 18 The The Commission has expressed a preference In November's presentation, we focused on the role 19 of supplemental coverage. Specifically, we used one of the 20 co-payment packages from October and combined it with the 21 three alternative policies related to supplemental coverage. 22 They varied in the degree to which Medicare's cost sharing 5 1 can be filled in by supplemental insurance. 2 The Commission at that time expressed a strong 3 preference for imposing an excise tax on supplemental 4 insurance rather than regulating supplemental benefits. 5 Today's presentation has two main parts. 6 First, we'll go over two benefit packages with a 7 revised set of co-payments. 8 beneficiary cost-sharing liability roughly the same as 9 current law (but program spending would be higher), whereas 10 the second package would keep the Medicare program spending 11 about the same (but beneficiary cost sharing would 12 increase). 13 The first package would keep In the second part of the presentation, we'll 14 overlay an excise tax on supplemental insurance on these two 15 benefit packages. 16 November as your preferred policy option toward supplemental 17 coverage. 18 This was the guidance you gave us in There are a couple of issues discussed in November 19 that are not on this outline. 20 analysis related to income and the value of insurance are 21 not forgotten and will be presented in the future. 22 We want to note here that the If you recall, the fee-for-service benefit package 6 1 from November had a co-payment structure of the type more 2 common under Medicare Advantage. 3 neutral" because it had approximately the same average cost- 4 sharing liability as the current fee-for-service. 5 It was labeled "MA- Today, we present two illustrative packages that 6 show key tradeoffs between some design elements. We'll 7 first go over what these packages look like. 8 $5000 out-of-pocket maximum and a combined deductible for 9 Part A and Part B services. Both have a Remember that under the current 10 benefit package, there's no out-of-pocket cap, and 11 beneficiaries face a high deductible of over $1000 for Part 12 A and a low deductible for Part B. 13 private plans, we differentiated co-payments for primary 14 care and specialist visits. 15 all physician services have a 20 percent coinsurance. 16 As typical in most In contrast, under current law, MA plans also told us that they have a co-payment 17 for advanced imaging services. 18 included here is similar to what MA plans charge and almost 19 equal to the average cost sharing for those services under 20 current law. 21 22 The $100 co-payment we Co-payments for skilled nursing facility services are higher compared to current law, which currently has no 7 1 2 cost sharing for the first 20 days. Finally, our cost sharing on home health is 3 consistent with the $150 co-payment per episode the 4 Commission has included in the March report from last year. 5 These two packages are quite similar in their 6 structure, but the levels of co-payments do differ because 7 they were set to meet two quite different budget 8 constraints. 9 Under the first package, average beneficiary cost- 10 sharing liability would be about the same as the current 11 law, but Medicare program spending would be higher. 12 package is referred to as the "beneficiary-neutral" package 13 on the slide. 14 This Under the second package, average beneficiary 15 liability would be higher compared to current law, but 16 Medicare spending would be roughly equal to current law. 17 This package is referred to as the "program-neutral" package 18 on the slide. 19 Comparing the two columns, the second has a higher 20 deductible -- $750 versus $500 -- and a higher co-payment on 21 skilled nursing facility days -- at $100 versus $80. 22 differences are highlighted in yellow. These 8 1 Generally, compared to current law, the 2 distribution of cost-sharing liability across type of 3 service is quite different under the two alternative 4 packages. 5 with very low Medicare spending would pay higher cost 6 sharing overall compared to current law. 7 hospital admission, on the other hand, would pay lower cost 8 sharing. 9 see their cost sharing go down substantially because of the 10 11 Because of the combined deductible, beneficiaries Those with a Those with very high Medicare spending would also out-of-pocket maximum. This is the chart that you've seen before. It 12 shows the results of simulating changes in out-of-pocket 13 spending and premiums for 2009 if the alternative benefit 14 package had been in place. 15 correspond to the two benefit packages we just described. 16 The first one reproduces the distribution from November. 17 The second and third bars So let's start with the one in the middle that is 18 the beneficiary-neutral package. The bottom part of the bar 19 -- that's the blue and green -- shows that 9 percent of 20 beneficiaries would see their out-of-pocket spending go down 21 by $250 or more under the new benefits. 22 are the beneficiaries who reach the out-of-pocket maximum. Generally, these 9 1 On the other hand, at the top of the bar -- that's 2 the orange and red -- about a little over 20 percent of 3 beneficiaries would see their out-of-pocket spending go up 4 by $250 or more. 5 are spending more out-of-pocket due to the deductible. 6 Mainly, these are the beneficiaries who The distribution under the program-neutral package 7 is shown in the third bar. Recall that this package has a 8 higher deductible and overall cost sharing is higher 9 compared to the beneficiary-neutral package. And this 10 difference you can see in the top part of the bar: 11 percent of beneficiaries compared to about 20 in the 12 beneficiary-neutral package would see their out-of-pocket 13 spending increase by $250 or more. 14 about 30 We want to point out here how the picture might 15 look different over a longer period of time. These charts 16 show the distribution in a given year. 17 pointed out in November, people's needs and preferences for 18 insurance change over time as they age. But as Karen has 19 For example, even though a little over 20 percent 20 of fee-for-service beneficiaries have at least one hospital 21 admission in any given year, that number goes up to more 22 than half if we look over five years. That means that many 10 1 more beneficiaries would see some years in which their out- 2 of-pocket spending would be lower under the new benefit over 3 a longer period. 4 If you recall, in November's presentation, we 5 looked at three policy options that restricted what 6 supplemental insurance can and cannot do. 7 Commission expressed a strong preference for the excise tax 8 approach over the regulatory approach. 9 was that under the tax approach, risk-averse beneficiaries At that time, the Its main argument 10 who wish to buy first-dollar coverage or reduce the 11 uncertainty in their out-of-pocket spending through 12 supplemental insurance should be allowed to do so. 13 of restricting how supplemental coverage can fill in 14 Medicare's cost sharing, the tax would charge the insurer 15 for at least some of the added costs imposed on Medicare of 16 having such comprehensive coverage. Instead 17 As a result, in today's presentation, we show an 18 illustrative example of a simple 20 percent tax on Medigap 19 and employer-sponsored retiree plans. 20 annual premiums, this 20 percent tax translates into $420 21 per year on Medigap plans and $200 per year on retiree 22 plans. Assuming the average 11 1 2 3 There are two main effects of a tax on supplemental policies: First, the tax would provide revenues to help 4 recoup some of the additional Medicare spending associated 5 with supplemental coverage. 6 Second, as the insurers pass the tax along by 7 raising premiums, the tax may provide incentives for 8 beneficiaries to switch or drop supplemental insurance. 9 For modeling changes in the take-up of 10 supplemental insurance in response to higher premiums, we 11 consulted the Actuarial Research Corporation. 12 Unfortunately, there is very little data on this question. 13 The conventional assumption seems to be that the response to 14 a premium increase among those who have purchased Medigap 15 policies would be minimal, at least in the short term. 16 17 The Commission may wish to consider several policy design questions with respect to the tax: 18 What should be the appropriate tax rate? 19 Should the tax be imposed only on plans above a 20 21 22 certain threshold of generosity? Should the tax apply only to newly purchased plans rather than all supplemental plans? 12 1 As mentioned in the previous slide, we made a very 2 simplistic assumption that a 20 percent tax would mean that 3 beneficiaries' expenses would be $420 higher for those with 4 Medigap and $200 higher per year for those with retiree 5 benefits, even before we consider the changes in their cost- 6 sharing liability. 7 in this chart, where we see a noticeably bigger change 8 compared to the previous slide, which was Slide 8). You can see the effect of this reflected 9 Focusing on the top part of each bar, we see that 10 the majority of beneficiaries would see their total out-of- 11 pocket spending go up by $250 or more -- 70 percent under 12 the beneficiary-neutral package and 75 percent under the 13 program-neutral package. 14 beneficiaries keep their supplemental coverage and pay the 15 tax on their supplemental insurance. For this chart, we assumed that 16 This slide summarizes the relative change in 17 annual Medicare program spending under the four policy 18 combinations we presented today -- the two benefit packages 19 with and without a 20 percent tax. 20 For example, if we hold beneficiary cost-sharing 21 liability roughly equal to current law, program spending 22 would increase by about 1 percent. That's because of the 13 1 catastrophic protection for high-cost beneficiaries. 2 tax revenues would offset the increase by 1.5 percent -- 3 that's the second row -- leaving a net budgetary effect at a 4 half percent in savings. 5 But In contrast, under the program-neutral package, 6 Medicare spending would be about the same, but 7 beneficiaries' cost sharing would be higher. 8 percent in revenue offsets, the net budgetary effect under 9 this option would be about 1.5 percent in savings. 10 We want to reiterate several caveats and With the 1.5 11 limitations of our modeling because it represents an 12 imperfect approximation of the policies outlined. 13 First, our results provide a one-year snapshot of 14 relative changes. 15 budget scores, which will have to take into account 16 additional factors. 17 budgetary effect of alternative policy options, which are 18 useful to keep in mind as you consider and weigh different 19 aspects of the benefit design. 20 We want to emphasize that these are not But the results do show the relative Also, the scope of our modeling excludes dually 21 eligible beneficiaries because we assumed that Medicaid 22 would fill in any changes under the alternative benefit 14 1 package and would keep the cost sharing the same for those 2 beneficiaries. 3 As we discussed in November, our results are 4 sensitive to the behavioral assumptions underlying the 5 model. 6 simplifying assumptions on supplemental coverage. In addition, our model contains some important 7 Finally, we want to point out that throughout our 8 analysis, our numbers do not capture the value of insurance 9 that risk-averse people get when they insure against 10 undesirable outcomes. 11 important for many beneficiaries. 12 This value of insurance is real and To ensure that a new benefit design is flexible 13 enough to respond to changes in medical evidence or delivery 14 system reform, you may want to identify elements of the 15 benefit design that are fixed and those that can vary. 16 For instance, you may want to create incentives to 17 encourage the use of high-value services and discourage the 18 use of low-value services. 19 give the Secretary the authority to reduce cost-sharing for 20 services that medical evidence has identified as high value. 21 22 One way to do this would be to For your discussion, we present an illustrative example of possible recommendations. IN this example we 15 1 tried to capture the main threads of the discussions to 2 date. 3 First, direct the Secretary to develop a new fee- 4 for-service benefit design with: 5 combined deductible for Part A and Part B services, co- 6 payments that may differentiate by type of service and 7 provider, such as primary care versus specialist visits. 8 Second, authorize the Secretary to reduce cost 9 out-of-pocket maximum, sharing on high-value services where there's evidence. 10 The next bullet point is more of a direction for 11 implementing the first two. 12 combined two constraints in putting the benefit package 13 together, keeping the beneficiary liability neutral versus 14 keeping the program cost neutral. 15 number you wish to work toward in designing the new benefit? 16 17 18 19 In today's presentation, we What's the ultimate Third, establish an excise tax on supplemental coverage. Finally, that concludes our presentation, and we look forward to your discussion. 20 MR. HACKBARTH: 21 So today let's do our customary two rounds, round 22 Okay. one being clarifying questions. Thank you, Julie. Karen, do you want to lead 16 1 round one? 2 DR. BORMAN: Julie, if you could go to Slide 6, I 3 just want to make sure I'm correct. 4 visit we're talking about here would be like emergency 5 department or hospital outpatient department? 6 we're capturing there? 7 DR. LEE: 8 DR. BORMAN: 9 Is that what That's correct. I just want to be sure I'm interpreting that correctly. 10 The outpatient per MR. GRADISON: Great. Thank you. Under the current circumstances 11 without a cap, a catastrophic benefit, do we have any idea 12 what actually happens with regard to those rather large 13 sums, that is, whether they're written off or whether 14 they're paid or who they affect in terms of socioeconomic 15 factors? 16 DR. LEE: So with respect to socioeconomic 17 factors, people who have a higher income tend to have 18 supplemental coverage, so they would get catastrophic 19 protection in that way. 20 very high level of liability, we know that it is pretty 21 stable in that it's about 5 percent of the beneficiaries, 22 but as to exactly who they are, that kind of detail we do In terms of who is reaching that 17 1 not -- or the kind of detail that one would wish, we do not 2 have that. 3 It's very general. There would be some part of that beneficiary 4 responsibility that would not be paid, but actually I do not 5 know at this point to what extent that would be. 6 MR. GRADISON: Thank you. 7 DR. MARK MILLER: Let me just draw a distinction. 8 We have a model built specifically for the purposes of doing 9 some of this estimation. I think your question could be 10 approached differently using a different data set, and we've 11 done some work a few years back where we were looking at 12 distributions of beneficiaries and who was in the extreme 13 tail of the distribution. 14 could revisit and bring back to you, but I think Julie is 15 sort of speaking more to, I think, what's in this model 16 where it is not highly detailed to answer the question that 17 you're asking. 18 19 20 DR. LEE: Again, that's something that we Certainly on the socioeconomic variables, we really do not have that much information. DR. MARK MILLER: Right. But we could approach 21 that question through a different data set. So, Bill, we'll 22 try and dig that -- and we've actually done it a few years 18 1 back, and we can dig some of that back up and bring it back 2 to you. 3 MR. GRADISON: I'd appreciate that. I still have 4 wounds that haven't healed since the Medicare catastrophic 5 dispute of a few days ago. 6 DR. CASTELLANOS: Just on Slide 13, you mentioned 7 creating appropriate incentives. I totally agree. We need 8 to discourage low-value services and encourage high-value 9 services. I see the problem being to be able to at this 10 point in our discussions to clearly identify high- and low- 11 value services. 12 Maybe Mike will help us. 13 MR. HACKBARTH: 14 we get to the round two discussion. 15 questions? 16 I don't know how you plan to do that. DR. STUART: Okay. Thank you. Ron, why don't we get to that when Bruce, clarifying Yeah, I don't have the slides, but it 17 was one of the earlier ones showing the characteristics of 18 the new benefits -- not the effect but the -- yeah, right. 19 I'm curious about outpatient visits at $100 a pop 20 because that seems to be directly contrary to what we talked 21 about yesterday in terms of equalizing payments across site 22 of service. So I think we open ourselves up to criticism by 19 1 having such a recommendation. 2 illustrative, it doesn't strike me as being a particularly 3 good illustration here. 4 MR. HACKBARTH: Even if it's only That's a good thing for us to keep 5 our eye on. 6 yesterday was about equalizing payment for E&M services, and 7 outpatient departments are providing a wide range of other 8 services. 9 I would note, though, that our recommendation But good thought. MR. GEORGE MILLER: Yes, thank you. On Slide 11, 10 as I understand the presentation, if we chose to make this 11 effective only on new purchase plans, would that change the 12 budgetary implications of this slide going forward? 13 DR. LEE: Can you repeat the question 14 MR. GEORGE MILLER: Yeah. I believe you said in 15 the presentation we could choose the option of only making 16 this applicable to new purchase plans going forward. 17 we chose that option, if I understood that correctly, then 18 would that change the value of the slide? 19 20 DR. LEE: So if Yes, that will change the revenue offsets from the tax. That will decrease significantly. 21 MR. GEORGE MILLER: 22 MR. HACKBARTH: Thank you. Mitra, any clarifying questions? 20 1 MS. BEHROOZI: Yeah, I think a couple related to 2 what Bill was asking, and I think we've asked them before, 3 and I forget. 4 Of those 5 percent of beneficiaries who reach the 5 catastrophic level, is it the same approximately 10 percent 6 who don't have supplemental coverage? 7 percent of all beneficiaries who are exposed, who are, in 8 fact, exposed to the high out-of-pocket? 9 DR. LEE: Or is it the 5 So in very broad terms, people who just 10 have Medicare only, they tend to be younger and healthier. 11 So in terms of the probability of getting really sick, it 12 tends to be a little lower. 13 reflected. So that's kind of what's 14 Now, I actually do not remember -- we did look at 15 this cut of the data before, but I actually cannot remember 16 what that was. 17 direction. 18 But it was not actually noticeable in that MS. BEHROOZI: So of the 10 percent who don't have 19 supplemental coverage of some kind, it's probably around the 20 same 10 percent of that group who reaches the -- who has no 21 -- I'm sorry, 5 percent of that group who reaches the out- 22 of-pocket level? I'm trying -- you know what I'm trying to 21 1 get at? 2 population who, in fact, is exposed to the catastrophic 3 cost. 4 The percentage total of the overall Medicare Not to minimize it. DR. LEE: I just want to get an idea. So I think you are referring to the 5 actually cost-sharing liability, not necessarily their out- 6 of-pocket spending, right? 7 coverage would have -- they could have a very high 8 liability, but, in fact, be paid very small. 9 MS. BEHROOZI: 10 DR. LEE: Okay. 11 MS. BEHROOZI: The people with supplemental That's not who I'm talking about. So you said the liability -I'm talking about the actual expose 12 -- the actual liabilities -- 13 DR. LEE: So the liability, that reflects the 14 overall spending. 15 a relatively smaller portion of that group that will be 16 reaching that liability. 17 18 So if you are younger and healthier, it's MS. BEHROOZI: So it's less than 10 percent of the 10 percent? 19 DR. LEE: Yeah. 20 MS. BEHROOZI: Okay. Thank you. Now just leaving 21 that alone then, the 10 percent of people who don't have 22 supplemental coverage, how has that changed over time? 22 1 That's what I understand to be sort of the present snapshot, 2 but like ten years ago, what was the ratio of people who 3 didn't have supplemental coverage to those who did? 4 DR. LEE: So this is one of the things that -- 5 it's from year to year in the chart book. 6 where -- Dan? Dan would know. 7 DR. ZABINSKI: 8 MR. HACKBARTH: 9 10 [off microphone] What's that? The question is whether the share of Medicare beneficiaries with some form of supplemental coverage has been relatively constant over time. 11 12 I don't know DR. ZABINSKI: [off microphone] Yes, it has. Around 10. 13 MR. HACKBARTH: 14 MS. BEHROOZI: Yeah. Thank you. And one more question. 15 Sorry. 16 in premiums, so this assumes that the premiums will go up 17 because the plans will be covering the excise tax? 18 how you get to -- 19 20 21 22 On Slide 10, the changes in out-of-pocket spending DR. LEE: That's Yes, and we assume that the entire tax will be passed on to the beneficiary. MS. BEHROOZI: Right, and it didn't assume any change in pricing behavior or whatever by the insurance 23 1 company? 2 DR. LEE: So we did look at it -- so under 3 beneficiary-neutral package, we did not assume any changes 4 in the premiums because the overall cost-sharing liability 5 under the new package was roughly equal to current law. 6 Under program-neutral package where cost-sharing 7 liability is higher compared to current law, we do believe 8 that both the Medigap and retiree plans, there will be 9 change in the premium. For Medigap we did look at how much 10 of -- you know, where the premium increased, that that will 11 shift the distribution of this chart a little bit higher. 12 What is presented here, we did not assume premium changes 13 for this particular chart, but we did look at for Medigap. 14 DR. NAYLOR: Just to make sure I understand, when 15 you define the characteristics of the program on Slide 3, 16 you talk about reducing exposure to unexpected out-of-pocket 17 spending. 18 understanding is one way in which this has been modeled on 19 Slide 11 is expected increases in Medicare spending for 20 catastrophic under the beneficiary. 21 don't understand is how these models are mindful of effects 22 on people in poor health. I don't want you flipping through slides. My But the one that I 24 1 DR. LEE: To the extent that they are the ones who 2 will incur the higher cost-sharing liability, they would be 3 more likely to benefit from the catastrophic cap. 4 Now, to the extent any special provisions 5 addressing that third point, our current -- what we 6 presented today did not make that distinction. 7 DR. NAYLOR: I guess one way to do it is thinking 8 about if cost sharing gets to lower-value services, and 9 value means higher quality at reduced cost, that might be. 10 But I wasn't clear in how -- 11 DR. LEE: So all the discussions today with 12 specific benefit packages, we are mindful of the effects on 13 low-income beneficiaries, but what we have modeled has not 14 distinguished those characteristics. 15 DR. BERENSON: I'm following up on Mitra's 16 questions about the interaction between the benefit changes 17 and then what happens with supplemental insurance, so go to 18 Slide 10. 19 having higher cost-sharing obligations and maybe 6 percent 20 being significantly benefitted, it doesn't look like a good 21 tradeoff from that point of view. 22 want to understand a little more what the response of the Just looking at this, with 75 percent of people But what happens -- I 25 1 Medigap plans would be. 2 catastrophic expenses. They don't have to reserve for 3 catastrophic expenses. I assume they reallocate that into 4 covering most cost sharing. 5 They no longer have to pay for the So isn't it close to a wash? In fact, if people 6 still were buying Medigap insurance, they wouldn't actually 7 be paying this $310 on average. 8 roughly the same premiums and having the third party picking 9 up their cost sharing. 10 DR. LEE: They would be paying Is that not right? So that was the rationale we used for 11 beneficiary-neutral package because under that package we 12 rejiggered the cost sharing so that by putting on extra 13 benefit with the out-of-pocket maximum, there were other, 14 you know, increases in cost sharing. 15 value of that package was -- in terms of cost sharing, that 16 was roughly equal to current law. So the kind of average 17 Now, if we look at program-neutral package where 18 overall cost sharing has increased, so even though Medigap 19 supplemental plans might not have to now pay the 20 catastrophic level, there have been enough increases in cost 21 sharing at the lower end of it to increase the overall cost 22 sharing. So we would think then Medigap premium will -- the 26 1 cost of that will increase. 2 3 DR. BERENSON: What order of magnitude of an increase? 4 DR. LEE: So the average cost-sharing liability 5 for program-neutral package was about $130 higher a year 6 compared to current law, using 2009 data, and that was -- so 7 if we assume the average admin load of 25 percent, that 8 translates into about $160 per year in Medicare -- Medigap 9 premiums. 10 So that was kind of a very simplistic back-of- the-envelope calculation we did. 11 DR. BERENSON: 12 DR. HALL: That helps. Thank you. Julie, that was a very clear 13 presentation you gave. 14 bit more to say in round two, but you just mentioned very 15 briefly one sentence, that it's very difficult to predict 16 what sort of behavioral changes would occur in one model 17 versus another, particularly in terms of out-of-pocket 18 expenses. 19 the scope of this Commission? 20 know, I'll wait until round two to come back to that. 21 22 I have a little Difficult, but is it impossible? DR. LEE: opinion. I appreciated it. Or is it beyond If the answer is we don't We have been asking experts for their It's a lack of data that I think has been the 27 1 challenge. 2 but they always qualify it with, "Well, this is just my 3 personal opinion." 4 So they are happy to express what they think, MR. HACKBARTH: So, Bill, are you focused on 5 patient response to cost sharing at the point of service or 6 how beneficiaries might respond to an excise tax? 7 DR. HALL: Yeah, how beneficiaries would respond 8 to what they see as various incentives and disincentives to 9 be insured or not be insured; and, conversely, how willing 10 would recipients be to accept lower costs of one sort or 11 another based on behavioral changes -- the sort of thing 12 that's being done in industry. 13 14 15 16 17 18 MR. HACKBARTH: But is it their decision to purchase insurance or their decision to get medical care? DR. HALL: No, no, I think it's to buy insurance like Medigap. MR. HACKBARTH: Okay. I'm just trying to understand what part you're focused on. 19 DR. HALL: Medigap would be a good example. 20 DR. MARK MILLER: And I want you to know that 21 there have been -- I think the sense that one way to put 22 your question is I have now this clearer schedule of co- 28 1 payment, and as a patient I might then say, "Well, look at 2 the value of the insurance and this increase and the 3 premium, and do I want to continue to purchase it?" 4 And so we've had a number of focused on 5 conversations with people out in the field, and they really 6 -- we have really had a hard time getting a good estimate or 7 sense of when you hit this price point, people will start to 8 change their behavior. 9 saying, that it's going to be very hard for us to say this And that's what I think Julie is 10 is it, now the behavior will change. 11 anticipate a lot more precision there, although you seem 12 to... 13 DR. LEE: And I wouldn't So where people seem to be more 14 comfortable is if you already have Medigap insurance and 15 whether the incremental increase in premium is, you know, 16 $20 a month or $40 a month, it's much more difficult for 17 them to drop that Medigap insurance. 18 Now, people did point out that if you are actually 19 coming to that decision whether to buy Medigap or not as you 20 age into Medicare, so you currently do not have it but it's 21 that initial decision, then the higher price might have a 22 bigger effect. So there are various factors that can change 29 1 that decision, but for the current holder of Medigap 2 policies, there is that inertia. 3 DR. HALL: Thank you. 4 DR. CHERNEW: What I was going to say is most of 5 the time people assume that there's loosely going to be some 6 continuous response, so it's not like no one responds for a 7 nickel and no one responds for a time, and then all of a 8 sudden it's 15 cents and everybody responds. 9 magnitudes are hard, although it could be investigated in a Getting the 10 whole number of ways. 11 of course, is you have to decide what employers are doing. 12 So a lot of people are also getting their supplemental stuff 13 subsidized by employers, and there's a bunch of other 14 reasons why there's changes going on in the employer market. 15 One of the challenges regarding this, My general sense is over time you would also see 16 people shifting to less generous -- they're still buying 17 supplemental coverage, but, remember, in the individual 18 market there's the letters -- you know, C and F are the 19 biggest ones of the supplemental, and so they'll switch to 20 lower premiums in that regard. 21 is although people wouldn't want to give you a number, I 22 would be astounded if in the absence of evidence the right But I guess what I would say 30 1 assumption is no behavioral change to a price change. 2 then the question is what level, you know, a conservative or 3 a less conservative assumption or some version of that. 4 so far there has been incredibly stable participation in 5 supplemental insurance coverage, and I would be surprised if 6 that stayed. 7 people have moved to Medicare Advantage is because this gets 8 more -- the other place people run when the premiums go up 9 is they run to Medicare Advantage, which has a whole 10 11 And But In fact, I think a lot of the reason why separate set of benefit design issues. MR. HACKBARTH: It seems to me that this is a 12 really important question because if you were to assume that 13 the excise tax had no effect on purchasing behavior and 14 people just continued to fill in all of the cost sharing, 15 then basically what we've done through the restructuring of 16 the benefit package is nullified largely because people 17 aren't seeing those incentives, and it's just a question of 18 how good is the excise tax as a new revenue source for the 19 Medicare program. 20 the Part B premium. 21 constant amount, whereas an excise tax would vary according 22 to health care costs in different markets. And, you know, you compare it to, say, The Part B premium is a national And some people 31 1 might think that's a good thing or a bad thing. 2 premium has some income-related elements. 3 a straight excise tax would not. 4 different ways of raising revenue for the Medicare program. 5 The Part B Excise tax, just And so you'd evaluate the If, on the other hand, you think that the excise 6 tax will affect purchasing behavior of supplemental coverage 7 and, therefore, the reformed benefit package has a different 8 effect, then you've got a whole different set of questions. 9 So this is really a central issue in this whole 10 effort, and, unfortunately, there's not a lot of evidence, 11 the experts are telling us, to make an assessment. 12 DR. MARK MILLER: I agree with all of that, and 13 the other way to think about what's happening here, I think, 14 if you guys could track this carefully, is we're sort of 15 assuming at this level of taxation you don't get a lot of 16 movement, which is also a conservative assumption about the 17 effect. 18 not counting a lot of additional effect beyond that. 19 think -- this is correct, right? 20 looking for. 21 is, you know, we didn't hear a lot of information from the 22 environment to be more bold in saying, okay, I'm going to I mean, you're basically counting the revenue, but And I One nod, that's all I'm And I think what we feel a little strapped on 32 1 assume this utilization effect because, you know, both on a 2 hired basis and just making a phone call basis, we didn't 3 get a lot of people saying you're going to get a big effect 4 here. 5 DR. CHERNEW: There are estimates of the 6 elasticity of demand for insurance from the literature. 7 question then becomes this is a different population and 8 it's subsidized insurance and you could sort it out. 9 the assumption of zero as a default assumption for lack of 10 any information strikes me as not -- "we don't really know 11 so we're going to assume no response" doesn't strike me as 12 the right assumption. 13 The But I would start with an assumption of the literature 14 says in this other population you get this, and now we're 15 going to try and figure out, we're going to shade that a 16 little bit one way or another for this population. 17 18 19 DR. MARK MILLER: response. And I don't think it's no I think we have like very small -DR. LEE: That's correct. 20 percent dropping at 20 percent tax. 21 MR. ARMSTRONG: There was about 3 22 Just a question on this point. the analysis isn't just about the sensitivity to different So 33 1 pricing structure for a Medigap policy. 2 relative value of the basic benefit, too, and we're changing 3 both. 4 benefit to make it more valuable, that too will have an 5 influence over people's choices. 6 It's also about the And so if we're changing the value of the basic So, I mean, I think even more strongly to Mike's 7 point, people are going to move. 8 you know, how do you predict how much. 9 MR. HACKBARTH: The question is really, Let me play back what I thought I 10 heard, Scott. The underlying structure of the Medicare 11 benefit package will logically affect the propensity to buy 12 supplemental coverage, but it is, again, difficult to know 13 exactly how much. 14 package is simpler and more readily understood by 15 beneficiaries and maybe reduces some reservations they have 16 about co-insurance when they -- you know, it's going to be 17 20 percent of some bill that they have no idea what it might 18 be, and that creates uncertainty. 19 provide some greater confidence and simplicity and 20 understandability in the benefit. 21 provide reassurance. 22 purchasing decision on supplemental coverage is going to be But to the extent that the benefit Fixed co-payments may The catastrophic may And so the effect of a tax on the 34 1 also influenced by what the underlying benefit package is. 2 You've got multiple variables at play here that really 3 complicate the analysis, I think. 4 5 So let's get back to our round one clarifying questions. 6 DR. DEAN: How much has the premium on 7 supplemental policies varied over, say, the last few years? 8 Has it followed what private insurance does, or has it been 9 more stable? 10 DR. HARRISON: Generally it follows the cost of 11 the Medicare benefit because it's a wrap-around and that's 12 what the insurance companies would be liable for, is the 13 cost sharing on Medicare. 14 spending. 15 16 17 DR. DEAN: So it tends to track Medicare So has that gone up in the same way that individual insurance has gone up? DR. HARRISON: It has certainly gone up as 18 Medicare has gone up. 19 data is from 2009. 20 changes since 2009 because Medicare spending has been 21 growing pretty closely. 22 DR. DEAN: Now, the last couple years -- this I don't think there have been big But it has gone up some. And I take it that the rises have not 35 1 affected the take-up of the insurance; in other words, the 2 proportion of people that are buying the policies has stayed 3 pretty steady. 4 DR. HARRISON: 5 DR. DEAN: 6 DR. CHERNEW: 7 DR. HARRISON: 8 DR. DEAN: 9 DR. HARRISON: 10 DR. CHERNEW: No, I think Medigap has fallen off. It has fallen off? And they moved to MA. That's right. Oh, okay. Right. To some extent, Medicare Advantage 11 payments haven't necessarily risen. For example, the 12 actuaries at CMS have predicted a drop in Medicare Advantage 13 enrollment because the generosity of payment for Medicare 14 Advantage plans has been less, to my understanding. 15 see Medicare Advantage plans are rising, and part of the 16 reason is the Medicare Advantage market is connected to this 17 market. 18 you want to do -- the other thing that's going on is the 19 employers are changing what they're doing in ways that we 20 don't know a lot about. 21 coverage, you're deciding between getting your gaps filled 22 through a Medicare Advantage plan or through this plan -- But you So when you're trying to make choices between what But if you don't have employer 36 1 and people shift between the markets. 2 know a lot about those shifting, and so the overall amount 3 with total supplemental coverage has changed. 4 has definitely changed over the past few years as Medicare 5 Advantage enrollment has changed. 6 DR. DEAN: 7 MR. HACKBARTH: 8 9 I see. And I agree, we don't But the mix Thank you. Mike, did you have any further clarifying questions? DR. CHERNEW: No. I think I'm halfway through my 10 clarifying and a quarter of the way through [off 11 microphone]. 12 MR. BUTLER: My questions were around price 13 sensitivity, too. 14 we know anything about the profile, the socioeconomic 15 profile of the people in each of the categories now, the 10 16 percent that have no supplemental insurance, those that are 17 employer sponsored, those that are picking the typical 18 supplemental and those that are picking Medicare Advantage? 19 I'm wondering if we knew their profiles that would help us 20 understand their behaviors today. 21 DR. HARRISON: 22 set to do that. Another way to look at it is to say, Do Can we match that? We don't have the data in this data We hear things, but Medicare Advantage 37 1 enrollees tend to be sort of in the middle-income brackets. 2 They're heavier there. 3 to be -- the most well-off people tend to have employer- 4 sponsored coverage, and the Medicare-only are probably lower 5 income but not Medicaid. 6 The employer-sponsored people tend DR. MARK MILLER: And this goes back to some of 7 the exchange there again. This data set has certain 8 characteristics that allow us to do some of these things, 9 most notably bringing together the supplemental coverage for 10 the beneficiary in a place where we can do this kind of 11 modeling. 12 out looking at different data and going -- I'm thinking 13 these are the same people, but not precisely within this 14 data set. 15 These kinds of questions probably require going We don't have the demographics. MR. BUTLER: So I suspect that the healthy 25- 16 year-old that says, "I'm going to just take a catastrophic 17 and pay a low premium" is one thing. 18 a lot of those that are 65 and healthy and saying, "You know 19 what? 20 money to pay for the supplemental, I'm not going to pay for 21 it." 22 what you're saying, in that category. There probably aren't I've run my numbers, and even though I've got the I don't know that, but there probably aren't many, is 38 1 MS. UCCELLO: Okay. I'm going to ask for a 2 clarification of your clarification to Bob, who clarified 3 Mitra. 4 program-neutral case, the Medigap premiums are going to go 5 up somewhat. 6 premium increases included in these slides? On Slides 8 and 10, the assumption is that for the But I'm not sure I heard exactly -- are those 7 DR. LEE: 8 MS. UCCELLO: 9 Or do they only reflect the excise tax part? 10 11 [off microphone] The premium -- DR. LEE: part. What's here reflects only the excise We did run -- 12 MS. UCCELLO: 13 DR. LEE: 14 MS. UCCELLO: But you're including -- Yeah. Okay. I just wanted to make sure I 15 knew what I was looking at. Okay. And now I'm going to 16 open up a can of worms. 17 to set this excise tax, we're kind of thinking of -- we 18 could think about it in a couple of ways. 19 about it as we're trying to, you know, set it high so that 20 people drop this coverage and in combination with the 21 incentives in the other parts. 22 kind of accurately reflect the extra cost that these You know, when we think about where We could think Or we could set it to more 39 1 beneficiaries are incurring. And those extra costs not from 2 selection but given that -- meaning that, you know, people 3 who think they're going to use more get Medigap, but the 4 increase in cost due to the cost sharing -- due to filling 5 in the cost-sharing rates. 6 DR. CHERNEW: Increase in Medicare costs. 7 MS. UCCELLO: Increase in Medicare costs, yes. 8 I'm trying to get a sense of how does that 20 percent 9 So compare to what we think those extra costs are, and then 10 that relates back to what we think the revenue offsets are 11 going to be. 12 maintaining that coverage? 13 people drop it? 14 savings to Medicare? 15 And do those revenue offsets depend on people And how does that change if then Is that going to come up with the same DR. LEE: So even using very conservative 16 estimates of what the induced demand from having that first- 17 dollar coverage from supplemental insurance, that if you 18 translate that difference in spending and then to base that 19 to the Medigap premiums, then it becomes a very high number. 20 So it's definitely over 20 percent. 21 22 So if you kind of think of it as what's the revenue source versus if somebody drops it and then loses 40 1 that induced demand from having the first-dollar coverage, 2 so if 20 percent is lower, then the behavior, in fact, will 3 be higher than the revenues. 4 5 MS. UCCELLO: helpful. 6 Okay. So -- all right. That's Thank you. MR. HACKBARTH: Julie, what I heard you say was 7 that, in fact, if you set the excise tax at the level to try 8 to cover the cost of the induced demand, it would have to be 9 a much higher percentage than 20 percent. 10 How much higher, roughly? 11 DR. LEE: Roughly it was over 50 percent. 12 MR. HACKBARTH: 13 DR. MARK MILLER: 14 our conversations it was 60 percent. 15 DR. LEE: 16 DR. MARK MILLER: Yeah, I thought even in some of [Laughter.] 17 Okay. [off microphone] -- 50. Actually, just to spell this out 18 a little bit more -- I'm sorry, Mike -- what you're doing is 19 you're saying there's this utilization effect, and now I'm 20 going to build it into a premium that is designed to wrap 21 around. 22 really quickly. And that's why the percentages get eye-popping 41 1 And just to follow up on the second half of your 2 question, but what if they didn't have that, and then what 3 would be the effect on utilization of facing this new cost- 4 sharing structure, now obviously that we could model, but, 5 you know, that's the tradeoff you're trying to set up in 6 this conversation. 7 an excise tax you are not recovering, you know, the full 8 effect of the utilization -- the utilization effect of the 9 first-dollar coverage. 10 DR. CHERNEW: And the punch line is at this level of I'm sorry. The reason why the percentages seem 11 high is because Medicare is such a large share and the 12 Medigap portion is such a small share. And so -- 13 MR. HACKBARTH: A lot of leverage 14 DR. CHERNEW: 15 is what makes it high. 16 in response to that that I think is really important, I 17 don't view this excise tax as a tax because you're really 18 just charging people for something they're getting from the 19 -- you're sort of pricing it correctly. 20 it in some way. Right, exactly. And the leveraging But the only other thing I would say You're not taxing I don't know if that makes -- 21 DR. BAICKER: You're offsetting a subsidy. 22 DR. CHERNEW: You're right. There's this big 42 1 subsidy that's going on that these plans are getting that 2 you're kind of making them just pay their full price, which 3 is slightly different than the way you would think of a tax. 4 And the tax language has some -- might have some burden with 5 that word. 6 [Laughter.] 7 DR. CHERNEW: But if anyone cared about the 8 language, I think recognizing that what we're really doing 9 is pricing the plans correctly for the full cost they're 10 impose is a different activity than taxing. 11 DR. MARK MILLER: 12 MR. HACKBARTH: 13 14 More correctly. Yes, and that's an important point. DR. BAICKER: So I'll want to focus a little more, 15 shockingly, on the insurance value in round two, but in 16 thinking about that, we've talked before about different 17 ways of parameterizing the gain that people are getting, 18 even talking about sort of beneficiary-neutral, it looks 19 like -- it looks like they're paying more and getting 20 nothing if you just look at the numbers, whereas we know 21 they're getting this nebulous insurance value. 22 It sounded like you didn't have enough 43 1 longitudinal data to be able to say how much is your 2 individual variance reduced under this plan. 3 way to capture how much the variance is reduced either for a 4 cohort or, you know, for synthetic cohorts to say imagine 5 that the 66-year-old we observe today is actually one year 6 from of the 65-year-old we observe today, which is obviously 7 not quite true, but can you use a model like that to try to 8 capture how much variance is reduced over a span of time for 9 someone and put a number on it to balance against what looks 10 like, well, wait, beneficiaries are just paying more and not 11 getting anything else? 12 to parameterize reduction in variance? 13 DR. LEE: But is there a What other mechanisms are available So in terms of just looking at Medicare 14 utilization and spending, we can create a fairly large panel 15 for that. 16 information on their supplemental status, so that's -- and 17 the effect of a new benefit does depend on your out-of- 18 pocket spending, which is a function of the supplemental 19 coverage. 20 simulating what the effect would be, how much variance is 21 reduced under a new package. 22 The limiting factor is we do not have the So that is kind of the limiting factor for In just terms of how your overall spending varies 44 1 over time, that we can get, but I don't think that is what 2 you are getting at. 3 reduction in variance under an alternative package, right? 4 It's how much of the variance -- DR. BAICKER: So you don't have individual level - 5 - for the individuals you observe, you don't observe their 6 supplemental coverage. 7 coverage of the cohort so you could make some assumptions 8 about let's just tag people with coverage based on the 9 distribution of coverage for the whole group. 10 DR. LEE: You have a sense of the supplemental For one year we have individual 11 information, so that's why we've been just looking at a one- 12 year analysis. 13 DR. BAICKER: That also seems like it might be an 14 important input to these other conversations about how 15 Medigap coverage might change under different models 16 because, you know, Glenn was saying, oh, the certainty of a 17 co-pay versus a co-insurance, but I would think that the 18 bigger effect would be you now have an insurance product 19 that's more valuable. 20 you had a reduced variance Medicare package, that should 21 come with a different premium, which should then affect the 22 value of the Medigap plans. If you were paying for Medicare and You're getting a different 45 1 item, a less valuable item in Medigap because some of the 2 variance has been taken care of by Medicare. 3 to parameterize how much the variance is reduced seems like 4 an input into lots of different calculations to be able to 5 say we've just given you a more valuable Medicare package, 6 you're going to pay less for Medigap because -- you're going 7 to be willing only to pay less for Medigap because it's now 8 worth less because it's not picking up that variance. 9 10 MR. HACKBARTH: Kate. So being able I didn't understand all that, We need to translate that into lawyer language. 11 DR. BAICKER: 12 [Laughter.] 13 MR. HACKBARTH: But it was so lucid. But one thing that I thought I 14 heard you allude to was looking at this over time and not 15 just a year at a time in terms of evaluating the benefit of 16 the coverage that people get. 17 time, especially in a population of seniors, the likelihood 18 that people are going to incur high costs and benefit from 19 the catastrophic presumably grows as you look at the 20 analysis -- not one year at a time but five years at a time. 21 Is that -- 22 DR. BAICKER: And so if you look at it over Actually, having multiple years of 46 1 data would let you do some additional analysis, but what I 2 was actually hoping to capture doesn't necessarily hinge on 3 having multi-years of data. 4 risk are you exposed to this year. 5 MR. HACKBARTH: 6 DR. BAICKER: What I want to know is how much But only one year. And one good way to figure that out 7 is to look at paths over time and see what might happen to 8 you over a wider window. 9 across people and say, "I could be the lucky person who But another way would be to look 10 stays healthy or I could be the unlucky person who's 11 diagnosed with an expensive disease," and look at a cohort 12 of people who are representative of the many different 13 things that might happen to you. 14 have to look over time, although that's a very helpful way 15 to capture that. 16 trying to think of other ways to say the package is more 17 valuable even if your individual spending goes up by $10, if 18 your risk of exposure to $1,000 has gone down, even if that 19 isn't realized today, you still got some insurance value. 20 And I'd love to be able to put some numbers on that to 21 balance against the, well, wait, average spending went up, 22 why isn't that just a bad deal for everyone? So you don't necessarily I was being cognizant of data constraints, We don't have 47 1 the number that helps put weight on that. 2 DR. MARK MILLER: I guess the last thing I -- can 3 we go to 8, I think? So if we were able to do that -- and 4 notice the real careful choice of words, because I'm also 5 just trying to visualize -- I also now understand how you're 6 saying -- you know, I know it's an over-time phenomenon, but 7 perhaps cross-sectionally you could -- okay, I see that. 8 Not necessarily how to do it, but I see the concept. 9 you would actually have -- you think you could derive a Then 10 dollar value that you would then say put in -- because, I 11 mean, in the end, people on the Hill are going to be saying, 12 "So how does this happen?" 13 a chart like this? 14 DR. BAICKER: Would I really overlay that into That's risky because to put a dollar 15 value on the insurance value, you have to put in a bunch of 16 assumptions about how people feel about risk and how -- you 17 know, it doesn't come out of just summary statistics of the 18 data. 19 weren't layered on and, you know, looking like apples and 20 apples when they're not. 21 know, the share of people whose risk was increased versus 22 decreased or -- I mean, everybody's risk should be going But you could create some parallel exhibits that It would say something like, you 48 1 down with this, but something to help visually capture and 2 quantitatively capture, even if it's not directly overlaid 3 with this, the fact that people are gaining some protection. 4 MR. HACKBARTH: You can't do this slide just for 5 multiple years of data? 6 want the one where you're not including the effect of the 7 premiums, sort of the first one where you showed the 8 different benefit packages and how they -- 9 10 So this isn't trying -- actually, I DR. MARK MILLER: MR. HACKBARTH: That's the [off microphone]. That is this one? Okay. So you 11 can't just do this over time and say that if you look at a 12 five-year span this is how many beneficiaries would 13 experience higher out-of-pocket costs and lower out-of- 14 pocket costs? 15 DR. LEE: We can do that for their cost-sharing 16 liability, but we do not have individual supplemental 17 coverage information. 18 19 MR. HACKBARTH: the liability. 20 DR. LEE: 21 MR. HACKBARTH: 22 That's what I'm saying, just for Liability we can. Yeah, and how -- I would think that over time, if you look at it over, that it would show 49 1 that the catastrophic coverage is more valuable. 2 that be an interesting exhibit to have? 3 DR. LEE: Wouldn't So the kind of simple proxy we tried to 4 do was to look at we know that your overall cost sharing is 5 lower compared to current law under the alternative packages 6 if you have at least one hospitalization. 7 at, you know, the number of years that you had at least one 8 hospitalization over a five-year period. 9 in any given year of a hospitalization is about 20 percent, And so we looked So the probability 10 and that goes up to about 50 percent if you look at 50 11 years. 12 time this one-year snapshot might look different. So that was kind of a way to see, you know, how over 13 DR. BAICKER: I think that that would be very 14 helpful for two reasons. One, it would capture that, in 15 fact, this one year may not be a good proxy for how you 16 benefit over time, and just looking at Medicare, not with 17 Medigap, is then capturing the value of this program. 18 doesn't tell you how individual exposure is changing, but 19 it's telling you how the value of the benefit package is 20 changing. 21 in risk protection that you see because there are some 22 things that are idiosyncratic year to year and there are It If anything, that's understating the improvement 50 1 some things that are persistent. 2 you also don't know if you're going to be diagnosed with an 3 expensive chronic disease that's going to show up year after 4 year or if you're going to have an idiosyncratic 5 hospitalization that's not going to show up year after year. 6 That will smooth away one kind but not the other, but it's 7 at least getting you partway there. 8 MR. ARMSTRONG: 9 And so the persistent -- So, Julie, I'm not sure if this is for you or for Kate, but is "parameterize" really a verb? 10 [Laughter.] 11 DR. BAICKER: 12 MR. ARMSTRONG: 13 14 15 Oh, yes. I mean, Julie, you just didn't flinch. DR. BAICKER: And then "parameterization" is a noun, and keep going. 16 MR. ARMSTRONG: 17 [Laughter.] 18 MR. ARMSTRONG: The Harvard talk. My question is much simpler. 19 Actually, it goes back to some questions raised earlier. 20 The outpatient per visit out-of-pocket $100 deductible, I'm 21 familiar with this really being targeted at emergency room 22 visits, and that often it's much higher than that. This 51 1 seems to be kind of generic hospital outpatient. 2 wondering if you looked at, modeled higher costs for 3 emergency room visits and discarded that, or if that's 4 something still worth considering. 5 DR. LEE: I'm So for physician services, we went back 6 and actually got much more granular data. Now, for 7 outpatient services, we have not done that, so we just have, 8 you know, all different kinds of outpatient visits, 9 including surgeries to emergency room visits, that are just 10 in that number of outpatient visits. 11 granular information so that we can make such distinctions, 12 but we have not done it yet. 13 DR. BERENSON: We do plan to get more I'm going to jump in, Scott, 14 because we were talking a couple of days ago on this very 15 issue, and maybe you can help us, or anybody else. 16 concern that I have is that we're modeling a traditional 17 Medicare benefit package on a good HMO-type benefit package, 18 and we're going to basically go from 20 percent cost sharing 19 for physicians to a co-payment for a visit. 20 wondering if there are models, benefit designs that actually 21 do have more -- that move away from just 20 percent of 22 something you don't understand, but recognize that there are The And I'm just 52 1 physician services that cost thousands of dollars in many 2 cases. 3 So my question to you is: Is there a way to do 4 this without forgoing 20 percent of revenue and at the same 5 time not creating huge barriers to that care? 6 MR. HACKBARTH: And let me just piggyback on that, 7 Scott, and others who are familiar with current approaches 8 to structuring packages. 9 One particular question that we were asking one 10 another about is what happens today in HMOs when a patient 11 is hospitalized and using physician services, perhaps a 12 surgeon, a radiologist, anesthesiologist, various other 13 consultants. 14 it, there's no cost sharing for the patient for all of those 15 physician services. You're just paying the $750 co-pay for 16 the hospital visit. Is that the approach currently used by 17 Group Health and other plans or something -- 18 Under this benefit package, as we understand MR. ARMSTRONG: Yes, that's correct. When you are 19 hospitalized, it's an all-inclusive bill, if you will. We 20 have a lot of examples where we are really differentiating 21 the out-of-pocket costs for some of these outpatient 22 services that are not associated with an actual admission to 53 1 the hospital, and we would be happy to describe those. 2 I think most notable, the advanced imaging and the emergency 3 room services are those that tend to be low value, 4 discretionary, and we place a high out-of-pocket cost on 5 those in particular. 6 But But there could be more to that. I think -- and this would be a point I'll make in 7 round two -- part of the issue we're dealing with here is 8 that the HMO plans don't just use out-of-pocket financial 9 incentives. I mean, they have these whole care management 10 infrastructure and all sorts of other things that are really 11 complementary to this, and that is, I think, part of the 12 issue we're dealing with. 13 MR. HACKBARTH: I think that's part of the issue 14 here. We're talking about adopting an HMO-type benefit 15 structure for an open-ended fee-for-service plan that 16 doesn't have those care management factors, and so it's 17 mirroring two different types of ideas. 18 Okay, round two comments, questions? 19 MR. GRADISON: This is going to be a little stream 20 of consciousness but it won't take long. I lived through 21 this, through Medicare catastrophic, and there are a few 22 things I'm proud of from serving in public office. One is I 54 1 went down with the ship. 2 I was convinced that, on balance, it would be beneficial to 3 the participants in the program. 4 I never voted to repeal it because They didn't see it that way, and I've been trying 5 to understand why it is so hard to bring about change in the 6 Medicare program, and that's a pretty good example of what 7 was happening in that time -- resistance to change. 8 9 I would summarize my experience that the losers know they're losers, but the winners don't know they're 10 winners. 11 in insurance. 12 you're going to have catastrophic expenses until it's too 13 late to do anything about it and go out and buy the 14 insurance. 15 not talking about dollars, but most beneficiaries are 16 losers. 17 paying more than they pay today under the present system. 18 And the winners do have a value. I know that. There is a value But you really don't know that But our own numbers show that most people -- I'm We show that. They actually are going to end up And so I asked myself, well, why is this? And 19 this is a pretty broad generalization, but I've been trying 20 to figure out why are we basically stuck with a 1965 design. 21 That's the screen at which I look at it. 22 it was a prevailing design at that time; it was based And you say, well, 55 1 largely on the Federal Employees Health Benefit Plan as it 2 existed at that time. 3 today this kind of overall structure doesn't exist in 4 nature, so to speak. 5 but the world has passed it by -- not only with HMO plans 6 but with changes through POS and so forth. 7 And the world has changed. I mean, I don't mean it doesn't exist at all, So I think that as far as bringing about these 8 changes is concerned, I still am trying to grapple -- I was 9 trying this in my first round -- with learning a little bit 10 more about the potential winners and losers. 11 My hunch is that newer beneficiaries would be much 12 more comfortable with this than existing beneficiaries, and 13 you could say, well, of course, because the older ones are 14 used to this pattern and they would just resist change. 15 I think it's deeper than that. 16 have been covered under plans which don't look anything like 17 this. 18 least, towards the experience that they've had. 19 But I think newer beneficiaries And so we're moving closer, in a small sense at You can justify and say, "What are you driving 20 at?" And I'd have to say it's sort of a stream of 21 consciousness, but I'm really concerned whether we can -- I 22 like what we're doing here, but I'm not sure we can pull it 56 1 off. 2 MR. HACKBARTH: Bill, let me just pick up on your 3 point about newer beneficiaries might be more comfortable 4 with this than older beneficiaries. 5 possible approach would be to do this on an age-in basis, 6 that -- and obviously I haven't thought through 7 administrative implications of this, but say that, you know, 8 for people joining the program after some date certain, 9 there's a new benefit package and one that will look Does that imply that a 10 familiar to them based on recent experience with private 11 insurance? 12 MR. GRADISON: I think that's well worth thinking 13 about, because up until now we've talked about for everybody 14 -- which I still think makes sense. 15 MR. HACKBARTH: I hope that's clear. Again, the politics of this are 16 difficult. 17 decades, and they don't gain political traction. 18 instance they did. 19 repealed. 20 that history doesn't exist. 21 22 These ideas have been around, you know, for In one There was a strong backlash and it was And, you know, it's folly just to pretend that DR. STUART: I had the same reaction that Bill did in the sense that in a way we have gone through this before 57 1 with the Medicare Catastrophic Coverage Act, and there are 2 obviously differences between what you're proposing or -- 3 I'm not saying proposing -- what you're examining and what 4 was actually enacted in 1988. 5 very wise to go back and look at that experience and not 6 just the numbers part of the experience but the reaction by 7 those who knew that they were losing. 8 9 But I think that it would be If you go to Slide 10, the main concern I have with this is a Medicare catastrophic concern. You tell me - 10 - you guys are coming up with a proposal that is going to 11 make 85 percent of Medicare beneficiaries worse off? 12 face validity of this thing I think is going to be really 13 problematic. 14 The Having said that, I agree with Michael that this 15 outcome is highly unlikely, and I say that for a reason. 16 I'm looking at it from the standpoint not so much of those 17 with employer coverage but those who are currently buying 18 one of the Medigap plans, and I add up the extra prices on 19 this, and using a Laffer curve approach by saying, you know, 20 if the price gets to a certain point, then, you know, you 21 get no revenue and the market just completely collapses, and 22 here are the four parts. 58 1 The first part is Kate's part. If you change the 2 benefit so that you provide catastrophic coverage, you 3 improve the insurance value of the Medicare program, which 4 means that you reduce the insurance value of the Medicare 5 catastrophic coverage. 6 know the direction. 7 We may not know how much, but we Secondly, if you have a 20 percent excise whatever 8 you call it, it's still something that people have to pay, 9 and so the price jumps immeasurably. 10 Third, the Medigap policies already have extremely 11 high loading, and if you look at the demand for Medigap 12 policies, people have said, well, why do people buy this 13 insurance anyway because it's such a lousy value. 14 And then, fourth -- and this is a point that 15 Michael also raised -- is that when you raise the price of 16 this alternative, this 20 percent premium on a good that has 17 less value than it did before, then the alternatives look 18 better and the MA plans look much, much better. 19 And so if I put all four of those things together, 20 my best bet is that the Medigap market would just disappear. 21 You know, I don't know whether that's true, but I'd be 22 willing to put bucks on -- not $10,000 but a few bucks on a 59 1 bet -- 2 [Laughter.] 3 DR. STUART: -- that said if we were able to try 4 this, my guess is that people just wouldn't buy Medigap 5 policies anymore. 6 the MA plan -- and this is the other thing. 7 to be a supply response to this thing. 8 expect that the PPO part of the MA plans would really 9 expand. They'd just go right to the MA plans, and There's going I would really In fact, the insurers are the same people. They're 10 the ones that are offering the current Medigap plans, that 11 are offering the PPOs, you know, and it's not that difficult 12 to move from one product line to another. 13 So there are two parts to this: the perceptual 14 part, which I think is really problematic, and I would 15 strongly urge you to have something in the language of this 16 that recognizes that we've been through something like this 17 before, and so we don't want to make that mistake; and then 18 you have to pay attention to behavioral responses somehow. 19 And the last point I'll make is that we've 20 concentrated here on kind of the price response to value in 21 the insurance product, but I think we're on much firmer 22 ground in terms of making behavioral assumptions about what 60 1 will happen when we change the nature of the product. 2 other words, we have much better research about the 3 elasticity of demand for Medicare services than we do for 4 the responsiveness to the insurance price. 5 MR. HACKBARTH: In Bruce, you've made a number of 6 good points. I want to go back and just be the devil's 7 advocate on the very first thing you said. 8 this, and a significant, large majority of Medicare 9 beneficiaries end up paying more, and, you know, isn't that You look at 10 -- I can't remember your exact words, but isn't that sort of 11 a non-starter as an idea, and it may well be. 12 But if you step back and look at this in a bigger 13 policy context, one of the fundamental issues facing the 14 program is if, in fact, we need to lower Medicare costs or 15 slow the rate of growth so they aren't consuming as much of 16 the national resources as they have been, how do we 17 distribute that across the different participants in the 18 system? 19 can't take it all out of us because you're going to 20 adversely affect access to care, quality care, and we need 21 to figure out ways to share the responsibility of lowering 22 Medicare costs." Increasingly, you know, providers are saying, "You So the question is: To what extent, if 61 1 any, should Medicare beneficiaries share in lowering the 2 costs of the program? 3 alternatives for increasing the beneficiaries' share? 4 And this whole conversation introduces some And what are the different policy 5 potential elements in terms of how you structure the benefit 6 package or whether you right-size the premiums for 7 supplemental coverage by having beneficiaries pay more for 8 supplemental coverage, and then in turn each of those 9 possibilities has pros and cons in terms of the incidence of 10 the burden and which beneficiaries bear a higher cost. 11 I don't think we should, in other words, just stop 12 at your initial observation and say, "Oh, beneficiaries on 13 average are paying more; therefore, it's a political non- 14 starter." 15 to the Congress. 16 If we do that, I'm not sure we're doing a service DR. STUART: Actually, that wasn't my point. My 17 concern here is that if you show that slide, the 18 conversation stops and that whatever else we are trying to 19 do here just doesn't go further because people are not going 20 to -- that was my point. 21 MR. HACKBARTH: 22 DR. STUART: I understand your subsequent -- I'm certainly not against analysis. 62 1 I think the whole point of this is that we think that there 2 are significant problems with the supplementary insurance 3 that Medicare beneficiaries currently have, and what we'd 4 like to do is to move them into something else. 5 trying to get them so that they don't have anything. We'd 6 like to move them into some other kinds of products. I 7 mean, that's in part why I think it's important that we take 8 on the issue of value-based insurance design here. 9 got to say something. 10 We're not We've I mean, that's where the returns are, is getting rid of things that have low value. 11 If we get rid of everything, then, you know, 12 again, I think we're at a point where we're going to hit 13 this wall, and then the policymakers are just not going to 14 listen to us. 15 That's my fear. MR. GEORGE MILLER: First of all, this has been a 16 very helpful discussion in hearing everybody's point of 17 view. 18 educational and beneficial as we try to provide additional 19 service to the Medicare beneficiaries. This is not one of my strong suits, so this is very 20 I'm struck by a couple of things. On Slide 10, it 21 seems to me that it would be very helpful -- I'm sorry, on 22 Slide 7. It seems to me it would be very helpful as we have 63 1 had the discussion maybe to identify an ED visit separate 2 from an outpatient visit, to push that as a point of trying 3 to move folks from abusing the ED versus just the outpatient 4 line, which I think would be embedded in there, and then try 5 to move folks away from using the ED as a place, especially 6 probably low-income beneficiaries using that as a point of 7 entry to get health care benefits. 8 The other thing that I think as we listened to 9 this, if we think about just applying this to newly 10 purchased plans, that may have some traction, but I 11 certainly would like to see the analysis of what impact that 12 would have on the numbers. 13 answered earlier. 14 crazy about the term "tax" -- "tax benefit" -- I forgot the 15 name of it. I think that had the question I'm not sure who said it, but I'm not too 16 MR. HACKBARTH: "Excise tax." 17 MR. GEORGE MILLER: "Excise tax." If we could 18 come up with a better name for that, a correct -- I think 19 Michael said it, correctly pricing that benefit, it would 20 just eliminate the political negativity about the word "tax" 21 in that design. 22 MS. BEHROOZI: So I think I have two pages here. 64 1 I'll try not to take two pages' worth of time. 2 to distill it into, you know, conceptually how to organize 3 these thoughts. 4 I'm trying I'm trying to go back to the beginning, which is 5 that I feel like we're doing two different things here. 6 is providing the catastrophic protection, and the other 7 quite separate thing is trying to build a better benefit 8 design to help control excessive spending in the program. 9 And they're really quite separate things. 10 One I think thinking about the 10 percent who don't 11 have supplemental coverage I feel like puts into sort of 12 sharp relief how mushing these two things together might 13 produce some not so great results, because as you said, 14 among that 10 percent probably a smaller share of them will 15 actually get to that true out-of-pocket catastrophic level 16 than among the general population. 17 general Medicare population it's 5 percent in a year, so 18 it's less than 5 percent of that 10 percent will actually be 19 protected strictly by the Medicare program being redesigned. 20 And where will the burden fall? 21 have catastrophic coverage who are lower income, as you 22 said, and with a combined deductible in particular, they're So you said among the On those others who don't 65 1 not going to be able to access their Medicare benefits. 2 They're not going to be able to afford to get to the point 3 where their Medicare benefits will kick in. 4 combined deductible -- and I said this, you know, when we 5 were talking about this in prior years or last year, or 6 whatever -- I don't think it's a good idea. 7 people to go to the doctor when they're sick, and the lower 8 Part B deductible is designed, I think, somebody thought, 9 it's better that people get to the doctor sooner rather than I think a I think we want 10 waiting until they can somehow come up with, you know, $500 11 over the course of a year, or thinking that they can't and 12 so they never go. 13 So, you know, there's a reason that different 14 types of services have different thresholds, cost 15 thresholds, before you access them, right? 16 17 18 MR. HACKBARTH: Could you tell us a little bit about how your plan is structured in that regard? MS. BEHROOZI: I can't talk about our main plan 19 because we have no point-of-service costs because we use all 20 those other management tools that Scott talked about and we 21 have a very low cost, but it's the all-in cost. 22 We do have a couple of plans that have some point- 66 1 of-service cost sharing, but we don't have deductibles. 2 It's just co-payments, because I still don't get the point 3 of deductibles. 4 far as I can tell, and, you know, from conversations with 5 Mike, it's just a way of keeping the premium down or keeping 6 other costs down that you otherwise might shift. 7 think it's kind of the worst -- it's the least thoughtful, 8 shall we say, the least targeted way of shifting costs. 9 if we're worried about the 10 percent who don't have That's just an insurance pricing feature as But I So 10 insurance coverage -- or supplemental coverage, I think 11 we're hurting them the most by covering that smaller share 12 of them with Medicare dollars. 13 For everybody else who does have some kind of 14 supplemental coverage, it's just a question of which pocket 15 it comes out of or how it comes out of their pocket. 16 clearly, the people who have chosen Medigap plans because 17 they value the insurance protection, the risk aversion, 18 they've chosen to pay a premium not at the point of service 19 -- they don't like, you know, showing up and either not 20 knowing how much they're going to pay or having to shell out 21 dollars at that time. 22 once a month or, you know, an automatic deduction from their And, They like being able to write a check 67 1 checking account. 2 employers, they may be also contributing toward that during 3 the course of their working time by forgoing raises or 4 whatever. 5 And, yeah, those who get it through It does seem like more people have a preference, 6 to the extent that they can afford it, to kind of spreading 7 their costs via premiums, via forgone wages, and that 8 instinct I don't think is necessarily reflective of a 9 conscious choice, but it tends to align with what I think we 10 see coming out in the evidence that, you know, the whole 11 issue of whether point-of-service costs deter unneeded or 12 needed care. 13 as though the demand wouldn't be there but for the coverage. 14 Well, maybe the utilization wouldn't be there but for the 15 coverage and people would forgo needed care and then there 16 would be -- they would suffer consequences. 17 suffer poorer health status. 18 later on, and I've cited the research before, but, you know, 19 there's slowly more research coming out demonstrating a cost 20 offset in terms of acute costs later because of forgone 21 care, drug coverage and physician coverage costing more at 22 the front end ends up in some cases -- well, it was the You know, we keep talking about induced demand They would Or there would be higher costs 68 1 Medicare population, Medicare Advantage population in 2 California I think that was studied that showed a hospital 3 offset, you know, longitudinally, three years later or 4 something like that, the hospital costs presented a higher 5 cost to the plan overall. 6 So whether people know that when they're 7 purchasing that coverage or they just have an aversion to 8 risk or whatever, I think we should pay attention to that 9 signal; and if we are, like I said, worried about that 10 10 percent who don't have coverage -- and I am -- I really 11 don't think this is the right thing to do to them. 12 that putting so much cost up front in the deductible is the 13 wrong thing to do. 14 I think As far as taxing plans, frankly, when we were 15 talking about initially, I didn't think of it only as 16 applying to the new benefit design. 17 applying it to the current benefit design. 18 suggest there because people do have the choice to go to MA, 19 they should also be able to make a choice about their 20 Medigap plans, and they can choose to pay more for more 21 coverage or choose to pay less for lower coverage. 22 the way it is now. I thought of it But I would just That's I think, you know, as Mike says, it's 69 1 right, load more of the actual cost into those plans. But I 2 think we've got the example in PPACA of the so-called 3 Cadillac plan tax going into effect in 2018. 4 apply to all plans; otherwise, it kind of doesn't make sense 5 because you're trying to, you know, recover the revenue, 6 give people time to choose whether they'll stay with their 7 current plan that's going to be, whatever, 30 percent more 8 or 40 percent more, or go to another plan that might be 10 9 percent more because it offers less generous coverage -- I It has to 10 think it should track the generosity of the plan -- or move 11 to MA. 12 Maybe they'll squeeze some more out of their load, as Bruce 13 says, to be able to compete to keep the business and not 14 pass along all of that additional cost, whatever you want to 15 call it -- the excise thingy -- to their beneficiaries. And maybe you'll see insurance behavior changing. 16 I think that's it. 17 MR. HACKBARTH: Those are my two pages. Let me just nail down one thing, 18 Mitra. What I hear you saying is you're not sold on the 19 combined deductible, even, you know, if we stipulate -- 20 MS. BEHROOZI: 21 MR. HACKBARTH: 22 [off microphone] Not sold. If we stipulate that, you know, there's going to be significant cost sharing because there 70 1 is an existing Medicare benefit package, you are 2 particularly worried about the effect of a combined 3 deductible on access to physician services and would rather 4 see separate deductibles or no deductibles at all, although 5 I suspect that that would imply very, very high co-pays, 6 huge co-pays. 7 8 And so -- DR. MARK MILLER: You would be pushing the catastrophic cap way out, the co-payments up. 9 MS. BEHROOZI: Exactly, and it's not that I want 10 anybody to suffer the higher catastrophic spending, but it 11 is a very small number of people, and, you know, I think the 12 question -- I'm sorry. 13 happens -- I think Bill asked what happens to those people. 14 I doubt that it's like in the general population where more 15 than 50 percent of personal bankruptcies are driven by 16 medical costs. 17 have that high catastrophic costs among that share of people 18 who are lower income or whatever. 19 unreimbursable and gets reimbursed by DHS probably. 20 you know. 21 22 Somebody on this side asked what I don't think that's what happens when you MR. ARMSTRONG: this. I bet that ends up in the I mean, Glenn, just one brief comment on So we have plans now that have deductibles, but we 71 1 exclude from the deductible preventative visits or a whole 2 series of visits around 12 chronic illnesses where we want 3 those patients to have no cost out-of-pocket in order to get 4 in to see those providers. 5 to have both a deductible but to not create the 6 disincentives, Mitra, that you're talking about. 7 just may be something for us to look into. 8 9 MS. BEHROOZI: And so there is a way, actually, So that Yeah, I think that's kind of the problem with a fee-for-service, you know, unmanaged package 10 where -- and then to load more into the deductible without 11 being able to do those things, I mean, you know, unless 12 Medicare wants to administer that. 13 That would be really good. 14 15 16 DR. NAYLOR: That would be cool. Nothing new to add to this rich conversation. DR. BERENSON: I don't have much to add. This has 17 been very interesting. I'm still formulating my own views 18 on a lot of this, but I'm encouraged that we're doing the 19 right thing here and that we actually can make a case for 20 the kind of package that we're putting together. 21 in a couple of areas we need more work, like trying to 22 figure out what this sort of hybrid is between a tightly I do think 72 1 managed HMO and just sort of a traditional indemnity-based 2 20 percent across-the-board, whether we can do something 3 more creative on the physician side. 4 in some notions of value-based benefits like with 5 identifying services that are excluded from the deductible, 6 I think we should take advantage of. 7 Wherever we can build I guess what I want to talk about a little bit is 8 are we going to just get into trouble by proposing something 9 that's smart but is -- I guess you said had no face 10 validity, I think, Bruce, was your -- or was sort of dead on 11 arrival, or whatever. 12 answer on that. 13 don't think we would be proposing this as a stand-alone, 14 let's redo catastrophic from '88 and just say let's adopt 15 this plan. 16 years is going to be subject to intense review and scrutiny 17 and that there will be proposals and at some point some 18 agreement on some restructuring. 19 way towards a premium support model or puts expenditure 20 targets in the program or whatever it might be, in the midst 21 of that kind of a restructuring, this isn't the only -- this 22 could be sort of presented also as part of it. And I don't certainly have a good My sense is that -- well, not my sense. I I'm assuming that Medicare over the next few Whether that goes all the In other 73 1 words, this is not the stand-alone, we're going to 2 redistribute towards people who have catastrophic illnesses 3 from low-income people who now may have difficulty finding - 4 - I mean, that's not what we want to do. 5 I think what we want to do is have a series of 6 options available, having analyzed them, that it just makes 7 sense if there's a program restructuring that we do take on 8 the benefit structure, which is not a very good one right 9 now. So I'm really thinking aloud here. 10 On a number of these issues, I don't think there's 11 a correct answer. I can argue on the deductible. There's 12 pros and cons of going to the combined deductible and not 13 going to the combined deductible. 14 beneficiary-neutral versus program-neutral as a policy 15 decision. 16 surcharge, not an excise tax -- maybe that's a more neutral 17 term -- versus regulating first-dollar coverage as a design 18 issue. We've identified We've talked about what I guess I'll call a 19 I'd just throw out the possibility that rather 20 than sort of having a defined package that we vote up or 21 down on, that we actually do a very good job of laying out 22 design options, the advantages and disadvantages, and I 74 1 think present a pretty clear picture of where policy would 2 come out, but don't presume to judge the politics of it. 3 So I just throw that out as a possibility that we 4 might want to talk through, but I guess the major thing I 5 want to say is I don't think we want to be positioning 6 ourselves to endorse a specific package and say we think the 7 Congress should legislate this tomorrow. 8 9 10 11 DR. MARK MILLER: that. I agree and I want to build on So if I could get you just to flip to the last slide, there's a couple of points that you could imagine. I think Bob is absolutely correct that we wouldn't 12 be coming along in a report with a recommendation that says 13 here's the benefit package. 14 We're cranking through this for a lot of analytical reasons: 15 to make sure people understand impacts, how levers work, all 16 the rest of that. We're not doing that. Okay? 17 Notice this slide where you could imagine a set of 18 statements that the Commission says -- and, Mitra, obviously 19 there's at least one in here that you don't agree with, but, 20 you know, we think there should be an out-of-pocket maximum, 21 we think there should be designs and benefit that work like 22 this, co-payments instead of co-insurance. I'm deftly 75 1 moving past the flashpoint with Mitra. 2 You know, the notion of trying to focus on high- 3 value services and at least giving the flexibility to deal 4 with those as evidence arises, you could imagine statements 5 like that. 6 chapter, but you're not saying it is $101 for this service. 7 Bob has even, I think, a point off of that, which 8 says say here are the issues, design issues, pros and cons, 9 and the way you could think about them. And then you have illustrations through the And so I want to 10 just take this opportunity to focus you on a final product 11 and make sure that you understand. 12 is it," you know, because there's detail underneath this 13 that just really needs to be thought out past that. We wouldn't say, "This 14 I'm hoping that was building on your point. 15 DR. BERENSON: 16 DR. HALL: Absolutely building. Absolutely. So what I've taken away from the 17 discussion so far is that life is hard, and I already knew 18 that, but that the implications of a lot of these decisions 19 about price point and where you actually interject the price 20 point, we don't know that much about it, and particularly we 21 don't know much about how it influences human behavior, 22 maybe a little bit more about the impact on health. 76 1 But it seems to me it's so important because this 2 business of price points isn't the same thing as to whether 3 I buy a Kindle or an iPad 2. 4 of future Medicare expenses if people make wrong choices, 5 and it also has implications in terms of quality of life if 6 we want to throw that into it. 7 It has implications in terms So the reason I ask the question of is there 8 anything known about the psychology of choice that we can 9 even interject into this that makes any sense, or at least 10 any proposals we've put forward, say that at least somebody 11 ought to take a look at this, so I have a little challenge 12 here for Mike and for Kate. 13 but I've been sort of dabbling in some of this business of 14 "Slow and Fast Thinking" by Daniel Kahneman and a few of his 15 experiments in how people view various medical procedures. 16 Isn't there still -- let me give it one more try I'm way out of my league here, 17 on this. Isn't there still some way we could interject this 18 kind of thinking into how we judge the relative value of the 19 different options we put forward? 20 MR. GRADISON: Bill, one possible way would be to 21 look at the choices actually made in the Medigap market 22 among the defined plans. The statute has defined what can 77 1 be offered, and I'm not sure where this would lead us, but 2 there's real information there somewhere about how people 3 deal with the choices that are available. 4 DR. CHERNEW: Two quick things about that. One of 5 them is prior to, of course, standardization, the view was 6 the market was pretty dysfunctional, and you had to 7 standardize it to support the choices. 8 and maybe Kate can -- where the more cutting-edge research 9 of people I know would be going would be to try and quantify And I would say -- 10 the mistakes that people made for a whole bunch of reasons. 11 So I think you used to have a paradigm that if you let the 12 market work, people make a bunch of choices. 13 telling you something about their preferences, which is 14 really great if you're getting through econ grad school. 15 They're It turns out that, I think, increasingly people 16 are thinking, looking at a set of choices, realizing how 17 sensitive those choices might be to seemingly minor choice 18 setting modifications and asking, boy, we need to think more 19 strongly about how we set this up, and simple things that 20 would have been the standard answer, we'll just give them 21 information, aren't nearly sufficient to try and modify 22 those, although what I would say in response to Bill's 78 1 comment is that's exactly right. I don't think -- and, 2 again, Kate can comment. 3 anywhere near close enough to figuring out exactly how you 4 would do that, because one thing that is true is you have to 5 have a good set of values for what you think other people 6 should do before you start fiddling around with some of 7 these things. 8 might have, but it is a challenge, particularly in this 9 population, as challenging as any population, to see that I don't think the science is And that's a harder lift than one otherwise 10 they're making the right choices. 11 know what "right" means, but choices you think are 12 reasonable in the Medigap market. 13 There's a lot of misinformation. 14 problems. 15 And I'm not even sure I There's a lot of inertia. There's a lot of other Kate can... DR. BAICKER: Just super briefly, I don't want to 16 get us off track, but my reading of the evidence on this is 17 that the strongest empirical evidence comes from 18 prescription drug use in part because the setting is 19 amenable to people playing with the choice architecture and 20 the nudges that we think might influence behavior day in and 21 day out as opposed to major procedures where there isn't 22 that same opportunity, and that there are a set of findings 79 1 that help us understand when people are likely to 2 underutilize or underadhere and when they're likely to 3 overutilitize and that price is one factor, but all these 4 other choices -- all these other choice architecture, nudge, 5 contextual factors are really important in those and that 6 there's less information on how those factors affect 7 insurance choice, but where we see the most evidence is from 8 take-up of public programs like Medicaid where it's hard to 9 write down a rational model that explains people not taking 10 up very low cost benefits that are available to them through 11 public insurance programs, and that we can learn some things 12 from there about how people make choices, but it's a 13 different population and we don't have that same kind of 14 data for a Medicare population, in part because in a good 15 outcome most people who are eligible for Medicare are taking 16 it up, and then you're left with these choices that are very 17 constrained in the Medigap market or in the employer wrap- 18 around coverage market that don't give us as strong evidence 19 as we have from some of those other contexts. 20 DR. DEAN: What do we know about the group of 21 people that do not buy supplemental coverage? Do we know -- 22 I mean, I assume we know -- is it because -- is it a group 80 1 that's basically healthy and doesn't think they need it? 2 is it a group that feels they can't afford it? 3 I assume, some of both? 4 some of these questions. 5 DR. LEE: Or Or is it, as I wonder how much that helps with So both of those factors are relevant. 6 If you are young and healthier so you perceive less need for 7 supplemental insurance, then you would not get it. 8 even for those who feel they need it, but then affordability 9 can play. 10 11 But then So exactly, you know, the relevance of those two factors we don't know. DR. DEAN: I would certainly think that we should 12 do everything we can to move toward these value-based 13 structures, because one of the things that I found 14 frustrating over the years is people -- to oversimplify, "If 15 Medicare pays for it, I'll do it; if Medicare doesn't pay 16 for it, I'm not going to do it." 17 I've recommended that I really thought were valuable to an 18 individual and for individuals that I knew full well had the 19 resources to cover it, even so I've had a number of 20 situations, like I said, which I found very frustrating, 21 where people decided they weren't going to do it just 22 because there's this sort of mind-set, like I said, "If Even for services that 81 1 Medicare pays for it, I'll do it; if Medicare doesn't pay 2 for it, don't bother." 3 And so the more that we can move to the kind of 4 structures that Scott just mentioned, I think it helps to 5 get past, you know, what I consider to be pretty irrational 6 decisions. 7 the beneficiary. 8 9 But it benefits, I think, both the program and DR. CHERNEW: I think part of the challenge here is that there are two main goals and they don't always fit 10 well together. The first one I take as our goal is to 11 essentially save money or reduce revenue -- increasing 12 revenue or save money by reducing the distortion associated 13 with supplemental coverage in the supplemental coverage 14 market, sort of this idea of right-pricing, and inherently 15 that's going to shift the burden to consumers one way or 16 another. 17 made, Glenn, that, of course, the alternative is some of the 18 other stuff that we really dislike, and so at least in 19 situations where you think there's a distortion, ignoring 20 the politics, I think we certainly have to think about how 21 to do that. 22 think is the first thing, and then the tension in that And I am 100 percent on board with the comment you And you could do that without -- so that I 82 1 general array seemed to have been do we prohibit or do we 2 tax. 3 the word "taxing," so I didn't -- by "taxing," I meant 4 something else. 5 appealing than "prohibiting" because there are people that 6 might want something. 7 scenario is what you said, Bruce, all right, no one is going 8 to buy. 9 prohibited it one way or another. 10 I tend to find taxing more appealing, and I don't like But I tend to find "charging" more It strikes me as the extreme case So that just gets us to where we would have DR. STUART: No, actually, I think the point that 11 I wanted to make is that there would be a market response to 12 that and that this would drive the market into MA, and it 13 would be the least managed of the MA because those are the 14 easiest for the insurers to develop. 15 that's a place that we want to go. 16 DR. CHERNEW: So we might agree that But, I mean, there's odd things that 17 one has to think through because, of course, the MAs get 18 paid based on the fee-for-service cost or the extent that 19 people are left in the fee-for-service cost but now they're 20 not covered as much in the fee-for-service -- so there's 21 some complicated dynamics as to how that works, but all of 22 that plays through some notion of we want to get the pricing 83 1 right to avoid this distortion that I think we all agree 2 exists in the supplemental coverage market, and we want to 3 do that basically for financial reasons because we have pain 4 that has to be shared, we have to fix the SGR, there's a lot 5 of payment that's going on to physicians, for providers that 6 potentially could be problematic over time under current 7 law, and we need to think about that. 8 what you said, Glenn, about that. 9 And so I agree with The second thing that we're trying to do which I 10 view as a reasonably separate thing is improve the value of 11 the package the way it's designed in one way or another to 12 get more value, and that includes a range of goals. 13 them, of course, is we want to improve the insurance 14 component of it with the cap. 15 simplify it in various ways, which I think also matters. 16 And the most important part, I believe, is essentially we 17 want to encourage good behavior change, and by "good," I 18 mean drive people to use the services we want them to get 19 and get them not to use the services we don't want them to 20 get. 21 22 One of The second part is we want to Ron pointed out that that's a lot easier to say than do, although I think there are a lot of areas where at 84 1 least at a minimum we could identify what those things are, 2 and I think there will be more if we get half of what we 3 hope to get out of some ongoing research areas. 4 think a little more emphasis on those -- we've had more 5 emphasis in other versions of this type of chapter. 6 Oftentimes these ideas go under the rubric of value-based 7 insurance design, but they're kind of different. 8 them of course, is what Scott said, which is the -- or 9 different versions of the same basic philosophy: And I do One of the 10 carveout within the deductible. 11 Mitra is. 12 people they're just a tax on being sick, and for healthy 13 people they often discourage them from getting the services 14 that you want. 15 there's a lot of waste going on in certain types of 16 services. 17 would want to think about carveouts the way that I think 18 Scott would. 19 So I actually am where I think deductibles aren't useful. For sick But they could be useful if you think But if you were going to impose them there, you We've had, I think, very useful discussions of 20 least costly alternative type programs or other more nuanced 21 benefit designs, and I think there is the potential for 22 doing that. 85 1 The one thing that I don't understand -- so this 2 is a round one question. I don't know exactly the 3 restrictions that managed care plans are under if they 4 wanted to charge you for a low-value service one way or 5 another. 6 there's a certain amount that the fee-for-service system 7 would pay for a given service, are you allowed to charge 8 patients more for that, Scott? Is the rule that if Medicare has a certain -- 9 So say there was one of these proton -- and Ron 10 would probably say this is high value, but at least in my 11 circles, everyone uses this proton beam, whatever these 12 things are. They're big machines. 13 DR. STUART: 14 Expensive machines. DR. CHERNEW: Big, expensive machines, and what I 15 understand from the evidence, they're considered not that 16 great value. 17 that even if Medicare covers the cost? 18 19 Are you allowed to charge people a lot for MR. ARMSTRONG: So you're asking me as a provider or as a health plan? 20 DR. CHERNEW: 21 MR. ARMSTRONG: 22 DR. CHERNEW: As a health plan. So I can just not cover it. Even if Medicare covers it? 86 1 DR. BAICKER: Don't you get flowback then? 2 DR. CHERNEW: And are you allowed to cover it but 3 just at a higher copay, or do you actually have to not cover 4 it at all? So if you wanted to -- 5 MR. ARMSTRONG: 6 DR. CHERNEW: 7 Well --- do a least costly alternative version, for example, could you do that? 8 MR. ARMSTRONG: 9 DR. CHERNEW: I think so -- 10 MR. ARMSTRONG: 11 DR. CHERNEW: Anyway, without going through the -- 12 I don't know --- the clarifying questions to the non-presenter -- 13 [Laughter.] 14 DR. CHERNEW: -- knowing what their limits are, 15 knowing what flexibility the MA plans have to do these 16 better designs and making sure there aren't barriers to them 17 doing them, I think, is important. 18 DR. HARRISON: Yes. Plans cannot do -- cannot 19 make coverage rules that are discriminatory against sick 20 people or things like that. 21 propose different fee schedules. 22 put some copays on home health that are not currently in So when they bid, they can They can, for instance, 87 1 Medicare, but they have to be careful that, like their total 2 SNF copays are no more than the total SNF copays under fee- 3 for-service. 4 but the total has to be the same. 5 actuarial equivalents rules, but -- 6 MR. HACKBARTH: So they could change the structure of them, And then there's some [Off microphone.] So the rules 7 issued within the last several years about the structure of 8 cost sharing on, for example, very expensive oncology drugs, 9 and my recollection is that at a point in time, plans were 10 really charging very high -- some plans were charging very 11 high copays on oncology drugs and Congress moved to restrict 12 that, or maybe it was CMS by rural moved to restrict that. 13 DR. HARRISON: Yes, so Part B drugs, 20 percent in 14 fee-for-service. 15 than 20 percent and that -- I believe CMS said they couldn't 16 do that anymore, but there was also some general legislation 17 that talked about nondiscriminatory -- 18 19 There were plans that were charging more MR. HACKBARTH: So there is flexibility, but it is also constrained -- 20 DR. CHERNEW: 21 MR. HACKBARTH: 22 DR. BAICKER: Flexibility, subject to review. Yes. But I do think that's a different 88 1 question than what Mike asked. 2 impose a copayment on all oncology treatments, you're 3 discriminating against people with cancer. 4 discriminating against sick people. 5 costly alternative. 6 cancer treatment, but you'll pay for it only at the rate of 7 regular old radiation therapy and not proton beam therapy. 8 That, I believe the Medicare Advantage plans can do. 9 10 I mean, you know, if you You're Mike asks about least It's not that you won't cover prostate DR. CHERNEW: And so my broader comment is think – DR. MARK MILLER: This is what's kind of been 11 bothering me about this conversation, too. 12 example, the managed care plan could have a higher copayment 13 on the proton beam whatever we're talking about here. 14 15 DR. HARRISON: So in that They would have had to have written it into their original bids, but -- 16 DR. MARK MILLER: 17 DR. HARRISON: 18 CMS would have reacted to that -- 19 [Inaudible.] -- service, I'm not quite sure how DR. MARK MILLER: Because it's a matter of 20 actuarial equivalency, and then how far out of line it 21 appears in review that you might be discriminating against 22 some set of patients. 89 1 2 DR. HARRISON: it through coverage rather than through copay differences. 3 4 My guess is that they would handle MR. HACKBARTH: Actually, it may even be neither. It may just be handled through clinical decision making. 5 DR. CHERNEW: 6 MR. HACKBARTH: 7 likely outcome. 8 things. 9 Right. That would be, to me, the most it. 10 They're not changing copays for different They're just saying, our clinicians don't recommend DR. CHERNEW: So I think the broader point I would 11 make is separating out the first goal of this supplemental 12 coverage distortion from the second goal of figuring out how 13 to design the benefit package in order to encourage 14 beneficiaries to use those things we want them to use and 15 not use the things that we don't want them to use, I think, 16 is a worthwhile discussion and one that we have done in 17 other chapters. 18 VBID workshop stuff and there was the least costly 19 alternative. 20 in much the spirit of what Bob said, thinking through those 21 and how they might work in the Medicare traditional type 22 program and how they could work outside of that in terms of There was the whole -- Joan ran the whole There was a whole set of options. And I think 90 1 flexibility, I think, is an important exercise for the 2 chapter to get into. 3 The last thing I would say is all of this, the 4 behavioral assumptions are going to be crucial, so it 5 depends a lot on how much help you could get from various 6 people would depend a lot on what your time line is for 7 doing some of this. 8 advice on assumptions that at least -- you're still going to 9 get the same -- Julie's same comment of, well, this is our But I think you could get reasonable 10 opinion, but I think there can be opinions guided in at 11 least where the literature is, you know, and we can help 12 with that. 13 DR. CASTELLANOS: Can I comment just for a second, 14 because we're all looking at 30,000 feet down to the ground 15 and where Karen is, where I am, and where Tom is and where 16 Bill is, we're on the ground. 17 road. 18 practitioner some ability to have appropriate guidelines and 19 you can protect that person from liability if he or she 20 decides based on good medical evidence, a lot of these 21 issues can be dealt with on the ground level. 22 looking at the 30,000-feet level, but I think we need to -- We're where the tire hits the And we all have to work together. If you give the Now, we're 91 1 in our discussions, as we've done over the last four or five 2 or six years, we really need to kind of put this all 3 together in that respect, too. 4 MR. BUTLER: So I struggle still a little bit with 5 what problem are we trying to solve, and I think in the end 6 it's engaging the beneficiary in part of the reform of the 7 program in a way that we haven't. 8 comments I would make, but it does get discouraging when we 9 kind of spend 90 percent of our effort in this Commission on Glenn made some of the 10 the provider side and often not just looking at ways to 11 influence their behavior and often accusing them of being 12 mischievous or illegal or evil, you know, we almost end up 13 there in some aspects of this when we're trying to kind of 14 do the right thing for an awful lot of committed people that 15 are delivering care. 16 So when I step back and look at this issue, we're 17 also trying to dive right into some of the -- as a 18 Commission, we also like to be as specific as we can in 19 recommendations because just floating concepts doesn't do a 20 lot of good. 21 into the copays and the deductibles like, okay, that's our 22 comfort zone. So I'm struggling here, because we dive right We like to talk about all those things, yet 92 1 we can't really model them. 2 terms of what we pass along here is I think we need to think 3 about. 4 And so what's the sweet spot in And I wouldn't underestimate the importance of 5 just educating at the front end and kind of resetting the 6 dial and really saying this is about engaging a player in 7 this, the beneficiary in a very positive way. 8 to say we understand that the traditional Medicare has been 9 wildly popular as a government program and that supplemental And we need 10 insurance has provided incredible security for so many 11 people. 12 And we understand that. But we also do need to say very explicitly, 13 there's a price to that, and as Glenn said, one option is 14 just to charge everybody the true costs of that. 15 that, you can keep the same system you have now, the 16 unfettered kind of access and so forth, but there is a price 17 to that and we really need to be clear about that as we 18 introduce the chapter because there's no free lunch in that. 19 And I don't think that even the legislators really If you do 20 understand that very basic, you know, this front end, front 21 dollar coverage, because we wouldn't end up with a health 22 reform package that covers the uninsured now with, guess 93 1 what, the same first dollar coverage in many of these newly 2 insured under Medicare. 3 repeat some of the same ways of not engaging the consumer. 4 And so we're kind of going to Similarly, it's been two years, I think, since 5 we've looked at shared decision making, for example, in our 6 Commission, and yet it's, you know, it's such an important 7 concept in the way we're going to help reform the package. 8 9 So I caution us from getting too -- the right balance is tricky. I like Bob's idea of give them some 10 options and some samples, but make sure that we're not so 11 rigid and use this formula, because they'll lift that 12 formula as opposed to missing this opportunity to kind of, 13 hopefully kind of provide education at a little bit higher 14 level, because there should be more pages in our work on 15 this kind of issue than we have taken on in the past. 16 MR. HACKBARTH: So, Peter, I hear some similarity 17 between your comments and Bob's comments. So the track that 18 we were on was framing this issue as a benefit redesign 19 issue and working towards broad principles, not a specific 20 benefit package, as a recommendation. 21 hear both Bob and Peter saying is even that may be too tight 22 a focus and the framing maybe should be more engaging of But what I think I 94 1 beneficiaries in making the system more efficient and 2 effective and what are the policy options, try to frame some 3 of the policy options for doing that. 4 far as to link it even to the discussion about premium 5 support as another type of mechanism for trying to engage 6 beneficiaries. 7 tight knit on just benefit design. 8 you correctly? 9 You know, Bob went so So it's sort of a brad frame as opposed to DR. BERENSON: Am I hearing the two of I mean, I think Peter's notion 10 needs to be included. 11 that be the frame, which is that this is about consumer or 12 patient engagement, because there are benefit design issues 13 here which -- so, I mean, I think we could figure out how to 14 do both. 15 I don't know whether I would have But in terms of stepping back a little bit and 16 laying out a context for all of this, I think that -- 17 exactly what those words would be, I'm not sure, but I think 18 we wouldn't want to lose the context by jumping right into 19 here's the principles. 20 is it that we're trying to achieve here, and I think -- 21 MR. BUTLER: 22 DR. BERENSON: We want to spend some time on what I think that's my main point -Yes. 95 1 MR. BUTLER: -- the context in a way that, whether 2 it's the beneficiary, the legislator, or whatever, it kind 3 of sets a bigger context. 4 design. 5 concepts. 6 We can get into the specific Otherwise, we won't be given them much other than So -DR. DEAN: Just yesterday in the Times there was a 7 story about -- which really relates to what Peter just said 8 -- about the Fairview system in Minneapolis was using an 9 instrument to measure what I think they called patient 10 activation, and they had surveyed something like 25,000 of 11 their patients and then categorized them into those folks 12 that really understood and were involved in decision making 13 and those that weren't, and then were using that to focus 14 some of their advice and counseling and -- I don't know. 15 didn't get into all of the details. 16 invested a substantial amount of money in trying to figure 17 out which of their patients really were using the system 18 properly and had -- they didn't have all the -- they weren't 19 clear through the assessment of this, but their sense was 20 that those that were, quote, "activated," were -- they could 21 sort of make the decisions on their own. 22 also a fairly large group that didn't meet those criteria I But at least they had And then there was 96 1 that really probably would benefit from some assistance and 2 some help in making the decisions. 3 that was a good investment, so -- 4 MS. UCCELLO: At least in their view, I think we've had a lot of really 5 great comments so far, and I really like the way that Bob 6 talked about making sure that we're fairly comprehensive in 7 the chapter on discussing various approaches, even if we 8 come to a consensus of something else. 9 I want to suggest making sure we include a couple of things. And so to that end, 10 In terms of coordinating changes in the fee-for- 11 service design with supplemental coverage, we are strongly 12 supporting this surcharge or whatever we want to call it. 13 But I think we just want to make sure that we discuss in the 14 chapter the regulatory approach and its potential 15 implications. 16 be useful to include looking at adding the out-of-pocket cap 17 on a true out-of-pocket approach, that being one of the 18 options that's discussed. 19 And as part of that section, I think it would Also, with respect to adding the out-of-pocket 20 cap, the way we're looking at this is looking at the 21 neutrality in terms of the beneficiary or the program. 22 what we're really trying to do, we're adding this out-of- But 97 1 pocket cap and figuring out how to pay for it and we're 2 paying for it really just through the other cost sharing 3 provisions. 4 that out-of-pocket cap, including adding a premium, an extra 5 premium. 6 And there are actually other ways to pay for That may not be something we want to pursue, but I 7 think it's something worth talking about, and it somewhat 8 gets at what Mitra was talking about a little bit. 9 kind of somewhat philosophical of do we want to be spreading 10 costs over everyone or do we want to be spreading costs more 11 on those who are actually spenders, the ones who are getting 12 the care, because you're going to get kind of different 13 effects and different people are going to be targeted 14 depending on which approach is chosen. 15 It's A couple other things that are a little more in 16 the weeds but I think we need to think about, and I'm 17 comfortable with using a combined deductible, but what would 18 that mean for those who are only in Part A? 19 continue under the current cost sharing rules? 20 of things would happen there? 21 22 Would they What kinds Also, I think we need to think about whether there are any impacts on the trust funds. Are we actually 98 1 shifting costs from Part B to Part A, or -- is that the way? 2 Yes. 3 about what -- because that has implications for the funding 4 of the program generally. 5 Yes, between them. I think we need to kind of think In terms of when we're looking at levying a 6 surcharge on these supplemental plans, the issue of whether 7 we levy it on only the new people coming in or everyone, 8 it's a slightly different case than the Cadillac tax under 9 ACA, I think. With the Cadillac tax, it's on employer plans 10 that are really looking at -- they're doing their pricing on 11 a year-by-year basis. 12 have a little bit more prefunding. 13 be a little higher in the early years and their premium 14 increases somewhat lower than they would be normally just to 15 reflect the increases in cost over age. 16 -- if we were thinking about doing something on a regulatory 17 approach, I think it would be more difficult, I think, to 18 implement changes on everybody because of that. 19 would be more appropriate to make changes just on the new 20 folks. 21 22 The Medigap plans have -- they can So their premiums could So that prefunding And it In the surcharge case, it may be less of an issue, but I think we still need to think through whether or not 99 1 there are implications for some of that prefunding and 2 whether people who then drop out, maybe already paid for 3 something they're not continuing on -- I'm not sure it's as 4 big of a deal, like I said, but I think it's something we 5 need to think about. 6 And I don't know if this is helpful, but I was 7 thinking of, when Bruce was speaking, we're looking at these 8 tables on 8 and 10, looking at what the effects on 9 beneficiaries are going to be given our assumptions of their 10 switching behavior. 11 going to happen to them given most of them are just going to 12 stay, it sounds like. 13 choices that would save them money -- if they dropped their 14 plan and save more money, then I'm less worried about an 15 increase that they would incur if they're staying with what 16 they have. 17 prototypical person basis and laying out their choices and 18 looking at, well, if they stay with what they have or what 19 their other choices are and if we see what their potential 20 gains and losses are as opposed to just looking at their 21 gains and losses assuming they stay or change. 22 Yes. But the issue is, well, this is what's But if they actually have other So I'm wondering if looking at things on a So I think that's my list. Does that -- 100 1 Oh, one more thing. 2 [Laughter.] 3 MS. UCCELLO: One more. Bob mentioned this a little, and I 4 think I may have brought it up before, but maybe one of 5 these -- if we're thinking very comprehensively, we also 6 need to include whether or not there should be a public 7 Medigap type of plan, and that, again, feeds into -- that's 8 on that premium support continuum. 9 DR. BAICKER: So I think there are two separate 10 issues conceptually that interact with each other, so I 11 understand why we may want to bring them together in a 12 chapter like this, but I'd at least like to lay out very 13 separately the issue of good insurance value and value-based 14 insurance design, even though they have a couple of words in 15 common. 16 And by the first, I mean we're inherently offering 17 an inferior insurance product. It doesn't have an out-of- 18 pocket stop loss. 19 large part is to get the protection that you would think a 20 good insurance policy would provide against catastrophic 21 expenses. 22 these policies that have these spillover effects back to the The reason people buy Medigap policies in And so a consequence of that is that people buy 101 1 main Medicare program that aren't right priced in the 2 Medigap premium and we could fix all of that or improve it 3 by offering a better insurance product with better back-end 4 protection that would then make these alterative policies 5 less attractive, and if they were right priced, that would 6 also make them less attractive, but they would be less 7 important because people would have better protection 8 through the main product. 9 And that seems like just an improvement to the 10 program that could be done in their revenue-neutral kind of 11 way or not, but that would improve the insurance value and, 12 in essence, shift something that's being provided through 13 this inefficient Medigap market into being provided through 14 the main benefit and payments could shift accordingly. 15 that could be done in a revenue-neutral way. 16 So I'd love to make that point more strongly, that 17 the basic benefit is not providing as much insurance value 18 as it could, and we talked in round one in the past about 19 ways to capture that in a more vivid way to juxtapose 20 against what looks like a figure where 80 percent of people 21 are losing out in the transaction, and I would argue 22 strongly that they're not. Even if some of their costs are 102 1 going up this year, they're still gaining valuable 2 protection. 3 policies shows that they value that protection. 4 And the fact that they were buying Medigap So if there's a way to make that point more 5 strongly, even in the language throughout, that I don't want 6 to call it neutral. 7 neutral when they have better risk protection. 8 I know what you mean by that, I think that kind of language 9 minimizes the insurance protection that we think that people I don't want to call it beneficiary- 10 should be getting from the basic benefit. 11 Even though I'd really like to make more strongly. 12 So that point, And then there's the second point of value-based 13 insurance design where we know that a more innovative 14 insurance product would charge less for services that were 15 really of high value, would encourage high-value use, 16 discourage low-value use, and all of that could be 17 modernized in a way that we're seeing some private insurers 18 experimenting more with and that we have some evidence. 19 That's a bit of a separate issue in that that's not about 20 insurance protection. 21 within a given risk profile, or within a given insurance 22 package. That's about driving high-value use 103 1 Now, I see that they kind of go together in that 2 if you're going to provide this catastrophic coverage 3 protection and you want to make it revenue neutral, then 4 you're going to increase payments in some ways, and if 5 you're going to start messing with copays, you might as well 6 do it in a value promoting kind of way. 7 the end product might be a blend of the two, but I'd almost 8 like to start with the concepts being more separate and then 9 think about what the program's goals are in terms of the So I can see how 10 insurance product that we're delivering and then how to put 11 those resources to the best possible use. 12 MR. HACKBARTH: So, Kate, just to pick up on that, 13 so the first part about what constitutes a good insurance 14 product, that's a more static idea, whereas the people, to 15 use your point, said it ought to cover catastrophic costs 16 and protect people against that. 17 basic insurance design. 18 That's an element of good The value-based piece of it is a more dynamic 19 process that is changing based on available evidence about 20 different things. 21 terms of who the decision maker is in the policy world. 22 can imagine Congress writing a piece of legislation that So I always think of these things in You 104 1 redesigns the basic structure of the benefit package. 2 find it difficult to imagine Congress repeatedly changing 3 that to incorporate value-based principles, and that's 4 something that it seems to me would be more appropriately 5 delegated to the Secretary within boundaries established by 6 the Congress. 7 DR. BAICKER: Yes. I At first when you said 8 "dynamic," I thought you were going more towards the effect 9 on future costs and all of that -- 10 MR. HACKBARTH: 11 DR. BAICKER: 12 Yes. Yes. No. No. -- and abstracting from all of that. In some ways, the way this recommendation example -- 13 MR. HACKBARTH: 14 DR. BAICKER: Right. -- is laid out is consistent with 15 that in that we're sure that insurance protects you against 16 potentially very high expenses. 17 need more evidence on. 18 MR. HACKBARTH: 19 DR. BAICKER: This isn't something we That's right. This isn't something -- we can all 20 agree that better protection against very high expenses is 21 higher insurance value and that we could design a package 22 that improves that. 105 1 DR. CHERNEW: [Off microphone.] 2 DR. BAICKER: Oh, you. 3 [Laughter.] 4 DR. BAICKER: -- expenses -- But the challenge, then, is in 5 something where we're getting all -- improves evidence over 6 time or changing sands and you don't want to try to write 7 down, these services should have this copay, these services 8 should have that copay. 9 promote this through more innovative structures or more And there, you could say we want to 10 flexibility in beneficiary copays over those services and 11 that could be a separate piece of flexibility, not a 12 prescription. 13 MR. HACKBARTH: Yes. Scott. 14 MR. ARMSTRONG: Yes. At this point in this 15 conversation, which I think has been really terrific, most 16 of the points I would want to make have been made. 17 would, I think, reiterate a couple of things. 18 I just First, despite the fact that Bill's wounds are 19 still healing, I'm really glad that we're pushing this and I 20 think it's an important agenda for us. 21 that's been articulated, I thought, very well, of balancing 22 our ability to influence the design of payments to providers This whole idea 106 1 with the incentives that are in the benefits structure for 2 the beneficiaries themselves, I think, is just an expanse -- 3 an area of work for MedPAC that I really look forward to us 4 spending some real time in. 5 I also -- I thought Kate did a great job of 6 articulating a point I would have made a little more simply 7 and with simpler verbs -- 8 [Laughter.] 9 MR. ARMSTRONG: I think the Medigap industry is a 10 symptom of a deficient benefit and that I'm really glad that 11 we're diving into some of those issues. 12 One other point I would make is that I do think 13 what we're talking about, in fact, is consistent with and 14 potentially has more alignment with many of the other 15 principles behind other policy proposals that we've been 16 putting forward, you know, the whole idea of bundled 17 payments for post-acute services or ACOs and other things. 18 I look forward to an opportunity for us to think about, 19 well, how could you actually begin to marry benefit 20 structure with provider payment structure and really create 21 even more influence over behaviors by doing some of those 22 kinds of things. I think until we get into this, we just 107 1 won't even be able to start asking, let alone answering, 2 some of those questions. 3 And I think that's all I'll say. 4 MR. HACKBARTH: 5 Anything that you want to ask about, Mark, to get clarification for next steps? 6 DR. MARK MILLER: No. I think what we're going to 7 have to do is huddle and sort of outline how we would walk 8 across the range of issues that were put on the table here. 9 And then once we have -- and there are some additional 10 analyses that we have got to bring. 11 And then I think I've got to talk to you and Bob 12 about this, you know, how principles versus -- and then how 13 to come back to the Commission on that continuum is sort of 14 that. 15 And I'm not being articulate, obviously. One thing that Cori said on that front in 16 following up is whether the Commission just ends up saying 17 there are six major design principles and here's the pros 18 and cons and here's how you could go, or whether the 19 Commission sort of says, and on this one, we lean this way, 20 or we more than lean, we stand. 21 bring together both your thought and are we going to lean -- 22 so that's kind of what I was thinking, even though I can't That might be a way to 108 1 say it. 2 MR. HACKBARTH: Yes. And I'll be equally 3 inarticulate about it, but that was sort of my sense, as 4 well. 5 I have a different sense of this than I did when it started, 6 and part of it is that we need to broaden the frame a little 7 bit and have a run-up that frames the issues. 8 also conclude with some statements, directional statements 9 about here's how we might be inclined to resolve some of This has been a very helpful conversation for me and I hope we can 10 those issues. 11 And highlighting just, you know, there are, as Bob said, 12 pros and cons on many of these different things and trying 13 to educate before we get to, oh, there ought to be a 14 redesigned benefit package that looks like this, I think is 15 important. 16 But I think the broader framing is important. DR. BORMAN: Just briefly, Glenn, and a piece of 17 this issue that I think maybe we're not -- we haven't talked 18 about yet or I haven't heard yet and that ultimately needs 19 to be brought in. 20 reflects sort of the bad benefit -- or the limitations of 21 the benefit that we provide, but it also is a measure of our 22 inability to communicate to people at a time when they're The market as we have it now not only 109 1 better able to make rational decisions about their choices 2 and how to think about them. 3 So whatever package or options or whatever we 4 recommend going forward, we may need to give some thought 5 about do we need a more structured educational program, if 6 you will, because right now, as you watch the ads, for 7 example, on TV, you can see how very quickly people get 8 sucked into sort of the fear factor or other ways that they 9 use to make a decision about something that would -- the 10 kind of education about this, for example, needs to start 11 way before you've picked up your Medicare card, and we need 12 to think about how do we better share choosing options. 13 realize that comes very close to a perilous boundary between 14 what's public and what's private obligation here, but I 15 think that whole education part and communication to help 16 the beneficiary make smart choices is something that we 17 should carry forward wherever we go in terms of the 18 specifics of the issue. 19 20 21 22 MR. HACKBARTH: [Off microphone.] Okay. I Thank you, Julie and Scott. So our next session is on the mandated report on rural issues. 110 1 [Pause.] 2 DR. AKAMIGBO: Good morning. Our aim today is to 3 provide a recap of our findings on the four areas MedPAC was 4 mandated to study, and clarify our guiding principles for 5 each area. 6 Today, Jeff and I will summarize those findings, 7 discuss payment adequacy as it pertains to rural providers, 8 and we look forward to your guidance on the plan as we 9 complete the rural report this spring. 10 So for a quick overview, this report was managed 11 by the Patient Protection and Affordable Care Act and 12 requires us to examine four specific issues. 13 access to health care services, which we discussed last 14 February. 15 October of 2011. 16 which was separately discussed in detail in each sector back 17 in December but we’ll summarize again today to incorporate 18 access and quality when Jeff talks about the adequacy of 19 Medicare payments later in the presentation. 20 was payment adjustments to rural payment rates, which we 21 discussed last September. 22 The first is Second is quality of care, which we presented in The third is adequacy of rural payments, The rural report is due June 2012. The last issue 111 1 We have found, as others have, that there are 2 fewer physicians per capita in rural areas. 3 is in specialists. 4 more acute. 5 serious challenge for many rural communities. 6 The difference Non-primary care specialists are even And recruitment of physicians continues to be a However, despite the differences in the number of 7 physicians, we showed back in February that access to care 8 is relatively equal in both rural and urban areas as 9 measured by volume of hospital services, outpatient visits, 10 and utilization of skilled nursing, home health, dialysis 11 and pharmacy services. 12 Now in some cases, rural beneficiaries may have to 13 drive further distances to access care, but travel times 14 were relatively similar for urban and rural beneficiaries. 15 In addition, our analyses of different surveys confirm that 16 Medicare beneficiary satisfaction with their access in rural 17 and urban areas were relatively equal. 18 With Commissioners’ guidance over the past year, 19 we have developed guiding principles to examine rural health 20 care for Medicare beneficiaries. 21 to care posits that rural beneficiaries should have 22 equitable access to health care services. The principle for access Equity and access 112 1 can be measured by volume of services, number of visits, 2 prescriptions, as well as beneficiaries’ reports of their 3 experience once they interact with the health care system. 4 And when we discuss equity and access, we 5 recognize that some rural beneficiaries may drive longer 6 distances, although not necessary travel for longer periods 7 of time, than their urban counterparts. 8 9 The quality findings we presented in October, which inform these principles, are summarized on the slide. 10 Overall, we found that quality of care in rural and urban 11 areas is similar as measured in each setting, namely skilled 12 nursing facilities, home health agencies, and dialysis 13 facilities. 14 On the other hand, hospital quality across rural 15 and urban areas is mixed. 16 equal but mortality rates are worse in rural areas and were 17 only partially explained by volume. 18 measures, as reported on Hospital Compare, were also 19 generally worse for rural providers and tended to worsen as 20 providers became more rural. 21 22 Readmission rates are roughly Clinical quality Now on to guiding principles for rural quality of care. First, the quality of non-emergency care delivered in 113 1 rural areas should be equal to that of urban areas. 2 reflects the reality that non-emergency care, where there is 3 a choice of whether to treat the patients locally or 4 transport them to a larger urban facility, should be held to 5 the same standards as the urban facility. 6 facility should be as good as the alternative site of care. 7 This The small rural However, emergency are is different. There may be 8 no alternative and small rural hospitals are obligated to 9 treat those patients. In these emergency situations, our 10 expectation for outcomes in small rural hospitals may not be 11 as high as they are for larger facilities. 12 expectations, therefore, should reflect the inherent 13 limitations that may exist in small rural hospitals compared 14 to large urban hospitals. 15 Our Finally, most hospitals are currently evaluated on 16 the care they provide to Medicare beneficiaries and their 17 performance is public reported on Hospital Compare. 18 However, some critical access hospitals have been exempted 19 from some quality reporting requirements. 20 As the Commission has stated, providers should be 21 evaluated on all the services they provide. This includes 22 measures common among rural and urban providers, as well as 114 1 measures that are specific to rural providers, such as 2 timely communication of patient information after a 3 transfer. 4 To allow equal access to information for rural and 5 urban patients, all hospitals should be subject to public 6 disclosure of their performance scores. 7 accountability and hopefully improve the quality of care 8 delivered in small facilities. 9 10 11 This may improve Jeff will now pick up with our discussion of payment adequacy and special rural payments. DR. STENSLAND: In your mailing materials, and 12 during our December meeting, the analysts for each sector 13 discussed the adequacy of rural Medicare payments. 14 this session, I’m going to try to bring together the 15 different findings on rural payment adequacy from the 16 different chapters and the different presentations you heard 17 in December. 18 adequacy as we do for overall payment adequacy, examining 19 variables such as volume of services, payments, and costs. 20 However, we will also examine whether rural payments are 21 adequate relative to urban payments, given our mandate. 22 we have discussed throughout the process, we will compare So in We will use the same criteria for rural As 115 1 different types of rural areas, including the more sparsely 2 populated rural counties. 3 When we discuss rural physician payment adequacy, 4 we have limited financial data so we tend to focus on access 5 to care indicators. 6 payment show that adequacy is good and similar to urban. 7 Our survey indicates that rural and urban beneficiaries 8 report that they have similar ability to obtain physician 9 appointments when they want those appointments, and to In general, those access indicators of 10 obtain new physicians when they’re looking for a new 11 physician to accept them into their panel. 12 Consistent with that, when we look at claims, we 13 see that rural beneficiaries have roughly equal volumes of 14 visits as urban beneficiaries. 15 Similarly, rural home health agencies appear to 16 have adequate payments. The number of home health services 17 per capital is similar in rural and urban areas. 18 addition to that, we tend to see much more regional 19 variation in home health use than we do rural/urban 20 variation. 21 margins. 22 any systematic differences that would require additional In Quality scores are also similar, as are profit Given the rural/urban similarities, we do not see 116 1 assistance to all rural home health agencies. 2 The SNF story is similar to the home health story. 3 There is a similar number of SNF services received by rural 4 and urban beneficiaries per capita. 5 similar and margins tend to be high in all types of rural 6 areas. 7 concerns for the different types of rural counties. 8 9 Quality scores are also Hence, we do not see any rural payment adequacy For hospice we do see some rural/urban differences. Hospice use is a little bit lower in rural 10 areas but the rate of hospice use is growing. 11 rural providers tend to have smaller hospices and have 12 slightly lower margins, as Kim discussed last December. 13 However, given the Commission’s discussion that there is a 14 need to broadly refine the hospice payment system, the small 15 rural/urban differences we see should be seen as a secondary 16 concern to the larger effort of refining the hospice payment 17 system. 18 In addition, In addition, we had a recommendation to adjust 19 hospice payments to pay more for the first days of the stay 20 and more for the last days of the episode. 21 tend to benefit rural hospices. 22 changes to the hospital payment system that would benefit And that would So given the potential 117 1 rural providers, and given that most rural counties still 2 have healthy profit margins under the current system, there 3 does not appear to be a pressing need for additional payment 4 changes that would target solely rural providers. 5 Now when we turn to IRFs, it’s a little bit 6 different because all areas do not have IRFs, and in some of 7 those markets patients will receive care in other settings 8 such as SNFs. 9 geographic areas is a less useful measure. 10 So patient volumes in IRFs for different When we look at profit margins, we have a more 11 complicated story than we’ve seen in the last few slides. 12 In general, IRFs with more discharges per year tend to have 13 higher profit margins, as you heard yesterday. 14 margins for urban IRFs, who see an average of 942 patients 15 per year across all payers was 9 percent while the margin 16 for the handful of IRFs in the rural areas next to urban 17 areas, who saw an average of 104 patients per year was 18 negative 5.6 percent. 19 economies of scale issue. 20 In the Certainly, part of that is the However, the story is mixed in the most rural 21 IRFs, the ones that are located in counties not adjacent to 22 urban areas and that don’t have a city of 10,000 people had 118 1 a margin of 16.1 percent. 2 relatively few patients per IRF. 3 This was despite still having There’s a few things I would like to mention about 4 this. First, notice on the first row, you’ll see the 5 numbers we are talking about here are very small. 6 adjacent to urban areas, that’s only 13 IRFs we are looking 7 at in our sample. 8 17. 9 20 margin is not uncommon, a positive 20 margin is not In the non-adjacent to urban, it’s only There’s a lot of variation in IRF margins. 10 uncommon. 11 In the A negative so few. 12 So take it with a grain of salt because there are The second, if you look at the ones that are 13 adjacent to urban areas, they have 104 patients per year. 14 But their census is only about three. 15 to run a business with only three patients in there, you can 16 see where you’d have some real economies of scale problems. 17 So if you’re trying The next one, the average census is four. And so 18 maybe from three to four that still sounds small but from a 19 percentage basis it’s a pretty big difference and it might 20 explain part of the differential. 21 22 Finally, probably due to pure chance, there’s more free-standing IRFs in the last column than there is in the 119 1 third column and that can also affect the numbers. 2 trying to give you some flavor on why you see the difference 3 between those two different categories of urban areas. 4 lot of it might be the small numbers, some of it economies 5 of sale, and some of whether you’re free-standing or non- 6 free-standing. 7 So I’m A The current policy for IRFs is that they all 8 receive an 18.4 percent add-on if they’re in a rural area. 9 There is some concern that this policy is provided to all 10 IRFs and not targeted just to low volume IRFs that are 11 necessary for access. 12 The payment for dialysis appears adequate as 13 capacity has grown by 4 to 5 percent per year in rural 14 areas. 15 lower than urban margins, a negative 3.7 percent compared to 16 3.4 percent due to lack of economies of scale, starting in 17 2011 small dialysis facilities with fewer than 4,000 18 services per year will receive an 18.9 percent increase to 19 their payments. 20 no matter what your distance is from another facility. 21 22 While rural Medicare margins have historically been And it goes to all low volume facilities, So first of all, this raises some concerns about it not being targeted, if we’re giving this addition to two 120 1 dialysis facilities that are both one mile apart from each 2 other, that might not be an efficient use of Medicare 3 dollars. 4 Also, because the low volume adjustment will have 5 a significant affect on rural profit margins, we have a plan 6 to really look at this again next year and once again look 7 at rural profit margins once they start receiving this low 8 volume adjustment. 9 The other factor we looked at is quality, and for 10 dialysis patients, the process and outcomes measures do 11 appear to be similar in urban and rural locations, and the 12 rates of hospitalization are slightly lower in rural areas 13 but the differences are very small and not statistically 14 significant. 15 The share of hemodialysis patients who receive 16 adequate dialysis virtually the same across rural and urban 17 areas. 18 who have a catheter, where we generally think rates are 19 better -- though not always, as Karen has said -- also show 20 that the rates are similar across urban and rural areas. 21 22 And for new to dialysis beneficiaries in 2009, those With micropolitan areas, those micropolitans being the rural areas with a city between 10,000 and 50,000 121 1 people, show the best rates of all the groups. 2 So I would say, in summary, the rural story is 3 kind of lets wait and see until we have this extra rural 4 add-on in the mix. 5 Rural and urban patients also receive similar 6 numbers of hospital services during the year, as we’ve 7 discussed in our February meeting. 8 quality of care is mixed when we look at hospitals, with 9 readmissions being equal for rural facilities. We also found that But the 10 smallest facilities tend to have higher mortality rates, as 11 Adaeze just mentioned. 12 When we examine profitability, the most rural 13 facilities tend to have the highest margins. This reflects 14 the special payments provided to rural facilities, which 15 we’ll discuss later. 16 is a relatively new phenomenon, where rural hospitals have 17 higher Medicare margins than urban hospitals. As we can see on the next slide, this 18 I think the purpose of this slide is to explain 19 why the story might change in our 2012 report compared to 20 what we said in 2001. 21 and then listen to this discussion and say why has MedPAC 22 changed their story? Somebody might read our 2001 report I think the reason is that the 122 1 underlying story has changed, and that’s reflected in how 2 the data has changed. 3 If you look at the slide, you can see back in 2000 4 rural margins were much lower than urban margins. We 5 mentioned that in our report. 6 recommendations to change payments to make them more 7 equitable in rural and urban areas. 8 were largely adopted. 9 urban margins. We made a couple of Those recommendations Rural margins improved relative to And then there was some additional payment 10 adjustments that came into play that further helped rural. 11 And that’s how we ended up with the rural margins now better 12 than the urban margins. 13 Now we can examine some of those special payments 14 that we talked about and also discuss the issue of special 15 payments, which is part of our mandate. 16 some of the special adjustments that were made to increase 17 rural hospital payments. 18 this slide. 19 adjustments. This slide lists There’s really just two points to We won’t go through each one of these 20 First, I want to note that there’s many 21 adjustments, and this explains part of the graphic that you 22 just saw on the prior slide. Second, we want to reiterate 123 1 that some of these adjustments were, in fact, MedPAC 2 recommendations and they did fit some of the principles 3 we’ve discussed in terms of targeting low volume hospitals 4 that need special assistance. 5 adjustments that were added have not fit our principles that 6 we discussed in our prior meetings. 7 However, some of the other This slide looks at some of the special payments 8 for the other sectors. First, we could say that many of 9 these adjusters are not targeted. They’re either going to 10 all the providers in a frontier state, in some cases to all 11 rural providers, or in other cases to all low volume 12 providers. 13 isolated providers. 14 And they generally aren’t targeted to low volume Just to go through an example that we can use to 15 show how one of these adjusters might not fit the principles 16 that you discussed in some of our prior meetings on special 17 adjusters this past fall, we look at the hospital low volume 18 adjustment. 19 will be entitled to an additional low volume adjustment and 20 that will stay in place through 2012. 21 it will widen the difference between the rural and urban 22 margins. Starting in 2011, this year, small hospitals What this will do is 124 1 As I’ve said before, I think there’s generally 2 three problems with this low volume adjustment. 3 not targeted to isolated hospitals. 4 is not empirically based and it uses just Medicare 5 discharges to determine whether you’re low volume rather 6 than total discharges across all payers. 7 community hospital and Medicare dependent hospital programs 8 cause some duplication of payments with the low volume 9 adjustment program. 10 First, it’s Second, the adjustment Third, the sole If we added the new low volume adjustment to our 11 2010 margins, we would see those simulated margins would 12 rise to a projected 14 percent, which would be far above the 13 margins of other hospitals. 14 So in summary, while a low volume adjuster may be 15 needed for some providers, in this case the adjuster isn’t 16 targeted and it isn’t empirically justified. 17 So now I’ll discuss some of the guiding principles 18 for the special payment adjusters. 19 summarize what we heard from you during the last time that 20 we talked about special payment adjusters in the fall. 21 22 And we’re trying to The first principle is that low volume adjustments should be targeted to isolated providers. The general idea 125 1 here is it doesn’t make sense to provide a low volume 2 adjustment to two competing providers when they’re both one 3 mile apart or five miles apart and both suffering from low 4 volume. 5 Second, after we identify who is eligible for 6 additional payments, the question is how much should they 7 get? 8 justified. We want the amount of adjustments to be empirically 9 Third, with respect to low volume adjustments, the 10 adjustment should be tied to the total volume of patients, 11 not just Medicare volume. 12 Fourth, the low volume adjustments should not 13 duplicate other adjustments, as they did in the case of the 14 current hospital low volume adjustment. 15 Finally, it’s important to think about the 16 financial incentives associated with the adjustment. 17 Different ways of payment can carry different incentives 18 and, all else equal, prospective payment adjustments tend to 19 have stronger incentives for cost control than cost-based 20 payments. 21 22 So that’s a summary of the four topics that are mandated under our required report. For each of those, 126 1 we’ve tried to summarize first, what we found; and second, 2 the principles which were tried to be a summary of what you 3 have discussed at the prior meetings on payment adequacy, 4 quality, access, and special payments. 5 Now we’d like to hear from you any comments or 6 questions you have on the report, any suggestions you have 7 for changes to the principles, or any other guidance you 8 have as we start to draft the report this spring. 9 10 MR. HACKBARTH: Herb, why don’t we start with you. 11 12 Okay, thank you. MR. KUHN: one? Are we going to do two rounds or just How do you want to do that? 13 MR. HACKBARTH: 14 MR. KUHN: Let’s do two. Two quick questions, just on 15 clarification. I think I’ve asked this in meetings before, 16 and if I have, excuse me for this, I forget the answer. 17 when we look at the documentation and coding adjustment 18 that’s out there, there has been some conversation that it 19 is not as powerful an adjustment in the rural areas because 20 of the types of cases where there’s the opportunity to code 21 more accurately is -- they don’t see a lot of those cases, 22 perhaps, in some rural hospitals. But 127 1 Is there evidence that there is a differential 2 here, as a result of DCI, that would -- as we make the 3 adjustment, that we might be disadvantaging rural hospitals 4 because we’re suppressing their payments further. 5 DR. STENSLAND: There isn’t evidence of that, and 6 there’s two ways to look at it. At the really granular 7 level, you can look at the different MS-DRGs. 8 what we saw is we saw a movement in all the MS-DRGs, up from 9 where more of them are showing higher rates of complications And basically 10 and major complications. 11 across the board, your pneumonia admissions, your heart 12 failure admissions, as well as your more complex type things 13 that go to tertiary care hospitals. 14 So what we’re saying is it’s going So because it happens at all the different DRGs, 15 those ones that are treated at small hospitals and those at 16 large hospitals, it’s not surprising that when we see the 17 change in case-mix, we see roughly the same change in the 18 small hospitals as we do in the large hospitals. 19 MR. KUHN: Thank you. 20 The second question, when you were talking about 21 quality, the observation from the data that you shared is 22 that it’s pretty much the same, for post-acute care 128 1 hospitals it’s a bit mixed. 2 seeking care in a rural area and you went to one of the 3 compare websites, would you be able to differentiate that 4 from the Compare websites? 5 that goes beyond what’s available on Compare? 6 DR. STENSLAND: If you were a Medicare patient Or is this a different analysis It’s a different analysis, and 7 this was one of the things we talked about. The way the 8 Hospital Compare does its analysis is it uses what it would 9 call a random effects model. The important thing to 10 remember about the model, the key thing that’s driving it, 11 is it has this kind of prior assumption that everybody’s at 12 the mean. 13 it’s going to assume basically that you’re at the mean and 14 it’s going to create a weighted value for you on the CMS 15 Compare website that’s basically almost completely weighted 16 at the mean and very little weighted on your own data. 17 And if you have a really low volume of cases, So if you look at the Hospital Compare website, 18 all the really small hospitals, they’re going to look like 19 they’re really close to the mean no matter what their actual 20 performance was because what essentially CMS is saying is if 21 it’s different from the mean, because the number is so 22 small, we’re really not confident on whether that’s random 129 1 luck or really differences in quality. 2 to say they’re average. 3 So we’re just going Now what we did is we didn’t look at the 4 individual observations. 5 has 15 discharges in CHF. 6 hospital’s mortality rate is. 7 volume problems. 8 9 So we didn’t say oh, this hospital Let’s exactly know what that That would cause a lot of low What we did is we said let’s combine the data from a whole bunch of these small hospitals. Like we combined 10 the data from all 1,300 CAHs. 11 1,300, you have a much bigger n and you have much more 12 confidence in the n, and you don’t have to move to these 13 kind of random effects models like CMS did. 14 So once you combine it from, So then we can now, looking at this big group, how 15 does the big group compare to this other big group. 16 then we see that the smaller hospitals tend to have higher 17 mortality than the bigger hospitals, and we also see that 18 rurals tend to have a little worse mortality than urban. 19 And those findings aren’t like some news flash. 20 also the same things you will see in the literature for the 21 past 10 years. 22 DR. NAYLOR: Great report. And Those are 130 1 On Time Table 15 can you -- and I know you put the 2 time of each of these MedPAC recommendations and so on -- 3 but can you help to clarify the relative impact of say the 4 three MedPAC recommendations on what we seen in terms of 5 changes? 6 give it a sense of this was instituted and this was the 7 impact? 8 9 I don’t know that that’s important or not, but to I don’t know if that’s possible. DR. STENSLAND: Off the top of my head, so this is really rough, first increasing the rural base rate up to the 10 urban base rate was a pretty big move. 11 order of a percent difference in your margins. 12 used to be structured is when they originally set up the 13 system, it looked like rural providers had lower costs, for 14 some unknown reason, than urban providers. 15 we kind of got better at measuring case-mix and other 16 things. 17 there’s really no justification for having a lower base rate 18 for a rural provider than an urban provider. 19 their wages are the same and the case-mix is the same, they 20 should get the same. 21 22 I’m thinking on the The way it But over time, And when MedPAC looked at this in 2001, they said So MedPAC recommended that. Meaning, if And that was a material bump up in the rural profitability when Congress 131 1 adopted that. 2 The DSH payments was also a material bump up. 3 That was pretty significant. 4 much lower DSH payment even though they had the same share 5 of low income people than urban folks. 6 basically said that wasn’t equitable. 7 adjustment. 8 9 The rural folks used to get a In 2001, we We made that It made a significant change in rural payments. The third was the low volume adjustment. What happened here is when we estimated this, I think we saw a 10 low volume effect closer up to 500 discharges or so. 11 way it was enacted was CMS, the Secretary, was given 12 discretion on how to enact it. 13 they said they were only going to give a low volume 14 adjustment to folks with 200 or fewer discharges. 15 The And they looked at it and And what happened over time was a lot of these 16 really small ones just became CAHs. 17 adjustment, as it stood, had almost no effect. 18 there just isn’t very many with 200 discharges that aren’t 19 CAHs. 20 21 22 So this low volume Because Those are our three policies and how much they affected the margin. DR. NAYLOR: Thank you very much. 132 1 [Pause.] 2 MR. GEORGE MILLER: Yes, on Slide 5, first of all, 3 I want to state I think that no matter what the quality 4 parameters should be, they should be of high standards and 5 really no difference than any other provider. 6 curious on the mortality rate. 7 function of maybe travel time and distance, why the 8 mortality rate is lower? 9 would not be at a higher level? I'm just Do you know if that's a Do we know the reasons why they I think you said they're 10 mixed, and mortality is one of the reasons they're mixed. 11 Do we know the reason why? 12 DR. STENSLAND: Well, when we looked at this, the 13 two things that we highlighted were pneumonia discharges and 14 CHF discharges. 15 there's probably an extra 10 minutes, on average, travel to 16 the hospital. 17 admission, we don't think that's really going to do it. 18 didn't focus on the AMI admissions in our analysis. 19 20 And if you do look at the travel times, But for a CHF admission or a pneumonia MR. GEORGE MILLER: Folks that are AMI, but you're telling me this includes both CHF and pneumonia? 21 DR. STENSLAND: 22 MR. GEORGE MILLER: We Yes. All right. Ten minutes 133 1 wouldn't make the difference on those. 2 about the AMI. 3 4 All right. DR. STUART: I'm just wondering Thank you. I think you've done a great job. Thank you. 5 DR. CASTELLANOS: Yes, on Slide 17, you describe a 6 15 percent add-on to physician payments billed by the 7 critical access hospital. 8 community that's seeing the same patient, doing the same 9 service, but not being billed by critical access? 10 How about the physician in the Does he or she get the 15 percent? 11 DR. STENSLAND: 12 billing to the hospital. 13 DR. CASTELLANOS: 14 DR. STENSLAND: No. They have to assign their Why? I'm not sure what the rationale 15 there is. 16 integration or what they were doing, but that's the policy. 17 I don't know if they were trying to promote DR. CASTELLANOS: Well, that's similar to a 18 discussion we had yesterday. 19 that's an equity issue. 20 MR. HACKBARTH: 21 MR. GEORGE MILLER: 22 MR. HACKBARTH: I don't think it's -- I guess Is that statutory? [off microphone] Yeah. It's written that way in the 134 1 statute. 2 This isn't by regulation. MR. GEORGE MILLER: It is statutory, and I think 3 the reason is it was a recruitment issue to areas. 4 just a recruitment issue, how to get more physicians in the 5 rural areas, if I remember correctly the rationale for that 6 being -- 7 MR. HACKBARTH: 8 MR. GEORGE MILLER: 9 MR. HACKBARTH: 10 It was I think Ron's point is -I understand his point. DR. CASTELLANOS: Okay. Slide 8, the limited financial 11 data, now that, you know, physicians are employed by 12 hospitals greater than 40 percent, is that data available 13 through the hospital side? 14 physicians' practices, you know, now that physicians are 15 employed and it's greater than 40 percent, I would think 16 some of that data may be available on the hospital side. 17 DR. STENSLAND: Limited financial data on I would have to think about that. 18 I don't think the cost reports are really designed to pick 19 up all of that. 20 reports, so we can look at the clinic. 21 we wanted to only look at the clinic either when we started 22 to look at what the physicians' expenses are, if we thought We have the clinic part of the cost And I'm not sure if 135 1 there was some difference in relative efficiency between the 2 hospital-based clinics, which are a small share, and the 3 free-standing, which are a big share. 4 about that more to get you a better answer. 5 DR. CASTELLANOS: Okay. I'll have to think Then on Slide 6, just the 6 first one, services provided by the rural hospitals that 7 they choose to deliver. 8 go to a rural hospital one day a month. 9 hospital doesn't have that quality the other 29 days. 10 Now, obviously that Maybe I'm making a mountain out of a molehill. 11 12 You know, there's some of us -- I DR. AKAMIGBO: I'm not sure I understand the question. 13 DR. CASTELLANOS: Well, you're talking about 14 quality of care should be equal for non-emergency services. 15 In the areas where the provider chooses to deliver that 16 service. 17 DR. MARK MILLER: Ron, where this came from in the 18 discussions -- and this is something that some of the 19 Commissioners were saying -- if a rural hospital decides 20 it's going to do hip replacement or some procedure like 21 that, the thought process is that's a choice and that we 22 should expect quality outcomes to be the same. If somebody 136 1 arrives at your doorstep after a car accident and it takes 2 30 minutes to get people into the hospital and take care of 3 that patient, you might reasonably anticipate that there's 4 some difference in quality in that circumstance. 5 6 That's what this is trying to say. getting at -- 7 8 It may not be DR. CASTELLANOS: I guess I'm trying to ask whether it -- I think we best just leave it like it is. 9 MR. GEORGE MILLER: I think, Ron, for the days you 10 provide care at that rural hospital, the standard of care 11 should be the same as any other place while you're there. 12 The days that you don't provide it, then that service is not 13 available and there's no prerequisite for the standard of 14 care when you're not there. 15 DR. CASTELLANOS: 16 MR. ARMSTRONG: Thank you for clarifying that. Do we know, when we look at the 17 overall Medicare program, how much we spend on what we would 18 define as rural care versus everything else? 19 magnitude? 20 DR. STENSLAND: Just order of So the rural people are probably 21 about 20 percent and maybe the rural providers are going to 22 be a little bit less than 20 percent. 137 1 MR. ARMSTRONG: So we would generally think this 2 chapter is covering something less than 20 percent of the 3 overall spend for the Medicare program? 4 DR. STENSLAND: Correct. 5 MR. ARMSTRONG: Okay, great. 6 I just wanted to get... 7 MR. BUTLER: So this is an older than average 8 population, if my memory serves me right, and I'm still 9 trying to get the profile of the post-acute care services in 10 the average rural community. 11 Go to Slide 11. You've done a nice job kind of slide by slide 12 walking through the post-acute sectors, but just as an 13 example, you know, it looks like about 40 percent of the 14 hospices in the country are in rural areas, yet you say the 15 use is lower. 16 disproportionate number of hospices but they're not being 17 used. So, obviously, there are a lot of, a They could be dinky hospices. 18 So my own visual sense of this is that there's 19 probably more dependency on SNFs than in urban populations, 20 less dependency certainly on IRFs, a little bit of 21 underutilization of hospice, and sporadic availability of 22 home care. But I'm trying to get the -- it might be as you 138 1 roll this up, you almost look at per capita spending by each 2 of these segments and compare it to the urban areas, and 3 just to get a sense of the dependency on the various post- 4 acute sectors in a rural area versus an urban area, because 5 I think it tells a little bit bigger story than just showing 6 the number of facilities and then one at a time the 7 utilization. 8 the web of services are or are not available in the 9 community. 10 It doesn't paint the bigger picture of what DR. STENSLAND: And I think we can -- that's a 11 good idea, to paint that bigger picture, and the picture 12 here we're painting is the provider picture, so you do see a 13 lot of home health, a lot of SNF use. 14 side picture we can present, which is the patient's picture, 15 and I think the patient's picture is a little different. 16 For some things, like you don't see LTCHs in the rural 17 areas, but rural people in Louisiana use LTCHs because they 18 get transferred out to the LTCH in the urban area. 19 there's kind of two pictures there that we have to paint: 20 the picture of the rural providers and then the picture of 21 the providers that rural patients get their services from, 22 which are a little bit different. There's kind of a So 139 1 MR. BUTLER: Well, as Tom would remind us, if 2 you've seen a rural area, you've only seen one rural area. 3 They're all a little bit different. 4 that general, am I right that SNF is a bigger participant in 5 general in post-acute care? 6 DR. DEAN: But I'd ask him, is I think that's a fair -- I mean, of 7 course, you know, I'm looking at it from a very sparsely 8 populated area, and it isn't necessarily typical. 9 area that probably presents some special problems, but much I'm in an 10 of what is in this category defined as rural is very, very 11 different from where I live. 12 answer to your question is yes. 13 MR. HACKBARTH: But where I live, yes, the Along, I think, the same lines as 14 Peter's comment about painting a picture, in terms of use of 15 physician services -- and Peter focused on the post-acute 16 care, but can we say something about where rural 17 beneficiaries go for physician services? 18 has been that they probably get more services from hospital 19 outpatient departments than people in urban areas. 20 recollection from 2001 is that when you look at the 21 ambulatory visits, the rates are similar between rural and 22 urban, but the mix is somewhat different. But my impression Rural My 140 1 beneficiaries are somewhat more likely to see advanced 2 practice nurses as opposed to MDs. 3 some more texture to the description, I think that would be 4 helpful. 5 MR. GEORGE MILLER: So if we can sort of add I, Glenn, to your point, I 6 think if I remember correctly, also the rural population is 7 older, they're sicker, they're poorer, and have other 8 attributes that probably their urban counterparts would not 9 have as well, and to add that context to the chapter might 10 be helpful as well. 11 DR. STENSLAND: We should clarify. You know, we 12 did talk about the demographics, I think in February, of the 13 population, so there are some differences in what we have so 14 far in that mailing that we think will end up in the final 15 report that we should discuss if people have issues with 16 that. 17 mixed bag depending on which measure you used. 18 healthier on HCC scores. 19 morbidities, less of some co-morbidities, and we didn't see 20 a big rural/urban difference on the sickness. 21 regional difference, like if you look in the south-central 22 states -- Kentucky, Alabama, Mississippi -- there you see First, whether they're sicker is we really found a They look They have more of some co- We saw a big 141 1 people poorer and sicker. 2 there was some work by the USDA on that where they said if 3 you look at raw income, it's a little bit lower; but if you 4 adjust for the cost of living, it's actually higher. 5 then if you look at wealth, especially farmers are way 6 wealthier. 7 simpler story that we always hear. 8 9 10 11 And then on the income side, And So it's more complicated than some of the We will have all that in detail in the report, and you'll have a chance to comment on it, but I just wanted to be clear that it might differ from the standard story. DR. DEAN: Just on that point, after the report, I 12 went back and looked at some of that stuff, and it did 13 appear that there is a difference between the general rural 14 population and the rural Medicare population, and the 15 statistics that I saw at least were actually better for the 16 Medicare population. 17 true for what I looked at if you looked at the rural 18 population in general. 19 Medicare population, what Jeff says, it wasn't as dramatic. 20 The thing that George just stated held But when you looked at just the DR. MARK MILLER: [off microphone] And that's one 21 of the points we've been trying to make, is that sometimes 22 there's that basic description of the rural people that then 142 1 just kind of carries over into the policy conversation, and 2 there is this distinction -- in part there is this 3 distinction because that's what Medicare set out to do. 4 5 DR. DEAN: In some ways it's reassuring. I mean, it testifies probably to the value of Medicare. 6 DR. CHERNEW: First of all, I want to support what 7 Peter said, even though he's running away, about the 8 population focus to some extent. 9 You've structured this where you've gone through our normal But the question I had is: 10 payment thing that we just went through on our long march 11 yesterday, but you didn't have anything about managed care 12 payment. 13 about payment adequacy? 14 15 16 17 Is that outside of the scope of how you think DR. STENSLAND: It has been so far. For this report we haven't -DR. MARK MILLER: There's a little more advanced answer to that. 18 [Laughter.] 19 DR. CHERNEW: 20 21 22 It was not a normative question. It was a clarifying questions. DR. MARK MILLER: considers managed care. Yeah. The Commission certainly If you read the mandate, we've read 143 1 this to be asking us to pretty much focus on fee-for- 2 service. 3 blanking a little bit on the language. 4 reading is tell us about fee-for-service, is sort of the way 5 we saw it. It's providers of services -- Jeff and Adaeze, I'm 6 MR. HACKBARTH: 7 focus for this. 8 But your first plain And I think that is a sensible questions? 9 Tom, do you have any further clarifying DR. DEAN: I just wanted to follow up a little 10 more on the mortality rates. 11 mean, I guess the thing that comes to mind is that we see a 12 fair number of people with complicated illness who have 13 received a majority of their treatment in major centers, and 14 when the implication -- or when the effect of that treatment 15 has sort of run its course, they come back and we take care 16 of them. 17 in our facility. 18 It isn't huge numbers. 19 numbers. 20 phenomenon that we see, and I wonder if that is reflected in 21 these numbers. 22 Are those risk-adjusted? I And if it's a terminal illness, they tend to die And so they, you know, come home to die. Of course, we don't have huge It's a very small facility. DR. STENSLAND: But it is a We're using the AHRQ risk 144 1 adjustment method, which is based on the 3M method, so it is 2 risk-adjusted. 3 of omitted variable where there's some sort of severity of 4 illness which isn't picked up in that risk adjuster which 5 also sorts people that rural folks are more likely to go to 6 the hospital to die than urban folks? 7 But I guess you're asking is there some sort MR. GEORGE MILLER: It is possible. I think what Tom is asking is: 8 Is there a way to track where that person got care? 9 got the majority of their care in an urban area and then 10 chose to come back and die -- they didn't choose to, but 11 they came back to the rural facility and died there, the 12 majority of the care was given in an urban area. 13 mechanism or a way to track the census track or another way 14 to track where they got the majority of their care? 15 that's the genesis of Tom's question. 16 DR. DEAN: If they Is there a I think I think if I -- and, I mean, you can 17 tell me if the statistical logic is right, but I think that 18 would tend to -- as a proportion of our total activity, the 19 total care we give, I think that would make our mortality 20 rates look higher. 21 22 MR. GEORGE MILLER: Correct. Isn't that the genesis why the MS-DRGs were changed so that the severity of 145 1 the care -- yeah. 2 MR. HACKBARTH: [off microphone] -- would this be 3 beyond what you would capture through that sort of severity 4 adjustment. 5 DR. BORMAN: I certainly don't know any data 6 source for that, but having been on the receiving end at the 7 intake to come for the big care, I guess I would say back to 8 you that it's oftentimes extraordinarily hard to get those 9 people back to their originating care area. So that while I 10 absolutely accept that in your practice and experiences 11 you've seen the traffic in that direction, I assure you that 12 it also stays -- there's a fair amount of them that stay. 13 So I would hope that, by and large, it's a wash 14 because I think this would be incredibly hard to tease out 15 the motivation for a transfer. 16 figure out people that made -- a family or a patient made a 17 decision to withhold or withdraw care we can't really tease 18 out very well from the mortality. 19 incredibly hard to ascertain, and I think it goes both ways. 20 MR. HACKBARTH: It's kind of like trying to I think this would be Well, that's what I was just going 21 to say. There's also potentially some sorting not detected 22 by severity adjustments going the other direction. There 146 1 are patients that are especially difficult and challenging 2 that, you know, on paper it looks like the same MS-DRG, but 3 the clinicians are saying, "Boy, they really need to go to 4 the urban hospital." 5 DR. DEAN: I'm saying it's those patients who have 6 had their care in the urban center and have reached a point 7 where the decision has been made -- 8 MR. HACKBARTH: 9 10 I understand that. You're saying that the rural mortality may look artificially high because of that sorting -- 11 DR. DEAN: Because they're coming back after -- 12 MR. HACKBARTH: But it also could be true that 13 there are patients in a rural hospital that are sent to 14 urban centers because of the difficulty and they die there. 15 That makes the rural mortality look lower. 16 DR. DEAN: As Peter said, you know, I'm sure it 17 works both ways, but at least in my experience we certainly 18 see the former. 19 DR. HALL: No comments. 20 MR. KUHN: I want to thank both of you for a good 21 year of work, and this is a wonderful summation of kind of 22 where we are. And I think picking up on something that Jeff 147 1 said at the end of his presentation that the underlying 2 story has changed from the report that was produced in 2001 3 to the one we're going to produce in 2012. 4 we sit through and listen to the recitation of all the 5 reports and a summary of the data that we've seen in the 6 past, I think there's a good story to be told here that 7 Medicare has brought a lot of stability and a lot of 8 predictability to the rural areas and rural care that's out 9 there. 10 And I think as And I can see a report that could be constructed 11 where we could catalogue the changes that have occurred over 12 the last decade and perhaps not a lot of recommendations. 13 We might have some. 14 dialysis. 15 folks that thought this might be a report that's real robust 16 with a whole new set of recommendations that will be the 17 road map for the next decade of all these changes in rural 18 health care, that might not necessarily be the case given 19 some of the data that we've looked at and some of the areas 20 that were there. We see the information here on There might be some others. But rather than 21 But having said that, you know, one area that 22 might be helpful for us to look at in this report as we 148 1 continue to go forward is as a little bit of predictive of 2 what we might see in rural care over the next decade and, in 3 particular, access, picking up a little bit of something 4 that Glenn had made an observation to one person's question 5 or comment, is that, you know, we look at the data in terms 6 of on access, and we know we have a deficit of providers, 7 particularly physicians in rural areas, but care is 8 basically equal because people are willing to travel to get 9 care that's out there. 10 One of the things I know I raised when you 11 presented that data the first time is was there a way we 12 could stratify that data by physician age, the notion being 13 that perhaps a lot of these physicians that are practicing 14 in rural areas are a lot older than in urban areas. 15 done some of that work in Missouri, and it is indeed the 16 case. 17 over the next decade, and could that lead us to a position 18 where we might have even greater access issues that are out 19 there? 20 We've And so we will probably start to see some retirements Picking up a little bit on what Glenn said in 21 terms of primary care access with APNs and PAs, you know, I 22 go back to some of the data we looked at in the December 149 1 meeting when we were looking at access, when we were looking 2 at physician payment, and the survey we did of the Medicare 3 beneficiaries and those that were below Medicare, but fully 4 a third were now getting primary care from an APN or a PA. 5 So as we go through this report, if it's not going 6 to be a real major robust report with a whole lot of policy 7 recommendations because that may be directionally where we 8 don't need to go, one of the things that might be helpful, I 9 think, for the rural community overall but for the Medicare 10 program is if there is any kind of predictive modeling we 11 can do in some of these areas, particularly in access, and 12 look at what might be out there, at least give some 13 policymakers some things to begin thinking about. 14 different ways that we might want to think about medical 15 education or different aspects to make sure that we deal 16 with that inevitability that's going to occur in that space. 17 MR. HACKBARTH: Are there So as we go around in the second 18 round, one thing that would be helpful to get people to 19 react to is the plan for this report, and our sense right 20 now is that we will report data responding to the specific 21 questions asked in the mandate. 22 but not have bold-faced recommendations on which we vote. We will outline principles 150 1 So that's our plan to this point, and I'd invite people to 2 react to that. 3 Then just a question about the physician age 4 issue. Are there reliable national sources of information 5 on physician age that could be used to look at that? 6 DR. BORMAN: A huge number of the specialty 7 organizations are already reporting that for various 8 purposes, and I suspect the AMA Master File also would be 9 places. But there's aging of physicians and a whole bunch 10 of things both in terms of location and in terms of 11 specialty. 12 So I think those data are pretty accessible. MR. HACKBARTH: Well, I have often heard the point 13 made that Herb just made, that this is a particular issue in 14 rural areas. 15 we can shed some light on that on a national basis, that 16 might be an important contribution. 17 It sounds like you've done it in Missouri. DR. NAYLOR: If Let me just echo much of what Herb 18 has just said. I think that this is a good story for 19 Medicare, and I would highlight where you can, it's a good 20 story for MedPAC, and maybe thinking about superimposing 21 when recommendations went into play right on top of where we 22 began to see these migrations in margins I think would be 151 1 2 terrific. I think Peter's point about this being a story 3 about post-acute care and where there are lessons learned 4 here about the capacity to substitute other services, so not 5 having IRFs doesn't mean that people can't get access, 6 because your big story is about quality. 7 the quality measures and satisfaction measures look good. 8 So does that mean that there are opportunities here for a 9 continuum of post-acute services, et cetera? 10 And I think that I do think, picking up on the hospice, both access 11 on one hand and little utilization on the other is an area 12 for further exploration. 13 And the recommendations about, you know, 14 predictions of the future and future directions I think is 15 really important because in your proposed efforts around 16 targeting payment where we're continuing to see maybe great 17 opportunities that are on the one hand unmet and 18 opportunities for savings where we have duplication of 19 payment, you know, options or two hospitals right next to 20 each other I think are very, very important. 21 22 I would also say this is a good-news story for Medicare and HRSA because HRSA has placed a tremendous 152 1 amount of its attention on building the primary care 2 workforce, and one of the reasons we have 33 percent APNs 3 and PAs but many more proportionally in rural areas is 4 because of the kind of combination of efforts on the payment 5 and workforce delivery effort. 6 to highlight that as well. 7 MS. BEHROOZI: So I think it would be great I think you've done a great job of 8 organizing the discussion and stating the principles in ways 9 that I think we can all support. And I know that this is 10 not something that you can change -- you know, the horse is 11 already out of the barn on this one, but, you know, kind of 12 consistent with Scott's question, it's not really a 13 comparison between rural versus urban. 14 everybody else. 15 amount of variation both in terms of where people live and 16 what their access to services is. 17 that, you know, this was a congressional mandate and there 18 were important reasons that MedPAC made the recommendations 19 that it did, lots of which had to do with access and quality 20 for beneficiaries living in areas where they didn't have 21 sufficient levels of access and quality. 22 all move forward, I wonder if there will be an opportunity It's rural versus And in that 80 percent there is a huge And I fully recognize But as we, you, 153 1 to consider truly urban areas where there are problems of 2 access and quality, there are problems of low health status 3 because of socioeconomic status, because of conditions that 4 people live in, and really separate out the high-density, 5 low-socioeconomic status areas from, you know, greater 6 metropolitan statistical areas and suburban areas and places 7 like that that I think probably bring that overall so-called 8 urban, the everybody else level up, and it's great to bring 9 urban -- I'm sorry. It's great to bring rural to that 10 level, but it would also be really good to bring true urban 11 up to that level. 12 Thank you. 13 MR. GEORGE MILLER: Yes, I concur with everything 14 that has been said. 15 mandated, it's still a very good report and I think it is 16 fair and balanced and it certainly deals with what Mitra 17 just illuminated, dealing with the issues of access and 18 quality for rural areas, and I certainly want to thank the 19 staff for the efforts that have been done to date. 20 I think this is, although it's I would agree with Herb in that I'm not sure there 21 necessarily needs to be recommendations, but certainly fine 22 tune around the edges. And I also agree that this is a 154 1 success story not only for the Medicare program but for 2 MedPAC recommendations. 3 we are today, and again, except for tweaking around the 4 edges, I think there are successes, and particularly to the 5 point about mid-level practitioners. 6 we live and die with mid-level practitioners to certainly 7 help augment and support the physicians who provide care in 8 the area. 9 we can't recruit physicians, mid-levels have stepped into If you look back to 2000 and where In most rural areas, That also is a success story in many ways. Where 10 that gap, still supervised and overseen by the physicians, 11 but still provide a valuable service. 12 I think the other thing as we move down this path, 13 what Mitra said, is that -- and I agree with her, because 14 she hit right in my sweet spot, dealing with disparities and 15 inequities and pockets, whether they be in urban areas, 16 rural areas, or wherever they be. 17 attention to those beneficiaries that may fall through the 18 cracks or are not meeting that same standard of access and 19 quality no matter where that beneficiary is, and certainly 20 at some point we'll focus on that, as well. 21 22 DR. CASTELLANOS: We certainly should pay So thank you. I just want to comment that I totally agree with what's said. I just go back to 155 1 yesterday's discussion. 2 example of a targeted approach to solving some problems in 3 the Medicare system and we may want to consider that in the 4 safety net hospitals, et cetera. 5 concern yesterday concerning care, as George said and as 6 Mary said, as Mitra said. 7 MR. GRADISON: I think maybe this is a good I know that was a real I think it's wise to stick with the 8 principles rather than recommend specific changes in the 9 programs. However, having said that, I think this 10 information will not only be responsive to the Congressional 11 request, but also extremely helpful to us as we have to deal 12 each year with decisions by the silos, because it gives us a 13 little more color, a little more flavor and detail about the 14 rural, frontier, and the other categories of non-urban 15 institutions or non-urban settings than we might otherwise 16 have. 17 it here, but it may not be so much from just looking at 18 rural as it is looking at SNFs or looking at hospice or 19 other categories. 20 So I think we're going to get a lot of value out of DR. BORMAN: After almost six years, I'm sort of 21 out of adjectives and adverbs to say what really fine staff 22 work this represents, and it is just another example of 156 1 that. We should celebrate that ongoing work. 2 I think we also celebrate this as a success story 3 of identifying a problem that interventions -- that we 4 recommended things. 5 perhaps in a patchwork quilt way, as so often health care 6 policy is, and yet we have a good outcome, and we should 7 celebrate that in this instance, the process will work well. 8 I think that maybe the view to take of where we go 9 next is really more a phase of refinement and to think of it Interventions were undertaken, albeit 10 in that way, and that because of the success, we can now 11 focus on refinement. 12 mentioned. 13 monitor for the future? 14 may be, so that we don't end up so far behind the eight-ball 15 that we did that led us into this era, this phase of 16 activity that we've had. 17 So refinement is things like Herb What are the things that are high points to Where do we think the flash points So that would be one. Secondly, I think the targeting the dollars to the 18 best effect is something that we've tried to look at 19 pervasively in all the areas of the program, and I think to 20 the extent that we can comment on that in this particular 21 report in a very positive way -- we've done the experiment 22 here where we've done a number of kinds of programs. Where 157 1 have been the best results and how best do we target to 2 ensure that? 3 And then, finally, I think that as we're in this 4 refinement phase that we've shown, somewhat surprisingly in 5 some areas, perhaps, that, by and large, we are providing 6 equivalence as a result of these good interventions. 7 the driving metric, it seems to me, is quality. 8 we've defined, what are the appropriate -- or how do we 9 appropriately say what applies in these different population 10 efforts, and I think I continue to support the urgent versus 11 non-urgent services as good a break as we can come up with - 12 - there will be no perfect one -- that now we -- as long as 13 we monitor that access is still out there, then I think the 14 important thing that we follow is being more rich or more 15 deep about quality, making sure we've got that part right, 16 making sure that we're incentivizing as we look to targeted 17 payments and all the other things that we do, make sure that 18 those now incentivize quality appropriately in these areas. 19 That would be the driver, I think, that we would want to 20 conclude, also, for going forward. 21 work. 22 MR. ARMSTRONG: So now And as But this is really great So I, too, I don't have the six 158 1 cycles that Karen has of going through this, but would 2 affirm that the report that you've pulled together is really 3 outstanding. 4 I also would echo that it seems that we had an 5 issue. 6 we've had a real impact, and I think we have a lot to learn 7 from that. 8 9 We targeted solutions through payment policy and I would, however, just acknowledge that the payment structures we have for rural markets are really 10 complicated and kind of confusing and that I have to believe 11 that there's inefficiencies in it and that the principle 12 that you have with respect to avoid duplicating payments, I 13 think, is a little light. 14 to where we are right now, in the years ahead, I think we 15 should aspire to rationalizing and simplifying without 16 screwing up what we've done. 17 that point. I think that while we've gotten So I just would want to add 18 And then, second, our point of view around rural 19 services seems to be how can we through our payment policy 20 keep it from being -- keep our beneficiaries from being, 21 like, harmed by the special concerns that care delivery 22 providers have in rural communities. And at some point, I 159 1 would hope we could flip that a little bit and recognize 2 that care delivery systems in rural communities actually 3 achieve spectacular results given limited resources and that 4 there's probably a lot to be learned from that. 5 I mean, Tom and others have talked about the 6 creative use of mid-level providers, you know, kind of 7 actually replacing post-acute services, one for the other, 8 depending upon what's available and what really works. 9 mean, some of those kinds of things, I think in the years I 10 ahead, could actually offer insight into the ways in which 11 we evolve our payment policy in urban areas, and so just a 12 thought on that. 13 DR. BAICKER: I agree with Scott that the 14 principles make a lot of sense and could even be dialed up, 15 that we want to, even better than paramaterize, rationalize, 16 as you said, the payment structure. 17 payments can't be efficient and aren't particularly good at 18 targeting. 19 providers, and I might add in, with the explicit goal of 20 guaranteeing access or promoting access, that it's not just 21 because you're one, it's because you're what stands between 22 the beneficiary and not getting access to care of acceptable These overlapping As you pointed out, the goal is to target sole 160 1 quality and that that acceptable quality may vary across 2 circumstances, emergency versus non-emergency. 3 harmonize those payments so that we are effectively 4 targeting that particular goal would make the policy much 5 more effective. 6 MS. UCCELLO: Yes. But to Agreeing with Scott and Kate, 7 although there's a lot of good news in here, we have shown 8 that we're not -- we could target better. 9 things so they are more empirically justified, you know, 10 base things better. 11 worry about here. 12 We could refine and so we need to keep that in mind. 13 So it's not like there's nothing to There's still some room for improvement, But I agree in general with just sticking to these 14 principles and laying those out and I think the ones that 15 you've listed here are really excellent. 16 And just adding to the chorus of everyone here, 17 thank you so much for the tremendous work you've done over 18 the past year-and-a-half or longer working on this stuff. 19 It really shows. 20 MR. BUTLER: 21 congratulations. 22 final draft -- Okay. Enough of this I want to set the bar even higher for the 161 1 [Laughter.] 2 DR. MARK MILLER: 3 [Off microphone.] It looks like we're out of time. 4 [Laughter.] 5 MR. BUTLER: But let me tell you why. I'm 6 reminded frequently that we're working for Congress, and in 7 an area like this, this was not just the normal business. 8 This was a specific request by a lot of important members of 9 Congress, Senators that represent primarily rural States, 10 Representatives that sit in primarily rural areas. 11 so more than ever, think about the final draft as something 12 that is not only addressing the points, but something they 13 can use for a communication vehicle for their 14 Representatives, or if they were to come out and sit with a 15 SNF administrator or a hospital administrator or a community 16 group, it would kind of make sense to them. 17 I think, And that's kind of one of the reasons I mentioned 18 the post-acute profile. Make it -- this is a great 19 opportunity to cast just a little different light on it and 20 make sure it can be used effectively in many forums, because 21 these are a lot of important people in Congress and the more 22 you can kind of -- I won't say "dumb it down," even though I 162 1 just did, because I don't really mean that because it's 2 complicated and it does need to include the kinds of points 3 Scott made, as well, if you want to dive deep around there. 4 But I think it is a great opportunity to make the best of a 5 chapter and go beyond addressing a technical policy issue 6 and communicating a message. 7 DR. DEAN: I certainly agree with the principles 8 that were laid out. 9 appropriately the direction that our inquiry should go. I think that they really direct 10 I appreciate, obviously, what Scott just said, but 11 I'd make the point that -- and one of the things that I have 12 found appealing about rural practice is there are things you 13 can do in small systems that are very much more difficult in 14 larger systems, and the net effect of that is that they are 15 tremendously dependent on whatever capabilities or 16 leadership or whatever is there in that particular setting. 17 And if you have those capabilities, you can do great things 18 relatively easily. 19 real bad in a hurry. 20 If you don't have it, things can get For instance, in our own setting, every morning, 21 we meet with all the physicians, the nursing staff, physical 22 therapy, and social service. We all get together and we go 163 1 over every patient in the hospital. 2 rate is somewhere around four or five percent and I think 3 it's due to that issue. 4 there's a problem, we'll -- you can't do that in most 5 facilities. 6 And our readmission We all know what's going on. If I mean, it's just out of the question. So I think that we have some opportunities that 7 have nothing to do with the particular capabilities but just 8 it's possible to do things in that kind of a setting that 9 you can't do in bigger facilities. 10 I guess having said that, I do have some concerns 11 about some of the statements. I'm not sure that just 12 measuring number of services is really the answer in terms 13 of defining access. 14 measure, but it doesn't say anything about the timeliness of 15 those services and how quickly people reach. 16 instance, I would say that, basically, everybody in most 17 rural areas that has an MI eventually gets the care. 18 question is, do they get there in ten or 15 minutes or do 19 they get there two or three hours later? 20 of ways you can do it. 21 access, I think we need to tie it to some kind of outcome 22 measures, as well, to validate that and to see if it's I think it certainly is an important I mean, for The There's all kinds But I think to really determine 164 1 important. 2 Also, as Peter said, there really is tremendous 3 variation and I think we probably need to get to look at the 4 range of variation across the country, because there are a 5 lot of geographical differences and we've tended to look at 6 averages, which I've mentioned before, which frustrate me, 7 but which are sometimes useful, but I think also can hide 8 variation. 9 in each one of those measures, from the best to the worst, We need to look at how much variation there is 10 and determine if that's a narrow span, then the 11 generalization is clearly justifiable. 12 span, then I have a little more concern about it. If there's a wide 13 I think we looked at some of the cost issues 14 related to Critical Access Hospitals, and as I've said 15 before, the CAH program has been a tremendous boon and I 16 think has really stabilized care in many rural areas. 17 the same time, it has been criticized, legitimately, because 18 there probably are a number of facilities that qualified 19 where you could really question whether they met the basic 20 intent. 21 22 At Also, I think sometimes we jump to the conclusion that, well, if they have cost-based reimbursement, they're 165 1 automatically financially stable. The numbers are that a 2 substantial portion of CAHs also have negative margins, 3 something like 40 percent of them was the number I saw. 4 That's not immediately an issue, but I think we just can't 5 assume that they are automatically living easily, or that 6 it's an automatic sort of gravy train, because there are 7 still stresses out there. 8 raised that that takes away any pressure for cost control. 9 I don't think it's that simple. And I think the concern has been Certainly, I know in our 10 setting, our administration does worry about cost 11 considerably because they do have negative margins about 12 half the time. 13 Finally, I guess the other thing which I have 14 raised before, in some of these, especially the home health 15 area, which I'm sure you're tired of hearing me talk about, 16 we really need to look at the total spectrum of home health 17 agencies and not just freestanding ones. 18 and you've heard me say this before, there's two-thirds, at 19 least, of our home health facilities are associated with 20 Critical Access Hospitals and aren't even included in the 21 analysis that we do in terms of margin. 22 there are problems with measurement, but to make the In South Dakota, Now, I understand 166 1 generalization that rural home health agencies have the same 2 margin is, I don't think, accurate. 3 accurate in the areas that I'm most familiar with. 4 not exactly sure how you do that, but in South Dakota, 5 freestanding facilities only exist in about ten percent of 6 the counties and the rest of the State is all dependent on 7 provider-based facilities. 8 deficiency. 9 concern. 10 At least, it's not So I'm So I just think there is a I'm not sure just how to solve it, but it is a Overall, I agree. There's a lot of very useful, 11 and I think there's been a lot of progress. 12 the recommendations that have been made have moved the whole 13 state of rural health care forward considerably and it's, in 14 general, much more solid than it was when I started practice 15 quite a few years ago. 16 And clearly, The recruitment issue is a real worry, but it's a 17 worry across the country. 18 uniquely rural, although it is a particular concern because 19 we are so dependent on primary care physicians and they are, 20 obviously, as everybody knows, the group that's in most 21 demand and in smallest supply. 22 That's not a real -- isn't So, anyway, I'll stop there. 167 1 MR. HACKBARTH: Tom, the point about variation, 2 which I don't disagree with, for me, the challenge is 3 figuring out how to deal with it. 4 just talking about rural averages is to subdivide rural into 5 subcategories -- micropolitan, adjacent to urban, not 6 adjacent to urban. 7 classification to try to give a flavor for the varieties of 8 ruralness that exist. 9 So one approach to not We sometimes use the frontier Another more sort of statistically-minded approach 10 to variation is to, with everything, report not just the 11 median or the mean, but the standard deviation and all that 12 stuff. 13 readers. 14 it's very meaningful to them. 15 meaningful to me. 16 statistic reported, you've got this incredibly challenging 17 to present and read document. 18 19 20 I'm not sure that that really helps a whole lot of There are certain readers that look at that and It isn't particularly And if you have to do that for every So I agree with your basic point. I don't know how to solve it. DR. DEAN: I guess I would respond that we have 21 done this in some just reporting 25th and 75th percentiles 22 or something along those lines, in terms of whether it's 168 1 utilization or whatever. 2 cut and that may or may not qualify with these other 3 categories, you know. 4 sparsely populated, the worse it would be. 5 that's clearly true, but obviously, in other cases, it's not 6 true, and -- 7 I think that's sort of the first We sort of assume that the more MR. HACKBARTH: 8 put up Slide 11? 9 [Off microphone.] Oh, I'm sorry. In some cases, Jeff, can you these tables. 10 DR. DEAN: 11 MR. HACKBARTH: So it could be any one of Yes. So if we now subdivide, in this 12 case, the rural into three categories and show -- are these 13 means or medians here? 14 Medicare margin numbers? 15 DR. STENSLAND: 16 17 Are these medians, Jeff, the Aggregates, because we're adding together all the payments – MR. HACKBARTH: Oh, that's right, because we're 18 just combining them all together. 19 approach -- I'm statistically challenged here. 20 even report the 75th and the 25th percentile here. 21 the aggregate of the whole pool. 22 DR. STENSLAND: Yes. Yes. So with that So you can't This is We could also present the 169 1 75th and 25th as a separate way of doing it. 2 triple all the numbers in here. 3 MR. HACKBARTH: It would just And that's the point that I'm 4 headed to. 5 that, do we have a document that is readable? 6 help me out? 7 If you imagine for every table that we're doing DR. CHERNEW: Mike, can you Well, it might be useful, and maybe 8 not for all of these, to have a sort of scatterplot or some 9 other type of graph that just shows the range, because what 10 you really want to know -- I think what the comment is, is 11 if there's big outliers one way or another that's pulling 12 these averages down. 13 same but there's a few in some weird place, you could 14 actually look at them. 15 So if you saw they're basically the I don't think that's useful for the report, but 16 what it might be useful to do is to look at so then when 17 someone asks a question, you can say, this looks like 18 distributions that are just shifted, which is sort of what 19 you get when you think about it, or they're pretty much on 20 top of each other with a few outliers, particularly the one 21 that has the 17 and the 13 observations and you can't tell. 22 If you looked at where they were distributed, you would get 170 1 some sense of whether they're really just systematically 2 different distributions or there's just something going on 3 with the sampling. 4 substantively, I would then simplify it for the 5 presentation. 6 DR. DEAN: But once you know what the answer is Yes, I would agree with that. I think 7 as a test to see, for instance, these three criteria, 8 micropolitan, adjacent, and non-adjacent, is that really 9 telling us what we need to know? I think as an example, in 10 the chapter we looked at, I guess it was last time, about 11 home health, there was that table that listed 23 of 25, or 12 whatever it was, most expensive areas were all rural. 13 that was -- they were also all located in a very close 14 geographic area, and I think the problem was not that they 15 were rural, it was that they were in a particular geographic 16 area because -- so I think we -- it's just as a measure, 17 because I think if you looked at those same data, there was 18 a wide spread in terms of cost. 19 MR. HACKBARTH: The point about the home health 20 margins, I very much agree with you. 21 a little different issue -- 22 DR. DEAN: Well, Yes -- To me, that seems like 171 1 MR. HACKBARTH: 2 DR. DEAN: 3 -- and to me, that's one -- MR. HACKBARTH: [Off microphone.] Yes. Well, actually, I think it 4 is a major issue, and my comment when it got around to me 5 was going to be that I think one of the things that should 6 be really highlighted here is the artificiality of this 7 urban-rural distinction, and there is a huge amount of 8 variation. 9 regional variation swamps the urban-rural variation. For example, this home health statistic, the 10 DR. DEAN: [Off microphone.] 11 MR. HACKBARTH: That's right. And we have these policy 12 categories that are driving lots of dollars that in some 13 ways are artificial. 14 15 MS. BEHROOZI: Well, and then isn't it the same thing on the rural side -- 16 MR. HACKBARTH: 17 MS. BEHROOZI: 18 MR. HACKBARTH: 19 MS. BEHROOZI: On the urban side. I'm sorry, on the urban side. Yes. So if you're going to compare 20 apples to apples, I mean, you're still going to get the 21 washout of urban being a bigger group, but then you should 22 also do 25th to 75th on the urban side. You can't just do 172 1 25th to 75th on the rural side and compare it to the average 2 on the urban side. 3 MR. HACKBARTH: 4 [Off microphone discussion.] 5 MR. HACKBARTH: 6 Right. Okay. Right. Right. Yes. Were you finished, Tom? Any other points, Bill? 7 DR. HALL: Pass. 8 MR. HACKBARTH: 9 DR. BERENSON: Bob? I was going to make the same point 10 about the regional variations being very striking and not to 11 minimize that. 12 looking at the variation is very revealing of something, 13 then it goes in, but not routinely. 14 have -- based on everything I've heard, I have great 15 confidence that you guys have control over this paper and 16 would be able to identify those instances where that kind of 17 an analysis would actually add something versus -- and just 18 not doing it routinely, just as a matter of statistics. 19 I think this is in very good shape. 20 MR. HACKBARTH: 21 22 I guess what I'd say is where a sub-analysis Okay. In other words, where I Thank you very much. But good work. Very And we'll now have our public comment period. 173 1 [No response.] 2 MR. HACKBARTH: 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 you all. Okay. I think we are done. See you in March. [Whereupon, at 11:52 a.m., the meeting was adjourned.] Thank