AGENCY:

Centers for Medicare & Medicaid Services (CMS), HHS.

ACTION:

Final rule.

SUMMARY:

The statute requires aggregate reductions to state Medicaid Disproportionate Share Hospital (DSH) allotments annually beginning with fiscal year (FY) 2020. This final rule delineates the methodology to implement the annual allotment reductions.

DATES:

These regulations are effective on November 25, 2019.

FOR FURTHER INFORMATION CONTACT:

Stuart Goldstein, (410) 786-0694 and Richard Cuno, (410) 786-1111.

SUPPLEMENTARY INFORMATION:

I. Executive Summary

A. Purpose

Section 2551 of the Patient Protection and Affordable Care Act of 2010 (Pub. L. 111-148, enacted March 23, 2010), as amended by the Health Care and Education Reconciliation Act of 2010 (Pub. L. 111-152, enacted March 30, 2010)) (the ACA) amended section 1923(f) of the Act by setting forth aggregate reductions to state DSH allotments annually from FY 2014 through FY 2020. In the September 18, 2013 Federal Register (78 FR 57293), we published the “Medicaid Program; State Disproportionate Share Hospital Allotment Reductions” final rule (herein referred to as the “2013 DSH allotment reduction final rule”). In the 2013 DSH allotment reduction final rule, we finalized a DSH Health Reform Methodology (DHRM), as required by statute, to implement annual allotment reductions that would have been in place only for FY 2014 and FY 2015. Prior to the implementation of allotment reductions, legislation was signed into law delaying the start of the reductions.^[1] Subsequent legislation delayed the start of these reductions, modified the reduction amounts, and extended the fiscal years subject to reductions.^[2] Under current law, annual allotment reductions start in FY 2020 and run through FY 2025. In July 28, 2017 Federal Register (82 FR 35155), we published the “Medicaid Program; State Disproportionate Share Hospital Allotment Reductions” proposed rule (herein referred to as the “the July 2017 proposed rule”), in which we proposed to establish a DHRM applicable for all fiscal years subject to allotment reduction that would account for relevant data that was unavailable to CMS during prior rulemaking for DSH allotment reductions originally set to take place for FY 2014 and FY 2015. In this final rule, we are finalizing the DHRM as proposed, with limited exceptions identified below.

B. Summary of the Major Provisions

The statute as amended by the ACA, directs the Secretary of Health and Human Services (the Secretary) to implement the annual DSH allotment reductions using a DHRM. This final rule amends 42 CFR 447.294 by establishing the DHRM for FY 2020 and subsequent fiscal years, which incorporates factors identified in the statute. We are finalizing § 447.294(a) and (e) to remove language referring to specific federal fiscal years (FY 2014 and FY 2015) when calculating annual state DSH allotment reductions. We are finalizing § 447.294(b) to add the definition of “Total hospital cost.” We are modifying this definition from the proposed in order to give the term the same meaning as it is defined in § 447.299(c)(20). We believe that cross-referencing the existing provision is clearer, less likely to result in any confusion or ambiguity, and is not intended to be a substantive difference in meaning from that of the proposed definition. This rule finalizes § 447.294(d) to clarify state data submission requirements by simplifying the language and removing language related to the submission of data for previous state plan rate years (SPRY) already provided to CMS. We are finalizing § 447.294(e)(3)(i) to utilize total estimated Medicaid service expenditures in the calculation of the Low DSH adjustment factor (LDF) for the applicable year. In this rule, we are finalizing revisions to § 447.294(e)(5)(i) through (iii) to adjust the weighting of statutorily defined factors required to be included in the DHRM. Additionally, this rule finalizes revisions to § 447.294 to revise paragraph (f) by removing references to specific fiscal years in regulation.

C. Impacts

The DHRM will generate a state-specific DSH allotment reduction amount for each fiscal year in accordance with the requirements specified in section 1923(f)(7) of the Act. The total of all DSH allotment reduction amounts in a specific fiscal year will equal the aggregate annual reduction amount identified in the statute for that year. To determine the effective annual DSH allotment for each state, the state-specific annual DSH allotment reduction amount will be applied to the unreduced DSH allotment amount for the state.

II. Background

A. Introduction

In anticipation of lower uninsured rates and lower levels of hospital uncompensated care, the ACA modified the amounts of funding available to states under the Medicaid program to address the situation of hospitals that serve a disproportionate share of low-income patients, and therefore, may have uncompensated care costs. Under sections 1902(a)(13)(A)(iv) and 1923 of the Act, states are required to make payments to qualifying DSHs (DSH payments). Section 2551 of the ACA amended section 1923(f) of the Act, by adding paragraph (7), to provide for aggregate reductions in federal funding under the Medicaid program for such DSH payments for the 50 states and the District of Columbia. DSH allotments are not provided for the five US territories.

Section 1923(f)(7)(A)(i) of the Act requires that the Secretary implement the aggregate reductions in federal funding for DSH payments through reductions in annual state allotments of federal funding for DSH payments (state DSH allotments), and accompanying reductions in payments to each state. Since 1998, the amount of federal funding for DSH payments for each state has been limited to an annual state DSH allotment in accordance with section 1923(f) of the Act. The addition of section 1923(f)(7) of the Act requires the use of a DHRM to determine the percentage reduction in annual state DSH allotments to achieve the required aggregate annual reduction in federal DSH funding. The statutory reductions apply to all states and the District of Columbia, except the State of Tennessee. Under section 1923(f)(6)(A)(vi) of the Act, notwithstanding any other provision of section 1923(f) of the Act, or any other provision of law, the DSH allotment for Tennessee is established at $53.1 million per year for FY 2015 through FY 2025. Therefore, Tennessee's DSH allotment is not subject to reduction under section 1923(f)(7) of the Act. For purposes of this rule, references to the reduction for “each state” means “each state subject to a DSH allotment reduction” (that is, the 50 states and the District of Columbia, except, for periods before FY 2026, Tennessee).

Section 1923(f)(7)(B) of the Act establishes the following factors that must be considered in the development of the DHRM. The methodology must:

• Impose a smaller percentage reduction on low DSH States;
• Impose the largest percentage reductions on:

++ States that have the lowest percentages of uninsured individuals during the most recent year for which such data are available;

++ States that do not target their DSH payments on hospitals with high volumes of Medicaid inpatients;

++ States that do not target their DSH payments on hospitals with high levels of uncompensated care; and

• Take into account the extent to which the DSH allotment for a state was included in the budget neutrality calculation for a coverage expansion approved under section 1115 of the Act as of July 31, 2009.

In section II.B. of the July 2017 proposed rule, we described the principles we intended to apply when calculating the annual DSH allotment reduction amounts for each state through the DHRM.

B. Legislative History and Overview

The Omnibus Budget Reconciliation Act of 1981 (OBRA'81) (Pub. L. 97-35, enacted on August 13, 1981) amended section 1902(a)(13) of the Act to require that Medicaid payment rates for hospitals take into account the situation of hospitals that serve a disproportionate share of low-income patients with special needs. Over the more than 35 years since this requirement was first enacted, the Congress has set forth in section 1923 of the Act payment targets and limits to implement the requirement and to ensure greater oversight, transparency, and targeting of funding to hospitals.

To qualify as a DSH under section 1923(b) of the Act, a hospital must meet two minimum qualifying criteria in section 1923(d) of the Act. The first criterion is that the hospital has at least two obstetricians who have staff privileges at the hospital and who have agreed to provide obstetric services to Medicaid beneficiaries. This criterion does not apply to hospitals in which the inpatients are predominantly individuals under 18 years of age or hospitals that do not offer nonemergency obstetric services to the general public as of December 22, 1987. The second criterion is that the hospital has a Medicaid inpatient utilization rate (MIUR) of at least 1 percent.

Under section 1923(b) of the Act, a hospital meeting the minimum qualifying criteria in section 1923(d) of the Act is deemed as a DSH if the hospital's MIUR is at least one standard deviation above the mean MIUR in the state for hospitals receiving Medicaid payments, or if the hospital's low-income utilization rate (LIUR) exceeds 25 percent. States have the option to define DSHs under the state plan using alternative qualifying criteria as long as the qualifying methodology comports with the deeming requirements of section 1923(b) of the Act. Subject to certain federal payment limits, states are afforded flexibility in setting DSH state plan payment methodologies to the extent that these methodologies are consistent with section 1923(c) of the Act.

Section 1923(f) of the Act limits federal financial participation (FFP) for total statewide DSH payments made to eligible hospitals in each federal FY to the amount specified in an annual DSH allotment for each state. Although there have been some special rules for calculating DSH allotments for particular years or sets of years, section 1923(f)(3) of the Act establishes a general rule that state DSH allotments are calculated on an annual basis in an amount equal to the DSH allotment for the preceding FY increased by the percentage change in the consumer price index for all urban consumers for the previous FY. The annual allotment, after the consumer price index increase, is limited to the greater of the DSH allotment for the previous year or 12 percent of the total amount of Medicaid expenditures under the state plan during the FY. Allotment amounts were originally established in the Medicaid Voluntary Contribution and Provider Specific Tax Amendments of 1991 based on each state's historical DSH spending.

Section 1923(g) of the Act also limits DSH payments by imposing a hospital-specific limit on DSH payments. Specifically, a DSH payment must not exceed a hospital's uncompensated care costs for that year (that is, it must not exceed the costs of providing inpatient hospital and outpatient hospital services to Medicaid patients and the uninsured, minus payments received by the hospital by or on the behalf of those patients). FFP is not available for DSH payments that exceed the hospital-specific limit.

The statute, as amended by the ACA, required annual aggregate reductions in federal DSH funding from FY 2014 through FY 2020. However, subsequent legislation extended the reductions, modified the amount of the reductions, and delayed the start of the reductions, which now begin in FY 2020. The most recent related amendments to the statute were through the Bipartisan Budget Act of 2018 (Pub. L. 115-123, enacted February 9, 2018) (BBA 18). Currently, the aggregate annual reductions are set to begin in FY 2020, and the annual reduction amounts are specified in section 1923(f)(7)(A)(ii) of the Act:

• $4,000,000,000 for FY 2020.
• $8,000,000,000 for FY 2021.
• $8,000,000,000 for FY 2022.
• $8,000,000,000 for FY 2023.
• $8,000,000,000 for FY 2024.
• $8,000,000,000 for FY 2025.

To implement these annual reductions, the statute requires that the Secretary reduce annual state DSH allotments, and payments to states, based on a DHRM specified in section 1923(f)(7)(B) of the Act. The proposed DHRM relied on five statutorily-identified factors collectively to determine a state-specific DSH allotment reduction amount to be applied to the allotment that is calculated under section 1923(f) of the Act prior to the reductions under section 1923(f)(7) of the Act.

In the May 15, 2013 Federal Register (78 FR 28551), we published the “Medicaid Program; State Disproportionate Share Hospital Allotment Reductions” proposed rule. The rule proposed a DHRM that relied on the statutory factors and solicited comments regarding whether state decisions to extend Medicaid coverage to low-income adults under section 1902(a)(10)(A)(i)(VIII) of the Act (the Medicaid expansion population) should be accounted for in the reduction methodology. We received several comments in support of accounting for Medicaid coverage expansion and numerous comments in opposition.

In the September 18, 2013 Federal Register (78 FR 57293), we published the “Medicaid Program; State Disproportionate Share Hospital Allotment Reductions” final rule (herein referred to as the “2013 DSH allotment reduction final rule”). In the 2013 DSH allotment reduction final rule, we decided to finalize a DHRM that would be in place only for FY 2014 and FY 2015 to allow time for revaluation of the methodology with improved and more recent data and information about the impact of the ACA on levels of coverage and uncompensated care. As a result of our reevaluation, we subsequently proposed to modify the DHRM factor weights and to use improved data sources where possible.

III. Summary of the Provisions of the July 2017 Proposed Rule and Responses to Public Comments

In the July 2017 proposed rule, we proposed to amend § 447.294 by establishing the DHRM for FY 2018 and subsequent fiscal years, incorporating factors identified in the statute. We received approximately 140 public comments on the proposed rule from organizations, individuals, health care providers, advocacy groups, and states. In the sections that follow, we describe each proposed provision, summarize any public comments received on each provision, and provide our responses to the comments.

A. General Comments

In addition to the comments we received on the July 2017 proposed rule's discussion of specific aspects of the State DSH Allotment Reductions (which we address later in this final rule), commenters also submitted the following more general observations on the reductions. The following is a discussion of these comments.

Comment: Many commenters urged delaying the implementation of the annual aggregate reductions to State DSH allotments. The commenters provided various reasons for the requested delay.

Response: The statute directs the Secretary to develop a DHRM to implement annual Medicaid DSH allotment reductions. Various legislation, including most recently the BBA 18, delayed the start of the reductions until FY 2020. We have no flexibility administratively to delay the start of the statutory reductions.

Comment: Multiple commenters expressed concern that unreduced DSH allotments under section 1923(f) of the Act are inequitable. Some of these commenters recommended modifications to the method for determining the unreduced allotments and some commenters indicated a belief that the proposed DHRM would exacerbate the alleged inequities of the unreduced allotments.

Response: Section 1923(f)(7) of the Act specifies the five factors for the DHRM, but does not authorize modifications to the statutory formula for calculating unreduced state DSH allotments under section 1923(f) of the Act. While the statute does not direct the Secretary to modify the formula for unreduced DSH allotments through the DHRM, the DHRM does take into account the size of the existing state DSH allotments in determining annual allotment reduction amounts. Most notably, the Low DSH Adjustment Factor (LDF) requires the imposition of smaller percentage reductions on low DSH states that historically have received lower DSH allotments relative to their total Medicaid expenditures than non-low DSH states.

Comment: One commenter inquired as to when the reduced 2018 DSH allotments will be available as cuts were to begin October 1, 2017.

Response: The BBA 18 delayed the start of annual DSH allotment reductions until FY 2020, which begins on October 1, 2019. We intend to make final FY 2020 reduction amounts available to states once finalized data necessary to calculate these reductions are available, which CMS anticipates will be on or before October 1, 2019.

Comment: A number of commenters expressed concern that the DSH allotment reductions will cause financial distress to hospitals.

Response: We understand the commenters' concerns. However, the statute requires annual aggregate reductions in DSH allotments starting in FY 2020 and the use of a DHRM to determine the percentage reduction in annual state DSH allotments to achieve the required aggregate annual reduction amounts. We are finalizing a DHRM that is consistent with statutory direction and does not affect the considerable flexibility afforded states in setting DSH state plan payment methodologies to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations.

Comment: One commenter stated that those affected by drug addiction and mental health issues will be hurt by the DSH reductions.

Response: We recognize the importance of health care services for substance use disorders and behavioral health issues. However, section 1923(f)(7)(A)(i) of the Act requires the Secretary implement aggregate reductions in federal funding for DSH payments through reductions in annual state DSH allotments. Moreover, these statutorily-required annual state DSH allotment reductions do not directly affect payment rates for services, including services related to substance use disorders or behavioral health, or otherwise directly affect reimbursement to providers that do not receive DSH payments.

Comment: A few commenters suggested that CMS finalize the rule for a limited period of time to allow for reevaluation and refinement to strengthen the DHRM in future years.

Response: We recognize the importance of the DHRM to states, hospitals, and other stakeholders. Therefore, we will monitor and reevaluate the DHRM and its application throughout implementation. If necessary, we will undertake future rulemaking to make modifications to the DHRM.

Comment: Multiple commenters suggested that the DHRM does not take into consideration that Medicaid reimbursement rates are considerably lower than private insurance.

Response: Section 1923(f)(7) of the Act specifies the five factors for the DHRM, but does not direct the Secretary to consider specifically the levels of Medicaid reimbursement rates as compared to private insurers. However, the DHRM does consider Medicaid coverage and payment levels by imposing the largest percentage DSH allotment reductions on states that do not target their DSH payments on hospitals with high volumes of Medicaid inpatients and states that do not target their DSH payments on hospitals with high levels of uncompensated care, which includes Medicaid shortfall.

Comment: Some commenters expressed concern that the Congress passed Medicaid DSH allotments reductions expecting that hospitals would care for fewer uninsured patients as a result of health care coverage expansion related to the ACA. Commenters also stated that projected increases in coverage have not been fully realized for a variety of reasons and some noted that some providers in Medicaid expansion states are still experiencing significant losses for serving Medicaid beneficiaries. Some commenters also expressed concern that increases in the number of insured individuals has not decreased the need for DSH payments.

Response: We appreciate the comments, but the statute directs the Secretary to develop a DHRM to implement annual Medicaid allotment reductions. We have no administrative flexibility to delay the start of the statutory reductions or to reduce the aggregate reduction amounts specified in statute. We believe that the final DHRM distributes DSH allotment reduction amounts among the states in an equitable manner, consistent with statutory requirements.

Comment: Several commenters stated that the hospital industry greatly opposes CMS' policy for the treatment of third party payments when calculating the hospital-specific DSH limit, stating it is a misinterpretation of the Medicaid statute.

Response: CMS' policy regarding the treatment of third party payments when calculating the hospital-specific DSH limit is outside the scope of this rule.

Comment: A few commenters indicated there are unresolved legal questions related to the DSH audit process that are the subject of pending litigation; therefore, CMS should delay finalizing the DSH reduction methodology. One commenter expressed concern that the DSH audit and reporting data may not be consistent with federal Medicaid law.

Some commenters recommended that CMS delay the final rule until stakeholders have had ample opportunity to replicate and evaluate the proposed DHRM and that CMS should provide requisite data sets and sufficient technical information before issuing a final rule. The commenters requested that if that is not possible, then CMS should finalize the DHRM for FY 2018 only and provide an adequate comment period, requisite data sets, and refined technical information with a proposed rule for FY 2019. The commenters noted that, given the complexity of the DHRM and the destabilizing effect that statutorily-required annual state DSH allotment reductions may have on safety net hospitals, a longer comment period and more transparency would be warranted.

Response: We do not believe that there is any need to delay finalizing the July 2017 proposed rule. The statute directs the Secretary to develop a DHRM to implement annual Medicaid DSH allotment reductions, and the intent of this rule is to provide the methodology used to calculate the statutorily-required Medicaid DSH allotment reductions. While a number of issues related to Medicaid DSH payment calculations currently are the subject of litigation, the statutorily-required allotment reductions and the DHRM are not among them, and we are bound by statute to adopt a rule to implement the DSH reductions. With this final rule, we are doing so according to our view of the best interpretation of the DSH statute and will utilize the most recent data available to us that is consistent with applicable laws and regulations.

The BBA 18 delayed the start of the reductions until FY 2020. Accordingly, concerns with respect to how a DHRM might have applied with respect to prior fiscal years, including FY 2018 and FY 2019, are moot. We have no flexibility to delay the start of the statutory reductions. Finally, we intend to publish a separate DHRM technical guide that provides information regarding the DHRM calculation and associated data sources in order to be fully transparent with states and other stakeholders.

Comment: Several commenters expressed concern with the 30-day comment period and the availability of data used in the illustrative model during the comment period and noted that a 60-day comment period would have been more appropriate. Another commenter suggested a second comment period prior to when the DSH allotment reductions for FY 2018 are published to allow states to accurately estimate the impact of the proposed methodology on the state.

Response: We believe the 30-day comment period was appropriate and are not providing an additional comment period. Section 1923(f)(7)(B) of the Act, establishing the five factors that must be considered in the development of the DHRM, was enacted in statute in 2010. Additionally, we signaled our intent to pursue a similar methodology in future rulemaking when publishing the final 2013 DSH allotment reduction rule.

Comment: One commenter indicated that research has shown that residents of Medicaid expansion states are less likely to experience financial barriers to healthcare access than residents of states that have not expanded Medicaid coverage.

Response: This comment is outside the scope of this rule.

Comment: One commenter encouraged CMS to consider that Medicaid is the single largest payer to children's hospitals and suggested that the regulation will impose a greater burden to these hospitals, which already face significant financial challenges due to inadequate Medicaid reimbursement rates. Another commenter expressed concern that the reductions will have a negative impact on hospitals in the commenter's state, given that there is not a sufficient number of privately insured patients to offset losses from Medicare and Medicaid, which pay significantly less than private insurers.

Response: We appreciate the important role that children's hospitals play in serving Medicaid beneficiaries. This rule provides the methodology used to calculate the statutorily-required Medicaid DSH allotment reductions and does not affect the flexibility afforded to states when setting DSH state plan payment methodologies, to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable laws and regulations. States retain flexibility to direct Medicaid DSH payments to qualifying hospitals in the state, including children's hospitals, in the manner the state determines most appropriate under the conditions in the state. In addition, we are finalizing a DHRM that would equitably allocate the statutorily-required annual reductions based on the factors specified in section 1923(f)(7) of the Act. Changes to Medicare and non-DSH Medicaid payment rates are outside the scope of this rule.

Comment: One commenter stated that the statute requiring DSH allotment reductions is unethical, particularly in that it would reduce payments to hospitals.

Response: We appreciate the concerns that the rule may have an impact on hospitals. However, the statute as amended by the ACA and subsequent legislation directs the Secretary to implement annual DSH allotment reductions using a DHRM, which is specified in this final rule.

Comment: One commenter noted their work for an institution that served mostly Medicaid patients and that the institution may not be able to continue to provide services to all individuals if DSH payments are reduced. Additionally, the commenter expressed concern that future Congressional action in health care might result in additional uninsured or underinsured patients.

Response: We appreciate the important role that DSHs play in providing health care to low-income individuals and vulnerable populations. The statute specifies the annual amount of aggregate DSH allotment reductions and directs the Secretary to develop a methodology which takes into consideration the required statutory factors for allocating a reduction amount to each state. This final rule does not affect state flexibility to develop methodologies as described in section 1923(c) of the Act for payments to qualifying hospitals, provided the methodology complies with all applicable statutory and regulatory requirements.

Comment: One commenter recommended that CMS carve out most non-DSH supplemental payments made to inpatient hospitals and add the funding into the state's DSH allotment, to better support essential hospitals by ensuring payments flow through one central distribution program.

Response: Non-DSH Medicaid supplemental payments and the method for calculating unreduced DSH allotments in section 1923(f) of the Act are outside the scope of this rule.

Comment: One commenter suggested CMS consider that Medicaid is about to embark on a new phase of payment and delivery reform, and the DSH reductions could disrupt those efforts.

Response: This rule does not address potential future payment and delivery reform, and does not affect state's flexibility under section 1923 of the Act to establish DSH payment methodologies.

Comment: Many commenters recommended that CMS mitigate the impact of reductions on specific hospital types, including rural hospitals, safety net hospitals, critical access hospitals, and academic medical centers. One commenter recommended that CMS mitigate reductions based on community needs to ensure individuals have access to care and that DSH funding is available for medically necessary services. Another commenter expressed concern for low and moderate income families having access to care and suggested that hospitals be required to meet basis standards related to charity care, billing, and collections to receive DSH payments.

Response: This rule only addresses the aggregate DSH allotment reductions under section 1923(f)(7) of the Act. The statutory requirements for DSH payment methodologies are specified in section 1923(c) of the Act and are outside of the scope of this rule. However, we believe that the DHRM reduces DSH allotments, at the state level, in an equitable manner that is consistent with the statute. Accordingly, we designed the DHRM to preserve the considerable flexibility afforded states in setting DSH state plan payment methodologies, to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations.

Comment: One commenter recommended that CMS consider Medicaid shortfalls, charity care, and bad debt in the distribution of funds from uncompensated care pools approved under section 1115 demonstrations. In addition, the commenter recommended that CMS consider all lines of a hospital's business in the DHRM for hospitals experiencing negative margins to better account for the overall financial situation of hospitals.

Response: This regulation does not address the distribution of payments under section 1115 demonstrations; it only addresses the statutorily-required Medicaid DSH allotment reductions. Changes affecting the distribution of payments under section 1115 demonstrations are outside the scope of this rule. Additionally, the hospital-specific limit under section 1923(g) of the Act only considers costs incurred for furnishing hospital services to individuals who are either Medicaid beneficiaries or uninsured. Consistent with the DSH statute's overall focus on these populations, the statutory DHRM targeting factors also require smaller reductions be imposed on states that target their DSH payments to hospitals with high volumes of Medicaid inpatients and high levels of uncompensated care (excluding bad debt). As such, we did not propose and are not finalizing consideration of other lines of a hospital's business for purposes of the statutorily-required Medicaid DSH allotment reductions.

Comment: Many commenters expressed concerns regarding the possibility of revisions to or repeal of the ACA and recommended that the DHRM include a provision for reversal of reductions if future legislation affecting section 1923(f)(7) of the Act is enacted.

Response: We appreciate the commenters' concerns but are statutorily-bound to implement the DSH allotment reductions beginning with FY 2020. This final rule does not prevent CMS from following future statutory provisions, including any revisions to the applicable statute pertaining to Medicaid DSH allotment reductions. We will undertake future rulemaking as may be necessary to ensure that the regulations continue to implement statutory requirements appropriately.

Comment: We received several comments related to the Medicare Inpatient Prospective Payment System (IPPS) rules.

Response: Comments on the Medicare IPPS rules are outside the scope of this rule.

Comment: One commenter expressed concern that the proposed methodology will exacerbate current inequalities in Medicare IPPS and jeopardize the existence of hospitals already experiencing negative margins.

Response: The Medicaid and the Medicare programs are distinct programs authorized under different titles of the statute and the Medicare and Medicaid DSH rules have somewhat different purposes and statutory directives. Section 1923(f)(7)(B) of the Act establishes five factors that must be considered in the development of the DHRM. While we appreciate the commenter's concern, considerations related to the Medicare IPPS are not included in the factors Congress has specified to be considered in the DHRM. However, states will continue to have considerable flexibility in setting DSH state plan payment methodologies, to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations.

Comment: One commenter expressed concern that California's estimated DSH reductions are more than double those estimated in the proposed rule released in 2013.

Response: The aggregate DSH allotment reductions shown for FY 2018, as included in the illustrative model included in the July 2017 proposed rule, were greater for all states (except Tennessee) than the aggregate DSH allotment reduction amounts in the illustrative example for the 2013 DSH allotment reduction proposed rule. This was the result of the magnitude of the reductions shown in the illustrative example in the July 2017 proposed rule, which were $2 billion, while the reductions shown in the 2013 proposed rule were $500 million. Additionally, the state-specific DSH allotment reductions included in both proposed rules were part of illustrative examples to show how the DHRM would work, and were not estimated reduction amounts. Under current law FY 2018 would not be subject to annual allotment reductions which will now begin in FY 2020 and run through FY 2025.

Comment: One commenter questioned whether state-specific DSH allotment reductions for each fiscal year will increase proportionately as the annual aggregate DSH allotment reductions increase.

Response: Each state's annual DSH allotment reduction will be determined annually based on the DHRM.

Comment: One commenter stated that 50 percent of all hospitals are DSH and expresses concern that the reductions may be unevenly allocated.

Response: We believe that the DHRM will determine state DSH allotment reductions in an equitable manner consistent with statutory requirements. States will continue to have considerable flexibility in setting DSH state plan payment methodologies, to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations.

B. DHRM Data Sources

The statute establishes parameters regarding data and data sources for specific factors in the development of the DHRM. In the July 2017 proposed rule, we proposed that the DHRM would rely, wherever possible, on data sources and metrics that are consistent with the statute, transparent, and readily available to CMS, states, and the public, such as: DSH MIUR data; Medicaid DSH data reported as required by section 1923(j) of the Act; United States Census Bureau (Census Bureau) data; existing state DSH allotments; and Form CMS-64 Medicaid Budget and Expenditure System (MBES) data. We proposed to utilize the most recent year available for all data sources and proposed to align the state plan rate year (SPRY) of data sources whenever possible. Selected data sources are discussed in greater detail below, including our responses to comments regarding particular data sources.

1. MIUR Data

To ensure that all hospitals are properly deemed disproportionate share in accordance with section 1923(b) of the Act, states must determine the mean MIUR for hospitals receiving Medicaid payments in the state and the value of one standard deviation above the mean. States are currently required to provide this data to CMS annually under § 447.294(d) (CMS-R-266, Office of Management and Budget (OMB) 0938-0746). We proposed to utilize MIUR data from the year that corresponds to the DSH audit SPRY used in the calculation of each state's DSH allotment reduction.

2. Medicaid DSH Audit and Reporting Data

We also proposed to rely on data derived from Medicaid DSH audit (CMS-R-266, OMB 0938-0746) and reporting data (CMS-R-266, OMB 0938-0746). The data is reported by states as required by section 1923(j) of the Act and the “Medicaid Disproportionate Share Hospital Payments” final rule published on December 19, 2008 (73 FR 77904) (and herein referred to as the 2008 DSH audit final rule) requiring state reports and audits to ensure the appropriate use of Medicaid DSH payments and compliance with the hospital-specific DSH limit imposed at section 1923(g) of the Act. This is the only comprehensive data source for DSH hospitals that identifies hospital-specific DSH payments and uncompensated care costs in a manner consistent with Medicaid DSH program requirements.^[3]

To date, we have received rich, comprehensive audit and reporting data from each state that makes Medicaid DSH payments. To facilitate the provision of high quality data, we provided explicit parameters in the 2008 DSH audit final rule and associated policy guidance for calculating and reporting data elements. As the data elements are based on hospital costs reports and are subject to audit, the data elements are not due to CMS until the end of the calendar year 3 years following the end of each SPRY. Additionally, state submitted audit and reporting data is subject to detailed CMS review to ensure quality and accuracy and requires significant resources to compile and prepare for use in the DHRM. This means that the data used for the methodology may not be the most recently submitted data, but instead the most recent data available to us in usable form. For the reductions scheduled for FY 2020, we anticipate utilizing SPRY 2015 DSH audit and reporting data, which was due to CMS from states on December 31, 2018. We considered utilizing alternative uncompensated care cost data and Medicaid utilization data from sources such as the Medicare Form CMS-2552 (OMB 0938-0050), which we explained in more detail in the 2013 DSH allotment reduction final rule. The DSH audit and reporting data, however, remains the only comprehensive reported data available that is consistent with Medicaid program requirements.

3. United States Census Bureau Data

As required by the statute, the DHRM must impose the largest percentage DSH allotment reductions on the states that have the lowest percentages of uninsured individuals. Although other sources of this information could be considered for this purpose, the statute explicitly refers to the use of data from the Census Bureau for determining the percentage of uninsured for each state. As with the 2013 DSH allotment reduction final rule, we identified and considered two Census Bureau data sources for this purpose: The American Community Survey (ACS); and the Annual Social and Economic Supplement to the Current Population Survey (CPS). In consultation with the Census Bureau, we proposed to use the data from the ACS for the following reasons. First, the ACS is the largest household survey in the United States; in that regard, the annual sample size for the ACS is over 30 times larger than that for the CPS—about 3 million for the ACS versus 100,000 for the CPS. The ACS is conducted continuously each month throughout the year, with the sample for each month being roughly 1/12th of the annual total, while the CPS is conducted in the first 4 months following the end of the survey year.

Finally, although the definition of uninsured and insured status is the same for the ACS and the CPS, the CPS considers the respondents as uninsured if they are uninsured at any time during the year whereas the ACS makes this determination based on whether the respondent has coverage at the time of the interview, which are conducted at various times throughout the year. For these reasons, and with the recommendation of the Census Bureau, we determined that the ACS is the appropriate source for establishing the percentage of uninsured for each state for purpose of the DHRM.

We received a number of public comments on our proposals regarding DHRM data sources in the July 2017 proposed rule. A discussion of these comments, with our responses, appears below.

Comment: One commenter expressed support for the DSH audit and reporting data being the source for uncompensated care cost data for the DHRM.

Response: We thank the commenter for the support for the proposal and are finalizing the use of the DSH audit and reporting data as the source of uncompensated care cost data for the DHRM.

Comment: Several commenters expressed concern regarding the use of DSH audit and reporting data for the DHRM. The commenters cited various concerns regarding the DSH audit data, including the use of out-of-date data which causes a lag between DSH policy and programmatic changes that is not reflected in audit and reporting data. One commenter indicated that use of the DSH audit data penalized states because it is not reflective of DSH payment policy changes that have been made in later time periods following the audit year. Many commenters requested that CMS provide states with at least 4 years advance notice of its intent to utilize DSH audit data for reductions based on payment targeting to give states proper time to consider adjustments to their programs. One commenter expressed concern that the timeliness of the DSH audit data undermines the incentive for states to target DSH payments because states have to wait 5 years, which the approximate lag time between a particular SPRY subject to audit and when related data for that year becomes available for use in the DHRM, to see the benefits of targeting hospitals with high Medicaid utilization and high uncompensated care costs. Some commenters recommended that CMS use uniform data in the DHRM wherever possible among all hospitals. Other commenters recommended that we consider initiating a separate survey to determine uncompensated care costs for a more recent year than the DSH audit data we propose to use in the DHRM. Several commenters recommended that CMS revise the DHRM if a source of timelier data become available.

Response: The Medicaid DSH audit and reporting data is the most comprehensive reported data available that is consistent with Medicaid program requirements. To date, we have received audit and reporting data from each state that makes Medicaid DSH payments. To facilitate the provision of high quality data, we provided explicit parameters in the 2008 DSH final rule and associated policy guidance for calculating and reporting data elements. The 2008 DSH final rule included a transition period in which states and auditors could develop and refine audit and reporting techniques. Moreover, states have had ample time to implement DSH payment methodologies that could mitigate DSH allotment reductions related to the DSH payment targeting factors, which have been codified in statute since March 23, 2010, and prior rulemaking as finalized in the 2013 DSH allotment reduction rule and as discussed in the July 2017 proposed rule. This final rule will not affect the considerable flexibility afforded to states with regard to establishing DSH state plan payment methodologies to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations.

We currently have no plans to develop a separate survey to serve as a timelier source of uncompensated care costs. However, we do not believe a timelier source of high quality data could be developed given that cost reports used to calculate uncompensated care costs may not be settled for 2 or more years following the end of a fiscal year. Moreover, an additional time period would be needed to allow for review and/or audit of this data to ensure its quality and accuracy. This would impose administrative burden on states, hospitals and us by essentially doubling effort relating to DSH auditing and reporting. As such, we are finalizing reliance on existing DSH audit and reporting data in the DHRM because it represents the best available data that is consistent with existing program requirements without imposing duplicative and otherwise unnecessary burden. Notwithstanding, we will continue to monitor the reduction methodology after implementation and will consider whether the development of a timelier data source is warranted, which we would undertake through future rulemaking, as necessary.

Comment: Several commenters recommended that CMS modify DSH audit requirements to rely on estimated costs in calculating hospital-specific limits instead of relying on actual costs to allow for more recent data to be included in the DHRM. Two commenters suggested that this approach would also minimize the financial burden that conducting independent certified DSH audits places upon states.

Response: While we recognize that states must use estimates to determine DSH payments in a given Medicaid SPRY, the independent certified DSH audits are statutorily-required under section 1923(j) of the Act to verify the extent to which such estimates are reflective of the actual costs and that resultant payments do not exceed the limitations on DSH payments imposed by Congress.

Comment: One commenter expressed concern that the DSH audit and reporting data included negative values for uncompensated care.

Response: Negative values for uncompensated care costs occur where hospitals receive payments by or on the behalf of Medicaid patients and the uninsured for inpatient and outpatient hospital services that exceed the costs of providing inpatient hospital and outpatient hospital services to such individuals.

Comment: One commenter recommended that CMS modify the DSH reporting requirements to collect total hospital costs from the Medicare cost report for all hospitals that receive DSH payments.

Response: We confirm that as part of the DSH audit submission, states are currently required to report total hospital costs, meaning the total annual costs incurred by the hospital for furnishing inpatient hospital and outpatient hospital services, for each in-state hospital that receives a DSH payment, per § 447.299(c).

Comment: One commenter requested a detailed explanation of how CMS derived Massachusetts' HMF and HUF reduction or the HMF and HUF reduction for any state missing hospital-specific DSH payments.

Response: As of the publication of this final rule, we have not calculated FY 2020 DSH allotment reductions. We will calculate FY 2020 reductions for Massachusetts and all other states by utilizing the final DHRM. States that do not make DSH payments may still receive a DSH allotment reduction.

C. DHRM Overview

The statute requires aggregate annual reduction amounts to be implemented through a DHRM designed by the Secretary consistent with statutorily-established factors. Taking these factors into account for each state, we proposed that the DHRM would generate a state-specific DSH allotment reduction amount for each applicable fiscal year for all states and the District of Columbia, with the exception of Tennessee, whose DSH allotment is defined in section 1923(f)(6)(A)(vi) of the Act to be $53.1 million, notwithstanding DSH allotment reductions in section 1923(f)(7) of the Act, for each FY from 2015 through 2025. The total of all DSH allotment reduction amounts would equal the aggregate annual reduction amount identified in statute for each applicable fiscal year. To determine the effective annual DSH allotment for each state, we proposed that the state-specific annual DSH allotment reduction amount would be applied to the unreduced DSH allotment amount for its respective state.

We proposed to calculate an unreduced DSH allotment for each state prior to the beginning of each FY, as we do currently. This unreduced allotment is determined by calculating the allotment in section 1923(f) of the Act prior to the application of the DHRM under section 1923(f)(7) of the Act. We proposed that the unreduced allotment would serve as the base amount for each state to which the state-specific DSH allotment reduction amount would apply annually. In the July 2017 proposed rule, we utilized estimated unreduced DSH allotments for FY 2017 for illustrative purposes. Moreover, we indicated that the illustrative estimate may rely on different data than what we proposed to use when calculating annual DSH allotment reductions for FY 2018, which is when reductions were scheduled to begin when we published the July 2017 proposed rule, and anticipated that more recent data would be available when calculating the final allotment reductions.

We proposed to apply the DHRM to the unreduced DSH allotment amount on an annual basis for the fiscal years specified in statute as subject to DSH allotment reduction. In developing the proposed DHRM, we considered the factors identified in the statute to determine each state's annual state-specific DSH allotment reduction amount.

We proposed a DHRM that utilizes the best available data at the time of the annual DSH allotment reduction calculations, and proposed that we would not recalculate the reduction amounts based on revised or late DSH audit reports, MIUR data, or other relevant data. The DHRM would also rely on a series of interacting calculations that result in the identification of state-specific reduction amounts that, when summed, equal the aggregate DSH allotment reduction amount identified by the statute for each applicable year. The proposed DHRM accomplishes this through the following summarized steps:

1. Separate states into two overall groups, non-low DSH states and low DSH states, to give effect to the statutory low DSH criterion. (States falling into each category were listed in Table 1 of the July 2017 proposed rule).

2. Proportionately allocate aggregate DSH funding reductions to each of these two state groups based on each state group's proportion of the total national unreduced DSH allotment amount.

3. Apply a low DSH adjustment percentage to adjust the non-low DSH and low DSH state groups' DSH funding reduction amount. This step maintains the combined aggregate DSH funding reduction for the low DSH and non-low DSH state groups by distributing a portion of the unadjusted low DSH state DSH funding reduction amount across the non-low DSH state group, as described in greater detail below.

4. Divide each state group's DSH allotment reduction amount among three statutorily-identified factors, the uninsured percentage factor (UPF), the high level of uncompensated care factor (HUF), and the high volume of Medicaid inpatients factor (HMF). We proposed to assign a 50 percent weight to the UPF and a 50 percent combined weight for the two DSH payment targeting factors (a 25 percent weight for the HUF, and a 25 percent weight for the HMF). This approach would assign equal weights based on the statutory structure under which the UPF is presented separately, in section 1923(f)(7)(B)(i)(I) of the Act, while the HMF and HUF are grouped together in section 1923(f)(7)(B)(i)(II) of the Act, at items (aa) and (bb). Additionally, compared to the approach taken in the 2013 DSH allotment reduction final rule, this weight assignment would place greater emphasis on the UPF to:

• Reduce the impact of the DSH allotment reduction for states with greater DSH need due to high uninsurance rates.
• Give greater weight to more recent data, since the UPF data relies on more recent data than the HUF and HMF.

We considered various alternative weight assignments prior to proposing equal weights for the UPF as specified in section 1923(f)(7)(B)(i)(I) of the Act and for the combined HMF and HUF as specified in section 1923(f)(7)(B)(i)(II) of the Act. We decided to propose the 50 percent weight for the UPF and a 50 percent combined weight for the two DSH payment targeting factors to reduce the impact of the DSH allotment reductions for states with high uninsurance rates, place a greater weight on more recent data, and reflect how these factors are specified in statute.

5. Limit the reduction to be applied to each state's total unreduced DSH allotment to 90 percent of its original unreduced allotment. Any excess reduction amounts called for under the DHRM which are limited by this reduction cap will be factored back into the reduction model and be redistributed among the remaining states that do not exceed the reduction cap based on the proportion of each remaining state's allotment reduction amount to the aggregate allotment reduction amount for its respective state group. This operation would be performed separately for each state group such that, for example, an excess reduction amount attributable to a low DSH state would be reapportioned only among other low DSH states and would not be reapportioned among any states in the non-low DSH state group. By limiting the overall amount by which each state's allotment may be decreased, we proposed to preserve at least 10 percent of each state's unreduced DSH allotment, thereby allowing all states to continue to make DSH payments. Placing limits on the reductions applied to each state's original unreduced allotments was a new proposal that was not considered in the 2013 DSH allotment reduction final rule. In view of the then-required aggregate DSH allotment reduction amounts and the DHRM under the 2013 DSH allotment reduction final rule, no state was in jeopardy of having its entire DSH allotment eliminated for FY 2014 or FY 2015 at the time that rule was promulgated. However, with the larger reduction amounts that were scheduled for FYs 2018 through 2025 under the statute as it was in effect at the time of the July 2017 proposed rule, and the reduction amounts currently scheduled for FYs 2020 through 2025, which are as high as $8 billion annually, states could experience the elimination of their entire DSH allotment without the inclusion of a reduction cap methodology in the DHRM. Although we did consider different reduction cap percentages, we believe the proposed 90 percent reduction cap strikes a balance between ensuring reduction amounts are determined based on the statutory DHRM factors and ensuring states maintain the ability to make an appreciable amount of DSH payments. Lower reduction caps would cause the reductions to approach even distribution among all states, instead of being based on the statutory DHRM factors. No cap might result in the complete elimination of some states' DSH allotments.

6. For each state group, determine state-specific DSH allotment reduction amounts relating to the UPF. To accomplish this, we will compare each state's uninsurance rate to the uninsurance rates of all states in relation to each state's unreduced allotment in proportion to its respective state group's total unreduced allotment to calculate each state's reduction. As required by statute, states with lowest uninsurance rates will receive largest percentage DSH reductions.

7. For each state group, determine state-specific DSH allotment reduction amounts relating to the HUF. By utilizing the most recently available Medicaid DSH audit and reporting data, we will determine the mean uncompensated care level for each state to determine the total payments each state makes to non-high uncompensated care level hospitals. We will then determine the HUF by dividing the total of each state's total payments made to non-high uncompensated care level hospitals by the total payments made non-high uncompensated care level hospitals for its respective state group.

8. For each state group, determine state-specific DSH allotment reduction amounts relating to the HMF. Again, by utilizing the most recently available Medicaid DSH audit and reporting data, we will determine the mean MIUR for each state to determine the amount of DSH payments each state makes to non-high Medicaid volume hospitals. We will then determine the HMF by dividing each state's total payments made to non-high volume Medicaid hospitals by the total payments made non-high volume Medicaid hospitals for its respective state group.

9. Apply a section 1115 budget neutrality factor (BNF) for each qualifying state. To apply this factor, we will not reduce any portion of a state's DSH allotment which was included in the budget neutrality calculation for a coverage expansion that was approved under section 1115 of the Act as of July 31, 2009. We will assign any qualifying states an average percentage reduction amount within its respective state group for diverted DSH allotment amounts that are not related to a coverage expansion in effect as of July 31, 2009 and for which the state does not have complete and/or relevant DSH payment data.

10. Identify the state-specific DSH allotment reduction amount.

11. Subtract each state's state-specific DSH allotment reduction amount from each state's unreduced DSH allotment to determine the state's available DSH allotment for the applicable year.

The manner in which we proposed that each of the five factors would be considered and calculated in the proposed DHRM is described in greater detail below.

The DHRM recognizes the variations in DSH allotments among states and the application of the methodology generates a lesser impact on low DSH states. The DHRM is designed to determine DSH allotment reductions in an equitable manner by grouping similar states together for purposes of applying the statutory reduction factors. Reductions assigned through the HMF and HUF would lessen the impact on states that have targeted DSH payments to hospitals that have high volumes of Medicaid inpatients and to hospitals that have high levels of uncompensated care, respectively, while incentivizing payment targeting for future DSH payments. As specified in statute, the DHRM would also take into account the extent to which the DSH allotment for a state was included in part or in whole in the budget neutrality calculation for a coverage expansion approved under section 1115 of the Act as of July 31, 2009 by excluding from DSH allotment reduction the amount of DSH that qualifying states continue to divert specifically for coverage expansion in the budget neutrality calculation. Any amount of DSH diverted for other purposes under the demonstration would still be subject to reduction by automatically assigning qualifying states an average percentage reduction amount within its respective state group for factors for which the state does not have complete and/or relevant DSH payment data.

We received the following comments regarding the overall approach to the DHRM and have responded to the comments below.

Comment: One commenter expressed concern that the proposed DHRM would result in a significant reduction for its state and recommended revising the proposed methodology to reduce the impact of the DHRM on the commenter's state.

Response: We are finalizing a DHRM that will reduce DSH allotments annually by an aggregate amount set in statute, using a methodology that is consistent with statutory factors that direct the allocation of the annual reduction amount among the states.

Comment: One commenter requested information regarding which data will be used to calculate the preliminary DSH allotments. Other commenters recommended that CMS be transparent about the data sources, including by identifying which states will have the BNF applied to their allotment reduction calculation. Many commenters recommended that CMS post all the data sets used to implement the FY 2018 DHRM on its website and post a more comprehensive explanation of the calculation for each component of each state's total reduction.

Response: Currently, we calculate preliminary unreduced DSH allotments based on data available around the August preceding the start of each fiscal year and publish an annual notice in the Federal Register with detailed information regarding the data sources used for each fiscal year. These data sources include the previous year's preliminary unreduced DSH allotment, the change in the previous year's consumer price index, and state budget estimates from MBES. In addition to publishing an annual notice in the Federal Register and updating MBES at the beginning of each FY to reflect each state's preliminary DSH allotment amount, we also inform states prior to the beginning of each FY of their preliminary DSH allotment via direct electronic communication. In this communication, we provide states with all relevant data utilized to calculate both the annual preliminary DSH allotment and IMD limits, which is analogous to the information that is provided and published in the Federal Register.

In the July 2017 proposed rule, we included a detailed description of the proposed DHRM methodology. We thoroughly reviewed and carefully considered public comments, and issued this final rule in a timely manner incorporating input from public comments. This final rule also provides a detailed methodological description of the DHRM. To ensure the use of most recent available data, we do not intend to calculate the FY 2020 DSH allotment reductions until after the publication of this final rule. Also, we intend to publish a separate DHRM technical guide that provides information regarding the DHRM calculation and associated data sources.

Comment: Several commenters noted concern with CMS' use of the FY 2017 DSH allotments, FY 2013 DSH audit data, and state-reported MIUR data to generate FY 2018 DSH allotment reduction amounts. Commenters stated that the data were not consistent with Medicaid statute, transparent, and readily available to the public during the notice and comment period and that the lack of transparency significantly hampered state governments' and stakeholders' ability to assess how the DHRM would affect their state DSH allotment, particularly for FY 2018, the first year that annual state DSH allotment reductions were scheduled to be implemented at the time of the July 2017 proposed rule. Additionally, the commenters requested that we identify a more comprehensive and reliable source for calculating the uninsured rate for each state and not rely upon survey sampling results.

Response: We believe that the data used in the DHRM as described in the July 2017 proposed rule is consistent with the statute, transparent and readily available to CMS and the public. The statute requires that the percentage of uninsured individuals is determined on the basis of data from the Census Bureau, audited hospital cost reports, and other information likely to yield accurate data, during the most recent year for which such data are available. For hospitals that receive DSH payments and are included in the DSH audit and reporting data (which CMS makes readily available to the public on an annual basis), we proposed and are finalizing the use of the most recent complete DSH audit and reporting data for purposes of the DHRM. For purposes of this rule, we intend to use the most recent DSH audit and reporting data available at the time of allotment reduction calculation based on the existing DSH audit and reporting process. Additionally, we intend to publish a separate DHRM technical guide that provides information regarding the DHRM calculation and associated data sources.

Comment: A few commenters suggested that due to the lack of timely and transparent data it would be difficult to fully assess CMS' proposal and noted that it would be irresponsible for CMS to move forward with DSH allotment reductions without resolving commenters' data transparency concerns and technical questions. One commenter stated that a delay is warranted so that CMS can address important deficiencies with transparency and outstanding legal questions impacting the data that, if not addressed prior to implementation, would have a material impact on the distribution of the reductions across states.

Response: More recent data will be available at the time CMS calculates annual reductions for FY 2020 (and thereafter) than was available at the publication of the July 2017 proposed rule. Therefore, we used an illustrative example to assist in transparency and provided the detailed DHRM, which we are statutorily-required to develop, to specify the methodology for determining the annual DSH allotment reduction amounts. As finalized, we believe the DHRM will use the timeliest, most transparent, and comprehensive reported data available that is consistent with Medicaid program requirements. As stated above, while a number of issues related to Medicaid DSH payment calculations currently are the subject of litigation, the statutorily-required allotment reductions and the DHRM are not among them, and we are bound by statute to adopt a rule to implement the DSH reductions. With this final rule, we are doing so according to our view of the best interpretation of the DSH statute and will utilize the most recent data available to us that is consistent with applicable laws and regulations. In an effort to be transparent in the application of the DSH allotment reductions, we intend to publish a separate DHRM technical guide that provides information regarding the DHRM calculation and associated data sources.

Comment: One commenter requested that CMS provide an opportunity for qualified stakeholders and consultants to confer directly with the CMS contractor that has performed work relating to the DHRM.

Response: We will not provide stakeholders with a formal process to confer directly with CMS contractors involved with calculations or other work relating to the DHRM. We are available to provide technical assistance to states regarding the DHRM following the publication of this final rule.

Comment: Several commenters suggested that the timeline of publication of preliminary DSH allotments does not support transparency, citing examples that the preliminary DSH allotments for FY 2016 were not public until late 2016 and the FY 2017 allotments were not expected to be made public until after 2018.

Response: We disagree and believe the rulemaking regarding proposed DSH allotment reductions has been timely. In addition, we notify states electronically and through MBES of their preliminary DSH allotments at the start of each federal fiscal year. We also finalize DSH allotment amounts as soon as all necessary information is available. The preliminary and final DSH allotment amounts are also published in the Federal Register. Moreover, we do not believe that knowledge of future preliminary unreduced DSH allotment amounts in necessary for evaluating the DHRM. In general, the DSH allotments for each state is increased by the consumer price index each year, so each state's unreduced DSH allotment remains constant in proportion to the total national DSH allotment.

Comment: One commenter stated that the methodology for calculating the state-specific cap on the annual DSH allotment reduction ignores what the commenter stated is an existing inequality across states in unreduced DSH allotments as established by the Balanced Budget Act of 1997 (Pub. L. 105-33, enacted August 5, 1997) which were based on each state's 1995 DSH spending levels. Several commenters supported a state-specific cap on annual reductions that will allow states to keep at least a portion of their DSH allotment. Commenters also recommended various modifications to the cap, and that CMS re-evaluate the cap based on experience. Some commenters recommended that states be permitted to retain more than 10 percent of their unreduced allotments, but did not recommend a percentage. One commenter suggested that CMS implement a reduction cap based on each state's cost coverage percentage determined by dividing each state's total uncompensated care by its respective unreduced DSH allotment. States with a cost coverage percentage below the national average would be subject to a cap on DSH allotment reductions with low-DSH states' reductions being capped at 5 percent reduction of their unreduced allotment, while non low-DSH states' reductions would be capped at 7 percent reduction of their unreduced allotment.

In addition, a few commenters did not support a state-specific cap on annual DSH allotment reductions that will allow states to keep at least a portion of their DSH allotment. One commenter indicated that a cap on DSH allotment reductions did not appear in the final 2013 DSH allotment reduction rule and should not be permitted to compete with the statutory obligations to implement the DSH allotment reductions. One commenter believes states can make their own determination regarding what level of funding is sufficient and that a cap on reductions shifts reductions away from states with lesser need to states with greater need for DSH funding.

Response: We believe that the DHRM, including the state-specific reduction cap methodology, calculates DSH allotment reductions in an equitable manner consistent with statutory requirements. We are finalizing our proposed state-specific cap that limits the reduction to be applied to each state's total unreduced DSH allotment to 90 percent of its original unreduced allotment because it strikes a balance between ensuring reduction amounts are determined based on the statutory DHRM factors and ensuring states maintain the ability to make an appreciable amount of DSH payments. Lower reduction caps might cause the reductions to approach even distribution among all states instead of being based on the statutory DHRM factors. No cap might result in the complete elimination of some states' DSH allotments and higher caps might result in states with an insignificant amount of DSH allotment with which to make DSH payments. We did not consider a state-specific reduction cap in the 2013 DSH allotment reduction rule since no state was in jeopardy of having its entire DSH allotment eliminated under the amounts designated under statute during that time. We will evaluate the reduction methodology after implementation and will consider whether modifications are warranted, which we would undertake through future rulemaking, as necessary.

Comment: Several commenters recommended that the DHRM reduce allotments by first applying it to unused state DSH allotments, then applying the factors set forth in the DHRM.

Response: Section 1923(f)(7) of the Act specifies the five factors for the DHRM, but does not distinguish between spent and unspent state DSH allotment amounts in directing that the allotments be reduced. Therefore, we did not propose and are not finalizing a policy to apply reductions first to unspent DSH allotment amounts before application of the DHRM. We believe that commenters' suggested method could serve to penalize unfairly states that do not currently expend their entire DSH allotment. Therefore, we are finalizing the structure of proposed DHRM that considers five factors identified by section 1923(f)(7)(B) of the Act when determining state-specific allotment reduction amounts.

Comment: One commenter noted concerned that CMS would use FY 2017 state DSH allotments to calculate allotment reduction amounts for FY 2018.

Response: As proposed, we will use the preliminary unreduced DSH allotment for each fiscal year to calculate DSH allotment reductions for the corresponding fiscal year. Specifically, we will utilize the preliminary unreduced FY 2020 DSH allotment amounts to calculate FY 2020 DSH allotment reductions.

Comment: One commenter expressed concern that the DHRM, by considering the five factors separately and summing the results, could create disproportionately large reductions for states with high levels of uninsured that are targeting hospitals with both a high volume of Medicaid inpatients and a high level of uncompensated care. The commenter stated this is in violation of the statutory intent.

Response: We disagree and believe the proposed methodology, which we are adopting in this final rule, supports the intent of the statute and the proposed rule, as it imposes smaller percentage reductions on low DSH states compared to non-low DSH states and, within each state group, imposes larger percentage reductions on states that have the lowest percentages of uninsured individuals and on states that do not target their DSH payments to hospitals with high volumes of Medicaid inpatients and high levels of uncompensated care. Further, the proposed DHRM takes into account the extent to which a state's DSH allotment was included in the budget neutrality calculation for a coverage expansion that was approved under section 1115 demonstration authority as of July 31, 2009.

We interpret the statute to require CMS to utilize both the UPF and the two targeting factors. We proposed to assign a 50 percent weight to the UPF and a 50 percent combined weight for the two DSH payment targeting factors (a 25 percent weight for the HUF, and a 25 percent weight for the HMF). We believe that this is an equitable approach for assigning factor weights, and appropriately implements the statutorily-required factors. This weight distribution does preserve more DSH allotment (that is, it imposes smaller allotment reductions) for states that may have greater DSH need due to high uninsurance rates while still incentivizing states to continue to target DSH payments to hospitals with both a high volume of Medicaid inpatients and high level of uncompensated care. Additionally, we proposed, and are finalizing, a weight of 50 percent for the UPF to rely more heavily on more recent Census Bureau data and to align the factor weights with how these factors are set forth in statute. We believe the proposed DHRM is an equitable method for calculating reduction amounts based on each state's rate of uninsurance and how well each state is targeting its DSH payments to hospitals with high volumes of Medicaid inpatients and high levels of uncompensated care.

Comment: Two commenters requested that CMS require states to allocate the reduction amount between Institutions for Mental Diseases (IMD) and all other hospitals proportionately so IMDs do not have to absorb a higher proportion of the DSH reductions.

Response: We will calculate the IMD DSH limit under section 1923(h) of the Act based on the state's DSH allotment after the reduction is applied, to ensure that the IMD DSH limit is subject to a reduction consistent with the overall reduction of the state's annual DSH allotment.

Comment: Another commenter suggested that CMS apply the DSH allotment reductions to the unreduced allotment and treat any DSH payments states make over the reduced allotment as an overpayment.

Response: We are finalizing a DHRM that will calculate annual reductions that will apply to unreduced DSH allotments. Additionally, section 1923(f) of the Act limits FFP for total statewide DSH payments made to eligible hospitals in each federal fiscal year to the annual DSH allotment for each state, which will be reduced annually through the DHRM for FYs 2020 through 2025. Any state claims for FFP in excess of the state's reduced annual DSH allotment are subject to potential disallowance as specified in 42 CFR 430.42.

Comment: Several commenters recommended that CMS allow for a process to revise the calculation of DSH allotment reductions. Some commenters suggested that CMS publish the underlying data and calculations for each factor included in the DHRM for each year so that states can validate the accuracy of the data and the calculations and work with CMS to make any corrections that might be necessary based on more up to date or corrected data related to DSH audit reports, MIUR, or other data.

Response: We will conduct a thorough review to ensure the quality and accuracy of all data and calculations. To promote transparency, we intend to publish a separate DHRM technical guide that will include all data source information and the underlying DHRM calculations. During the development and publication of this final rule, we have continued to work with states to ensure that we are utilizing accurate, complete data that is the most recent available, prior to calculating the FY 2020 DSH allotment reductions. Due to the timeframes associated with the publication of this final rule and the statutorily-required DSH allotment reductions scheduled to be applied to state FY 2020 DSH allotments, we will calculate the FY 2020 DSH allotment reductions using the most currently available data at the time we apply the DHRM to determine the allotment reductions, prior to October 1, 2019. In subsequent years, beginning with FY 2021, we anticipate that we will assemble necessary data and perform calculations to determine the DSH allotment reductions for the FY during the months of July, August, and September before the start of the FY, to enable us to publish the DSH allotment reductions prior to the start of the FY to which they will apply. Accordingly, for the annual DSH allotment reductions beginning with FY 2021, states must have submitted all revised and corrected data to CMS by July 1st of the FY prior to the FY for which reductions will be calculated and applied to each state's unreduced preliminary DSH allotment, so that the most recent data available to us at the time we apply the DHRM reflects all revisions and corrections determined by the state. For example, to be used in applying the DHRM for FY 2021, all corrected and revised data would be required to be submitted to us by July 1, 2020 (and meet applicable federal requirements) to be reflected in the DHRM calculations for the DSH allotment reductions scheduled to be applied to the FY 2021 unreduced preliminary DSH allotments. We anticipate that this schedule would be in effect for any years following FY 2020 for which DSH allotment reductions are to be applied under the statute.

Comment: Several commenters noted support for CMS' emphasis on targeting of DSH payments to hospitals with high volumes of Medicaid inpatients and hospitals with high levels of uncompensated care in the DHRM. One commenter urged CMS to incentivize states to target DSH payments to hospitals providing the highest share of care to low-income patients within each state.

Response: We believe that the proposed DHRM, incorporating the statutory factors identified in section 1923(f)(7)(B) of the Act, does incentivize states to target their DSH payments, both through the HMF and HUF, to hospitals providing care to low-income individuals, and have incorporated this method in the final rule.

Comment: Many commenters expressed concern with CMS' proposed increase of the UPF from a 33 percent weight, as finalized in the 2013 DSH reduction rule, to a 50 percent weight. Commenters stated that the 50 percent UPF weight would disadvantage states that have expanded Medicaid coverage under the ACA and create disincentives for states to continue to cover the Medicaid expansion population. One commenter noted support for the 50 percent UPF weight due to the opinion that this would minimize annual DSH allotment reductions for non-expansion states. Many commenters recommended that CMS revert back to the 33 percent weight for each of the core factors, the UPF, the HMF and the HUF. One commenter suggested that an equal weighting of the three core factors is appropriate in this period of market uncertainty. Commenters also variously recommended: That the UPF be weighted at 25 percent or less; that an 80 percent weight be placed on the UPF and a 10 percent weight on each of the targeting factors, the HMF and the HUF, to mitigate annual DSH allotment reductions for states that did not expand Medicaid; that a 60 percent weight be applied to the UPF and 20 percent to each of targeting factors, the HMF and the HUF; and that the weight assigned to the UPF be increased if other consideration were not given to mitigate the impact of the reductions on non-expansion states.

Response: We are finalizing our proposal to apply a weight of 50 percent to the UPF to rely more heavily on the more recent Census Bureau data (as it is more recent than DSH audit data and, therefore, likely more reflective of current circumstances than DSH audit data) and to align the factor weights with how these factors are set forth in statute. Section 1923(f)(7)(B)(i)(I) of the Act requires that the UPF be incorporated into the DHRM, while section 1923(f)(7)(B)(i)(II)(aa) of the Act requires that the HMF be incorporated into the DHRM and section 1923(f)(7)(B)(i)(II)(bb) of the Act requires that the HUF be incorporated into the DHRM. This structure of subclauses and items is consistent with a 50 percent weight being applied to the factor identified in section 1923(f)(7)(B)(i)(I) of the Act and an equal 50 percent weight being applied to the factors identified in section 1923(f)(7)(B)(i)(II) of the Act. The 50 percent UPF weight and combined 50 percent targeting factor weight will yield different results for both expansion and non-expansion states depending on each state's rate of uninsured and how well each state targets its DSH payments to hospitals with high volumes of Medicaid inpatients and uncompensated care. We believe that the weighting in the July 2017 proposed rule is a reasonable approach and have incorporated this methodology into the final rule.

Comment: One commenter recommended that the weight of the HMF be increased to provide consideration for states with high Medicaid enrollment.

Response: We disagree with the recommendation because we believe that the proposed DHRM reduces DSH allotments in an equitable manner that is consistent with the statute. The DHRM gives consideration to states with high Medicaid enrollment that target DSH payments to hospitals with high volumes of Medicaid inpatients. We believe that the proposed weighting is a reasonable approach to implementing the statutory requirements for the DHRM and are finalizing this methodology in § 447.294(e)(5) in this final rule.

D. Factor 1—Low DSH Adjustment Factor (LDF)

The first factor considered in the proposed DHRM is the Low DSH Adjustment Factor identified at section 1923(f)(7)(B)(ii) of the Act, which requires the DHRM to impose a smaller percentage reduction on “low DSH states” that meet the criterion described in section 1923(f)(5)(B) of the Act. To qualify as a low DSH state, total expenditures under the state plan for DSH payments for FY 2000, as reported to us as of August 31, 2003, had to have been greater than zero but less than 3 percent of the state's total Medicaid state plan expenditures during the FY. Historically, low DSH states have received lower DSH allotments relative to their total Medicaid expenditures than non-low DSH states.

We proposed to apply the LDF by imposing a greater proportion of the annual DSH funding reduction on non-low DSH states. To meet the statutory requirement to impose a smaller percentage reduction on low DSH states, the DHRM would create two state groups (low DSH states and non-low DSH states), then would apply the LDF when allocating reduction amounts to each state group. The LDF is calculated and applied as follows:

1. Separate states into two groups, non-low DSH states and low DSH states.

2. Divide each state's unreduced preliminary DSH allotment for the year for which the reduction is calculated by estimated Medicaid service expenditures for that same year. Currently, we create a preliminary DSH allotment based on the estimates available in August of the prior year and we issue a final DSH allotment once the federal FY ends.

3. For each state group, calculate the non-weighted mean of the value calculated in step 2 for states in the group.

4. Divide the average calculated in step 3 for the low DSH state group by the average calculated in step 3 for the non-low DSH state group.

5. Convert this number to a percentage. This percentage is the LDF.

6. Multiply the proportionately allocated DSH funding reductions for the low DSH state group by the LDF percentage to determine the aggregate DSH reduction amount that would be distributed across the low DSH state group.

7. Subtract the aggregate DSH reduction amount determined in step 6 from the proportionately allocated DSH funding reduction for the low DSH state group, and add the remainder to the aggregate DSH reduction amount that would be distributed across the non-low DSH state group.

We considered using various alternative proportional relationships to establish the LDF, including the proportion of each state group's annual Medicaid DSH expenditures to total Medicaid expenditures. However, we believe that this may benefit non-low DSH states that are unable to or otherwise do not spend their existing DSH allotment amount, which we believe is not the intent of the statute. Therefore, we proposed to calculate the LDF based on the proportion of each state group's DSH allotments to total Medicaid expenditures.

We received a number of public comments on the proposed Factor 1—LDF. A discussion of these comments, with our responses, appears below.

Comment: Several commenters suggested modifying the statutory definition of low DSH states in section 1923(f)(5)(B) of the Act.

Response: The statute directs the Secretary to impose a smaller percentage DSH allotment reduction on “low DSH states” that meet the criterion described in section 1923(f)(5)(B) of the Act, and we do not have the authority to modify this provision. We are implementing this statutory directive through the LDF.

Comment: In calculating the LDF, one commenter recommended that CMS use the median instead of mean to normalize non-low DSH state outlier values.

Response: We believe use of the mean, instead of the median, ensures arithmetically that the value representing each state's DSH allotment as a percentage of Medicaid service expenditures has an equal impact in determining the average used in step 2 of the LDF, regardless of whether the value is an outlier value (either very high or very low). We believe this is important as the values have a large variance and each value, including outliers, represents the situation of a state. Using the median, instead of the mean, would not adequately capture the variance among all the states.

Comment: One commenter stated that the proposed DHRM conflicts with section 1923(f)(7)(B)(iii) of the Act in that it could result in percentage reductions for certain low DSH states that are greater than the percentage reductions for certain non-low DSH states.

Response: We disagree that the reduction methodology conflicts with the statutory direction to impose “a smaller percentage reduction on low DSH States.” While the final DHRM includes the LDF to impose smaller percentage reductions on low DSH states, it is possible that the annual DSH allotment reduction percentage could be higher for one or more low DSH states than for one or more non-low DSH states based on the application of other factors identified by the statute. In this case, the annual DSH allotment reduction percentage for the low DSH state would be smaller than if the state were instead a non-low DSH state, due to the application of the LDF, consistent with section 1923(f)(7)(B)(iii) of the Act.

Comment: One commenter suggested that low DSH states be exempt from the reduction.

Response: The statute directs the DHRM to impose “a smaller percentage reduction on low DSH States,” but does not permit that low DSH states be categorically exempted from reduction. Consistent with the statute, the final DHRM imposes smaller percentage reductions on low DSH states, but does not exempt low DSH states from reduction. We believe that this methodology is consistent with the statute and is an equitable approach to allocating annual DSH allotment reductions.

Comment: One commenter expressed concerns that the LDF calculation is overly beneficial to low DSH states. The commenter stated that the formula exceeds the statutory requirements and recommended an alternative approach that would rely on calculating each group's proportion of annual Medicaid expenditures to total Medicaid expenditures.

Response: The proposed DHRM imposes smaller percentage reductions on low DSH states, which historically have received lower DSH allotments relative to their total Medicaid expenditures than non-low DSH states. This historical difference, between low DSH and non-low DSH state groups, serves as the basis for calculating the LDF value and addresses the statutory requirement to impose “a smaller percentage reduction on low DSH States.” Although we considered alternate methods for calculating the LDF, we believe that the proposed methodology for determining the LDF best addresses this historical difference while adhering to statutory direction. Furthermore, our proposed methodology is consistent with the statutory designation of low DSH or non-low DSH states. Therefore, we are finalizing the LDF as proposed.

Comment: One commenter stated that step 6 in the calculation should read “multiply the proportion of total unreduced allocations for the low DSH states group to total unreduced allocations for all states by the LDF percentage.”

Response: We believe that we have described the process accurately in calculating the total reduction amount for low DSH states once the LDF is applied. While the commenter's suggested language is accurate in describing the steps to calculate the revised percent of total weighting for the low DSH state group, our proposed language provides the steps to calculate the total reduction amount for the low DSH state group. We proposed to separate states into two overall groups, non-low DSH states and low DSH states, to give effect to the statutory low DSH criterion. Then, we proposed to proportionately allocate aggregate DSH funding reductions to each of these two state groups based on each state group's proportion of the total national unreduced DSH allotment amount. Next, we proposed to apply a low DSH adjustment percentage to adjust the non-low DSH and low DSH state groups' DSH funding reduction amounts. This step maintains the combined aggregate DSH funding reduction for the low DSH and non-low DSH state groups together, as specified by statute for the applicable FY, by distributing a portion of the unadjusted low DSH state DSH funding reduction amount to the non-low DSH state group.

Comment: Several commenters urged CMS to minimize annual DSH allotment reductions for states that have relatively low ratios of the unreduced annual DSH allotment to the number of uninsured individuals in the state. One commenter recommended that states that receive less than $125 in unreduced annual DSH allotments per uninsured individual should receive no more than a 5 percent annual DSH allotment reduction.

Response: The statute directs the DHRM to impose “a smaller percentage reduction on low DSH States,” which is described in paragraph 1923(f)(5)(B) of the Act where it defines low DSH states as states with total Medicaid DSH payments for FY 2000 between 0 and 3 percent of total (state and federal) Medicaid medical assistance expenditures. We do not have the authority to modify the statutory definition of a low DSH state in order to impose smaller percentage reductions on states that have low annual DSH allotments relative to the number of uninsured individuals in the state. Consistent with the statute, the final DHRM imposes smaller percentage reductions on low DSH states described in section 1923(f)(5)(B) of the Act. While we are statutorily-required to impose “a smaller percentage reduction on low DSH States,” the final DHRM does allocate reductions taking into account the size of the existing state DSH allotments prior to reduction in the UPF, which does give consideration to states that historically have smaller unreduced DSH allotments relative to similarly situated states with higher allotments.

Comment: One commenter stated that CMS did not provide total computable medical assistance expenditures used to calculate the LDF in the illustrative DHRM example in the July 2017 proposed rule. Further, the commenter stated that the proposed rule did not specify whether the denominator of the LDF includes or excludes DSH and whether it is total computable or Federal share.

Response: The July 2017 proposed rule included an illustrative example, not an actual DHRM calculation. For purposes of the final DHRM, we will exclude DSH expenditures from total computable Medical assistance expenditures described in § 447.294(e)(3)(i). The denominator for the value calculated in § 447.294(e)(3)(i) is the estimated Medicaid service expenditures. The denominator for the value calculated in § 447.294(e)(3)(iii) is the mean value of the ratio of each non-low DSH state's proportion of preliminary DSH allotment to estimated Medicaid service expenditures, calculated in § 447.294(e)(3)(ii). Additionally, we intend to publish a separate DHRM technical guide that provides information regarding the final DHRM calculation, including the additional information regarding data sources.

Comment: One commenter requested that CMS consider an alternative methodology for calculating the low DSH adjustment and stated CMS should consider a flat percentage rather than basing it on a factor ratio.

Response: We considered using various alternative proportional relationships to establish the LDF. However, we are finalizing the LDF as proposed without change to our proposal to use the LDF as currently codified in § 447.594(e)(3). The low DSH adjustment percentage is consistent with the statutory method used for classifying low DSH states at section 1923(f)(5)(B) of the Act by utilizing the proportion of each state group's DSH allotments to total Medicaid expenditures. Further, the proposed LDF percentage can evolve over time, respond to changes in state situations, and use better data as it becomes available while a flat percentage would remain static and not be responsive to state or data changes. Given that low-DSH states collectively receive lower DSH allotments relative to their total Medicaid expenditures than non-low DSH states, the LDF results in the application of a smaller percentage reduction to low DSH states.

E. Factor 2—Uninsured Percentage Factor (UPF)

The second factor considered in the DHRM is the UPF identified in section 1923(f)(7)(B)(i)(I) of the Act, which requires that the DHRM impose the largest percentage DSH allotment reductions on states that have the lowest percentages of uninsured individuals. The statute also requires that the percentage of uninsured individuals be determined on the basis of data from the Census Bureau, audited hospital cost reports, and other information likely to yield accurate data, during the most recent year for which such data are available.

To determine the percentage of uninsured individuals in each state, the DHRM relies on the total population and uninsured population as identified in the most recent “1-year estimates” data available from the ACS conducted by the Census Bureau. The Census Bureau generates ACS “1-year estimates” data annually based on a point-in-time survey of approximately 3 million individuals. For purposes of the DHRM, we would utilize the most recent ACS data available at the time of the calculation of the annual DSH allotment reduction amounts.

The UPF, as applied through the DHRM, has the effect of imposing the lowest relative DSH allotment reductions on states that have the highest percentage of uninsured individuals, and thereby mitigates the annual DSH allotment reductions for states with the highest percentage of uninsured individuals.

The UPF is determined separately for each state group as follows:

1. Uninsured Value—Using United States Census Bureau data, calculate each state's uninsured value by dividing the total state population by the number of uninsured in the state. (This is different than the percentage rate of uninsurance; the rate of uninsurance can be obtained by dividing 100 by this number.)

2. Uninsured Allocation Component—Determine the relative Uninsured Value for each state compared to other states in the state group by dividing the value in step one by the state group (low DSH state and non-low DSH state) total of step one values. The result will be a percentage, and the total of the percentages for all states in the state group will total 100 percent.

3. Allocation Weighting Factor—To ensure that larger and smaller states are given fair weight in the final UPF, divide each state's preliminary unreduced DSH allotment by the sum of all unreduced preliminary DSH allotments in the respective state group to obtain the allocation weighting factor, expressed as a percentage. The sum of all weighting factors will equal 100 percent. Then, take this percentage for each state and multiply it by the state's uninsured allocation component determined in step 2. The result is the allocation weighting factor.

4. UPF—For each state group, divide each state's allocation weighting factor by the sum of all allocation weighting factors. The resulting percentage is the UPF.

We would determine the UPF portion of the aggregate DSH allotment reduction allocation for each state by multiplying the state's UPF by the aggregate DSH allotment reduction allocated to the UPF factor for the respective state group. As with the prior factor, we proposed to utilize preliminary DSH allotment estimates to develop the DSH reduction factors, including the UPF. We received the following comments concerning this topic.

We received a number of public comments on the proposed Factor 2—UPF. A discussion of these comments, with our responses, appears below.

Comment: Many commenters supported the DHRM's identification of uninsured individuals based on 1-year estimates of the number of uninsured from the Census Bureau's ACS.

Response: We appreciate the support and are finalizing the use of 1-year estimates of the number of uninsured from the ACS in the DHRM, as discussed in the proposed rule and as described in the definition of “Uninsured population” in § 447.294(b).

Comment: Many commenters expressed concerns that the uninsured individual data used for the UPF may undercount the number of undocumented individuals as reported and estimated through the ACS. One commenter noted that this is particularly concerning, given the 50 percent UPF weight. Additionally, many commenters recommended that CMS work with Pew Research Institute, Census Bureau, and other researchers to develop a methodology that accounts for all uninsured individuals regardless of citizenship status.

Response: Section 1923(f)(7)(B)(i)(I) of the Act specifically requires that the percentage of uninsured individuals be determined on the basis of data from the Census Bureau, audited hospital cost reports, and other information likely to yield accurate data. According to the Census Bureau, the foreign-born population includes anyone who is not a U.S. citizen at birth. This includes two groups: (1) Naturalized U.S. citizens; and (2) noncitizens. Noncitizens include lawful permanent residents (immigrants), temporary migrants (such as foreign students), humanitarian migrants (such as refugees and asylees), and persons not lawfully present in the United States.

The Census Bureau collects data from all foreign-born individuals who participate in its censuses and surveys, regardless of legal status. Thus, unauthorized migrants are included in ACS estimates of the total foreign-born population. However, the Census Bureau only asks foreign-born respondents if they are naturalized U.S. citizens or noncitizens, so it is not possible to tabulate separate estimates of unauthorized migrants using the ACS. Accordingly, we believe the ACS data does account for uninsured individuals regardless of citizenship status and are finalizing our proposed use of ACS data without an adjustment in the uninsured data.

Comment: Several commenters noted support for CMS' goal of relying on the most recently available data for calculating the UPF, but expressed concern that CMS would use 2014 ACS data to calculate the FY 2018 DSH allotment reductions. Commenters recommended that CMS utilize more recent data when calculating final DSH allotments. One commenter recommended that CMS utilize ACS 5-year estimates for the uninsured to better align the years of the Census Bureau ACS data with the DSH audit and MIUR data.

Response: We are finalizing, as proposed, the application of a DHRM that utilizes the most recent year available for all data sources and aligns data sources whenever possible. That is, section 1923(f)(7)(B)(i)(I) of the Act requires the use of Census Bureau data, audited hospital cost reports, and other information likely to yield accurate data, for the most recent year for which such data are available. Therefore, with respect to annual DSH allotment reductions for FY 2020, we intend to use 2018 ACS data, which we anticipate will be the most recent year available at the time the DHRM is applied for FY 2020.

We will use the ACS 1-year estimates because it depicts the most current data on the uninsured population. The ACS 5-year estimates use 60 months of data. For example, 2013-2017 estimate is data collected from January 1, 2013 through December 31, 2017. This is the least current of the ACS estimates. The Census Bureau recommends using ACS 1-year when currency is more important.

Comment: One commenter expressed concern that the ACS data considers an individual's uninsured status based only on whether respondent has coverage at time of interview, and that ACS data may undercount the population of individuals experiencing homelessness. Another commenter recommended that CMS work with the Census Bureau to attain the point in time estimate as well as a determination of whether an individual was uninsured at any point in time during the past year.

Response: Section 1923(f)(7)(B)(i)(I) of the Act requires that CMS utilize data from the Census Bureau, from the most recent year for which data are available to calculate the UPF. Moreover, while the ACS data determine whether the respondent has coverage at the time of the interview, these interviews are conducted at various times throughout the year. The Census Bureau randomly selects addresses, through scientific sampling, to represent the total population. As such, we believe that the ACS 1-year estimates represent the best available data for use in determining the number of uninsured individuals in the states. Further, we understand that the Census Bureau works with organizations such as the National Coalition for the Homeless to help ensure a more accurate and comprehensive census, including with respect to individuals experiencing homelessness.

Comment: One commenter expressed concern that the DHRM assigns too much weight to the UPF and suggested that the UPF calculation methodology rely on state levels of insured individuals instead of percentages of uninsured individuals. Additionally, the commenter indicated the UPF and factor weighting would result in the DHRM penalizing Medicaid expansion states.

Response: The UPF, as applied through the DHRM, has the effect of imposing lower relative DSH allotment reductions on states that have higher percentage of uninsured individuals. Section 1923(f)(7)(B)(i)(I) of the Act specifies the “percentage of uninsured individuals,” not the level of insured individuals. To determine the percentage of uninsured individuals in each state, the DHRM relies on the total population and uninsured population as identified in the most recent “1-year estimates” data available from the ACS conducted by the Census Bureau. This approach is consistent with statutory requirements and mitigates the DSH allotment reductions for states with the highest percentage of uninsured individuals. Further, we believe that the final DHRM, including the factor weighting discussed above, distributes DSH allotment reduction amounts among the states in an equitable manner, consistent with statutory requirements and does not penalize Medicaid expansion states.

Comment: One commenter recommended that we rely on the Medicaid DSH definition of uninsured used for calculating hospital-specific DSH limits, adjusted also to include certain insured individuals who might be more likely to be associated with unpaid copayments and deductibles (such as individuals with high deductible plans), for purposes of defining uninsured individuals for the UPF.

Response: Section 1923(f)(7)(B)(i)(I) of the Act requires the use of Census Bureau data to determine the percentages of uninsured individuals. We are finalizing the use of 1-year estimates of the number of uninsured from the ACS conducted by the Census Bureau in the DHRM, as discussed in the proposed rule and as described in the definition of “Uninsured population” in § 447.294(b).

Comment: One commenter recommended that CMS distribute the entire available DSH allotment for all states based on its uninsured rate. Several commenters stated that statute does not require CMS to use both the UPF and the two targeting factors in the DHRM and suggested that CMS apply only the UPF in the determination of DSH allotment reductions. These commenters recommended this approach to mitigate reductions for both states that have not expanded Medicaid under the ACA and for states that have strict criteria for eligibility to receive DSH payments. One commenter suggested that, given the statutory language, CMS could and should use only the targeting factors (both the HUF and the HMF) in the DRHM, or that the UPF weight be lowered if both the UPF and the targeting factors are to be considered.

Response: Section 1923(f) of the Act specifies the manner in which each state's DSH allotment is determined. Moreover, section 1923(f)(7)(B) of the Act establishes the five factors that must be considered in the establishment of a DHRM to calculate the annual DSH allotment reductions. We interpret the statute to require CMS to utilize both the UPF and the two targeting factors. We proposed to assign a 50 percent weight to the UPF and a 50 percent combined weight for the two DSH payment targeting factors (a 25 percent weight for the HUF, and a 25 percent weight for the HMF). As described in detail in section III.C. of this final rule, we believe that this is an equitable approach for assigning factor weights, and appropriately implements the statutorily-required factors. This weight distribution does preserve more DSH allotment (that is, it imposes smaller allotment reductions) for states that may have greater DSH need due to high uninsurance rates while still incentivizing states to continue to target DSH payments to hospitals with both a high volume of Medicaid inpatients and high level of uncompensated care. Additionally, we proposed, and are finalizing, a weight of 50 percent for the UPF to rely more heavily on more recent Census Bureau data and to align the factor weights with how these factors are set forth in statute.

F. Factor 3—High Volume of Medicaid Inpatients Factor (HMF)

The third factor considered in the DHRM is the HMF identified in section 1923(f)(7)(B)(i)(II)(aa) of the Act, which requires that the DHRM impose the largest percentage DSH allotment reductions on states that do not target DSH payments to hospitals with high volumes of Medicaid inpatients. For purposes of the DHRM, the statute defines hospitals with high volumes of Medicaid inpatients as those defined in section 1923(b)(1)(A) of the Act. These hospitals must meet minimum qualifying requirements at section 1923(d) of the Act and have an MIUR that is at least one standard deviation above the mean MIUR for hospitals receiving Medicaid payments in the state. Every hospital that meets that definition is deemed a disproportionate share hospital and is statutorily-required to receive a DSH payment.

States that have been, and continue to, target a large percentage of their DSH payments to hospitals that are federally deemed as a DSH based on their MIUR would receive the lowest reduction amounts relative to their total DSH spending. States that target the largest amounts of DSH payments to hospitals that are not federally deemed based on MIUR would receive the largest reduction amounts under this factor. The current DSH allotment amounts are unrelated to the number of MIUR-deemed hospitals within each state and their DSH-eligible uncompensated care costs. By basing the HMF reduction on the amounts that states do not target to hospitals with high volumes of Medicaid inpatients as described below, this methodology incentivizes states to target DSH payments to such hospitals.

To ensure that all deemed disproportionate share hospitals receive a required DSH payment, states are already required to determine the mean MIUR for hospitals receiving Medicaid payments in the state and the value of one standard deviation above the mean. We proposed to rely on MIUR information for use in the DHRM that CMS collects from states on an annual basis under § 447.294(d). When a state or states do not submit this required MIUR information timely, for purposes of this factor, we would assume that the state(s) have the highest value of one standard deviation above the mean reported among all other states that did submit this information timely.

The calculation of the HMF will rely on extant data that should be readily available to states. The following data elements are used in the HMF calculation: The preliminary unreduced DSH allotment for each state; the DSH hospital payment amount reported for each DSH in accordance with § 447.299(c)(17); the MIUR for each DSH reported in accordance with § 447.299(c)(3); and the value of one standard deviation above the mean MIUR for hospitals receiving Medicaid payments in the state, reported separately.

The HMF is a state-specific percentage that is calculated separately for each state group (low DSH and non-low DSH) as follows:

1. For each state, classify each DSH that has an MIUR at least one standard deviation above the mean MIUR for hospitals receiving Medicaid payments in the state as a High Medicaid Volume hospital.

2. For each state, determine the amount of DSH payments to non-High Medicaid Volume DSH hospitals using data from the most recently submitted and accepted DSH audit template.

3. For each state, determine a percentage by dividing the state's total DSH payments made to non-High Medicaid Volume hospitals by the aggregate amount of DSH payments made to non-High Medicaid Volume hospitals for the entire state group. The result of step 3 is the HMF.

4. Determine each state's HMF reduction amount by applying the HMF percentage to the aggregate reduction amount allocated to this factor for each state group.

As a result of this methodology, there are a number of interactions that may occur for states among DSH payment methodologies, DSH allotments, and DSH allotment reductions. Most of these scenarios work in concert with this factor's established reduction relationship. For example, if a state paid out its entire DSH allotment to hospitals with high volumes of Medicaid inpatients, it would receive no reduction associated with this factor because all DSH payments were made only to hospitals that qualify as high volume. The results of this scenario would be consistent with the methodology because the state is incentivized to target DSH payments to high Medicaid volume hospitals.

Another example is a state that makes DSH payments up to the hospital-specific DSH limit to all hospitals with high Medicaid volume but also uses its remaining allotment to make DSH payments to hospitals that do not qualify as high Medicaid volume. In this example, the state would receive a reduction under this factor based on the amount of DSH payments it made to non-high Medicaid volume hospitals. Although the state targeted DSH payments to hospitals with high Medicaid volume, the existing size of its DSH allotment permitted it to make DSH payments to hospitals that did not meet the statutory definition of a hospital with a high volume of Medicaid inpatients. In that situation, we stated in the proposed rule that this allotment reduction would effectively reduce a state's existing DSH allotment if the allotment exceeded the maximum amount that the state could pay to hospitals that are high Medicaid volume. The resulting HMF reduction would be greater for states with DSH allotments large enough to pay significant amounts to non-high Medicaid volume hospitals. This helps ensure that states target DSH payments to high Medicaid volume hospitals and distributes the reductions in such a way as to promote the ability of all states to provide DSH funds to high Medicaid volume hospitals.

We described the HMF in greater detail in the July 2017 proposed rule (82 FR 35155). We received a number of public comments on the proposed Factor 3—HMF. A discussion of these comments, with our responses, appears below.

Comment: One commenter expressed concern that CMS will use DSH audit data and MIUR data from different years to calculate reductions based on the HMF. In addition, the commenter recommended that the DHRM rely on MIUR data from the audited Medicaid DSH audits and reports to improve accuracy of the DHRM.

Response: In the July 2017 proposed rule, we proposed, as a general principle, to utilize the most recent year available for all data sources and to align the Medicaid SPRY of data sources. The proposed DHRM relies on the most recent data for all data sources with one exception. For this exception, we believe the benefits of aligning the SPRYs of two data sources outweighs the benefits of using the most recent data. Specifically, the MIUR data required by § 447.294(d) used for the HMF may not be the most recent year available. We proposed to align and utilize MIUR data from the year that corresponds to the DSH audit SPRY used in the calculation of each state's DSH allotment reduction. Although more recent MIUR data might be available, we are aligning the MIUR data SPRY with the DSH audit SPRY for the HMF to ensure the universe of hospitals is the same and to ensure the DSH payment for a particular SPRY corresponds with the receiving hospital's MIUR for that same SPRY.

The Medicaid DSH audits and reports do not include the MIUR for all hospitals that receive a Medicaid payment. Therefore, we believe the DHRM is more accurate relying on MIUR information that we will collect from states on an annual basis as required under § 447.294(d).

Comment: Two commenters expressed concern that expansion states could receive relatively greater reduction through the HMF when many of their hospitals meeting MIUR-related deeming requirements defined in section 1923(b)(1)(A) of the Act have little or no uncompensated care costs, particularly due to the state targeting Medicaid supplemental payments to such deemed hospitals. One commenter suggested that CMS develop an alternative methodology for judging how well states target DSH payments to MIUR-deemed hospitals that recognizes that states may not pay in excess of the hospital-specific DSH limit.

Response: The proposed HMF would apply to states without regard to their Medicaid expansion status. Additionally, we understand that the proposed HMF reduction would be greater for states with DSH allotments large enough to pay significant amounts to non-high Medicaid volume hospitals, including in cases where states cannot target DSH payment to high volume Medicaid hospitals because they do not have significant uncompensated care costs. This helps ensure that states target DSH payments to high Medicaid volume hospitals and distributes the reductions in such a way as to promote the ability of all states to provide DSH funds to high Medicaid volume hospitals.

Comment: One commenter expressed concern that the DHRM could penalize some states that target deemed hospitals based on the LIUR. The commenter noted that about half of all deemed-DSH hospitals nationally qualify on the basis of their LIUR. The commenter suggested that the DHRM should either consider all payments made to deemed hospitals as being paid to high Medicaid volume hospitals, or DSH payments to LIUR-deemed hospitals should be excluded from the calculation of the HMF.

Response: We believe the DHRM as proposed will promote state targeting of payments to hospitals that qualify for DSH payments based on MIUR deeming requirements defined in section 1923(b)(1)(A) of the Act, consistent with section 1923(f)(7)(B)(i)(II)(aa) of the Act. The HMF targeting factor in the DHRM is consistent with the statutory direction to impose larger percentage reductions on states that do not target their DSH payments on hospitals with high volumes of Medicaid inpatients and do not target their DSH payments on hospitals with high levels of uncompensated care.^[4] The HMF provides mitigation of the state-specific DSH reduction amount for states that have targeted and do target DSH payments to these hospitals federally-deemed on the basis of their MIUR. We recognize the importance of hospitals with high LIURs and such hospitals may also experience high levels of uncompensated care costs. If those LIUR-deemed hospitals have high levels of uncompensated care, the HUF will provide mitigation of the state-specific DSH reduction amount for states that have targeted and do target DSH payments to those hospitals.

Comment: One commenter recommended that the demographics of the Medicaid population be taken into account when determining DSH allotment reductions. The commenter recommended that if a large percentage of the Medicaid expansion population represents individuals who shifted from other insurance coverage, that state should not have as large of a DSH allotment reduction as a state in which a larger share of the Medicaid expansion population was previously uninsured.

Response: The statute requires that larger percentage reductions be imposed on states that do not target their DSH payments on hospitals with high volumes of Medicaid inpatients and on hospitals with high levels of uncompensated care (excluding bad debt). The statutory requirements do not address the prior coverage status of Medicaid enrollees.

Comment: Several commenters expressed concern that many states had not submitted MIUR data to CMS, and therefore, CMS utilized proxy MIUR data for calculation of illustrative DSH allotment reductions. These commenters expressed concern that the use of proxy data may affect the distribution of DSH allotment reductions. One commenter recommended that CMS accept late MIUR submissions for FY 2018 and should consider accepting late MIUR submissions for subsequent years.

Response: Section 447.294(d) specifies the timeline according to which states are required to submit MIUR data to CMS. The example included in the July 2017 proposed rule was for illustrative purposes only. As specified in the final 2013 DSH allotment reduction rule (78 FR 57305), when a state does not timely submit this separately required MIUR information, for purposes of this factor, we will assume that the state has the highest value of one standard deviation above the mean reported among all other states.

Comment: One commenter suggested that CMS propose a standard definition of which hospitals should be included in each state's annual MIUR data submission. Another commenter suggested that the requested MIUR data is duplicative of data collected as part of the DSH audits.

Response: We believe the laws and regulations already provide a standard definition of hospitals with high volumes of Medicaid inpatients and which hospitals must be included in the annual MIUR submission required in § 447.294(d). Section 1923(f)(7)(B)(i)(II)(aa) of the Act defines hospitals with high volumes of Medicaid inpatients as those defined in section 1923(b)(1)(A) of the Act. Section 447.294(d) specifies that states must submit the MIUR for all hospitals receiving Medicaid payments in the State.

Although the DSH audits do contain MIUR data for each hospital that receives a DSH payment, the MIUR submission required under § 447.294(d) contains the Medicaid utilization for all hospitals that receive a Medicaid payment (including those that do not receive a DSH payment), which information is necessary to the calculation of the HMF.

G. Factor 4—High Level of Uncompensated Care Factor (HUF)

The fourth factor considered in the DHRM is the HUF identified at section 1923(f)(7)(B)(i)(II)(bb) of the Act, which requires that the DHRM impose the largest percentage DSH allotment reductions on states that do not target DSH payments to hospitals with high levels of uncompensated care (excluding bad debt). We proposed to rely on the existing statutory definition of uncompensated care cost used in determining the hospital-specific limit on FFP for Medicaid DSH payments.

As defined in section 1923(g)(1) of the Act, the state must calculate for each hospital, for each FY, the difference between the costs incurred by that hospital for furnishing inpatient hospital and outpatient hospital services during the applicable state FY to Medicaid individuals and individuals who have no health insurance or other source of third party coverage for the inpatient hospital and outpatient hospital services they receive, less all applicable revenues received for these hospital services. This difference, if any, between incurred inpatient hospital and outpatient hospital costs and associated revenues is considered a hospital's uncompensated care costs, or hospital-specific DSH limit.

We proposed to rely on this definition of uncompensated care costs for the calculation of the HUF, as reported by states on the most recent available Medicaid DSH audit and reporting data. For the proposed DHRM, hospitals with high levels of uncompensated care costs are defined based on a comparison with other Medicaid DSH hospitals in the state. Any hospital that exceeds the mean ratio of uncompensated care costs to total Medicaid and uninsured inpatient hospital and outpatient hospital service costs within the state is considered a hospital with a high level of uncompensated care. This data is consistent with the existing Medicaid DSH program definition of uncompensated care and is readily available to states and CMS.

The following data elements would be used in the HUF calculation:

• The preliminary unreduced DSH allotment for each state;
• DSH hospital payment amounts reported for each DSH in accordance with § 447.299(c)(17);
• Uncompensated care cost amounts reported for each DSH in accordance with § 447.299(c)(16);
• Total Medicaid cost amounts reported for each DSH in accordance with § 447.299(c)(10);
• Total uninsured cost amounts reported for each DSH in accordance with § 447.299(c)(14); and
• Total hospital cost amounts reported for each DSH in accordance with § 447.299(c)(20).

The statute also requires that uncompensated care costs used in this factor of the DHRM exclude bad debt. The DHRM relies on the uncompensated care cost data derived from Medicaid DSH audit and reporting required by section 1923(f) of the Act and implementing regulations. This uncompensated care data excludes bad debt, including unpaid copayments and deductibles, associated with individuals with a source of third party coverage for the service.

The HUF is a state-specific percentage that is calculated separately for each state group (low DSH and non-low DSH) as follows:

1. Determine each disproportionate share hospital's uncompensated care level by dividing its uncompensated care cost by total hospital cost. This data element would come from the most recently submitted and accepted Medicaid DSH audit and associated reporting.

2. For each state, calculate the mean uncompensated care level.

3. Identify all hospitals that meet or exceed the mean uncompensated care level as high uncompensated care level hospitals, and all hospitals with uncompensated care costs below this mean as non-high uncompensated care level hospitals.

4. For each state, determine the total amount of DSH payments to non-high uncompensated care level hospitals.

5. For each state, determine a percentage by dividing the state's total DSH payments made to non-high uncompensated care level hospitals by the aggregate amount of DSH payments made to non-high uncompensated care level hospitals for the entire state group. The result would be the HUF.

6. Determine each state's HUF reduction amount by applying the HUF percentage to the aggregate reduction amount allocated to this factor for each state group.

In previous rulemaking, we identified some potential scenarios, due to data limitations, where the DHRM finalized in 2013 could have produced some paradoxical outcomes when comparing hospital levels of uncompensated care for purposes of evaluating DSH payment targeting through the HUF. Specifically, in § 447.294(e), the 2013 DSH allotment reduction final rule, it was possible for a hospital not to have been considered to have a higher level of uncompensated care even though it provided a higher percentage of services to Medicaid and uninsured individuals and had greater total qualifying uncompensated care costs than another hospital that did qualify as having a high level of uncompensated care. This was due to the previous formula determining the level of uncompensated care by dividing uncompensated care costs by the sum of total Medicaid costs and total uninsured costs. We propose to resolve this problem at § 447.294(e) by determining the level of uncompensated care by dividing uncompensated care costs by the total hospital costs.

We sought comments on the proposed DHRM with respect to whether the proposed implementation of this factor is expected to be effective in tying the level of DSH reductions to the targeting of DSH payments to hospitals with high levels of uncompensated care. We believe that the proposed DHRM methodology, in using the mean uncompensated care cost level as the measure to identify hospitals with high levels of uncompensated care, captures a better balance in tying the level of DSH reductions to the targeting of DSH payments to such high level uncompensated care hospitals, imposing smaller annual state DSH allotment reductions on states that more effectively target DSH payments to hospitals with high levels of uncompensated care.

We described the HUF in greater detail in the July 2017 proposed rule (82 FR 35155). We received a number of public comments on the proposed Factor 4—HUF. A discussion of these comments, with our responses, is below.

Comment: One commenter suggested that the formula in the July 2017 proposed rule would disadvantage hospitals for their size and services provided to the insured by using the total hospital cost in the HUF denominator. The commenter requested that CMS not adopt the formula or adopt both the 2013 HUF calculation and the new formula and letting hospitals use the option that results in the higher UCC amount.

Response: We disagree that the policy reflected in the July 2017 proposed rule disproportionately harms hospitals with high uncompensated care costs related to the insured population and believe that the proposed formula, which we are adopting in this final rule, accurately and equitably calculates levels of uncompensated care costs. This rule specifies the methodology to be used to calculate the statutorily-required Medicaid DSH reductions. In the 2013 DSH allotment reduction final rule, we finalized a DHRM, which gave the HUF a 331/3 percent weight and that would be in place only for FY 2014 and FY 2015 to allow time for reevaluation of the methodology with improved and more recent data and information about the impact of the ACA on levels of coverage and uncompensated care. As a result of our reevaluation, in the July 2017 proposed rule, we proposed to modify the DHRM factor weights and to use improved data sources where possible, as discussed in this final rule. We believe this rule ensures the appropriate allocation of the DSH allotment reductions to those states that target their DSH payments to hospitals with high volumes of Medicaid inpatients and high levels of uncompensated care (excluding bad debt), as required under the statute. Therefore, we will only be using the policy reflected in the July 2017 proposed rule and this final rule, and we will not adopt the 2013 HUF calculation as an alternative option.

Comment: One commenter recommended that CMS include costs other than inpatient and outpatient hospital services, including physician services, transportation costs, and non-hospital services, in the calculation of the hospital-specific DSH limit. One other commenter recommended that CMS update the definition of uncompensated care to align with the definition under the Internal Revenue Code to determine community benefit, and that CMS require hospitals receiving DSH payments to report Medicaid shortfall, charity care, and bad debt to better understand the impact of DSH payments on hospitals.

Response: Consistent with statutory direction, the DHRM will use uncompensated care data that excludes bad debt, including unpaid copayments and deductibles associated with individuals with a source of third party coverage for the service. Changes to calculating the hospital-specific DSH limit are outside the scope of the July 2017 proposed rule. We are not addressing the calculation of hospital-specific DSH payment limits under section 1923(g) of the Act, or the DSH audit reporting requirements under section 1923(j) of the Act, through this rulemaking.

Comment: One commenter noted that the MIUR data do not appropriately account for state-created programs for low-income individuals that are funded by DSH payments, or were so funded prior to Medicaid expansion.

Response: We disagree. The DHRM relies on MIUR data as the data source specified in statute. Modifying the MIUR used in the DHRM to account for state-created programs would be inconsistent with statutory requirements.

Comment: Several commenters expressed concerns that the HUF does not properly address the statutory direction to impose larger percentage reductions on states that do not target their DSH payments to hospitals with high levels of uncompensated care because Medicaid DSH audit and reporting data does not include all hospitals in a state. These commenters noted that using only the hospitals identified on the DSH audit report creates a higher mean uncompensated care value than that of states with less strict criteria for eligibility for receiving DSH payments. One commenter suggested that the DHRM should account for states that have strict criteria for qualifying to receive DSH payments and recommended that CMS collect and utilize high LIUR values to consider hospitals targeted under the HUF. Another commenter suggested that for purposes of calculation reductions under the HUF, CMS cap each state's average uncompensated care level at the national mean plus one standard deviation. Yet another commenter suggested that CMS obtain average uncompensated care levels from all hospitals with Medicaid days, not just from those hospitals identified through DSH audit and reporting data.

Response: We recognize that the DSH audit and reporting data does not include uncompensated care information for all hospitals; however, the Medicaid DSH audit and reporting data represent the only existing uncompensated care cost data consistent with the existing statutory definition of uncompensated care cost used in determining the hospital-specific limit on FFP for DSH payments. We disagree that the HUF does not address the statutory direction to impose larger percentage reductions on states that do not target their DSH payments to hospitals with high levels of uncompensated care. The proposed and final HUF is designed to promote state targeting of DSH payments to hospitals with high levels of uncompensated care by imposing reductions based on the payments to non-high uncompensated care-level hospitals. We believe that the proposed calculation of the HUF represents an equitable method for comparing how states target payments to high uncompensated care hospitals, and therefore, we are not adopting the commenters' recommendations.

Comment: Many commenters noted support for total hospital cost in the denominator of the HUF. One commenter stated that using total hospital cost in the denominator of the HUF mitigates reductions for states that target deemed DSH hospitals.

Response: We believe this is an accurate and equitable method for calculating reductions under the HUF, and as such, we are finalizing the rule with the use of the total hospital cost as the denominator for purposes of calculating reductions under the HUF.

Comment: One commenter requested that CMS clarify the description of total hospital cost in the July 2017 proposed rule.

Response: The description of total hospital costs as it relates to the July 2017 proposed rule and this final rule is codified in § 447.299(c)(20). Total hospital cost is the total annual costs incurred by each hospital for furnishing inpatient hospital and outpatient hospital services.

Comment: One commenter suggested CMS use a standardized calculation for uncompensated care costs to promote more consistent results across all states, so that the states currently including third party payments for Medicaid eligible individuals in calculating uncompensated care cost for purposes of the hospital-specific DSH limit would not be disadvantaged.

Response: While a number of issues related to Medicaid DSH payment calculations currently are the subject of litigation, the statutorily-required allotment reductions and the DHRM are not among them, and we are bound by statute to adopt a rule to implement the DSH reductions. With this final rule, we are doing so according to our view of the best interpretation of the DSH statute and will utilize the most recent data available to us that is consistent with applicable laws and regulations. We believe the proposed DHRM relies on a standard definition of uncompensated costs for the HUF, which relies on data derived from Medicaid DSH audit and reporting data. Further, the DHRM, in using this data, imposes larger percentage reductions on states that do not target their DSH payments to hospitals with high levels of uncompensated care.

Comment: Several commenters noted support of CMS utilizing total hospital cost in the denominator of the HUF. Commenters expressed concern that the HUF should include an adjustment to account for the relative size of hospitals, and that utilizing total hospital costs in the denominator of the HUF disadvantages academic medical centers. The commenters noted that the need for academic medical centers to provide training, to maintain emergency standby capacity for rarely used hospital services, and to provide additional highly specialized services increases their total hospital cost compared to peer hospitals and, therefore, understates their HUF uncompensated care level compared to peer hospitals. One commenter expressed concern that CMS did not provide any data indicating which states would be impacted by this proposal.

Response: We disagree with this commenter that utilizing total hospital costs in the denominator of the HUF disadvantages academic medical centers and note that we received multiple comments in support of utilizing total hospital costs in the denominator of the HUF as opposed to our previous 2013 final rule approach of using only Medicaid and uninsured costs in the denominator. By using total hospital costs, we are accounting for the size of hospitals, therefore making an additional hospital size adjustment unnecessary. While we believe using total hospital costs in the denominator of the HUF represents a reasonable method for determining hospitals with high levels of uncompensated care costs, consistent with statutory requirements, we will monitor the application of this factor and the DHRM generally and may propose modifications if a better option avails itself in the future, nothing prevents CMS from readdressing the calculation of the HUF through future rulemaking, if appropriate.

H. Factor 5—Section 1115 Budget Neutrality Factor (BNF)

The statute requires that we take into account the extent to which a state's DSH allotment was included in the budget neutrality calculation for a coverage expansion that was approved under section 1115 demonstration authority as of July 31, 2009. These states possess full annual DSH allotments as calculated under section 1923(f) of the Act. Under an approved section 1115 demonstration, however, some states have limited authority to make DSH payments under section 1923 of the Act because all or a portion of their DSH allotment was included in the budget neutrality calculation for a coverage expansion under an approved section 1115 demonstration or to fund uncompensated care pools and/or safety net care pools. For applicable states, DSH payments under section 1923 of the Act are limited to the DSH allotment calculated under section 1923(f) of the Act less the allotment amount included in such a budget neutrality calculation. If a state's entire DSH allotment is included in such a budget neutrality calculation, it would have no available DSH funds with which to make DSH payments under section 1923 of the Act for the period of the demonstration.

Consistent with the statute, for states that include DSH allotment in budget neutrality calculations for coverage expansion under an approved section 1115 demonstration as of July 31, 2009, we proposed to exclude from the DSH allotment reduction, for the HMF and the HUF factors, the amount of DSH allotment that each state currently continues to divert specifically for coverage expansion in the budget neutrality calculation. DSH allotment amounts included in budget neutrality calculations for non-coverage expansion purposes under approved demonstrations would still be subject to reduction. Uncompensated care pools and safety net care pools are considered non-coverage expansion purposes for the BNF. For section 1115 demonstrations not approved as of July 31, 2009, any DSH allotment amounts included in budget neutrality calculations, whether for coverage expansion or otherwise, under a later approval would also be subject to reduction.

We proposed to determine for each reduction year if any portion of a state's DSH allotment qualifies for consideration under this factor. To qualify annually, CMS and the state would have to have included the state's DSH allotment (or a portion thereof) in the budget neutrality calculation for a coverage expansion that was approved under section 1115 of the Act as of July 31, 2009, and the coverage expansion would have to still exist in the approved section 1115 demonstration at the time that reduction amounts are calculated for each FY. If a state had a DSH allotment amount for coverage expansion approved under a demonstration under a section 1115 of the Act as of July 31, 2009 but subsequently reduced this amount, the approved amount remaining under the section 1115 demonstration would not be subject to reduction.

The proposed DHRM took into account the extent to which the DSH allotment for a state was included in the budget neutrality calculation for a demonstration approved under section 1115 of the Act as of July 31, 2009 by excluding from reduction under the HMF and HUF amounts diverted specifically for a coverage expansion and automatically assigning qualifying states an average percentage reduction amount (that is, the average HUF and HMF of the state's respective state group) for any DSH allotment diverted for non-coverage expansion purposes and any amounts diverted for coverage expansion if the section 1115 demonstration was not approved as of July 31, 2009. DSH allotment reductions relating to two DHRM factors (the HUF and the HMF) are determined based on how states target DSH payments to certain hospitals. Since states that diverted all or a portion of their DSH allotments would have limited or no relevant data for these two factors, we would be unable to evaluate how they spent the diverted portion of their DSH allotment for these targeting criteria. Accordingly, for diversion amounts subject to reduction, we proposed to maintain the HUF and HMF formula for DSH payments for which qualifying states would have available data. Because we would not have DSH payment data for DSH allotment amounts diverted for non-coverage expansion (or for coverage expansions not approved as of July 31, 2009), we proposed to assign average HUF and HMF reduction percentages for the portion of the DSH allotment that a state diverted for non-coverage expansion (or for coverage expansions not approved as of July 31, 2009) that it was consequently unable to use to target payments to disproportionate share hospitals. Instead of assigning the average percentage reduction to non-qualifying amounts, we considered using alternative percentages higher or lower than the average. However, these alternative percentages might provide an unintended benefit or penalty to these states for DSH diversions approved under a demonstration under section 1115 of the Act. We sought comment on the use of different percentages for the reductions to diversion amounts that do not qualify under the BNF and regarding alternative BNF methodologies that may be preferable.

We described the BNF in greater detail in the July 2017 proposed rule (82 FR 35155). We received a number of public comments on the proposed Factor 5—BNF. A discussion of these comments, with our responses, are below.

Comment: One commenter noted support for the BNF excluding diverted DSH allotment amounts, but stated that limiting this to waivers approved before July 31, 2009, unfairly limits the ability of some states to expand coverage using a model that has proven successful in the commenter's state. The commenter noted that if the rule is finalized as proposed, it could jeopardize their state's section 1115 demonstration program, which has currently been extended, but due to the statutory requirement that coverage expansion DSH diversion funding have been approved by July 31, 2009, its demonstration coverage expansion DSH diversion funding would not be excluded. The commenter stated this is contrary to the purpose of excluding DSH funds for coverage expansions from the DHRM, which the commenter noted is to ensure that DSH funds diverted to expand health coverage are insulated from reductions.

Response: The statute requires that we take into account the extent to which a state's DSH allotment was included in the budget neutrality calculation for a coverage expansion that was approved under section 1115 of the Act as of July 31, 2009, specifically. The ACA made non-DSH funds available to support Medicaid expansion and the purchase of private insurance for eligible individuals through Health Insurance Exchanges, which may have reduced the need for states to divert DSH funds through demonstration projects. In recognition of the reduced need for DSH diversion, the July 31, 2009 date, which predates the enactment of the ACA, serves to ensure that states could not newly divert DSH under demonstration projects to avoid allotment reductions. If a state's initial section 1115 demonstration was approved as of July 31, 2009 and later extended, the amount approved under the associated the waiver would still be excluded for purposes of the HMF and HUF factors from DSH allotment reductions in the DHRM. However, for section 1115 demonstrations not approved as of July 31, 2009, any DSH allotment amounts included in budget neutrality calculations, whether for coverage expansion or otherwise, under a later approval would be subject to reduction. We note that, in some cases, modifications made by amendment (including in connection with a renewal or extension) to a coverage expansion DSH diversion initially approved as of July 31, 2009 may be so significant that the DSH diversion is no longer appropriately considered the same coverage expansion DSH diversion program as was approved as of July 31, 2009. In such a case, we would cease excluding the diverted DSH amounts from reduction under the DHRM. We are finalizing the rule as proposed.

Comment: Several commenters urged CMS to take into account that there is no policy reason to differentiate DSH funding for a coverage expansion demonstration in relation to the July 2009 date, and noted that the only policy reason given by CMS was that subsequent to July 31, 2009, the ACA provided states with other, non-DSH funds for such coverage expansion, thus limiting the need for diverted DSH under demos. The commenters suggested that CMS did this because it did not want to provide financial relief to states that chose not to effectuate coverage through a mechanism other than Medicaid expansion through the ACA and that CMS has the legal authority to exclude funding approved after July 31, 2009. The commenters stated their belief that their state has the only section 1115 waiver approved after July 31, 2009 that diverted DSH allotment for coverage expansion, and states that choose to expand coverage through a section waiver 1115, rather than expanding Medicaid to the adult expansion population as permitted under the ACA, will save the federal government money. The commenters urged CMS to exclude from the DHRM any DSH funding diverted to support any section 1115 demonstration coverage expansion approved at any time between July 31, 2009, and the effective date of the new regulation, or at a minimum, to include such projects approved on or before July 31, 2012.

Response: Consistent with the statute, for states that include DSH allotment amounts in budget neutrality calculations for coverage expansion under an approved section 1115 demonstration as of July 31, 2009, we are excluding from the DSH allotment reduction, for the HMF and the HUF factors, the amount of DSH allotment that each state currently continues to divert specifically for coverage expansion in the budget neutrality calculation. To promote equitable DSH allotment reductions across states, other than this specified statutory exception implemented through this rule, the final DHRM does not provide additional relief to states that include all or a portion of their DSH allotment in their section 1115 demonstration budget neutrality calculation.

Comment: One commenter noted that CMS proposed to estimate the targeting of section 1115 payments not excluded from reductions under the BNF for states by using DSH data from other states as a proxy, but did not provide a timeline for replacing the proxy data with actual hospital-specific data. The commenter recommended that a better long term approach would be to collect hospital-specific data on these payments to calculate the DSH targeting factors for these states directly.

Response: DSH allotment reductions relating to two DHRM targeting factors (the HUF and the HMF) are determined based on how states target DSH payments to certain hospitals. States that diverted all or a portion of their DSH allotments either make limited or no DSH payments using this diverted DSH allotment amount; therefore, actual hospital-specific DSH payment data suggested by the commenter for use often does not exist. We are finalizing use of a proxy as proposed for calculating DSH allotment reductions for purposes of the HUF and HMF. We will assign any qualifying states an average percentage reduction amount within its respective state group for diverted DSH allotment amounts that are not related to a coverage expansion in effect as of July 31, 2009, and for which the state does not have complete and/or relevant DSH payment data. We believe this is a reasonable approach for determining reductions for the HUF and HMF factors given the absence of relevant hospital-specific DSH payment data for these payments.

Comment: Some commenters suggested that CMS should re-examine the definition of “coverage for expansion purposes” and as it applies to the BNF to include safety net care pools and Uncompensated Care pools to the extent that they are established or used as part of broader efforts to expand coverage. Additionally, the commenters stated that there is no rational basis and that it is in fact contrary to the statutory intent to automatically designate all safety net care pools and uncompensated care pools as not contributing to coverage expansion purposes, and the July 2017 proposed rule provided no discussion of or justification for CMS' decision. The commenters requested that the full amount of a state's diverted DSH allotment in effect on July 31, 2009, be excluded from reduction.

Response: Uncompensated care pools and safety net care pools are designed to pay providers directly for uncompensated care costs, do not provide or pay for health care coverage for individuals, and do not result in the expansion of Medicaid coverage. Accordingly, they are excluded from consideration as coverage expansion for purposes of this factor.

Comment: Multiple commenters noted that CMS' proposed methodology would exclude from the DSH allotment reduction, for the HMF and HUF, the amount of DSH allotment that each state uses for coverage expansion in the budget neutrality calculation and recommended that CMS change the way in which the BNF is applied to also exclude the amount of DSH allotment that each state uses for coverage expansion from the UPF to account for the level of uninsured in the state.

Response: The statute requires that we take into account the extent to which a state's DSH allotment was included in the budget neutrality calculation for a coverage expansion that was approved under a demonstration project under section 1115 of the Act as of July 31, 2009. The proposed DHRM takes into account the extent to which the DSH allotment for a state was included in the budget neutrality calculation approved under section 1115 demonstration as of July 31, 2009, by excluding amounts diverted specifically for a coverage expansion and automatically assigning qualifying states an average percentage reduction amount (based on the state group) for any DSH allotment diverted for non-coverage expansion purposes and any amounts diverted for coverage expansion if the section 1115 demonstration was or is approved after July 31, 2009. DSH allotment reductions relating to two DHRM factors (the HUF and the HMF) are determined based on how states target DSH payments to certain hospitals. Since states diverting their DSH allotments under section 1115 demonstration projects would have limited or no relevant data for these two factors, we would be unable to evaluate how they spent the portion of their DSH allotment that was diverted for non-coverage expansion, which is why we proposed and are adopting the proxy methodology of assigning an average percentage reduction amount. However, the data necessary to calculate the UPF is unaffected by whether a state has diverted its DSH allotment under a section 1115 demonstration. Therefore, we do not exclude the amount of DSH allotment that each state has diverted through a section 1115 demonstration for coverage expansion from the UPF. We believe that the proposed methodology is an accurate and equitable approach, and we are finalizing this method in this final rule.

Comment: Two commenters noted that CMS did not propose to change the regulatory language at paragraph (e)(12)(i), stating that the phrase “(without regard to approved amendments since that date)” within the regulatory language may be confusing and possibly lead to misinterpretation or uncertainty and requested that CMS clarify its proposal regarding the amount excluded under the BNF calculation.

Response: We agree that the regulatory language could be misinterpreted and we are clarifying our intent in this final rule. For section 1115 demonstrations not approved as of July 31, 2009, any DSH allotment amounts included in budget neutrality calculations, whether for coverage expansion or otherwise, would also be subject to reduction.

Comment: One commenter questioned whether certain hospitals involved with Medicaid demonstration programs are subject to DSH audit and reporting requirements. Additionally, the commenter requested information on the impact of the reductions on state demonstration programs in states that use both DSH payments and section 1115 demonstration payments to fund hospitals.

Response: The final rule relies on DSH audit and reporting data as submitted by states in accordance with section 1923(j) of the Act and implementing regulations. The implementing regulations and associated policy guidance specify all audit and reporting requirements, including which hospitals must be included in the audit and associated reporting. The DSH audit and reporting requirements apply to all hospitals receiving DSH payments under section 1923 of the Act. Moreover, the DSH audit and reporting requirements continue to apply to states with section 1115 demonstrations, unless requirements of that section are specifically identified as waived or inapplicable to expenditures under the demonstration. As the reductions are not in effect at the time of publication of this final rule, we cannot know the specific impact the reductions will have on state demonstration programs, which is also likely to be affected by states' policy decisions regarding their Medicaid programs. Other than states that have a qualifying coverage expansion under the BNF of the DHRM, we generally anticipate a similar impact of the reductions on states that utilize DSH payments and section 1115 demonstration payments to fund hospitals, as on states that do not make section 1115 demonstration payments to hospitals.

Comment: One commenter noted that states would like to know the impact of the July 2017 proposed rule on Medicaid demonstration programs, including those related to Medicaid DSH.

Response: The statute requires that we take into account the extent to which a state's DSH allotment was included in the budget neutrality calculation for a coverage expansion that was approved under section 1115 of the Act as of July 31, 2009. This final rule implements this requirement by excluding from DSH allotment reduction the amount of DSH that qualifying states continue to divert specifically for coverage expansion in the budget neutrality calculation. Any amount of DSH diverted for other purposes under the demonstration (or diverted for a coverage expansion approved after July 1, 2009) would still be subject to reduction by automatically assigning qualifying states an average percentage reduction amount within its respective state group for factors for which the state does not have complete and/or relevant DSH payment data. DSH allotment amounts included in budget neutrality calculations for non-coverage expansion purposes under approved demonstrations (or for a coverage expansion approved after July 1, 2009) would still be subject to reduction.

IV. Provisions of the Final Rule

As discussed in section III. of this final rule, this final rule generally finalizes the provisions as proposed in the July 2017 proposed rule. However, we are adding paragraph § 447.294(e)(14)(iv) to finalize a proposed state-specific cap that limits the annual DSH allotment reduction for each fiscal year to be applied to each state's total unreduced DSH allotment to 90 percent of its original unreduced DSH allotment for that fiscal year. This addition is a technical change to correct an unintentional omission of proposed regulatory text to implement this proposed policy, which was discussed in the July 2017 proposed rule.

V. Collection of Information Requirements

Beginning with each state's Medicaid state plan for rate year 2005, each state must submit to CMS (at the same time as it submits the completed DSH audit as required under § 455.304) the data specified under § 447.299 for each DSH hospital to which the state made a DSH payment. The reporting requirements which allows CMS to verify the appropriateness of such payments are currently approved by OMB under control number 0938-0746 (CMS-R-266). This rule does not impose any new/revised information collection requirements or burden pertaining to § 447.299.

Although mentioned in sections III.B and III.B.2. of this preamble, this rule does not impose any new/revised SPA or auditing requirements or burden nor any new/revised information collection requirements or burden associated with CMS-64 (control number 0938-1265) or CMS-2552 (control number 0938-0050).

Since this rule does not impose any new or revised “collection of information” requirements or burden, it need not be reviewed by OMB under the authority of the Paperwork Reduction Act of 1995 (PRA) (44 U.S.C. 3501 et seq.). For the purpose of this section of the preamble, collection of information is defined under 5 CFR 1320.3(c) of the PRA's implementing regulations.

VI. Regulatory Impact Analysis

A. Statement of Need

The ACA amended the statute by requiring aggregate reductions to state Medicaid DSH allotments annually from FY 2014 through FY 2020. Subsequent legislation extended the reductions, modified the amount of the reductions, and delayed the start of the reductions until FY 2020. The most recent related amendments to the statute were through the BBA 18. This final rule delineates the DHRM to implement the annual reductions for FY 2020 through FY 2025.

B. Overall Impact

We have examined the impact of this rule as required by Executive Order 12866 on Regulatory Planning and Review (September 30, 1993), Executive Order 13563 on Improving Regulation and Regulatory Review (January 18, 2011), the Regulatory Flexibility Act (RFA) (September 19, 1980, Pub. L. 96-354), section 1102(b) of the Act, section 202 of the Unfunded Mandates Reform Act of 1995 (March 22, 1995; Pub. L. 104-4), Executive Order 13132 on Federalism (August 4, 1999), the Congressional Review Act (5 U.S.C. 804(2)), and Executive Order 13771 on Reducing Regulation and Controlling Regulatory Costs (January 30, 2017).

Executive Orders 12866 and 13563 direct agencies to assess all costs and benefits of available regulatory alternatives and, if regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety effects, distributive impacts, and equity). Section 3(f) of Executive Order 12866 defines a “significant regulatory action” as an action that is likely to result in a rule: (1) Having an annual effect on the economy of $100 million or more in any 1 year, or adversely and materially affecting a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or state, local or tribal governments or communities (also referred to as “economically significant”); (2) creating a serious inconsistency or otherwise interfering with an action taken or planned by another agency; (3) materially altering the budgetary impacts of entitlement grants, user fees, or loan programs or the rights and obligations of recipients thereof; or (4) raising novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

A regulatory impact analysis (RIA) must be prepared for major rules with economically significant effects ($100 million or more in any 1 year). We estimate that this rulemaking is “economically significant” as measured by the $100 million threshold, and hence also a major rule under the Congressional Review Act. Accordingly, we have prepared a Regulatory Impact Analysis that to the best of our ability presents the costs and benefits of the rulemaking. Under the Congressional Review Act (5 U.S.C. 801 et seq.), the Office of Information and Regulatory Affairs designated this rule as a major rule, as defined by 5 U.S.C. 804(2).

C. Anticipated Effects

The RFA requires agencies to analyze options for regulatory relief of small entities, if a rule has a significant impact on a substantial number of small entities. For purposes of the RFA, small entities include small businesses, nonprofit organizations, and small governmental jurisdictions. The great majority of hospitals and most other health care providers and suppliers are small entities, either by being nonprofit organizations or by meeting the SBA definition of a small business (having revenues of less than $7.5 million to $38.5 million in any 1 year). Individuals and states are not included in the definition of a small entity. We are not preparing an RFA analysis because we have determined, and the Secretary certifies, that this final rule would not have a significant economic impact on a substantial number of small entities (including hospitals and providers) because states still have considerable flexibility to determine DSH state plan payment methodologies.

In addition, section 1102(b) of the Act requires us to prepare a regulatory impact analysis if a rule may have a significant impact on the operations of a substantial number of small rural hospitals. This analysis must conform to the provisions of section 604 of the RFA. For purposes of section 1102(b) of the Act, we define a small rural hospital as a hospital that is located outside of a Metropolitan Statistical Area for Medicare payment regulations and has fewer than 100 beds. We are not preparing an analysis for section 1102(b) of the Act because we have determined, and the Secretary certifies, that this final rule would not have a significant impact on the operations of a substantial number of small rural hospitals.

Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA) also requires that agencies assess anticipated costs and benefits before issuing any rule whose mandates require spending in any 1 year of $100 million in 1995 dollars, updated annually for inflation. In 2019, that threshold is approximately $154 million. This final rule would not mandate any requirements for state, local, or tribal governments, nor would it affect private sector costs.

Executive Order 13132 establishes certain requirements that an agency must meet when it promulgates a proposed rule (and subsequent final rule) that imposes substantial direct requirement costs on state and local governments, preempts state law, or otherwise has Federalism implications. Since this rule does not impose substantial direct costs on state or local governments, the requirements of Executive Order 13132 are not applicable.

C. Anticipated Effects

1. Effects on State Medicaid Programs

We anticipate, effective for FY 2020, that the DSH allotment reductions would have a direct effect on the ability for some or all states to maintain state-wide Medicaid DSH payments at FY 2017 levels. Federal share DSH allotments, which are published by CMS in an annual Federal Register notice and otherwise communicated to states and made publicly available on the Medicaid.gov website, limit the amount of FFP in the aggregate that states can pay annually in DSH payments to hospitals. This final rule would reduce state DSH allotment amounts, and therefore, would limit the states' ability to make DSH payments and claim FFP for DSH payments at FY 2017 levels. By statute, the rule would reduce state DSH allotments by $44,000,000,000 for FY 2020 through FY 2025. We anticipate that the rule would reduce total FFP claimed by states by similar amounts, although it may not equal the exact amount of the allotment reductions. Due to the complexity of the interaction among the DHRM methodology, state DSH allotments, DHRM data, future state DSH payment levels and methodologies for these years, we cannot provide a specific estimate of the total federal financial impact for each year.

The final rule utilizes a DHRM that would mitigate the negative impact on states that continue to have high percentages of uninsured and are targeting DSH payments to hospitals that have a high volume of Medicaid patients and to hospitals with high levels of uncompensated care, consistent with statutorily-required factors.

2. Effects on Providers

We anticipate that the final rule would affect certain providers through the reduction of state DSH payments that states would need to implement in order to comply with their reduced annual state DSH allotments. However, we cannot estimate the impact on individual providers or groups of providers. This final rule would not affect the considerable flexibility afforded states in setting DSH state plan payment methodologies to the extent that these methodologies are consistent with section 1923(c) of the Act and all other applicable statutes and regulations. States would retain the ability to preserve existing DSH payment methodologies, to the extent consistent with the state's reduced annual DSH allotment, or to propose modified methodologies by submitting state plan amendments to us. Some states may determine that implementing a proportional reduction in DSH payments for all qualifying hospitals is the preferred method to account for the reduced allotment. Alternatively, states could determine that the best action is to propose a methodology that would direct DSH payments reductions to hospitals that do not have high Medicaid volume and do not have high levels of uncompensated care. Some states could opt to take a different approach. Regardless, the rule would incentivize states to target DSH payments to hospitals that are most in need of Medicaid DSH funding based on their serving a high volume of Medicaid inpatients and having a high level of uncompensated care.

This final rule also does not affect the calculation of the hospital-specific DSH limit established at section 1923(g) of the Act. This hospital-specific limit requires that Medicaid DSH payments to a qualifying hospital not exceed the costs incurred by that hospital for providing inpatient and outpatient hospital services furnished during the year to Medicaid patients and individuals who have no health insurance or other source of third party coverage for the services provided during the year, less applicable revenues for those services.

Although this rule would reduce state DSH allotments, the management of the reduced allotments still largely remains with the states. Given that states would retain the same flexibility to design DSH payment methodologies under the state plan and that individual hospital-specific DSH payment limits would not be affected, we cannot predict whether and how states would exercise their flexibility in setting DSH payments to account for their reduced DSH allotment and how this would affect individual providers or specific groups of providers.

D. Alternatives Considered

The statute specifies the annual DSH allotment reduction amounts. Therefore, we were unable to consider alternative reduction amounts. However, we did consider various methodological alternatives to the DHRM discussed in individual sections above. Some of the various alternatives included using alternative weight assignments, utilizing various alternative data sources for uncompensated cost and uninsured data, and considering alternate methods for capping individual state allotment reductions. However, we decided to move forward with the approach specified in the proposed rule in an effort to pursue an equitable and reasonable approach in calculating the DSH allotment reductions while ensuring that the DHRM complies with federal statutory requirements.

E. Accounting Statement and Table

As required by OMB Circular A-4 (available at www.whitehouse.gov/​sites/​whitehouse.gov/​files/​omb/​circulars/​A4/​a-4.pdf), we have prepared an accounting statement table showing the classification of the impacts associated with implementation of this final rule. Table 1 provides our best estimate of the reductions to state Medicaid Disproportionate Share Hospital (DSH) allotments annually beginning with fiscal year (FY) 2020 based on the data.

F. Reducing Regulation and Controlling Regulatory Costs

Executive Order 13771, titled Reducing Regulation and Controlling Regulatory Costs, was issued on January 30, 2017, and requires that the costs associated with significant new regulations “shall, to the extent permitted by law, be offset by the elimination of existing costs associated with at least two prior regulations.” It has been determined that this final rule is a transfer rule that does not impose more than de minimis costs and thus is not a regulatory action for the purposes of Executive Order 13771.

In accordance with the provisions of Executive Order 12866, this regulation was reviewed by the Office of Management and Budget.

List of Subjects in 42 CFR Part 447

• Accounting
• Administrative practice and procedure
• Drugs
• Grant programs-health
• Health facilities
• Health professions
• Medicaid
• Reporting and recordkeeping requirements
• Rural areas

For the reasons set forth in the preamble, the Centers for Medicare & Medicaid Services amends 42 CFR chapter IV as set forth below:

PART 447—PAYMENTS FOR SERVICES

1. The authority citation for part 447 is revised to read as follows:

Authority: 42 U.S.C. 1302 and 1396r-8.

2. Section 447.294 is amended—

a. By revising the section heading;

b. By revising paragraph (a);

c. In paragraph (b), by adding the definition of “Total hospital cost” in alphabetical order;

d. By revising paragraphs (d), (e) introductory text, (e)(3)(i), and (e)(5)(i) through (iii);

e. By adding paragraph (e)(14)(iv); and

f. By revising paragraph (f).

The revisions and additions reads as follows:

Footnotes