Medicare Program, Proposed Changes to the Hospital Inpatient Prospective Payment Systems and Fiscal Year 2009 Rates
Score: 27 / 60
We are proposing to revise the Medicare hospital inpatient prospective payment systems (IPPS) for operating and capital-related costs to implement changes arising from our continuing experience with these systems, and to implement certain provisions made by the Deficit Reduction Act of 2005, the Medicare Improvements and Extension Act, Division B, Title I of the Tax Relief and Health Care Act of 2006, and the TMA, Abstinence Education, and QI Programs Extension Act of 2007. In addition, in the Addendum to this proposed rule, we describe the proposed changes to the amounts and factors used to determine the rates for Medicare hospital inpatient services for operating costs and capital-related costs.
There are twelve criteria within our evaluation within three broad categories: Openness, Analysis and Use. For each criterion, the evaluators assign a score ranging from 0 (no useful content) to 5 (comprehensive analysis with potential best practices). Thus, each analysis has the opportunity to earn between 0 and 60 points.
|1. How easily were the RIA , the proposed rule, and any supplementary materials found online?|
The link to the proposed rule containing the RIA is five clicks from the home page. Click on "regulations" (intuitive). The rest is unintuitive: click on "other regulations," expand the "detail" list, click on "CMS regulations," and then click on "quarterly provider updates - regulations" and select the update list from the period covering the date of the regulation. It can also be found on regulations.gov using RIN. A certain amount of regulatory and data analysis that one might expect to find in an RIA actually appears in the preamble to the rule.
|2. How verifiable are the data used in the analysis?|
Most data are from internal databases (eg, MedPAR) that are probably familiar to specialists who follow this type of regulation. The RIA portion does not explain where these data are or how outsiders might access them. The proposed rule includes a detailed section explaining how the public can obtain the data. An appendix provides tables with detailed results and underlying data.
|3. How verifiable are the models and assumptions used in the analysis?|
HHS apparently uses a simulation model to estimate how changes in parameters will alter payments to various types of hospitals. This may be well-accepted by the regulated community, but there is little to no documentation that would allow an outsider to assess or verify the model. Assumptions/inputs appear to be well-documented.
|4. Was the analysis comprehensible to an informed layperson?|
RIA essentially presents outputs of the simulation model for various regulatory changes. It is laden with acronyms (listed at the beginning) and technical jargon which is explained early in the preamble to the proposed rule. Charts used in the impact analysis are easy to follow given the thorough qualitative descriptions of their results. As a part of the final rule, however, the RIA is complicated by the fact that a reader is encouraged to read through the entire 412 pages to fully understand the impact analyses. The RIA could be improved if truncated descriptions of proposed policy changes were added and the analysis itself were separated from the proposed rule (23907).
|5. How well does the analysis identify the desired outcomes and demonstrate that the regulation will achieve them?|
|Does the analysis clearly identify ultimate outcomes that affect citizens’ quality of life?|
The analysis identifies efficient operations of hospitals, minimization of unnecessary costs, preservation of Medicare Trust Fund, and adequate compensation to hospitals for necessary costs. These sound important but are not clearly linked to outcomes that directly affect citizens.
|Does the analysis identify how these outcomes are to be measured?|
The only thing measured in the RIA is expenditures. One section of the proposed rule proposes 30 health outcome measures going forward, but these are used to evaluate hospitals rather than measure the benefits of the regulation.
|Does the analysis provide a coherent and testable theory showing how the regulation will produce the desired outcomes?|
The analysis does not explain how the projected level of expenditures accomplishes any of the above goals. Some judgment calls are explained in terms of efficiency or incentives. The RIA does a better job of explaining how the changes might affect outcomes that are not quantified. For instance, CMS claims that proposed changes to MS-DRG (Medicare severity diagnosis-related groups), the recalibration of relative weights and hospital wage indexes, IME (Indirect Medical Equation) adjustments, and changes to payment rates for excluded hospitals will better match Medicare payment rates to hospitals in many ways. Further, the establishment of programs like QualityNet furthers CMS’s continued commitment to quality improvement in both clinical care and patient safety (23651). What’s more, CMS claims that The Rural Community Hospital Demonstration Program aims to link hospital payments to their performance on quality measures while adjustments to physician self-referral provisions reduces program abuse by bringing more financial relationships within the scope of the physician self-referral law (23695). Finally, the proposed changes encourage fiscal discretion by instituting incentives to reduce avoidable readmissions to hospitals (which represents 15 billion of Medicare spending each year) (23673).
|Does the analysis present credible empirical support for the theory?|
Since most of the proposed changes are instituted based on statutory requirements, the RIA often refers back to proposed rules from past years. But effects of prior rules are not generally cited as reasons for the proposed changes. The RIA would be improved if it cited these references more clearly and widened the scope of its analysis by investigating similar medical payment system provisions beyond the scope of legal statute.
|Does the analysis adequately assess uncertainty about the outcomes?|
The analysis does not address this topic.
|6. How well does the analysis identify and demonstrate the existence of a market failure or other systemic problem the regulation is supposed to solve?|
|Does the analysis identify a market failure or other systemic problem?|
A statement of sometimes-conflicting desired outcomes might be taken as a statement of the problem the regulation is trying to solve, but there is no elaboration in this regard. Since the analysis is located within the proposed rule, it seems less likely that CMS would explore whether or not the Medicare price system itself allocates services as well as a competitive market would. However, since many arguments criticize the inefficiencies of non-market allocation, a discussion of CSM's justification for a systemic problem or market failure that necessitates Medicare would improve the RIA.
|Does the analysis outline a coherent and testable theory that explains why the problem (associated with the outcome above) is systemic rather than anecdotal?|
No; the RIA is written from a perspective that already assumes that Medicare solves a "systemic problem." This RIA would be improved it if outlined a testable theory that would explain the nature of the systemic problem (it does elaborate on the scope via Tables I-III) and why the proposed FY 2009 changes, and the mandates of the Medicare PPS outlined in the act more generally, solve or lessen said problems.
|Does the analysis present credible empirical support for the theory?|
Since there is no theory of an underlying problem, there is no empirical evidence either.
|Does the analysis adequately assess uncertainty about the existence or size of the problem?|
The analysis does not address this topic.
|7. How well does the analysis assess the effectiveness of alternative approaches?|
|Does the analysis enumerate other alternatives to address the problem?|
The analysis casually mentions in its "IX. Alternatives Considered" section that the "proposed rule contains a range of proposed policies (23921)." While the preamble of this proposed rule may provide (some) description of the many alternatives that were considered and why CMS chose the proposed change over outlined alternatives, the RIA would be much improved if it collected descriptions of the most relevant alternatives and explained their cost effectiveness in the "Alternatives Considered" section instead of forcing readers to try to locate them within the entire 412 page proposed rule.
|Is the range of alternatives considered narrow (e.g., some exemptions to a regulation) or broad (e.g., performance-based regulation vs. command and control, market mechanisms, nonbinding guidance, information disclosure, addressing any government failures that caused the original problem)?|
Alternatives are usually small deviations from a basic policy, e.g., which things count as preventable conditions acquired in the hospital? Among the many alternatives considered, the ones that are easily located in the proposed rule include: alternatives to CMS-DRG (23539), HSRV cost-weighting methodology for addressing the issue of charge compression, which was analyzed by the Rand Corporation (23593), different options proposed by MedPAC to compute wage indices (23618), lists of alternatives about using different sources of data that are electronically maintained (23663) nd alternatives to address “stand in the shoes,” or physician self-referral provisions (23686). Although wide ranging, most if not all of the alternatives considered are within the scope of statutory law of IPPS .
|Does the analysis evaluate how alternative approaches would affect the amount of the outcome achieved?|
A simulation model is run for the chosen option. There is sporadic discussion suggesting that many alternatives would have relatively small effects on payments, but not outcomes. In the preamble, there are some examples of data or consulting studies that shed light on the alternatives and seem to have influenced the decision.
|Does the analysis adequately address the baseline? That is, what the state of the world is likely to be in the absence of federal intervention not just now but in the future?|
CMS states that to illustrate the impacts of the proposed FY 2009 changes, its RIA begins with a "FY 2008 baseline simulation model using; the proposed FY 2009 update of 3.0 percent; the FY 2008 DRG GROUPER… the most current CBSA designation for hospitals based on OMB's MSA definitions; the FY 2008 wage index..." (23907). Each policy change, statutory or otherwise, is added to this baseline, finally arriving at an FY 2009 model incorporating all of the proposed changes. This simulation is represented in three main charts and it isolates the effects of each proposed change clearly (23908). The baseline uses data from a past year and explicitly assumes no changes in case mix, etc. This is a clear baseline but seems questionable given likely demographic changes as the population ages.
|8. How well does the analysis assess costs and benefits?|
|Does the analysis identify and quantify incremental costs of all alternatives considered?|
A simulation quantifies effect of each proposed change on federal expenditures, but only for the chosen alternative. CMS prepares impressive individual analyses of the marginal costs added by prospective FY 2009 total, operating and capital cost changes to the IPPS system. Table 1 breaks down the total effects of FY 2009 changes. Most importantly, the last column shows changes in payments from FY 2008 to FY 2009 including the proposed FY 2009 0.9% documentation and coding adjustments and the projected 1.8% increase in case-mix expected to occur in FY 2009 due to improvements in documentation and coding. This column incorporates all of the changes displayed in Columns 4, 5, 6, 7, 8 (the changes displayed in Columns 2 and 3 are included in Column 4). It also reflects the impact of the FY 2008 update, and changes in hospitals’ reclassification status in FY 2009 compared to FY 2008. The total percent change comes to 4.1% (23912). Table II represents the analysis of proposed changes for FY 2009 IPPS operating costs. For 3,528 hospitals CMS estimates that the average payment per case will increase from $9,144 to $9,519 from FY 2008 to FY 2009 (23913). Table III compares the total costs of capital payments per case from FY 2008 to FY 2009. These simulation results show that, on average, capital payments per case in FY 2009 can be expected to remain about the same as capital payments per case in FY 2008. Specifically, for the 3,528 hospitals, the average FY 2008 and FY 2009 payment/ case figure is 757 (23919).
|Does the analysis identify all expenditures likely to arise as a result of the regulation?|
All federal expenditures only. Based on an overall percent change in payments per case estimated using the CMS payment simulation model, it is estimated that total FY 2009 payments will increase 4.1% compared to FY 2008, largely due to the statutorily mandated update to the IPPS rates (23907). The market basket update to the IPPS rates required by the statute, in conjunction with other payment changes in this proposed rule, will result in an approximate $4 billion increase in FY 2009 operating and capital payments (23905). Specifically, the "Accounting Statement: Classification of Estimated Expenditures from FY 2008 to FY 2009" sets Annualized Monetized Transfers from the federal government to IPPS Medicare Providers at $3.967 billion (23921).
|Does the analysis identify how the regulation would likely affect the prices of goods and services?|
Only in the sense that the rulemaking determines what price the federal government will pay for services.
|Does the analysis examine costs that stem from changes in human behavior as consumers and producers respond to the regulation?|
The RIA repeatedly mentions a conclusion from another rulemaking that improved accuracy of coding and documentation will likely increase costs by 1.8 percent. But there is no recognition in the RIA of how changes in prices, or the prospective payment system generally, might affect incentives and behavior. The preamble notes that DRGs that group cases create some incentive for providers to specialize in high-margin cases, and asserts that changes to DRGs reduce this incentive, but the RIA does not address this issue. Similarly, the preamble discusses incentive effects of nonpayment for treatment of conditions patients acquired in the hospital, but the RIA does not seem to address this.
|If costs are uncertain, does the analysis present a range of estimates and/or perform a sensitivity analysis?|
CMS outlines several limitations in its analysis. For instance, it estimates the effects of individual proposed policy changes by estimating payments per case while holding all other payment policies constant, which is not realistic. Additionally, it claims to use the best data available but, generally, CMS does not attempt to make adjustments for future changes in variables that are likely to occur, such as admissions, lengths of stay or case-mix (23906). CMS notes several times that there are other changes for which they do not have data available and for which their payment simulation model cannot be utilized. For those changes CMS attempted to predict payment impact based on experience and other more limited data. Lastly, CMS mentions that over/under calculations and miscategorization is possible for the division of hospital sections in Tables I- III because various sources were used to categorize hospital data (23918). No sensitivity analysis is preformed.
|Does the analysis identify the alternative that maximizes net benefits?|
Since only federal expenditures are measured, net benefits are not calculated.
|Does the analysis identify the cost-effectiveness of each alternative considered?|
Cost effectiveness is only briefly described qualitatively and it is not easily located in the lengthy proposed rule. See comment 2C-1.
|Does the analysis identify all parties who would bear costs and assess the incidence of costs?|
Tables show how the changes would alter payouts to hospitals, broken down by geographic location, urban/rural, size, ownership, teaching status, etc. Tables I, II and III display the results of CMS’s analysis of the total proposed changes and the total operating and capital cost changes of FY 2009 by categorizing 3,528 hospitals and breaking up the respective costs/ benefits they receive in various ways. The main classifications are by geographic location and reclassification, based on their FY 2009 payment classifications, based on whether or not they have GME residency programs, or receive DSH payments, by the impacts of the proposed changes on rural hospitals by special payment groups (SCHs, RRCs and MDHs) and based on the type of ownership and the hospital’s Medicare utilization expressed as a percent of total patient days (23908, 23913, 23920).
|Does the analysis identify all parties who would receive benefits and assess the incidence of benefits?|
The analysis does not consider effects on taxpayers or patients.
|9. Does the proposed rule or the RIA present evidence that the agency used the analysis?|
The simulation model was used to calculate adjustments necessary to maintain budget neutrality where required by law. More accurate coding may increase payments, and an offset factor was adopted to counteract this. Changes in average hourly wage criteria are based on recent data. The RIA largely just reports the payments to hospitals that flow from the calculations using all these data. In that sense, the simulations in the RIA were used to develop the proposed rule. However, the proposed rule also contains many regulatory provisions that are not addressed in the RIA.
|10. Did the agency maximize net benefits or explain why it chose another alternative?|
Net benefits are not calculated. A large number of decisions seem to be driven by legislative mandates, or by judgments whose linkage to outcomes or costs is not explained in much depth.
|11. Does the proposed rule establish measures and goals that can be used to track the regulation's results in the future?|
The preamble lists 30 quality-of-care measures that could be used to evaluate the regulation's effects, and HHS could set goals for how the regulation should affect these measures and then see whether those goals are achieved. HHS proposes only to use them to evaluate hospitals, not the effects of the regulation.
|12. Did the agency indicate what data it will use to assess the regulation's performance in the future and establish provisions for doing so?|
The law requires HHS to reduce payments in FY 2010–12 to reflect any increase in FY 2008–09 payments that stems from case-mix changes that do not reflect real changes in severity of patient conditions. HHS plans to measure changes in case-mix for FY 2008 and 2009, then separate this change into a portion attributable to increased severity of patient conditions and a portion that is not. This is a forward-looking use of data, but it is not really an evaluation of the regulation's effects. Data on the 30 quality of care measures could be used to evaluate the regulation's effects, but HHS does not propose to do so.
|Total||27 / 60|
- Department of Health and Human Services
- Regulatory Identification Number
- Agency Name
- Department of Health and Human Services
- Rule Publication Date