[Federal Register: August 30, 1996 (Rules and Regulations)] [Page 46216-46266] From the Federal Register Online via GPO Access [wais.access.gpo.gov] [[pp. 46216-46266]] Medicare Program; Changes to the Hospital Inpatient Prospective Payment Systems and Fiscal Year 1997 Rates [[Continued from page 46215]] VI. Changes and Clarifications to the Prospective Payment System for Capital-Related Costs A. Consistent Cost Finding During the Capital Transition Period (Sec. 412.302(d)) Section 412.302(d) requires that during the transition period to full prospective payment for capital-related costs, a hospital must follow consistent cost-finding methods for classifying and allocating capital-related costs. Specifically, the regulation requires that unless there is a change of ownership, a hospital must continue the same cost-finding methods for old capital costs, including its practices for direct assignment of costs and its cost-allocation bases, that were in effect in the hospital's last cost-reporting period before becoming subject to payment under the capital prospective payment transition system. A hospital may request a change in its cost-finding methods for new capital, provided that the request is made in a timely fashion as provided in the regulation, the hospital provides justification for the change, and the intermediary determines that the justification is reasonable. It is important to note that, while the regulation does permit changes in cost-finding methods for new capital, such changes are only permitted where they do not involve any changes in cost-finding for old capital. In practice, this means that if a hospital claims any old capital, the intermediary cannot permit a change in any of the allocation bases on Worksheet B-1 of the cost report from the bases used in the last cost reporting period prior to the capital prospective payment system transition period. Otherwise, the consistency rule governing old capital cost-finding would be violated. In response to concerns expressed by the hospital industry about the costs of the recordkeeping required under the cost-reporting rules, HCFA has developed new cost reporting instructions, which will be released later this year, that permit hospitals to voluntarily adopt a simplified cost allocation methodology. This methodology reduces the number of statistical bases that a hospital is required to maintain. Under the new instructions for HCFA Form 2552-96 (the cost report instructions for FY 1996 cost reporting periods), hospitals may request the simplified cost allocation methodology. However, hospitals that elect this methodology must employ a prescribed list of statistical bases with no deviations. Hospitals may not pick and choose among the prescribed statistics for the combination that is most advantageous. The election of the simplified method cannot be used to shift costs inappropriately. Furthermore, a hospital that elects the simplified methodology must continue to use it for at least 3 years, unless a change of ownership occurs. In the proposed rule (61 FR 27478), we proposed to add a new paragraph (d)(4) to Sec. 412.302, to provide that hospitals may elect to adopt the simplified cost allocation [[Page 46215]] methodology, as will be provided in the instructions for HCFA Form 2552-96. Comment: One commenter agreed with our proposal to revise Sec. 412.302(d)(4) to allow for a simplified cost allocation methodology, but suggested that we make a technical change to existing Sec. 412.302(d)(1) to reflect the availability of the simplified methodology option. Response: We are adopting the commenter's recommended change to the regulations. Section 412.302(d)(1) will now read: "For cost reporting periods beginning on or after October 1, 1991 and before October 1, 2001, the hospital must follow consistent cost finding methods for classifying and allocating capital-related costs, except as otherwise provided in paragraph (d)(4) of this section." Comment: In response to our proposal on the simplified cost allocation methodology, one commenter argued that the general capital consistency rule is flawed. The commenter stated that a provider should be able to request that the fiscal intermediary reassign capital costs from the acute care hospital portion of a facility to exempt areas of the facility if the provider is using the space differently than it was used during the capital base year, such as using the space as a skilled nursing facility or a rehabilitation unit. Response: This comment concerns the underlying intent of the capital consistency rule itself rather than the subject of our May 31, 1995 proposed rule. In the August 30, 1991 final rule that implemented the capital prospective payment system (56 FR 43396), we explained the rationale for the capital consistency rule. We explained that the capital consistency rule is necessary: (1) to prevent cost shifting to outpatient departments through changes in cost finding methods, and (2) to provide consistency with the determination of the hospital-specific rate used in the base year. For these reasons, it is important that the hospital continue the same bases of cost allocation for old capital throughout the transition. Throughout the transition to a fully prospective payment system for capital, the provider must continue to allocate any space that was part of the acute care hospital in the base year in the same way. However, if the provider opens a new section of the facility as a skilled nursing facility or excluded unit, capital costs in those areas could be allocated directly to those areas. B. Possible Adjustments to the Capital Prospective Payment System Federal Rate and Hospital-Specific Rates (Secs. 412.308(b) and 412.328) In the proposed and final rules for FY 1996 (60 FR 29238-29239 and 60 FR 45830-45831), we discussed the effects of the expiration of the statutory budget neutrality provision on rates and aggregate payments under the capital prospective payment system. Under the budget neutrality provision, we set the capital-prospective payment system rates during FY 1992 through FY 1995 so that payments were projected to equal 90 percent of Medicare payments that would have been made on a reasonable cost basis for each fiscal year. As a result of the provision's expiration in FY 1996, the capital-prospective payment system rates and payments under the transition system increased significantly. The FY 1996 Federal rate is 22.59 percent higher than the FY 1995 Federal rate. We now estimate that aggregate capital payments will increase 27.5 percent in FY 1996 relative to FY 1995, and that payments will exceed capital costs by 8.8 percent in FY 1996. Under current law and regulations, we estimate that aggregate payments will further increase by 6.8 percent in FY 1997, for an increase of 36.1 percent over 2 years. We also estimate that payments will exceed capital costs by 7.5 percent in FY 1997. In the May 31, 1996 proposed rule, we stated that we continue to believe that such large increases in capital payments are neither necessary nor warranted. We identified several possible approaches for establishing a more appropriate level for the rates and discussed the options we considered in developing the proposed rule (61 FR 27479). These options included freezing the inflation updates for the rates in FY 1997 or making downward adjustments in the base rates, as discussed below: <bullet> Reduce the standard Federal rate by 7.38 percent and the hospital-specific rates by 9.48 percent to reflect revised data on base year costs used to determine the rates. <bullet> Implement the provision contained in the Administration's budget plan to reduce the base Federal and hospital-specific rates by 15.7 percent. As discussed in detail in the proposed rule, the rationale for reducing the base rate derives from an analysis of current data compared to data on which the rate was originally based. Under Sec. 412.308, HCFA determined the standard Federal rate, which is used to determine the Federal rate for each fiscal year, on the basis of an estimate of the FY 1992 national average Medicare capital cost per discharge. The FY 1992 national average Medicare capital cost per discharge was estimated by updating the FY 1989 national average Medicare capital cost per discharge by the estimated increase in Medicare inpatient capital cost per discharge. Section 13501(a)(3) of Public Law 103-66 amended section 1886(g)(1)(A) of the Social Security Act to require that, for discharges occurring after September 30, 1993, the unadjusted standard Federal rate be reduced by 7.4 percent. The purpose of that reduction was to reflect revised inflation estimates as of May 1993, for the increases in Medicare capital costs per discharge during FY 1989 through FY 1992. We now have extensive cost report data for FY 1992 that shows an audit-adjusted FY 1992 Medicare inpatient capital cost per discharge that is an additional 7.38 percent lower that the estimate on which the Federal rate is currently based. Accordingly, the rate could be reduced to reflect accurate FY 1992 capital cost per discharge data. Under Sec. 412.328, HCFA determined the FY 1992 hospital-specific rate by using a process similar to the process for determining the FY 1992 Federal rate. The intermediary determined each hospital's allowable Medicare inpatient capital cost per discharge for the hospital's latest cost reporting period ending on or before December 31, 1990. The intermediary then updated each hospital's FY 1990 allowable Medicare capital cost per discharge to FY 1992 based on the estimated increase in Medicare inpatient capital cost per case. As with the Federal rate updates, current data demonstrate that the estimates used to update the hospital specific rates from FY 1990 to FY 1992 were overstated. In order to adjust the hospital-specific rate to reflect actual FY 1992 data, the rates must be reduced by 9.48 percent. The reduction reflected in the President's budget plan is based on a different consideration. That reduction would build the budget neutrality adjustment for FY 1995 (0.8432, or -15.68 percent) permanently into the base rates, effectively using the FY 1995 base payment rate as the base for future years. The actual payment rates for future years would then be determined by applying the analytical update framework that we adopted in the final rule for FY 1996 (60 FR 45815- 45829). Our last analysis (60 FR 45826-45829) suggested that the estimated FY 1992 capital costs used to set the Federal and hospital- specific capital rates exceeded by approximately 28 percent the level that could be accounted for by known factors. This unaccounted for difference [[Page 46216]] in the rates justifies a 15.7 percent reduction to the rates. We seriously considered proposing one of these options in the proposed rule, and we invited public comment on their merits and on the advisability of implementing one or the other in the final rule, in the absence of legislative action. We received many comments on our discussion of possible adjustments to the capital Federal rate, and these comments and our responses are presented below. Although we continue to believe that any of these options is justified on the basis of current data and analysis, we are not implementing any freeze or reduction to the capital Federal rates in this final rule. The President's budget bill includes numerous proposals to reform the Medicare program, including a reduction to the capital prospective payment rate. At this time, we believe it would be more appropriate to adopt a change to the rate in the context of more global changes to the Medicare program than to implement this one specific provision of the President's budget through regulation. Therefore, we are not implementing any of the possible reductions to the capital Federal rate that were discussed in the proposed rule but instead are updating the capital rates in accordance with the capital update framework, as discussed in section III of the addendum to this final rule. In general, commenters opposed freezing or reducing the capital Federal rate as suggested in the proposed rule. Commenters cited various reasons why the suggested changes were inappropriate or unnecessary. One commenter, ProPAC, agreed that continued significant increases in capital payments are unjustified and supported reductions to the capital rate. ProPAC suggested several options for our consideration, such as using the FY 1995 rates as the base for future years, or rebasing the FY 1992 capital payment rates and updating them to the current year using an analytic framework. As explained earlier, although we agree with ProPAC that a reduction in the rates is warranted, we have decided not to proceed with reducing the rates by regulation at this time. We discuss the comments on the possible changes in more detail below. Comment: Some commenters contended that it would be illegal for HCFA to implement any of the identified reductions to the rates (including an efficiency adjustment) because HCFA does not have the authority to rebase the capital payment rate. Two commenters characterized the rate reduction options as thinly disguised attempts to rebase hospitals' base year capital costs, and asserted that Congress has not given the Secretary of Health and Human Services the authority to rebase hospital capital costs. One commenter stated that the rate revisions discussed in the proposed rule would violate a fundamental principle of prospective payment: that the system provide certain and predictable payment rates. Another commenter opposed any reduction in the capital Federal rate undertaken without legislative direction. Finally, one commenter noted that when Congress specified the 7.4 percent reduction in the Federal rate as part of OBRA 93, Congress referenced the capital Federal rate "as described in Sec. 412.308(c)." That regulation describes the methodology for defining the Federal rate. The commenter believes that the regulation does not contemplate the substitution of actual cost data for periods in which estimated data were used initially. The commenter believes that because Congress cited this section of the regulations, it implicitly approved the continued use of estimated data for setting the rates rather than the use of actual data. Response: Section 1886(g) of the Act states that "the Secretary shall, for hospital cost reporting periods beginning on or after October 1, 1991, provide for payments for [capital-related] costs in accordance with a prospective payment system established by the Secretary." The statute gives the Secretary wide discretion in determining the particular features of the prospective payment system for capital-related costs, including the appropriate level of payment rates. We believe that, consistent with this broad authority, it is appropriate to make prospective adjustments to the capital rates. We believe that any rate revision implemented prospectively would satisfy the principle of certainty and predictability under a prospective system. We have never contemplated a retroactive adjustment to payment rates used in prior years. The provision of OBRA 93 cited by the commenter does not indicate that we cannot make other adjustments to the capital Federal rate in future years. Section 412.308(c) describes the process for determining the Federal rate by adjusting the standard Federal rate by an update factor each year. We believe that Congress cited this section solely to identify the rate to which we applied the 7.4 percent reduction. Since the inception of the capital prospective payment system, rates have been set on the basis of FY 1992 capital costs. Since we set initially set rates before FY 1992 started, we necessarily had to project capital costs for FY 1992. We used FY 1989 costs as the basis for projecting FY 1992 costs because they were the latest cost report data available at that time. (Even the FY 1989 data required an estimated adjustment for the effect of audits not yet performed.) We applied estimated adjustment factors to the FY 1989 data to derive estimated FY 1992 capital costs. We used this estimated FY 1992 cost level to set rates beginning in FY 1992. When Congress legislated that the unadjusted standard Federal rate be reduced by 7.4 percent in 1993, the size of the adjustment was based on more recent data on FY 1992 costs available at that time. The latest available data now indicate an additional 7.36 percent reduction is appropriate. Again, although we are not implementing this adjustment, we believe that we have the authority to do so and that it would represent a logical extension of our policy of basing the capital Federal rate on FY 1992 capital costs. Comment: Several commenters stated that the discussion in the proposed rule of the possibility of implementing reductions to the capital Federal rate through the final rule did not constitute sufficient notice to the public of proposed regulatory changes. The commenters asserted that before implementing a reduction in the capital payment rates, HCFA was obligated to provide "formal" public notice and time for the public to respond. Response: As noted above, we do not intend to implement any reduction to the capital Federal rate at this time. However, we believe that the discussion in the proposed rule would have satisfied the requirements of the Administrative Procedure Act by (1) describing in some detail three potential options for cutting the capital rate, (2) informing the public that we might implement one of these options if Congress and the Administration did not act to cut the rate, and (3) soliciting public comment on the possible options. We stated that it was our intention to consider all of the options in light of the comments received. Moreover, in the FY 1996 proposed rule (60 FR 29238), we discussed in some detail and invited comments on two options for adjusting the Federal and hospital specific rate, to account for the overestimation of the FY 1992 Medicare inpatient capital cost per discharge, and to compensate for the effects of the expiration of budget neutrality. Finally, since FY 1992 we have printed seven discussions of the efficiency issue, and providers have long known that we [[Page 46217]] might make an adjustment in the rate to account for possible inefficiency. Comment: Some commenters stated that we should not adjust the Federal rate to reflect the actual level of FY 1992 capital spending because the FY 1992 level is lower than was projected. The commenters asserted that FY 1992 capital cost levels are lower than projected because hospitals responded in FY 1992 to the incentives of the prospective payment system and modified their capital spending behavior. Some commenters argued that hospitals responded to the possible implementation of a capital prospective payment system even prior to FY 1992. These commenters asserted that in order to get a true sense of the impact of the capital prospective payment system on hospital capital expenditure behavior, one must look further back to when hospitals believed implementation of such a system was imminent. One commenter explained that one reason actual increases in capital costs in FY 1992 were less than projected was because lengthy certificate of need (CON) approval processes prevented hospitals from beginning building projects as planned. The commenter also stated that if rates were reduced, hospitals in States with strict CON processes should not be subjected to the same rate reductions as hospitals in States without such processes. The commenter asserted that facilities in the commenter's State are undercapitalized relative to facilities in the rest of the country. Finally, some commenters believe that the overestimation of FY 1992 capital costs (discussed above) stems not from a forecast error in the FY 1992 capital cost per case but from a change in the treatment of allowable interest that was implemented in the first capital prospective payment system final rule published on August 30, 1991. Thus, they believe the overestimation resulted from a change in the rules regarding capital and that the proposed reduction based on a revised FY 1992 capital cost data is not justified. Response: Since the inception of the capital prospective payment system, we have based capital rates on FY 1992 cost levels. We believe it is appropriate for the rate to reflect actual FY 1992 capital spending, even if hospitals had modified capital spending behavior before the current system was implemented. We agree that the prospective payment system provides an incentive for hospitals to modify their capital spending behavior, and that it is likely that hospitals have done so. However, we do not believe that the magnitude of the difference between the projection for FY 1992 capital costs and the latest measurement of FY 1992 capital costs can be completely explained by changes in capital spending behavior caused by the incentives of the prospective payment system. First, most of the capital costs in FY 1992 would be attributable to capital acquired before FY 1992 that was still being depreciated. Second, most capital acquired in FY 1992 would have been planned and committed prior to FY 1992. Thus, only a small proportion of FY 1992 capital spending would have been impacted by the implementation of the capital prospective payment system. Consequently, the implementation of the prospective payment system would have had little, if any, effect on capital growth in FY 1992. Moreover, the anticipated onset of the prospective payment system for capital-related costs may have encouraged some hospitals to limit spending, but we are aware of several situations in which hospitals actually hastened building projects in order to qualify for possible old capital protections. We recognize that CON processes may well delay hospital building projects. However, the commenter does not explain why these effects would have been greater in FY 1992 than in previous years. Our data on the cumulative percentage change in capital-related cost per case, which we presented in the September 1, 1995 final rule (60 FR 45828), demonstrate that the growth of capital costs has slowed considerably in recent years, from a high of 19.9 percent per year in 1986 to a low of 2.9 percent per year in 1992. The most recent FY 1992 HCRIS data available show that hospitals' actual FY 1992 capital costs per discharge are an additional 7.36 percent lower than the estimate on which the capital Federal rate is currently based (taking into consideration the adjustment mandated by Public Law 103-66). We believe it is appropriate for the rate to reflect actual costs. In designing the prospective payment system for capital costs, we recognized the unique position of hospitals in States with CON programs by developing special rules with regard to obligated capital. Those special rules (see Sec. 412.302(c)(2), "Lengthy certificate-of-need process") are designed to ensure that hospitals in States with CON programs receive equitable treatment in terms of recognition "old capital costs." Essentially, this provision permits certain obligated capital costs in CON States to be treated in the same manner as actual capital expenditures in non-CON States. We believe these provisions adequately address the concerns of hospitals in states with CON processes. Finally, we do not agree that the August 30, 1991 final rule implemented any change in the treatment of allowable interest. Section 412.302(b)(2)(v), which defines old capital costs for purposes of the prospective payment system for capital-related costs, states that "Investment income, excluding income from funded depreciation accounts, is used to reduce old capital interest expense based on the ratio of total old capital interest expense to total allowable interest expense in each cost reporting period. "(Emphasis added.) The commenter apparently believes that this statement reflects a change in the treatment of allowable interest because Sec. 413.130(g)(2), which defines capital-related interest expense net of investment income (under our reasonable cost reimbursement rules), provides that in determining the proportion of investment income to be offset, the ratio is to be based on capital-related interest to total interest. However, Sec. 413.130(g) derives from Sec. 413.130(a)(7), and Sec. 413.130(a)(7) addresses only "allowable interest expense" (that is, interest expense as determined under Sec. 413.153), so the ratio expressed in Sec. 413.130(g) is reasonably interpreted to refer to "total allowable interest expense." Comment: Commenters also addressed the possible adjustment based in part on an efficiency analysis. A few commenters stated that higher than expected capital costs per case for FY 1992 were not the result of inefficient use of capital resources, but rather a reaction to pent-up demand in States that had restrictive certificate of need (CON) policies. Another commenter argued that no overexpansion of health facilities has occurred in the commenter's State, because it is highly regulated, and that the average age of hospitals' physical plants in the State is among the oldest in the country. This commenter too believes that it is inappropriate to apply a rate reduction equally in all States. Some commenters agreed with our statement that economic theory would suggest incentives for the overuse of capital during a period in which capital was paid on a cost basis while operating costs were paid on the basis of a prospective rate. However, the commenters contended that economic theory would also suggest that, if hospitals over purchased capital, they conversely had to under employ operating inputs. Thus, the commenters believe that reductions to the capital Federal rate to account for the [[Page 46218]] inefficient overuse of capital should be matched by increases in the operating rates to account for inefficient underutilization of operating inputs. Finally, one commenter suggested that we obtain an independent evaluation of HCFA's capital model and the factors that account for the known increase in costs per case, such as the inflation in capital input prices, quality enhancing intensity increases, and real case-mix growth, as well as the factors that may be responsible for the unexplained growth in capital costs per case. Response: As noted in our September 1, 1995 final rule in response to a similar comment (60 FR 45829), we agree that the conjunction of rate-based payment for operating costs and cost-based payment for capital costs encouraged hospitals to substitute capital inputs for labor and other operating inputs. However, we do not agree that an inefficiently high level of capital inputs under those conditions necessarily implies an inefficiently low level of operating inputs. Rather, the conjunction of rate-based payment for operating costs and cost-based payment for capital could also lead to the substitution of inefficient capital inputs for inefficient operating inputs. Indeed, our previous analysis of efficient operating costs for hospitals during FY 1985 through FY 1991 (57 FR 40014) indicates that operating prospective payments during that period were sufficient for the efficient and cost-effective delivery of quality care. In conjunction with the analysis of capital spending during FY 1985 to FY 1992, these results suggest that hospitals may indeed have responded to the existing incentives by substituting an inefficiently high level of capital inputs for inefficient operating inputs. Under these circumstances, it would not be appropriate to increase operating rates in conjunction with a decrease in capital rates. Decreased capital rates, along with the existing level of operating rates, would provide the appropriate incentives for hospitals to achieve efficient levels of both capital and operating inputs. As we stated in our September 1, 1995 final rule in response to a similar comment (60 FR 45828), our analysis suggests a significant measure of inefficiency in capital costs, and was based on national figures. Therefore, since we are evaluating an efficiency adjustment in the national Federal rate, our analysis does not consider regional differences, such as the existence of CON requirements in some States. The national Federal rate is based on an average; thus, we recognize that some States will have higher costs than the average and other States will have lower costs. We note, however, that although we did not make adjustments for CON policies for purposes of this particular analysis, Sec. 412.302(c)(2) does provide for differential treatment of hospitals in CON States in terms of the recognition of obligated capital (as discussed in more detail above). In response to the commenter's suggestion that a group of independent economists should evaluate the capital model and our theory about the possible cause of the unexplained growth in capital costs per case, we note that ProPAC has also analyzed the current capital rate and has discussed possible reductions to the capital rate, implicitly endorsing a reduction to the capital rate in the order of magnitude that we discussed in the proposed rule. Comment: A number of commenters contended that the reductions discussed in the proposed rule would jeopardize the ability of many hospitals to meet current obligations and reduce their ability to meet future capital needs. Response: Our data indicate that there is ample room to cut the capital rate without a major adverse affect on facilities in any region. Before the implementation of the prospective payment system for capital-related costs, facilities were paid only 85 percent of their capital costs. In the proposed rule, we estimated that payments would exceed capital costs by 9.6 percent in FY 1996 (61 FR 27479). We now estimate that capital payments will exceed capital costs by 8.8 percent in FY 1996 and 7.5 percent in FY 1997. C. Possible Adjustment to Capital Prospective Payment System Minimum Payment Levels Section 412.348(b) of the regulations provides that, during the capital prospective payment system transition period, a hospital may receive an additional payment under an exceptions process if its total inpatient capital-related payments under its payment methodology (that is, fully prospective or hold-harmless) are less than a minimum percentage of its allowable Medicare inpatient capital-related costs. The minimum payment levels are established by class of hospitals under Sec. 412.348(c). The minimum payment levels for portions of cost reporting periods occurring in FY 1996 are: <bullet> Sole community hospitals (located in either an urban or rural area), 90 percent; <bullet> Urban hospitals with at least 100 beds and a disproportionate share patient percentage of at least 20.2 percent and urban hospitals with at least 100 beds that qualify for disproportionate share payments under Sec. 412.106(c)(2), 80 percent; and, <bullet> All other hospitals, 70 percent. Under Sec. 412.348(d), the amount of the exceptions payment is determined by comparing the cumulative payments made to the hospital under the capital prospective payment system to the cumulative minimum payment levels applicable to the hospital for each cost reporting period subject to that system. Any amount by which the hospital's cumulative payments for previous cost reporting periods exceed its cumulative minimum payment is deducted from the additional payment that would otherwise be payable for a cost reporting period. Section 412.348(h) further provides that total estimated exceptions payments under the exceptions process may not exceed 10 percent of the total estimated capital prospective payments (exclusive of hold- harmless payments for old capital) for the same fiscal year. In the final rule implementing the prospective payment system for capital- related costs we stated that the minimum payment levels in subsequent transition years would be revised, if necessary, to keep the projected percentage of payments under the exceptions process at no more than 10 percent of capital prospective payments. In section III of the addendum to the proposed rule (61 FR 27499), we discussed the factors and adjustments used to develop the FY 1997 Federal and hospital-specific rates. In particular, we discussed the FY 1997 exceptions payment reduction factor. This factor adjusts the annual payment rates for the estimated percentage of additional payments for exceptions in FY 1997. In the proposed rule, we estimated that exceptions would equal 6.07 percent of aggregate payments based on the Federal rate and the hospital-specific rate. We indicated that it might be necessary to implement adjustments to the minimum payment levels in the final rule and that it will almost certainly be necessary to adjust the minimum payment levels for FY 1998. We therefore provided public notification that adjustments to the minimum payment levels were imminent, discussed our ideas on the most appropriate method for adjusting the minimum payment levels, and solicited public comment. We stated that, when it does become necessary to adjust the minimum payment levels, we intended to adjust each of the existing levels (that is, 90 percent for sole community hospitals, 80 percent for large urban DSH hospitals, and 70 percent for all other [[Page 46219]] hospitals) by 5 percentage point increments until estimated exceptions payments are within the 10 percent limit. Current estimates indicate that we will not reach the 10 percent exception limit in FY 1997. Therefore, we are not making adjustments to the minimum payment levels at this time; the minimum payment levels for exception payments will remain at the current levels. We received several comments regarding the necessity and methodology of adjustments to the minimum payment levels. Comment: Some commenters objected to the proposed method for handling necessary reductions to the minimum payment levels. One commenter suggested that we develop a more sophisticated methodology that would allow more refined adjustment of the minimum payment levels. Another commenter suggested a 1 or 2 percent reduction increment, rather than the proposed 5 percent increment. Response: As stated above, in this final rule the minimum payment levels for exception payments will remain at the current levels, since our current forecasts indicate that we will not reach the 10 percent limit in FY 1997. All comments received on this issue will be taken under advisement and considered at such time as it becomes necessary to make such an adjustment. Comment: Some commenters believe that HCFA's capital acquisition model (see appendix B to this final rule for a detailed discussion) projects excessive growth in exception payments. These commenters objected to any reduction in the capital minimum payment levels based on projected rapid growth in exceptions and requested further explanation. The commenters further stated that they could not understand why exception payments would be so large when average payments exceed costs. Response: Since payments under the capital prospective payment system are based on averages, not on an individual hospital's costs, some hospitals may receive payments exceeding their costs, while other hospitals may receive payments less than their costs. Even if aggregate payments exceed aggregate costs, some hospitals may have costs so much higher than payments that they qualify for large exceptions payments. It is these large exceptions payments that are driving the aggregate exception payments toward the 10 percent ceiling on exception payments. We have reviewed the cost reports for the first 3 years under the capital prospective payment system. The number of hospitals receiving exceptions payments and the aggregate amount paid for exceptions have increased each year. We expect this trend to continue throughout the transition period, as some hospitals' payments deviate even more from their actual costs. Our model is consistent with these findings. The model projects, as expected, that exceptions payments will continue to grow. "Low cost" hospitals are paid a blend of their hospital-specific rate, and a higher Federal rate. "High cost" hospitals are paid 85 percent of their old capital plus their ratio of new capital to total capital applied to the Federal rate. In both cases, the capital the hospitals had at the time the capital prospective payment system was implemented is addressed by the standard payments. Capital prospective payment rates for FY 1992 were designed to adequately address capital costs that existed at the time the prospective payment system began. Since then, hospitals have acquired additional capital, with some hospitals acquiring more than others. With each passing year, more additional capital is accumulated. In some cases, this additional capital is large, and the affected hospitals' capital costs greatly exceed their standard payments. Exceptions payments mitigate the financial impact on these hospitals. High cost hospitals are more likely to qualify for exceptions payments. Their old capital costs are encompassed in the hold harmless payments, while their new capital costs are reimbursed at a fraction of the Federal rate. If their new capital costs are high, these high cost hospitals will need the full benefit of the exceptions process. Since high cost hospitals will acquire more additional capital over time, more hospitals will qualify for exceptions payments. In fact, high cost hospitals showed rapid growth in exceptions in the first three years under the capital prospective payment system. We expect this rapid growth to continue. Comment: Regarding minimum payment levels, one commenter suggested we reconcile exceptions payments retrospectively and recoup any overpayments on a pro rata basis by reducing future payments to hospitals. The commenter recommends reductions in subsequent Medicare payments to hospitals. Response: Section 412.348(d) states that "Total estimated payments under the exceptions process may not exceed 10 percent of the total estimated capital prospective payments (exclusive of hold-harmless payments for old capital) for the same fiscal year." (Emphasis added.) We believe reconciling actual exceptions payments with estimated exceptions payments on a retroactive basis would fundamentally undermine the prospectivity of the system. Moreover, recouping "overpayments" on a retroactive basis may be potentially unfair to individual hospitals. An individual hospital that qualifies for an exception payment in one year may not also qualify for an exception in the later year in which a "retroactive" exception payment is to be made. Hospitals would not be able to predict the effects of retroactive adjustments to supposedly prospective payment rates. Go to Top VII. Changes for Hospitals and Units Excluded From the Prospective Payment Systems Application of Ceiling in Calculating Payment for Hospital Inpatient Operating Costs (Sec. 413.40 (d) and (g)) Section 1886(b)(1)(B) of the Act provides for an additional payment to a hospital excluded from the prospective payment system when the hospital's reasonable operating costs exceed its target amount. The additional payment is based on the lesser of 50 percent of the amount by which the operating costs exceed the target amount, or 10 percent of the target amount. The Medicare statute further provides that this comparison is made "after any exceptions or adjustments are made to such target amount for any cost reporting period." The regulations, at 42 CFR Sec. 413.40(d)(3), state that the total payment to the hospital for inpatient operating costs (including the additional payment described above) is based on the lesser of the following: the ceiling (target amount multiplied by the number of Medicare discharges) plus 50 percent of the allowable net inpatient operating costs in excess of the ceiling, or 110 percent of the ceiling. However, the regulations do not explicitly include the additional statutory requirement regarding the effect of exceptions or adjustments. As discussed in the proposed rule (61 FR 27481), we understand that there are questions about the calculation of the additional payment under the regulations, which require comparison of two amounts: the "ceiling" plus 50 percent of the difference between allowable costs and the ceiling, and 110 percent of the "ceiling." Specifically, where a hospital has received an adjustment to the target amount under Sec. 413.40(g), there has been confusion as to whether the "ceiling" used for purposes of calculating the additional payment under Sec. 413.40(d) is the unadjusted ceiling (the amount determined without consideration of [[Page 46220]] any adjustments granted to the hospital) or the adjusted ceiling. To address any confusion about these issues, we proposed to revise Sec. 413.40(d)(3) to indicate specifically that calculation of payments for hospital inpatient operating costs under that provision reflects the adjusted ceiling amount (the amount determined after an adjustment under Sec. 413.40(g)). This would apply to all adjustments, including adjustments based on a longer average length of stay in the hospital's rate year as compared to the base year and adjustments for increased routine services. We received only two comments on this proposal. Both commenters supported the proposal, and we will adopt as final the proposed changes to the regulations at Sec. 413.40(d)(3). Go to Top VIII. ProPAC Recommendations As required by law, we reviewed the March 1, 1996 report submitted by ProPAC to Congress and gave its recommendations careful consideration in conjunction with the proposals set forth in the proposed rule. We also responded to the individual recommendations in the proposed rule (61 FR 27482). The comments we received on the treatment of the ProPAC recommendations are set forth below along with our responses to those comments. However, if we received no comments from the public concerning a ProPAC recommendation, we have not repeated the recommendation and response in the discussion below. The update factors for inpatient operating costs and the update factor for hospitals excluded from the prospective payment system and distinct- part units (ProPAC recommendations 10 and 12, respectively) are discussed in Appendix E to this final rule. Capital payment rates (recommendation 11) are discussed in section VI of this final rule. Disproportionate share hospitals (recommendations 17 and 18) are discussed in section V of this final rule. The remaining recommendations on which we received comments are discussed below. A. Discharges From Hospitals to Other Facilities (Recommendation 19) Recommendation: Medicare payments should be modified to account for the shift in services from acute to postacute settings. Broadening the definition of transfer cases, however, is not an appropriate approach. Response in the Proposed Rule: In both the September 1, 1994 and September 1, 1995 final rules, we expressed our concern that the current trend of declining average lengths of stay as hospitals discharge Medicare patients into alternative health care settings (other than acute care prospective payment hospitals) in less time may result in a misalignment of payments and costs under our existing payment systems (59 FR 45362; 60 FR 29221). In particular, we expressed concern over the potential for hospitals paid under the prospective payment system to shift costs (for which they are compensated through the DRG payments) to alternative settings, which are in turn paid on a cost basis. Although we solicited comments on possible solutions to this problem, we did not propose any change in policy. The President's FY 1997 budget includes a proposal to redefine discharges from acute care hospitals to excluded hospitals and units and skilled nursing facilities as transfers for payment purposes. Currently, for cases transferred from one acute care hospital paid under the prospective payment system to another like hospital, the sending hospital is paid a per diem rate instead of the full DRG amount. For cases transferred to an excluded hospital or unit or to a skilled nursing facility (as well as cases discharged home or home with home health care), hospitals receive the full DRG payment amount, regardless of the length of stay in the hospital. Under the per diem transfer payment methodology, hospitals receive a per diem amount (doubled for the first day of the stay) until the full DRG amount is reached. Therefore, under the President's budget proposal, hospitals transferring patients to excluded facilities or skilled nursing facilities prior to the geometric mean length of stay for the DRG, minus one day (to account for the double per diem on the first day), would receive less than the full DRG amount for that case. The basis for ProPAC's opposition to this proposal is that it "* * * thinks this policy would discourage the use of postacute providers. Moreover, it could result in longer inpatient stays, which may not be desirable or cost effective in the long run." We acknowledge that the change in the definition of a transfer is not the ultimate solution to this health care trend. In response to immediate concerns about overpaying hospitals for the reduced services they are providing and the rate of increase in expenditures for postacute care services, however, we believe this is an appropriate interim measure while we continue to explore long-term policy alternatives that will better integrate our payment systems for care provided to Medicare beneficiaries across the acute and postacute care settings. Comment: We received several comments on this response. ProPAC repeated its concern that redefining transfers may not be the right approach, indicating that "(m)ore needs to be known about the relationships among these services before implementing a policy that assumes that hospitals are being overpaid for cases who use post-acute care." Two other commenters expressed their objections to the redefinition of transfers from acute care hospitals. Generally, both of these commenters agreed with ProPAC's assessment that this would lead to longer inpatient stays and discourage the use of postacute care. Also, both commenters objected to ProPAC's suggestion that HCFA bundle acute and postacute care payments. Finally, one commenter recommended that "the total Medicare funding for hospitals be reduced to recognize the shift of patient days away from the hospital setting." Response: We agree with ProPAC that a better understanding of this phenomenon is needed, and we are well aware of the improved efficiency claims made by those who advocate even greater use of postacute care. However, while we continue to explore potential refinements to reflect the shift in services from acute to postacute settings, we believe it is appropriate to concurrently explore interim measures for responding to the undisputed trends showing continuing declining lengths of hospital inpatient stays and increasing postacute care utilization, particularly for certain DRGs. The present overlaps between our acute and postacute payment methodologies demand immediate attention, given our responsibility for preserving the Medicare Trust Fund. We also understand the commenters' concerns about the transfer redefinition. In evaluating any such interim measures two fundamental questions need to be answered: Will this approach protect beneficiaries' access to quality, effective health care and will it adequately compensate the providers of that care for their costs? To the extent that increasing utilization of postacute care allows hospitals to release patients earlier, redefining transfers would better match payments with costs, as well as eliminate some of the potential incentive for premature discharges. With regard to the comments we received about ProPAC's suggestion that bundling might be a potential alternative, we intend to continue to evaluate all potential payment approaches. For example, implementing an offset to the hospital inpatient standardized amounts to reflect cost [[Page 46221]] shifting is another approach under examination. B. Prospective Payment for Postacute Care (Recommendation 20) Recommendation: Prospective payment systems should be implemented for all postacute services. The payment method for each service should be consistent across delivery sites. The Secretary should explore methods to control volume of postacute service use, such as bundling services for a single payment. Response in the Proposed Rule: We agree that HCFA should develop prospective payment systems for all postacute services, and we have made significant progress in this area. As we discuss in our responses to Recommendations 22 and 23, we have developed detailed implementation plans for interim prospective payment systems for skilled nursing facilities (SNFs) and home health agencies (HHAs) that do not require patient classification systems. Execution of these plans will, of course, require legislative action. Beyond our interim plan, we have developed a strategy for developing a full-fledged prospective payment system for SNFs. In the absence of legislation, we have been pursuing data that could be used to support a case-mix prospective payment system through our Multi- State Case Mix Demonstration Project. This demonstration project, now in its operational phase, is collecting data on patient case mix using a modified version of the minimum data set, the assessment tool SNFs use in developing patient care plans. Through the course of the demonstration, we hope to gather data on the full range of SNF resources needed for each resource utilization group. We are proceeding to require by regulation that all facilities provide resident assessment data. Consolidated billing of SNF services (that is, requiring SNFs to bill for all services furnished to their patients) and uniform coding of SNF services are also prerequisites for a SNF prospective payment system. Consolidated billing and uniform coding are needed to determine the appropriate payment for the ancillary services component of SNF services and to provide useful data on the range of services SNFs furnish. We have also been working on a strategy to develop a full-fledged prospective payment system for HHAs. We have funded a project to develop outcome measures for home care that can be used for an outcome- based quality improvement system. These measures will be based largely on a core standard assessment data set that includes items measuring sociodemographic, environmental, support system, health status, functional status, and health service utilization characteristics of patients. Many of the data items included in the core standard assessment data set are not only essential for assessing patient outcomes but are also critical for designing an adequate case-mix system for payment purposes. To test and refine Medicare's approach to outcome based quality improvement for home health care, HCFA is currently sponsoring the Medicare Quality Assurance and Improvement Demonstration, which uses this instrument. We plan to publish regulations identifying the required data elements and addressing the collection of information from the core standard assessment data set. We also plan to sponsor additional research that would lead to an appropriate case mix adjuster that can be used in a national prospective payment system. In addition to the developmental work underway on SNF and HHA prospective payment systems, we have begun work on the preliminary steps necessary for the development of a prospective payment system for hospital inpatient rehabilitation services. The biggest obstacle we have faced in this effort is the lack of appropriate patient classification systems for the types of patients treated by rehabilitation hospitals. We have recently contracted with the Rand Corporation to evaluate a rehabilitation coding system known as the Functional Independence Measure (FIM), which is a scoring system that measures the degree of functional independence of rehabilitation patients. These researchers will also evaluate the patient classification system known as function related groups (FRGs), which are based on the FIM, as a possible basis for a Medicare prospective payment system for rehabilitation services. If the research confirms functional status measures can be used to develop an appropriate patient classification system, we will begin the additional work necessary to put a prospective payment system into place. This would require collecting patient assessment data from Medicare rehabilitation hospitals and units and developing all the necessary components of the new payment system. It will take at least 3 years to design and implement such a system. To facilitate implementation, we are considering initiating collection of patient assessment data in advance of legislation establishing a prospective payment system. We will be seeking public input on whether to proceed with a requirement for patient assessment data in the absence of legislation and what data elements should be included in a core data set that could be used not only as the basis for a patient classification system but also to assess outcomes. We recognize that there are advantages to a coordinated approach in developing prospective payment systems for postacute services and we will be evaluating how to make them as consistent as possible. We also recognize that the demand for implementation of prospective payment systems for postacute services is sufficiently immediate so that there may not be time for the broad study, data collection, and research needed to develop a "unified" system using similar resource grouping principles. Most of the current legislative proposals, including the Administration's proposals, would require implementation dates within the next several years. It may not be feasible to develop a "unified" system within the time frames contemplated by the current legislative proposals. Trade-offs may be required between continuation of the interim payment systems versus the prospective payment systems on one hand, and the separate versus "unified" prospective payment systems on the other hand. Comment: One commenter strongly supported adoption of a prospective payment system for inpatient rehabilitation and believes that the RAND research project will likely produce such a system. The commenter noted that we are considering initiating the collection of patient assessment data in advance of legislation establishing a prospective payment system and urged us to begin collecting the data at the earliest possible date. The commenter believes that imposition of a reporting requirement based on the FIM should not be a great burden on the industry since rehabilitation hospitals and units are already using the FIM or similar patient evaluation measures. Systematizing collection of such data would expedite introduction of a prospective payment system based on FRGs and would considerably reduce the 3-year minimum implementation period suggested in HCFA's response in the proposed rule. The commenter also urged, as a means toward developing a payment system that is consistent across payment sites based on patient characteristics, that HCFA expand the RAND research project to determine the feasibility of using an FRG-based payment system for rehabilitation patients in skilled nursing facilities. [[Page 46222]] Response: Since the collection of patient assessment data in advance of legislation establishing a prospective payment system would expedite implementation of the system, we are exploring whether we can initiate the collection of data from rehabilitation facilities without legislative action. Our estimate of 3 years to design and implement a payment system includes beginning data collection at the earliest possible time and continuing the collection over a period sufficient to ensure the validity and stability of the components of a payment system, such as payment rates, relative weights of patient groups, outlier payments, and facility payment adjustments, in addition to ensuring the validity of coding within and across hospitals. We agree with the commenter that, as a step toward developing a payment system that is consistent across delivery sites, it would be desirable to explore the usefulness of FRGs in a payment system for rehabilitation services in skilled nursing facilities. We will, therefore, evaluate our ability to expand the RAND project given the limits of available resources. We note that we are also engaged in research on other case-mix measures for SNF and home health services and we will investigate the suitability of these measures for rehabilitation hospital services. C. Case-Mix Measures for Postacute Services (Recommendation 21) Recommendation: Reliable case-mix measurement is important in prospective payment systems to account for resource use and to analyze treatment patterns and costs across sites. The Secretary should coordinate case-mix research across postacute care settings, using consistent methods for measuring patient acuity and resource use. Response in the Proposed Rule: We are attempting to coordinate our work on case-mix adjustment for home health care, long-term and SNF care, and rehabilitative services. To develop a case-mix adjustment system for SNF care, time studies were conducted in order to measure resource utilization. Similarly, as noted above in response to Recommendation 20, we have funded a new home health case-mix study. In addition, in the case-mix work to date for both home health care and SNF care, dependence in activities of daily living is the biggest predictor of resource utilization. Some of the other predictors differ across SNF care and home health care due to differences in the treatment settings and the availability of information for a classification system. As also noted in the preceding response, researchers at the University of Pennsylvania have developed a classification system based on FIMs called Function Related Groups (FIM-FRGs). This system appears promising for use in a case-mix adjusted prospective payment system for rehabilitation and long-term care facilities, and we are working with the Rand Corporation on a research project to evaluate the suitability of FIM-FRGs for this purpose. We agree that a compatible cross-provider measure of resource use would be the best multiplier in any universal postacute system. We also believe that such measures do not now exist and to produce them would require the program to incur significant costs and impose significant data reporting and collection requirements on providers. We would prefer to obtain explicit legislative direction before we incur these costs and impose these burdens. Even so, we believe several years would be required to gather the data and develop the case-mix measures. For these reasons, we believe that interim prospective payment systems of the types contained in the President's FY 1997 budget should be put in place. Comment: One commenter agreed with ProPAC's recommendation to develop a unified case-mix prospective payment system for postacute care, but expressed concern that such a prospective payment system based on ICD-9-CM codes will require the development of uniform coding guidelines that do not currently exist. Response: We have not yet decided whether it would be appropriate to use ICD-9-CM codes in connection with a postacute prospective payment system. We will keep the overall concern of uniform coding guidelines in mind as we progress in our evaluation of postacute prospective payment. D. Update to the Composite Rate for Dialysis Services (Recommendation 24) Recommendation: The Secretary should develop methods to control total Medicare per capita expenditures for end stage renal disease (ESRD) beneficiaries. In the meantime, the composite rate should be updated by 2.7 percent for hospital-based dialysis facilities and by 2.0 percent for freestanding facilities for fiscal year 1997. The Secretary should also develop reliable measures of patient severity and outcomes to analyze the relationships among treatment processes, patient outcomes, and costs. These factors should be considered in evaluating the need for and the level of future payment updates. Response in the Proposed Rule: One of ProPAC's suggestions is that HCFA consider opening enrollment for ESRD beneficiaries to participate in Medicare risk programs. The reason for this recommendation is the rapid growth in total Medicare spending for ESRD beneficiaries. A large part of this increase is attributable to the expanding ESRD population, especially older patients who require more services. These beneficiaries are using more acute inpatient, skilled nursing and other dialysis-related services than ever before. ProPAC suggests that to control these expenditures, Medicare examine the possibility of adopting a capitation payment system for ESRD services, since capitation rates have been successful in controlling expenditure growth for other populations. At a minimum, they are recommending that utilization review or other managed care techniques be used to control the total volume of services provided to ESRD beneficiaries across all sites of care. Section 1876(d) of the Act currently prevents an individual with ESRD from enrolling in an HMO or a competitive medical plan. However, an individual who is enrolled in a prepaid health plan when he or she is determined to have ESRD may continue enrollment in that plan. A prepaid health plan may only disenroll a beneficiary as provided by regulations at Sec. 417.460. Congress addressed the issue of paying for ESRD services in a capitation setting in legislation. Section 13567(b) of the Omnibus Budget Reconciliation Act of 1993 (Pub. L. 103-66) (August 10, 1993) amended section 2355 of Public Law 98-369 by requiring the Secretary to include the integration of acute and chronic care management for patients with ESRD through expanded community care case management services in a social health maintenance organization (SHMO). Initial legislation required the Secretary to grant demonstration waivers for SHMOs that provide for the integration of health and social services at a fixed annual prepaid capitation rate. In the January 26, 1996 Federal Register, we published a notice informing interested parties of the opportunity to apply for funds for a cooperative agreement to operate an ESRD Managed Care Demonstration (61 FR 2516). Two of the demonstration's purposes would be to test whether ESRD beneficiaries can and should be given access to HMOs during open enrollment and whether the statewide capitation rate can and should be adjusted. The demonstration would adjust rates for treatment status (such as dialysis, transplant, or a functioning graft), age groups and the cause of renal [[Page 46223]] failure (for example, diabetes). As the legislation requires, rates would be based on 100 percent of the adjusted average per capita costs (AAPCC); additional non-Medicare-covered benefits would be offered by the provider to justify the additional 5 percent beyond the 95 percent of the AAPCC paid to Medicare risk-contracting HMOs on behalf of ESRD enrollees. Based on the results of this demonstration, we would make recommendations to Congress concerning the appropriateness of paying for dialysis services in a capitation setting. To improve the quality of care ESRD patients are receiving, we are in the process of developing proposed rules for ESRD conditions for coverage. The essence of the regulation is patient-centered and outcome-oriented. The proposed conditions for coverage will focus on facilities achieving an optimal level of health and well-being for all dialysis patients. The proposed rules will be published in Spring 1996 with expected implementation in late fiscal year 1997. While we share ProPAC's concern that payment rates be sufficient to assure quality care for ESRD patients, we do not believe there is sufficient evidence at this point to conclude that more money is needed to provide appropriate care. Currently, the University of Michigan, as part of a National Institute of Health grant, is examining the relationship between facilities' costs and the level of KT/V. Also the National Institute of Diabetes and Digestive and Kidney Diseases is sponsoring a study on the impact of increasing dialysis as measured by KT/V and the use of high-flux-dialysis on ESRD patients. The results of these studies should help us analyze the relationship between patient outcomes and costs, and thus provide us with a basis for recommending an appropriate payment rate increase. While we acknowledge that an increase in the composite rate may be appropriate in the next few years, we believe that any rate increase should be linked to implementation of the revised conditions for coverage. Moreover, any ESRD rate increase must be considered within the context of Medicare budgetary concerns and should have a direct link to improved patient outcomes. We will continue to monitor ESRD facility costs, and, if appropriate, we may recommend an update to the ESRD composite rate for FY 1998. We note that ProPAC's recommendation provides for an across-the- board rate increase for all renal facilities. However, data show that high volume independent facilities (over 6,000 treatments per year) account for about 85 percent of independent dialysis treatments. These high volume facilities report margins between Medicare payments and costs that are higher than average. Therefore, in proposing a future rate increase, we would want to examine the need to adjust payment increases for volume. In addition, we believe that any update to the composite rate should include an update to the wage index currently used to adjust the labor portion of the rate. We are currently using an outdated wage index which is a blend of 1980 Bureau of Labor Statistics (BLS) and 1984 prospective payment system wage data and does not reflect the MSA revisions resulting from the 1990 census. The Commission's final recommendation is that the Secretary closely monitor treatment patterns and patient outcomes to ensure that facilities use the payment increase to improve quality of care. The proposed ESRD conditions for coverage should address this issue. We expect the proposed rule to be published in the Federal Register before Summer 1996. Between the publication of the proposed and final rules, HCFA is planning to meet with the renal community to develop complete clinical data sets to monitor patient outcomes and medical conditions. These data will then be used to evaluate the quality of dialysis services furnished by individual facilities. Of course, this is a long- term project. In the short term, we are exploring the possibility of collecting limited patient outcome data such as KT/V and URR. Comment: One commenter and the Commission reiterated that ProPAC's recommended update framework was appropriate. According to ProPAC, its analysis suggests that input costs are rising and large productivity gains may no longer be possible. Consequently, renal facilities may be unable to continue to provide quality dialysis without some payment increase. Response: As discussed above, we recognize that an increase in the composite payment rate may be appropriate in the future, but we believe that any rate increase should be linked to implementation of the revised conditions for coverage for ESRD facilities. Until such implementation, we will continue to monitor facility costs and other factors to determine if it is appropriate to recommend a payment rate increase. At this time, the composite payment rate is set by statute. Go to Top IX. Other Required Information A. Paperwork Reduction Act The Paperwork Reduction Act of 1995 provides for notice and comment when a collection of information requirement is submitted to the Office of Management and Budget (OMB) for review and approval. In order to fairly evaluate whether an information collection should be approved by OMB, section 3506(c)(2)(A) of the Paperwork Reduction Act of 1995 requires that we solicit comment on the following issues: <bullet> Whether the information collection is necessary and useful to carry out the proper functions of the agency; <bullet> The accuracy of the agency's estimate of the information collection burden; <bullet> The quality, utility, and clarity of the information to be collected; and <bullet> Recommendations to minimize the information collection burden on the affected public, including automated collection techniques. Therefore, in the proposed rule, we solicited public comment on each of these issues for the information collection requirement discussed below. The only information collection or paperwork burden item contained in the FY 1997 proposed or final rules involves the requirement under Sec. 489.27 that a hospital furnish each Medicare beneficiary with a notice of discharge rights supplied by HCFA, that is, "An Important Message from Medicare." As discussed in section V of this preamble, we are revising the current requirement that a hospital must distribute the "Important Message" to each Medicare beneficiary at or about the time of admission. In order to permit hospitals more flexibility, but still ensure that beneficiaries are aware of their discharge rights, we are revising Sec. 489.27 to specify that a hospital must provide the notice of discharge rights "during the course of the hospital stay." We estimated that the paperwork burden associated with the requirement that hospital personnel distribute the "Important Message" to each Medicare beneficiary is approximately 1 minute per admission. Based on our most recent available data (1995 Data Compendium, HCFA Pub. No. 03364), there are approximately 11 million Medicare beneficiaries admitted to hospitals each year, resulting in an annual burden of approximately 183,000 hours. This paperwork burden is not effective until it has been approved by OMB. A notice will be published in the Federal Register when approval is obtained. [[Page 46224]] B. Requests for Data From the Public In order to respond promptly to public requests for data related to the prospective payment system, we have set up a process under which commenters can gain access to the raw data on an expedited basis. Generally, the data are available in computer tape format or cartridges; however, some files are available on diskette, and on the internet at HTTP://WWW.HCFA.GOV/STATS/PUBFILES.HTML. In our May 31, 1996 proposed rule, we published a list of data sets that are available for purchase (61 FR 27490). C. Waiver of Notice of Proposed Rulemaking We ordinarily publish a notice of proposed rulemaking for a rule to provide a period for public comment. However, we may waive that procedure if we find good cause that prior notice and comment are impracticable, unnecessary, or contrary to public interest. Most provisions of this final rule were directly addressed in the May 31, 1996 proposed rule (61 FR 27444) or were made in response to comments on that proposed rule. The only issue raised in this final rule for which we have not provided an opportunity for notice and comment concerns a recently enacted statutory provision. On April 26, 1996, Congress enacted the Omnibus Consolidated Rescissions and Appropriations Act of 1996. Among other things, the new statute requires that, for certain purposes, the Federal Government "shall deem accredited any postgraduate physician training program that would be accredited but for the accrediting agency's reliance upon an accreditation standard that requires an entity to perform an induced abortion or require, provide, or refer for training in the performance of induced abortions, or make arrangements for such training, regardless of whether such standard provides exceptions or exemptions." In this final rule, we are revising the regulations at Sec. 412.105 and Sec. 413.86 to conform the regulations to the new statutory provision. We find good cause to waive the procedure for notice and comment with respect to these conforming changes. We find that the procedure for notice and comment is unnecessary because these technical changes merely conform the regulations text to the express requirements of the statute and do not involve an exercise of agency discretion; moreover, delaying these technical changes would be contrary to the public interest because any perceived discrepancy between the regulations and the statute might cause confusion. List of Subjects 42 CFR Part 412 Administrative practice and procedure, Health facilities, Medicare, Puerto Rico, Reporting and recordkeeping requirements. 42 CFR Part 413 Health facilities, Kidney diseases, Medicare, Puerto Rico, Reporting and recordkeeping requirements. 42 CFR Part 489 Health facilities, Medicare. 42 CFR chapter IV is amended as set forth below: A. Part 412 is amended as follows: PART 412--PROSPECTIVE PAYMENT SYSTEMS FOR INPATIENT HOSPITAL SERVICES 1. The authority citation for part 412 continues to read as follows: Authority: Secs. 1102 and 1871 of the Social Security Act (42 U.S.C. 1302 and 1395hh). Subpart D--Basic Methodology for Determining Prospective Payment Federal Rates for Inpatient Operating Costs 2. In Sec. 412.63(s)(1), a new sentence is added at the end to read as follows: Sec. 412.63 Federal rates for inpatient operating costs for fiscal years after Federal fiscal year 1984. * * * * * (s) * * * (1) * * * The wage index is updated annually. * * * * * Subpart G--Special Treatment of Certain Facilities Under the Prospective Payment System for Inpatient Operating Costs 3. In Sec. 412.105, the introductory text of both paragraph (g)(1) and paragraph (g)(1)(i) is republished and a new paragraph (g)(1)(i)(D) is added to read as follows: Sec. 412.105 Special treatment: Hospitals that incur indirect costs for graduate medical education programs. * * * * * (g) Determining the total number of full-time equivalent residents for cost reporting periods beginning on or after July 1, 1991. (1) For cost reporting periods beginning on or after July 1, 1991, the count of full-time equivalent residents for the purpose of determining the indirect medical education adjustment is determined as follows: (i) The resident must be enrolled in an approved teaching program. An approved teaching program is one that meets one of the following requirements: * * * * * (D) Is a program that would be accredited except for the accrediting agency's reliance upon an accreditation standard that requires an entity to perform an induced abortion or require, provide, or refer for training in the performance of induced abortions, or make arrangements for such training, regardless of whether the standard provides exceptions or exemptions. * * * * * Subpart L--The Medicare Geographic Classification Review Board 4. In Sec. 412.246, paragraph (b) is revised to read as follows: Sec. 412.246 MGCRB members. * * * * * (b) Term of office. The term of office for an MGCRB member may not exceed 3 years. A member may serve more than one term. The Secretary may terminate a member's tenure prior to its full term. Subpart M--Prospective Payment System for Inpatient Hospital Capital Costs 5. In Sec. 412.302, paragraph (d)(1) is revised and a new paragraph (d)(4) is added to read as follows: Sec. 412.302 Introduction to capital costs. * * * * * (d) Consistency in cost reporting--(1) General rule. For cost reporting periods beginning on or after October 1, 1991, and before October 1, 2001, the hospital must follow consistent cost finding methods for classifying and allocating capital-related costs, except as otherwise provided in paragraph (d)(4) of this section. * * * * * (4) Hospitals may elect the simplified cost allocation methodology under the terms and conditions provided in the instructions for HCFA Form 2552. B. Part 413 is amended as follows: [[Page 46225]] PART 413--PRINCIPLES OF REASONABLE COST REIMBURSEMENT; PAYMENT FOR END-STAGE RENAL DISEASE SERVICES; OPTIONAL PROSPECTIVELY DETERMINED PAYMENT RATES FOR SKILLED NURSING FACILITIES 1. The authority citation for part 413 continues to read as follows: Authority: Secs. 1102, 1861(v)(1)(A), and 1871 of the Social Security Act (42 U.S.C. 1302, 1395x(v)(1)(A), and 1395hh). Subpart C--Limits on Cost Reimbursement 2. In Sec. 413.40, paragraph (d)(3) is revised to read as follows: Sec. 413.40 Ceiling on the rate of increase in hospital inpatient costs. * * * * * (d) * * * (3) Net inpatient operating costs are greater than the ceiling. For cost reporting periods beginning on or after October 1, 1991, if a hospital's allowable net inpatient operating costs exceed the hospital's ceiling (or the adjusted ceiling, if applicable), payment will be based on the lower of the-- (i) Ceiling (or the adjusted ceiling, if applicable) plus 50 percent of the allowable net inpatient operating costs in excess of the ceiling (or the adjusted ceiling, if applicable); or (ii) One hundred-ten percent of the ceiling (or the adjusted ceiling, if applicable). * * * * * Subpart F--Specific Categories of Costs 3. In Sec. 413.86, under paragraph (b), the definition of "Approved geriatric program" is revised and a new paragraph (4) is added to the definition of "Approved medical residency program" and a new sentence is added at the end of paragraph (g)(1) introductory text to read as follows: Sec. 413.86 Direct graduate medical education payments. * * * * * (b) Definitions. * * * * * Approved geriatric program means a fellowship program of one or more years in length that is approved by the Accreditation Council for Graduate Medical Education (ACGME) under the ACGME's criteria for geriatric fellowship programs. Approved medical residency program * * * (4) Is a program that would be accredited except for the accrediting agency's reliance upon an accreditation standard that requires an entity to perform an induced abortion or require, provide, or refer for training in the performance of induced abortions, or make arrangements for such training, regardless of whether the standard provides exceptions or exemptions. * * * * * (g) * * * (1) * * * For combined residency programs, an initial residency period is defined as the time required for individual certification in the longer of the programs. * * * * * C. Part 489 would be amended as follows: PART 489--PROVIDER AGREEMENTS AND SUPPLIER APPROVAL 1. The authority citation for part 489 continues to read as follows: Authority: Secs. 1102, and 1871 of the Social Security Act (42 U.S.C. 1302, and 1395hh). Subpart B--Essentials of Provider Agreements 2. Section 489.27 is revised to read as follows: Sec. 489.27 Beneficiary notice of discharge rights. A hospital that participates in the Medicare program must furnish each Medicare beneficiary, or an individual acting on his or her behalf, the notice of discharge rights HCFA supplies to the hospital to implement section 1886(a)(1)(M) of the Act. The hospital must provide timely notice during the course of the hospital stay. For purposes of this paragraph, the course of the hospital stay may begin with the provision of a package of information regarding scheduled preadmission testing and registration for a planned hospital admission. The hospital must be able to demonstrate compliance with this requirement. (Catalog of Federal Domestic Assistance Program No. 93.773, Medicare--Hospital Insurance; and Program No. 93.774, Medicare-- Supplementary Medical Insurance Program) Dated: August 23, 1996. Bruce C. Vladeck, Administrator, Health Care Financing Administration. Dated: August 23, 1996. Donna E. Shalala, Secretary. Go to Top [Note: The following addendum and appendixes will not appear in the Code of Federal Regulations.] Addendum--Schedule of Standardized Amounts Effective With Discharges On or After October 1, 1996 and Update Factors and Rate- of-Increase Percentages Effective With Cost Reporting Periods Beginning On or After October 1, 1996 I. Summary and Background In this addendum, we are setting forth the amounts and factors for determining prospective payment rates for Medicare inpatient operating costs and Medicare inpatient capital-related costs. We are also setting forth rate-of-increase percentages for updating the target amounts for hospitals and hospital units excluded from the prospective payment system. For discharges occurring on or after October 1, 1996, except for sole community hospitals and hospitals located in Puerto Rico, each hospital's payment per discharge under the prospective payment system will be based on 100 percent of the Federal national rate. Sole community hospitals are paid based on whichever of the following rates yields the greatest aggregate payment: the Federal national rate, the updated hospital-specific rate based on FY 1982 cost per discharge, or the updated hospital-specific rate based on FY 1987 cost per discharge. For hospitals in Puerto Rico, the payment per discharge is based on the sum of 75 percent of a Puerto Rico rate and 25 percent of a national rate (section 1886(d)(9)(A) of the Act). As discussed below in section II, we are making changes in the determination of the prospective payment rates for Medicare inpatient operating costs. The changes, to be applied prospectively, will affect the calculation of the Federal rates. In section III, we discuss changes we are making in determining the prospective payment rates for Medicare inpatient capital-related costs. Section IV sets forth our changes for determining the rate-of-increase limits for hospitals excluded from the prospective payment system. The tables to which we refer in the preamble to this final rule are presented at the end of this addendum in section V. Go to Top II. Changes to Prospective Payment Rates for Inpatient Operating Costs for FY 1997 The basic methodology for determining prospective payment rates for inpatient operating costs is set forth at Sec. 412.63 for hospitals located outside of Puerto Rico. The basic methodology for determining the prospective payment rates for inpatient operating [[Page 46226]] costs for hospitals located in Puerto Rico is set forth at Secs. 412.210 and 412.212. Below, we discuss the manner in which we are changing some of the factors used for determining the prospective payment rates. The Federal and Puerto Rico rate changes are effective with discharges occurring on or after October 1, 1996. As required by section 1886(d)(4)(C) of the Act, we must also adjust the DRG classifications and weighting factors for discharges in FY 1997. In summary, the standardized amounts set forth in Tables 1a and 1c of section V of this addendum reflect-- <bullet> Updates of 2.0 percent for all areas (that is, the market basket percentage increase of 2.5 percent minus 0.5 percentage points); <bullet> An adjustment to ensure budget neutrality as provided for in sections 1886(d)(4)(C)(iii) and (d)(3)(E) of the Act by applying new budget neutrality adjustment factors to the large urban and other standardized amounts; <bullet> An adjustment to ensure budget neutrality as provided for in section 1886(d)(8)(D) of the Act by removing the FY 1996 budget neutrality factor and applying a revised factor; and <bullet> An adjustment to apply the revised outlier offset by removing the FY 1996 outlier offsets and applying a new offset. A. Calculation of Adjusted Standardized Amounts 1. Standardization of Base-Year Costs or Target Amounts Section 1886(d)(2)(A) of the Act required the establishment of base-year cost data containing allowable operating costs per discharge of inpatient hospital services for each hospital. The preamble to the September 1, 1983 interim final rule (48 FR 39763) contains a detailed explanation of how base-year cost data were established in the initial development of standardized amounts for the prospective payment system and how they are used in computing the Federal rates. Section 1886(d)(9)(B)(i) of the Act required that Medicare target amounts be determined for each hospital located in Puerto Rico for its cost reporting period beginning in FY 1987. The September 1, 1987 final rule contains a detailed explanation of how the target amounts were determined and how they are used in computing the Puerto Rico rates (52 FR 33043, 33066). The standardized amounts are based on per discharge averages of adjusted hospital costs from a base period or, for Puerto Rico, adjusted target amounts from a base period, updated and otherwise adjusted in accordance with the provisions of section 1886(d) of the Act. Sections 1886(d)(2)(C) and (d)(9)(B)(ii) of the Act required that the updated base-year per discharge costs and, for Puerto Rico, the updated target amounts, respectively, be standardized in order to remove from the cost data the effects of certain sources of variation in cost among hospitals. These include case mix, differences in area wage levels, cost of living adjustments for Alaska and Hawaii, indirect medical education costs, and payments to hospitals serving a disproportionate share of low-income patients. Since the standardized amounts have already been adjusted for differences in case mix, wages, cost-of-living, indirect medical education costs, and payments to hospitals serving a disproportionate share of low-income patients, no additional adjustments for these factors for FY 1997 were made. That is, the standardization adjustments reflected in the FY 1997 standardized amounts are the same as those reflected in the FY 1996 standardized amounts. Under sections 1886(d)(2)(H) and (d)(3)(E) of the Act, in making payments under the prospective payment system, the Secretary estimates from time to time the proportion of costs that are wages and wage- related costs. Since October 1, 1990, when the market basket was last rebased, we have considered 71.4 percent of costs to be labor-related for purposes of the prospective payment system. As discussed in section IV of the preamble, we are using a rebased market basket effective for FY 1997. Based on the rebased market basket, we are revising the labor and nonlabor proportions of the standardized amounts. Effective with discharges occurring on or after October 1, 1996, we are establishing a labor-related proportion of 71.2 percent and a nonlabor- related proportion of 28.8 percent. The standardized amounts in Table 1a of section V of this addendum have been recomputed to reflect the revised labor-related and nonlabor-related proportions. (We are revising the Puerto Rico standardized amounts by the average labor share in Puerto Rico of 82.8 percent. We are also revising the discharged-weighted national standardized amount to reflect the proportion of discharges in large urban and other areas from the FY 1995 MedPAR file.) 2. Computing Large Urban and Other Averages Within Geographic Areas Section 1886(d)(3) of the Act requires the Secretary to compute two average standardized amounts for discharges occurring in a fiscal year: one for hospitals located in large urban areas and one for hospitals located in other areas. In addition, under sections 1886(d)(9)(B)(iii) and (C)(i) of the Act, the average standardized amount per discharge must be determined for hospitals located in urban and other areas in Puerto Rico. Hospitals in Puerto Rico are paid a blend of 75 percent of the applicable Puerto Rico standardized amount and 25 percent of a national standardized payment amount. Section 1886(d)(2)(D) of the Act defines "urban areas" as those areas within a Metropolitan Statistical Area (MSA). A "large urban area" is defined as an urban area with a population of more than 1,000,000. In addition, section 4009(i) of Public Law 100-203 provides that a New England County Metropolitan Area (NECMA) with a population of more than 970,000 is classified as a large urban area. As required by section 1886(d)(2)(D) of the Act, population size is determined by the Secretary based on the latest population data published by the Bureau of the Census. Urban areas that do not meet the definition of a "large urban area" are referred to as "other urban areas." Areas that are not included in MSAs are considered "rural areas" under section 1886(d)(2)(D). Payment for discharges from hospitals located in large urban areas will be based on the large urban standardized amount. Payment for discharges from hospitals located in other urban and rural areas will be based on the other standardized amount. Based on 1995 population estimates published by the Bureau of the Census, 56 areas meet the criteria to be defined as large urban areas for FY 1997. These areas are identified by an asterisk in Table 4a. Table 1a contains the two national standardized amounts that are applicable to all hospitals, except for sole community hospitals and hospitals in Puerto Rico. For a number of years, Table 1b had been used to set forth the 18 regional standardized amounts applicable for hospitals located in census areas subject to the regional floor. However, as provided in section 1886(d)(1)(A)(iii)(II) of the Act, the regional floor expires effective with discharges occurring on or after October 1, 1996. Therefore, all hospitals (except sole community hospitals and hospitals in Puerto Rico) will be paid solely on the basis of the national standardized amounts. Under section 1886(d)(9)(A)(ii) of the Act, the national standardized payment amount applicable to hospitals in Puerto Rico consists of the discharge- weighted average of the national large urban standardized amount and the national [[Page 46227]] other standardized amount (as set forth in Table 1a). The national average standardized amount for Puerto Rico is set forth in Table 1c. This table also includes the two standardized amounts that will be applicable to most hospitals in Puerto Rico. We note that on June 28, 1996, the Office of Management and Budget announced the designation of the Pocatello, Idaho MSA and the Jonesboro, Arkansas MSA. In addition, Chester County was added to the Jackson, Tennessee MSA. We have incorporated these changes in this final rule. 3. Updating the Average Standardized Amounts In accordance with section 1886(d)(3)(A)(iv) of the Act, we are updating the large urban and the other areas average standardized amounts for FY 1997 using the applicable percentage increases specified in section 1886(b)(3)(B)(i) of the Act. Section 1886(b)(3)(B)(i)(XII) of the Act specifies that, for hospitals in all areas, the update factor for the standardized amounts for FY 1997 is the market basket percentage increase minus 0.5 percentage points. The percentage change in the market basket reflects the average change in the price of goods and services purchased by hospitals to furnish inpatient care. The most recent forecast of the rebased hospital market basket increase for FY 1997 is 2.5 percent. For FY 1997, this yields an update to the average standardized amounts of 2.0 percent (2.5 percent minus 0.5 percent). (See section IV of the preamble to this final rule for a discussion of the market basket rebasing.) As in the past, we are adjusting the FY 1996 standardized amounts to remove the effects of the FY 1996 geographic reclassifications and outlier payments before applying the FY 1997 updates. That is, we are increasing the standardized amounts to restore the reductions that were made for the effects of geographic reclassification and outliers. After including the FY 1997 offsets to the standardized amounts for outliers and geographic reclassification, we estimate that there will be an actual increase of 1.8 percent to the large urban and other area standardized amounts. We note that the FY 1996 standardized amounts reflected a budget neutrality factor of 0.997575 to account for the change in transfer payment policy implemented in FY 1996. See 60 FR 45854. In the proposed rule we stated that "there will be no need for a further budget neutrality adjustment" (61 FR 27573), but we incorrectly suggested that the FY 1996 budget neutrality adjustment for transfers should be removed in setting the FY 1997 rates. The budget neutrality adjustment for the transfer policy is built permanently into the unadjusted rates. Although the update factor for FY 1997 is set by law, we were required by section 1886(e)(3)(B) of the Act to report to Congress on our initial recommendation of update factors for FY 1997 for both prospective payment hospitals and hospitals excluded from the prospective payment system. For general information purposes, we published the report to Congress as Appendix D to the proposed rule. That recommendation was based on an earlier forecast of the market basket increase. Our final recommendation on the update factors (which is required by sections 1886(e)(4)(A) and (e)(5)(A) of the Act) is set forth as Appendix D to this final rule. 4. Other Adjustments to the Average Standardized Amounts a. Recalibration of DRG Weights and Updated Wage Index--Budget Neutrality Adjustment.--Section 1886(d)(4)(C)(iii) of the Act specifies that beginning in FY 1991, the annual DRG reclassification and recalibration of the relative weights must be made in a manner that ensures that aggregate payments to hospitals are not affected. As discussed in section II of the preamble, we normalized the recalibrated DRG weights by an adjustment factor, so that the average case weight after recalibration is equal to the average case weight prior to recalibration. Section 1886(d)(3)(E) of the Act specifies that the hospital wage index must be updated on an annual basis beginning October 1, 1993. This provision also requires that any updates or adjustments to the wage index must be made in a manner that ensures that aggregate payments to hospitals are not affected by the change in the wage index. To comply with the requirement of section 1886(d)(4)(C)(iii) of the Act that DRG reclassification and recalibration of the relative weights be budget neutral, and the requirement in section 1886(d)(3)(E) of the Act that the updated wage index be budget neutral, we compared aggregate payments using the FY 1996 relative weights and wage index to aggregate payments using the FY 1997 relative weights and wage index. The same methodology was used for the FY 1996 budget neutrality adjustment. (See the discussion in the September 1, 1992 final rule (57 FR 39832).) Based on this comparison, we computed a proposed budget neutrality adjustment factor equal to 0.998509. Based on the final FY 1997 relative weights and wage index, the final budget neutrality adjustment factor is 0.998703. This budget neutrality adjustment factor is applied to the standardized amounts without removing the effects of the FY 1996 budget neutrality adjustment. We do not remove the prior budget neutrality adjustment because estimated aggregate payments after the changes in the DRG relative weights and wage index should equal estimated aggregate payments prior to the changes. If we removed the prior year adjustment, we would not satisfy this condition. In addition, we will continue to apply the same FY 1997 adjustment factor to the hospital-specific rates that are effective for cost reporting periods beginning on or after October 1, 1996, in order to ensure that we meet the statutory requirement that aggregate payments neither increase nor decrease as a result of the implementation of the FY 1997 DRG weights and updated wage index. (See the discussion in the September 4, 1990 final rule (55 FR 36073).) b. Reclassified Hospitals--Budget Neutrality Adjustment.--Section 1886(d)(8)(B) of the Act provides that certain rural hospitals are deemed urban effective with discharges occurring on or after October 1, 1988. In addition, section 1886(d)(10) of the Act provides for the reclassification of hospitals based on determinations by the Medicare Geographic Classification Review Board (MGCRB). Under section 1886(d)(10) of the Act, a hospital may be reclassified for purposes of the standardized amount or the wage index, or both. Under section 1886(d)(8)(D) of the Act, the Secretary is required to adjust the standardized amounts so as to ensure that total aggregate payments under the prospective payment system after implementation of the provisions of sections 1886(d)(8) (B) and (C) and 1886(d)(10) of the Act are equal to the aggregate prospective payments that would have been made absent these provisions. To calculate this budget neutrality factor, we used historical discharge data to simulate payments, and compared total prospective payments (including indirect medical education and disproportionate share payments) prior to any reclassifications to total prospective payments after reclassifications. In the proposed rule, we applied an adjustment factor of 0.994059 to ensure that the effects of reclassification are budget neutral. The [[Page 46228]] final budget neutrality adjustment factor is 0.993514. The adjustment factor is applied to the standardized amounts after removing the effects of the FY 1996 budget neutrality adjustment factor. We note that the proposed FY 1997 adjustment reflected wage index and standardized amount reclassifications approved by the MGCRB or the Administrator as of March 14, 1996. The final budget neutrality adjustment factor reflects the effects of all reclassification changes resulting from appeals and reviews of the MGCRB decisions for FY 1997 or from a hospital's request for the withdrawal of a reclassification request. c. Outliers.--Section 1886(d)(5)(A) of the Act provides for payments in addition to the basic prospective payments for "outlier" cases, cases involving extraordinarily high costs (cost outliers) or long lengths of stay (day outliers). Section 1886(d)(3)(B) of the Act requires the Secretary to adjust both the large urban and other area national standardized amounts by the same factor to account for the estimated proportion of total DRG payments made to outlier cases. Similarly, section 1886(d)(9)(B)(iv) of the Act requires the Secretary to adjust the large urban and other standardized amounts applicable to hospitals in Puerto Rico by the same factor to account for the estimated proportion of total DRG payments made to outlier cases. Furthermore, under section 1886(d)(5)(A)(iv) of the Act, outlier payments for any year must be projected to be not less than 5 percent nor more than 6 percent of total payments based on DRG prospective payment rates. Beginning with FY 1995, section 1886(d)(5)(A) of the Act requires the Secretary to phase out payments for day outliers (correspondingly, payments for cost outliers would increase). Under the requirements of section 1886(d)(5)(A)(v), the proportion of day outlier payments to total outlier payments is reduced from FY 1994 levels as follows: 75 percent of FY 1994 levels in FY 1995, 50 percent of FY 1994 levels in FY 1996, and 25 percent of FY 1994 levels in FY 1997. We estimated the FY 1994 proportion of day outlier payments to total outlier payments at 31.3 percent in our September 1, 1993 final rule (58 FR 46348). Thus, the proportion of day outlier payments to total outlier payments in FY 1997 will be approximately 8 percent (25 percent of 31.3 percent). For discharges occurring after September 30, 1997, the Secretary will no longer pay for day outliers under the provisions of section 1886(d)(5)(A)(I) of the Act. i. FY 1997 Outlier Payment Policies, Including Outlier Thresholds For FY 1996, the day outlier threshold is the geometric mean length of stay for each DRG plus the lesser of 23 days or 3.0 standard deviations. The marginal cost factor for day outliers (the percent of Medicare's average per diem payment paid for each outlier day) is 44 percent for FY 1996. The fixed loss cost outlier threshold is equal to the prospective payment for the DRG plus $15,150 ($13,800 for hospitals that have not yet entered the prospective payment system for capital- related costs). The marginal cost factor for cost outliers (the percent of costs paid after costs for the case exceed the threshold) is 80 percent. We applied an outlier adjustment to the FY 1996 standardized amounts of 0.949054 for the large urban and other areas rates and 0.9526 for the capital Federal rate. For FY 1997, we proposed to set the day outlier threshold at the geometric mean length of stay for each DRG plus the lesser of 24 days or 3.0 standard deviations. Section 1886(d)(5)(A)(iii) of the Act, as amended by section 13501(c)(3) of Public Law 103-66, provides that additional payments for day outlier cases may be reduced below the marginal cost of care to meet the requirements of section 1886(d)(5)(A)(v) of the Act. We also proposed to reduce the marginal cost factor for each outlier day from 44 percent to 35 percent in FY 1997. The thresholds that we are establishing in this final rule will be the geometric mean length of stay for each DRG plus the lesser of 24 days or 3.0 standard deviations. Based on updated simulations, we are establishing in this final rule a marginal cost factor of 33 percent for each outlier day in FY 1997. We estimate that these policies will reduce the proportion of outlier payments paid to day outliers to approximately 8 percent, in accordance with section 1886(d)(5)(A) of the Act. In the proposed rule, we proposed to maintain the marginal cost factor for cost outliers at 80 percent and proposed a fixed loss cost outlier threshold in FY 1997 equal to the prospective payment rate for the DRG plus $11,050 ($10,075 for hospitals that have not yet entered the prospective payment system for capital-related costs). In this final rule, based on simulations using updated data and a revised cost inflation factor (discussed below), we are establishing a fixed loss cost outlier threshold in FY 1997 equal to the prospective payment rate for the DRG plus $9,700 ($8,850 for hospitals that have not yet entered the prospective payment system for capital-related costs). We are also establishing a marginal cost factor for cost outliers of 80 percent, as proposed. We note that the FY 1997 cost outlier calculations are to be completed using the revised labor/nonlabor shares discussed above in section II.A.1 in this Addendum. The final FY 1997 cost outlier threshold reflects a revised cost inflation factor. As explained in the proposed rule, in setting the proposed FY 1997 cost outlier threshold, we used a cost inflation factor of 0.0 percent to simulate payments using FY 1995 hospital bills (61 FR 27497). That is, to determine when a case should qualify for cost outlier payments in FY 1997, we calculated FY 1997 "costs" for each bill in the FY 1995 MedPAR file by applying a cost inflation factor of 0.0 percent. We indicated that we would reevaluate this factor in developing the final rule. The latest available Medicare cost reports indicate that hospital cost per case decreased from FY 1993 to FY 1994 as well as from FY 1994 to FY 1995. Cost report data for 4,600 hospitals for cost reporting periods beginning in FYs 1993 and 1994 show that cost per case decreased 1.906 percent from FY 1993 to FY 1994. Preliminary data for cost reports beginning in FY 1995, which were unavailable when we developed the proposed rule, show that cost per case decreased 2.392 percent from FY 1994 to FY 1995. The latter figure is preliminary to the extent that it reflects only 1,800 hospitals and also reflects "as submitted" cost reports. Nevertheless, it suggests a continued trend in cost deflation. Accordingly, based on the more complete data for hospital cost reporting periods beginning in FYs 1993 and 1994, we have decided to use a cost inflation factor of minus 1.906 percent (a cost per case decrease of 1.906 percent) for purposes of setting the final FY 1997 outlier thresholds (as compared with our proposed FY 1997 cost inflation factor of 0.0 percent). We note that this is the first time we have deflated costs in making the outlier projection. The use of a negative cost inflation factor results in lower FY 1997 "costs" for the set of cases analyzed. For example, if a bill in the FY 1995 MedPAR file reflects FY 1995 "costs" of $1,000, the FY 1997 "costs" will be $1,000 x (1-0.01906) x (1-0.01906) (reflecting 2 years of cost deflation), or $962.24. These lower costs, in turn, result in a lower cost outlier threshold relative to a methodology using a positive or zero cost inflation factor (other things being equal). As stated above, the final FY 1997 cost outlier threshold is the DRG amount plus $9,700, rather than $11,050 as indicated in the proposed rule. [[Page 46229]] In accordance with section 1886(d)(5)(A)(iv) of the Act, we calculated outlier thresholds so that outlier payments are projected to equal 5.1 percent of total payments based on DRG prospective payment rates. In accordance with section 1886(d)(3)(E), we reduced the FY 1997 standardized amounts by the same percentage to account for the projected proportion of payments paid to outliers. As stated in the September 1, 1993 final rule (58 FR 46348), we establish outlier thresholds that are applicable to both inpatient operating costs and inpatient capital-related costs. When we modeled the combined operating and capital outlier payments, we found that using a common set of thresholds resulted in a higher percentage of outlier payments for capital-related costs than for operating costs. We project that the thresholds for FY 1997 will result in outlier payments equal to 5.1 percent of operating DRG payments and 5.2 percent of capital payments based on the Federal rate. The proposed outlier adjustment factors applied to the standardized amounts and the capital Federal rate for FY 1997 were as follows: ------------------------------------------------------------------------ Operating standardized amounts Capital Federal rate ------------------------------------------------------------------------ 0.948968.................................. 0.9476 ------------------------------------------------------------------------ The final outlier adjustment factors applied to the standardized amounts and the capital Federal rate for FY 1997 are as follows: ------------------------------------------------------------------------ Operating standardized amounts Capital Federal rate ------------------------------------------------------------------------ 0.948766.................................. 0.9481 ------------------------------------------------------------------------ As in the proposed rule, we apply the final outlier adjustment factors after removing the effects of the FY 1996 outlier adjustment factors on the standardized amounts and the capital Federal rate. ii. Other Changes Concerning Outliers Table 5 of section V of this addendum contains the DRG relative weights, geometric and arithmetic mean lengths of stay, as well as the day outlier threshold for each DRG. When we recalibrate DRG weights, we set a threshold of 10 cases as the minimum number of cases required to compute a reasonable weight and geometric mean length of stay. DRGs that do not have at least 10 cases are considered to be low volume DRGs. For the low volume DRGs, we use the original geometric mean lengths of stay, because no arithmetic mean length of stay was calculated based on the original data. Table 8a in section V of this addendum contains the updated Statewide average operating cost-to-charge ratios for urban hospitals and for rural hospitals to be used in calculating cost outlier payments for those hospitals for which the intermediary is unable to compute a reasonable hospital-specific cost-to-charge ratio. These Statewide average ratios will replace the ratios published in the September 1, 1995 final rule (60 FR 45922), effective October 1, 1996. Table 8b contains comparable Statewide average capital cost-to-charge ratios. These average ratios will be used to calculate cost outlier payments for those hospitals for which the intermediary computes operating cost- to-charge ratios lower than 0.24265 or greater than 1.28879 and capital cost-to-charge ratios lower than 0.013243 or greater than 0.19730. This range represents 3.0 standard deviations (plus or minus) from the mean of the log distribution of cost-to-charge ratios for all hospitals. We note that the cost-to-charge ratios in Tables 8a and 8b will be used for all cost reports settled during FY 1997 (regardless of the actual cost reporting period) when hospital-specific cost-to-charge ratios are either not available or outside the three standard deviations range. iii. FY 1995 and FY 1996 Outlier Payments In the proposed rule, we estimated that actual outlier payments for FY 1995 were approximately 3.7 percent of actual total DRG payments (lower than the 5.1 percent we projected in setting outlier policies for FY 1995). This percentage was computed by simulating payments using actual FY 1995 bill data available at the time of the proposed rule. Our current estimate is that actual outlier payments for FY 1995 were approximately 3.8 percent of actual total DRG payments. These estimates are based on simulations using the July 1996 update of the provider- specific file and the June 1996 update of the FY 1995 MedPAR file. In the proposed rule, we estimated that actual outlier payments for FY 1996 would be approximately 4.2 percent of actual total DRG payments (lower than the 5.1 percent we projected in setting outlier policies for FY 1996). We currently estimate that FY 1996 outlier payments will approximate 4.0 percent of total DRG payments. This current estimate is based on simulations using the July 1996 update of the provider- specific file and the June 1996 update of the FY 1995 MedPAR file. We used these data to calculate an estimate of the actual outlier percentage for FY 1996 by applying FY 1996 rates and policies to the FY 1995 bills. In the proposed rule, we discussed in detail our methodology for setting outlier thresholds, our periodic refinements to that methodology, and some possible explanations for the recent differences between projected and actual outlier percentages (61 FR 27496). We invited comments and suggestions for further refinements to the methodology. The comments on our outlier policies and methodology and our responses are set forth below. Comment: A number of commenters are concerned that the percentages of actual outlier payments for FYs 1995 and 1996 are lower than we projected when we set the respective thresholds for those years. Some commenters requested that we monitor outlier payments during a fiscal year, so that we can change the thresholds in the middle of the year in the event that projected actual outlier payments are not between 5 and 6 percent of projected actual total DRG payments. Other commenters requested that any difference between outlier payments and the amount set aside be used to offset the amount required in the next year. One commenter argued that it is fundamentally inequitable, even assuming that it is not illegal, not to make additional outlier payments after the end of the fiscal year to assure that we meet our 5.1 percent goal. The commenter cited historical figures on outlier payments from a pending court case in the United States District Court for the District of Columbia, County of Los Angeles v. Shalala, C.A. No. 93-0146 SSH (D.D.C). Response: We have responded to similar comments a number of times, including the final rules for FY 1993 (57 FR 39784), FY 1994 (58 FR 46347), FY 1995 (59 FR 45408), and FY 1996 (60 FR 45856). As we have explained before and as explained below, we believe our outlier policies are consistent with the statute and the goals of the prospective payment system and are not inequitable. In accordance with section 1886(d)(5)(A)(iv) of the Act, we set outlier thresholds before a fiscal year so that outlier payments for the fiscal year are projected to be 5.1 percent of total DRG payments. In doing so, we use the best available Medicare discharge data and hospital-specific data. Many of the factors used to set prospective payment amounts for a given fiscal year are based on estimates. These factors include not only the outlier thresholds, but also the market basket rate of increase used to establish [[Page 46230]] the update factors, the recalibration of the DRG weights, and the various required budget neutrality provisions. We do not believe that Congress intended for us to revise these factors in midyear. Similarly, we do not believe that Congress intended that the standardized amounts for a given fiscal year should be adjusted (upward or downward) to reflect any difference between projected and actual outlier payments for a past fiscal year. Payments for a given discharge in a given fiscal year are generally intended to reflect or address the average costs of that discharge in that year; that goal would be undermined if we adjusted prospective payment system payments to account for "underpayments" or "overpayments" in other years. Moreover, the midyear or retroactive adjustments contemplated by the commenters would be extremely difficult or impracticable (if not impossible) to administer. Hospital bill data with respect to a given fiscal year continues to be added to the MedPAR file for some time after the end of the fiscal year. (We update the MedPAR file for 2 full years after the end of the respective fiscal year.) Therefore, precise figures on actual outlier payments for a given fiscal year cannot be determined until well after that fiscal year ends. We do publish estimates of "actual" outlier payments for recent fiscal years, but those estimates are based on available bills (and sometimes based on simulations using bills for a previous year, adjusted for estimates of inflation). In short, we believe our outlier policies are consistent with the statute and the goals of the prospective payment system. In a recent court decision, the United States District Court for the Central District of California upheld the agency's interpretation of the statute as reasonable, writing in part that "[a]ny retroactive adjustment would be inconsistent with [prospective payment system] because the incentives for cost reduction and efficiency would be eliminated." Alvarado Community Hospital v. Shalala, Case No. CV 94- 0972 RMT (Ex) (C.D. Cal May 6, 1996), appeal filed, No. 96-55967 (9th Cir.). (There is pending litigation on the same issues in the U.S. District Court for the District of Columbia.) Finally, we do not agree that our outlier policies are fundamentally inequitable. As we discussed in the proposed rule, we believe that one reason outlier payments have been lower than expected is that hospital costs are not increasing at the rate we expected, and costs may even be decreasing. Available data show that, beginning in FY 1994, for the first time since the inception of the prospective payment system, hospitals are experiencing actual decreases in cost per case from one year to the next. This information is confirmed by ProPAC in its June 1996 Report to Congress "Medicare and the American Health Care System" (Table 3-3, Annual Change in PPS Operating Costs and Payments, First 11 Years of PPS, p. 65). These actual decreases in cost per case follow a period of several years in which the rate of increase in operating cost per case declined from one year to the next. The thresholds for a given fiscal year reflect a certain level of costs, so if hospitals are generally holding costs down, then fewer cases qualify for outlier payments and outlier payments are lower than expected. But if lower hospital costs result in lower than expected outlier payments, it also results in higher than expected "profits" (at least with respect to nonoutlier cases). Hospital, Medicare profit margins have rebounded to levels not seen since the middle of the 1980s. In the June 1996 report, ProPAC found the aggregate prospective payment system operating margin to be 6.0 percent for FY 1994 (Figure 3-2, Aggregate PPS Operating Margin, First 13 Years of PPS p. 68). ProPAC believes that aggregate prospective payment system margins are even higher for FYs 1995 and 1996. Therefore, we believe that "underpayments" for outliers are not fundamentally inequitable because one factor contributing to this result--lower hospital costs--results in "overpayments" with respect to the standard DRG payments. We do not make retroactive adjustments to the standard DRG payments to account for the effect of actual costs being lower than expected; similarly, we do not make retroactive adjustments to outlier payments. As we have stated previously, we believe the more appropriate action for addressing outlier payments is to continue to examine the outlier policy and try to refine our estimation methodology. Comment: Two commenters stated that, after modeling the outlier payments, they were able to replicate HCFA's result of 5.1 percent for operating outlier payments, but that their analysis yielded only 4.8 percent for capital outlier payments as compared with HCFA's result of 5.2 percent. Response: We have determined that the methodology used by the commenters contained several technical errors. Comment: Two commenters requested that we develop an econometric hospital cost model to help us predict the cost inflation factors used for purposes of setting outlier thresholds. Response: Currently, we calculate the cost inflation factor used to set outlier thresholds by analyzing hospital cost report data on cost per case for recent cost reporting periods. The nature of the econometric cost model contemplated by the commenters is not entirely clear to us, but we are interested in exploring such an approach and welcome specific suggestions for developing an econometric model. We believe such an approach might be helpful if the model could analyze data that are more recent than the data available in hospital cost reports. We did not receive any specific suggestions for refinements to our outlier estimation methodology. We note that one commenter believes that the 0.0 percent cost inflation factor reflected in the proposed rule is warranted. As explained above, in this final rule, we are using a cost inflation factor of minus 1.906 percent to further reflect the decreases in cost per case. B. Adjustments for Area Wage Levels and Cost of Living The adjusted standardized amounts are divided into labor and nonlabor portions. Tables 1a and 1c, as set forth in this addendum, contain the actual labor-related and nonlabor-related shares that will be used to calculate the prospective payment rates for hospitals located in the 50 States, the District of Columbia, and Puerto Rico. This section addresses two types of adjustments to the standardized amounts that are made in determining the prospective payment rates as described in this addendum. 1. Adjustment for Area Wage Levels Sections 1886(d)(3)(E) and 1886(d)(9)(C)(iv) of the Act require that an adjustment be made to the labor-related portion of the prospective payment rates to account for area differences in hospital wage levels. This adjustment is made by multiplying the labor-related portion of the adjusted standardized amounts by the appropriate wage index for the area in which the hospital is located. In section III of the preamble, we discuss certain revisions we are making to the wage index. This index is set forth in Tables 4a through 4e of this addendum. [[Page 46231]] 2. Adjustment for Cost of Living in Alaska and Hawaii Section 1886(d)(5)(H) of the Act authorizes an adjustment to take into account the unique circumstances of hospitals in Alaska and Hawaii. Higher labor-related costs for these two States are taken into account in the adjustment for area wages described above. For FY 1997, we are adjusting the payments for hospitals in Alaska and Hawaii by multiplying the nonlabor portion of the standardized amounts by the appropriate adjustment factor contained in the table below. Table of Cost-of-Living Adjustment Factors, Alaska and Hawaii Hospitals ------------------------------------------------------------------------ Alaska--All areas............................................. 1.25 Hawaii: County of Honolulu........................................ 1.225 County of Hawaii.......................................... 1.15 County of Kauai........................................... 1.20 County of Maui............................................ 1.225 County of Kalawao......................................... 1.225 ------------------------------------------------------------------------ (The above factors are based on data obtained from the U.S. Office of Personnel Management.) C. DRG Relative Weights As discussed in section II of the preamble, we have developed a classification system for all hospital discharges, assigning them into DRGs, and have developed relative weights for each DRG that reflect the resource utilization of cases in each DRG relative to Medicare cases in other DRGs. Table 5 of section V of this addendum contains the relative weights that we will use for discharges occurring in FY 1997. These factors have been recalibrated as explained in section II of the preamble. D. Calculation of Prospective Payment Rates for FY 1997 General Formula for Calculation of Prospective Payment Rates for FY 1997 Prospective payment rate for all hospitals located outside Puerto Rico except sole community hospitals = Federal rate. Prospective payment rate for sole
