Medicaid Program; Medicaid and Children's Health Insurance Program (CHIP) Managed Care

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Centers for Medicare & Medicaid Services

42 CFR Parts 438 and 457

[CMS-2408-F]

RIN 0938-AT40

Medicaid Program; Medicaid and Children's Health Insurance Program (CHIP) Managed Care

AGENCY:

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

ACTION:

Final rule.

SUMMARY:

This final rule advances CMS' efforts to streamline the Medicaid and Children's Health Insurance Program (CHIP) managed care regulatory framework and reflects a broader strategy to relieve regulatory burdens; support state flexibility and local leadership; and promote transparency, flexibility, and innovation in the delivery of care. These revisions of the Medicaid and CHIP managed care regulations are intended to ensure that the regulatory framework is efficient and feasible for states to implement in a cost-effective manner and ensure that states can implement and operate Medicaid and CHIP managed care programs without undue administrative burdens.

DATES:

Effective Date:

These regulations are effective on December 14, 2020, except for the additions of §§ 438.4(c) (instruction 4) and 438.6(d)(6) (instruction 7), which are effective July 1, 2021.

Compliance Dates:

States must comply with the requirements of this rule beginning December 14, 2020, except for §§ 438.4(c), 438.6(d)(6), 438.340, and 438.364. States must comply with §§ 438.4(c) and 438.6(d)(6) as amended effective July 1, 2021 for Medicaid managed care rating periods starting on or after July 1, 2021. States must comply with § 438.340 as amended for all Quality Strategies submitted after July 1, 2021. As § 438.340 applies to CHIP through an existing cross-reference in § 457.1240(e), separate CHIPs must also come into compliance with the requirements of § 438.340 as amended for all Quality Strategies submitted after July 1, 2021. States must comply with § 438.364 for all external quality reports submitted on or after July 1, 2021. Because § 438.364 applies to CHIP through an existing cross reference in § 457.1250(a), separate CHIPs must also come into compliance with the requirements of § 438.364 for external quality reports submitted on or after July 1, 2021.

FOR FURTHER INFORMATION CONTACT:

John Giles, (410) 786-5545, for Medicaid Managed Care provisions.

Carman Lashley, (410) 786-6623, for the Medicaid Managed Care Quality provisions.

Melissa Williams, (410) 786-4435, for the CHIP provisions.

SUPPLEMENTARY INFORMATION:

I. Medicaid Managed Care

A. Background

States may implement a managed care delivery system using four types of Federal authorities—sections 1915(a), 1915(b), 1932(a), and 1115(a) of the Social Security Act (the Act); each is described briefly in this final rule.

Under section 1915(a) of the Act, states can implement a voluntary managed care program by executing a contract with organizations that the state has procured using a competitive procurement process. To require beneficiaries to enroll in a managed care program to receive services, a state must obtain approval from CMS under one of two primary authorities:

• Through a state plan amendment that meets standards set forth in section 1932 of the Act, states can implement a mandatory managed care delivery system. This authority does not allow states to require beneficiaries who are dually eligible for Medicare and Medicaid (dually eligible), American Indians/Alaska Natives (except as permitted in section 1932(a)(2)(C) of the Act), or children with special health care needs to enroll in a managed care program. State plans, once approved, remain in effect until modified by the state.

• We may grant a waiver under section 1915(b) of the Act, permitting a state to require all Medicaid beneficiaries to enroll in a managed care delivery system, including dually eligible beneficiaries, American Indians/Alaska Natives, or children with special health care needs. After approval, a state may operate a section 1915(b) waiver for a 2-year period (certain waivers can be operated for up to 5 years if they include dually eligible beneficiaries) before requesting a renewal for an additional 2 (or 5) year period.

We may also authorize managed care programs as part of demonstration projects under section 1115(a) of the Act that include waivers permitting the state to require all Medicaid beneficiaries to enroll in a managed care delivery system, including dually eligible beneficiaries, American Indians/Alaska Natives, and children with special health care needs. Under this authority, states may seek additional flexibility to demonstrate and evaluate innovative policy approaches for delivering Medicaid benefits, as well as the option to provide services not typically covered by Medicaid. Such flexibility is approvable only if the objectives of the Medicaid statute are likely to be met, and the demonstration is subject to evaluation.

The above authorities may permit states to operate their programs without complying with the following standards of Medicaid law outlined in section of 1902 of the Act:

•

Statewideness

[section 1902(a)(1) of the Act]: States may implement a managed care delivery system in specific areas of the state (generally counties/parishes) rather than the whole state;

•

Comparability of Services

[section 1902(a)(10) of the Act]: States may provide different benefits to people enrolled in a managed care delivery system; and

•

Freedom of Choice

[section 1902(a)(23)(A) of the Act]: States may generally require people to receive their Medicaid services only from a managed care plan's network of providers or primary care provider.

In the May 6, 2016

Federal Register

(81 FR 27498), we published the “Medicaid and Children's Health Insurance Program (CHIP) Programs; Medicaid Managed Care, CHIP Delivered in Managed Care, and Revisions Related to Third Party Liability” final rule (hereinafter referred to as “the 2016 final rule”) that modernized the Medicaid and CHIP managed care regulations to reflect changes in the use of managed care delivery systems. The 2016 final rule aligned many of the rules governing Medicaid and CHIP managed care with those of other major sources of coverage; implemented applicable statutory provisions; strengthened actuarial soundness payment provisions to promote the accountability of managed care program rates; strengthened efforts to reform delivery systems that serve Medicaid and CHIP beneficiaries; and enhanced policies related to program integrity.

In the January 18, 2017

Federal Register

(82 FR 5415), we published the “Medicaid Program; The Use of New or Increased Pass-Through Payments in Medicaid Managed Care Delivery Systems” final rule (the 2017 pass-through payments final rule) that made changes to the pass-through payment transition periods and the maximum

amount of pass-through payments permitted annually during the transition periods under Medicaid managed care contract(s) and rate certification(s). That final rule prevented increases in pass-through payments and the addition of new pass-through payments beyond those in place when the pass-through payment transition periods were established in the 2016 final Medicaid managed care regulations.

In the November 14, 2018

Federal Register

(83 FR 57264), we published the “Medicaid Program; Medicaid and Children's Health Insurance Plan (CHIP) Managed Care” proposed rule (the 2018 proposed rule) which included proposals designed to streamline the Medicaid and CHIP managed care regulatory framework to relieve regulatory burdens; support state flexibility and local leadership; and promote transparency, flexibility, and innovation in the delivery of care. This 2018 proposed rule was intended to ensure that the Medicaid and CHIP managed care regulatory framework is efficient and feasible for states to implement in a cost-effective manner and ensure that states can implement and operate Medicaid and CHIP managed care programs without undue administrative burdens.

Since publication of the 2016 final rule, the landscape for healthcare delivery continues to change, and states are continuing to work toward reforming healthcare delivery systems to address the unique challenges and needs of their local citizens. To that end, the Department of Health and Human Services (HHS) and CMS issued a letter

1

to the nation's Governors on March 14, 2017, affirming the continued HHS and CMS commitment to partnership with states in the administration of the Medicaid program, and noting key areas where we intended to improve collaboration with states and move toward more effective program management. In that letter, we committed to a thorough review of the managed care regulations to prioritize beneficiary outcomes and state priorities.

1

Letter to the nation's Governors on March 14, 2017:

https://www.hhs.gov/sites/default/files/sec-price-admin-verma-ltr.pdf.

Since our issuance of that letter, stakeholders have expressed that the current Federal regulations are overly prescriptive and add costs and administrative burden to state Medicaid programs with few improvements in outcomes for beneficiaries. As part of the agency's broader efforts to reduce administrative burden, we undertook an analysis of the current managed care regulations to ascertain if there were ways to achieve a better balance between appropriate Federal oversight and state flexibility, while also maintaining critical beneficiary protections, ensuring fiscal integrity, and improving the quality of care for Medicaid beneficiaries. This review process culminated in the November 14, 2018 proposed rule. After reviewing the public comments to the 2018 proposed rule, this final rule seeks to streamline the managed care regulations by reducing unnecessary and duplicative administrative burden and further reducing Federal regulatory barriers to help ensure that state Medicaid agencies are able to work efficiently and effectively to design, develop, and implement Medicaid managed care programs that best meet each state's local needs and populations.

B. Medicaid Managed Care Provisions of the Rule and Analysis of and Responses to Public Comments

We received a total of 215 timely comments from state Medicaid and CHIP agencies, advocacy groups, health care providers and associations, health insurers, managed care plans, health care associations, and the general public. The following sections, arranged by subject area, include a summary of the comments we received and our responses to those comments. In response to the November 14, 2018 proposed rule, some commenters chose to raise issues that were beyond the scope of our proposals. In this final rule, we are not summarizing or responding to those comments.

1. Standard Contract Requirements (§ 438.3(t))

In the 2016 final rule, we added a new provision at § 438.3(t) requiring that contracts with a managed care organization (MCO), prepaid inpatient health plan (PIHP), or prepaid ambulatory health plan (PAHP) that cover Medicare-Medicaid dually eligible enrollees provide that the MCO, PIHP, or PAHP sign a Coordination of Benefits Agreement (COBA) and participate in the automated crossover claim process administered by Medicare. The purpose of this provision was to promote efficiencies for providers by allowing providers to bill once, rather than sending separate claims to Medicare and the Medicaid MCO, PIHP, or PAHP. The Medicare crossover claims process is limited to fee-for service-claims for Medicare Parts A and B; it does not include services covered by Medicare Advantage plans under Medicare Part C.

Since publication of the 2016 final rule, we heard from a number of states that, prior to the rule, had effective processes in place to identify and send appropriate crossover claims to their managed care plans from the crossover file the states received from us. Medicaid beneficiaries can be enrolled in multiple managed care plans or the state's fee-for-service (FFS) program. For example, a beneficiary may have medical care covered by an MCO, dental care covered by a PAHP, and behavioral health care covered by the state's FFS program. When a Medicaid managed care plan enters into a crossover agreement with Medicare, as required in § 438.3(t), we then send to that plan all the Medicare FFS crossover claims for their Medicaid managed care enrollees, as well as to the state Medicaid agency. When this occurs, the managed care plan(s) may receive claims for services that are not the contractual responsibility of the managed care plan. Additionally, states noted that having all claims sent to the managed care plan(s) can result in some claims being sent to the wrong plan when beneficiaries change plans. Some states requested regulation changes to permit states to send the appropriate crossover claims to their managed care plans; that is, states would receive the CMS crossover file and then forward to each Medicaid managed care plan only those crossover claims for which that plan is responsible. These states have expressed that to discontinue existing effective processes for routing crossover claims to their managed care plans to comply with this provision adds unnecessary costs and burden to the state and plans, creates confusion for payers and providers, and delays provider payments.

To address these concerns, we proposed to revise § 438.3(t) to remove the requirement that managed care plans must enter into a COBA directly with Medicare and instead would require a state's contracts with managed care plans to specify the methodology by which the state would ensure that the managed care plans receive all appropriate crossover claims for which they are responsible. Under this proposal, states would be able to determine the method that best meets the needs of their program, whether by requiring the managed care plans to enter into a COBA and participate in the automated claims crossover process directly or by using an alternative method by which the state forwards crossover claims it receives from Medicare to each MCO, PIHP, or PAHP, as appropriate. Additionally, we proposed to require, if the state elects to use a methodology other than requiring

the MCO, PIHP, or PAHP to enter into a COBA with Medicare, that the state's methodology would have to ensure that the submitting provider is promptly informed on the state's remittance advice that the claim has been sent to the MCO, PIHP, or PAHP for payment consideration.

The following summarizes the public comments received on our proposal to revise § 483.3(t) and our response to those comments.

Comment:

Many commenters supported the proposed addition of state flexibility to use alternate mechanisms to send crossover claims to managed care plans. Commenters stated that the changes would provide states and plans more flexibility while continuing to promote better coordination of benefits for dually eligible individuals and reducing burden on the providers who serve them.

Response:

We appreciate the support for the proposed change to the regulation.

Comment:

One commenter supported the proposed rule but added that it is necessary for CMS to ensure that any alternative state crossover methodology separates Medicare claims from Medicaid rate setting and actuarial soundness.

Response:

While Medicare Part A or Part B cost-sharing payments—which are Medicaid costs—must be factored into Medicaid rate setting if a Medicaid plan is responsible for covering them, we agree that other costs for the provision of Medicare covered services should not be. Nothing in our proposal or the revision to § 438.3(t) that we finalize here will impact the processes for Medicaid rate-setting or determination of actuarial soundness.

Comment:

One commenter supported the proposed changes but offered a note of caution relating to potentially opening the door to subpar manual processes that states might adopt that could incur additional costs and unnecessary complexity.

Response:

As discussed in this final rule, the regulations at § 438.3(t) finalized in 2016 established a crossover process in which providers only bill once, rather than multiple times. The revision to § 438.3(t) that we are finalizing here maintains a process in which providers only bill once (to Medicare), because the regulation only applies when the state enters into a COBA but allows greater state flexibility in how that claim is routed from Medicare to Medicaid and Medicaid managed care plans. We agree that automated processes are usually optimal and create efficiencies for states, plans, and providers. We encourage states that adopt alternate methodologies to use automated processes as appropriate to achieve efficient and economical systems. Regardless of the method chosen by a state, the provider role in the crossover claim submission process is not changed by this proposal.

Comment:

Some commenters noted concern that the proposed changes would have on plans and providers that operated across state lines. One commenter noted that Medicaid managed care plans that operate in multiple states would need to develop and maintain different processes in different states. Another commenter who supported the proposed changes noted that it may pose challenges for providers that furnish services in multiple states.

Response:

We carefully considered the benefits of national uniformity for Medicaid managed care plans and providers that serve multiple states. In this instance, we believe that states should have the flexibility to adopt the methodology that works best within their state to ensure that the appropriate MCO, PIHP, or PAHP will receive all applicable crossover claims for which the MCO, PIHP, or PAHP is responsible. This flexibility will allow states to maintain current processes and not incur unnecessary costs or burden to providers and beneficiaries to conform to a new mandated process. We note here that this revision to § 438.3(t) does not require states to change their current cross over claim handling processes; it merely provides states with an option. Regardless of which methodology a state chooses to implement, it should not have any effect on providers, who should be able to submit their claims once and have it routed to the appropriate MCO, PIHP, or PAHP for adjudication.

Comment:

One commenter who objected to the proposed changes requested clarification about how it intersects with a similar provision in section 53102(a)(1) of the Bipartisan Budget Act (BBA) of 2018 (Pub. L. 115-123, enacted February 9, 2018) concerning procedures for states processing prenatal claims when there is a known third party liability.

Response:

We do not believe that § 438.3(t), as amended here, conflicts with the third party liability requirements added by section 53102(a)(1) of the BBA of 2018. We note that section 53102(a)(1) of the BBA of 2018 applies when the provider bills Medicaid directly for a prenatal claim. As further discussed in our June 1, 2018 Informational Bulletin to states,

2

that provision requires states to use standard cost avoidance when processing prenatal claims. Thus if the state Medicaid agency has determined that a third party is likely liable for a prenatal claim, it must reject, but not deny, the claim returning the claim back to the provider noting the third party that the Medicaid state agency believes to be legally responsible for payment. If a provider billed Medicaid for a prenatal claim for a dually eligible individual, the state would be required to reject the claim but note that Medicare is the liable third party, as Medicare would be the primary payer.

2

See

https://www.medicaid.gov/federal-policy-guidance/downloads/cib060118.pdf.

By contrast, the regulation in § 438.3(t), which is triggered because the state enters into a COBA with Medicare, applies when the provider bills Medicare for any Part A or B service under Medicare FFS for a dually eligible individual for which there is cost-sharing covered by Medicaid. Medicare would generate the crossover consistent with the COBAs in place. If the state has elected to require its Medicaid managed care plans to enter into a COBA with Medicare, then Medicare would forward the crossover claims to the Medicaid managed care plan. If the state has elected under § 438.3(t) to use a different methodology for ensuring that the appropriate managed care plan receives the applicable crossover claims, then Medicare would forward the crossover claims to the state pursuant to the state's COBA with Medicare; the state would then forward that crossover claim to the Medicaid managed care plan for payment. If a state adopts an alternate methodology as provided in § 438.3(t), the state must ensure that the submitting provider is promptly informed on the state's remittance advice that the claims have not been denied, but instead, has been sent to the MCO, PIHP, or PAHP for consideration.

Comment:

One commenter requested that CMS add regulatory language at § 438.3(t) stating that “The coordination methodology also must ensure that dually eligible individuals are not denied Medicaid benefits they would be eligible to receive if they were not also eligible for Medicare benefits.”

Response:

We agree that it is essential that dually eligible individuals are not denied Medicaid benefits they would be eligible to receive if they were not also eligible for Medicare benefits; however, this is outside the scope of this regulation, which is limited to how states must ensure the appropriate managed care plan receives all

applicable crossover claims for which it is responsible. We note that there is no regulation in part 438 that authorizes denial of Medicaid-covered services for an enrollee who is eligible for those services based on the enrollee's eligibility as well for Medicare.

Comment:

One commenter suggested providing Medicare eligibility data to MCOs, PIHPS, and PAHPs through data feeds, file transfers, or an online portal for efficient coordination of benefits, to improve care coordination and outcomes. One commenter encouraged free and timely access to all clinical and administrative data to promote coordination among managed care plans. The commenter suggested creating a standardized process by which managed care organizations can receive timely claims and clinical data from both Medicaid and Medicare. The commenter noted that the proposed changes may limit full integration in instances where a beneficiary in a Medicaid managed long term care plan is enrolled in an unaligned (that is, offered by a separate organization) Medicare Advantage plan such as a Dual Eligible Special Needs Plan offered by a different organization than offers the Medicaid plan in which the person is enrolled.

Response:

We appreciate the value of making such data available to plans. We are separately exploring whether we have authority to do so within existing limits, such as those established under the Health Insurance Portability and Accountability Act of 1996 (HIPAA) (Pub. L. 104-191, enacted August 21, 1996). For the comment on enrollment in different managed care plans for Medicaid and Medicare, we note that the Medicare crossover process is limited to Original Medicare (Medicare Part A and B). Claims for cost sharing for a Medicare Advantage enrollee are beyond the scope of this regulation.

Comment:

One commenter requested that we itemize all claim inclusion and exclusion selection criteria for professional claim services.

Response:

We do not have the authority to make the requested change. The National Uniform Claims Committee (NUCC), which establishes the standards for 837 professional claims and the CMS-1500 form, has chosen not to array professional and DME claims by type of bill (TOB), as happens with institutional claims, which are under the purview of the National Uniform Billing Committee (NUBC).

Comment:

One commenter requested that we allow plans to continue their COBA with, and receive crossover claims directly from, Medicare in states where plans already did so as required under the 2016 final rule.

Response:

As we proposed the amendment and are finalizing it here, § 438.3(t) does not require any changes for states and MCOs, PIHPs, and PAHPs that are already complying with the 2016 final rule. This final rule amends § 438.3(t) to provide states with additional flexibility to adopt a different methodology to ensure that the appropriate MCO, PIHP, or PAHP will receives all applicable crossover claims for which the MCO, PIHP, or PAHP is responsible, subject to some limited parameters to ensure that the applicable provider is provided information on the state's remittance advice. This additional flexibility might result in states developing and using more efficient and economical processes for handling cross-over claims applicable to Medicaid managed care enrollees.

Comment:

One commenter who supported the proposed changes encouraged CMS to monitor states that adopt alternative methodologies to ensure that providers still receive payments in a timely manner.

Response:

While this regulation does not establish a timeframe for the state to forward the crossover claim to its managed care plan, we note that the existing regulations on timely claims payment in the Medicaid FFS context apply. Specifically, § 447.45(d)(4)(ii) specifies that the state Medicaid agency must pay the Medicaid claim relating to a Medicare claim within 12 months of receipt or within 6 months of when the agency or provider receives notice of the disposition of the Medicare claim. A state that uses an alternative methodology under § 438.3(t) and receives crossover claims from Medicare would need to ensure payment within this timeframe. To do so, the state would need to forward crossover claims to a Medicaid plan and ensure the plan pays it within 6 months of when the state initially received it.

Comment:

A few commenters who opposed the proposed changes recommended that, if the regulation is finalized as proposed and a state elects to devise its own system, the state be required to promptly educate participating health care providers about any ensuing changes in the state's updated remittance advice. One commenter expressed concern that some health care providers would not be promptly made aware of the new requirements to submit multiple claims to the managed care plan for payment consideration, resulting in unpaid claims through no fault of their own, and that this would be antithetical to CMS' “Patients Over Paperwork” initiative.

Response:

We believe that provider education is critical whenever a state implements changes to how crossover claims are routed to the Medicaid managed care plan responsible for processing them. We encourage states to conduct such education prior to implementing any process changes.

For the concern that without education, providers would not know where to submit claims for Medicare cost-sharing, this provision is designed to remove from providers the burden of having to identify the Medicaid managed care plan in which each dually eligible patient is enrolled, and submit the bill for the Medicare cost-sharing to the correct plan. Under our proposed regulation, the crossover claim is still routed to the Medicaid managed care plan. States may continue to require that plans enter into a COBA with Medicare to route crossover claims directly to the plan. In the alternative, states that elect to receive crossover claims from Medicare (or elect any other methodology than having the Medicaid managed care plans enter into COBAs with Medicare) would route the claims to the plan and issue remittance advice to the applicable provider. In both cases, the claims will be routed to the Medicaid managed care plan; there is no need for the provider to take any action to identify or submit the crossover claim to the plan. We believe this is fully in line with our “Patients over Paperwork” initiative.

We also sponsor an enhanced secondary COBA feed (also known as the “Medicaid Quality project”), which is available to states that have a COBA and participate in the Medicare crossover process. This secondary feed ensures that states receive a complete array of Medicare FFS adjudicated Part A and B claims for individuals that the states submitted to CMS on their eligibility file. The state must be in receipt of the normal crossover claims file to be eligible to receive the second enhanced COBA feed.

Comment:

One commenter who opposed the proposed changes expressed concern that any process in which there is an intermediary would create confusion and delay. The commenter noted that a smoothly operating crossover claim process reduces burden on providers, and may make them more willing to serve Qualified Medicare Beneficiaries and other dually eligible individuals. The commenter suggested simplifying and streamlining the procedures so that all crossover claims can be handled promptly by one entity, either the state

Medicaid agency or the MCO. The commenter also noted that when CMS adopted § 438.3(t), it allowed states time to have Medicaid managed care plans get COBAs in place, and that if more time is needed, the better course would be to extend the time for enforcement rather than to modify the regulation.

Response:

We share the preference for reducing the complexity of the crossover claim process and agree with the commenter that complexity in the crossover process can be a disincentive to serving dually eligible individuals. The § 438.3(t) regulatory language that we proposed and are finalizing requires that if a state uses an alternate methodology, it must ensure that the appropriate managed care plan (that is, the MCO, PIHP, or PAHP whose contract covers the services being billed on the claim) receives all applicable crossover claims, and ensure the remittance advice conveys that the claim is forwarded rather than denied. In either scenario, the provider is relieved of the burden of determining which entity—the state or Medicaid managed care plan—is liable for the Medicare cost-sharing.

Comment:

Several commenters opposed the proposed changes and requested we leave the current regulatory requirements in place. Many of these commenters noted that the flexibility in the proposed change could increase provider administrative burden and confusion when states indicate multiple denials on the state's remittance advice to providers (that is, when they forward a crossover claim to an MCO, PIHP, or PAHP) and that it would create further confusion when the Medicaid managed care plan then processed the claim and notified the provider on the plan's remittance advice. Some also expressed concern that alternate methodologies would increase provider practice costs—especially for small practices.

Response:

As noted in the proposed rule and in this final rule, an important factor prompting the proposed change was that some states and providers raised concerns after the original provision was finalized in the 2016 final rule requiring a state to abandon effective alternative processes would actually add to provider burden and increase risk of payment delays. We believe that state flexibility would permit carefully crafted alternative arrangements to continue in a way that benefits providers, plans, and states. We reiterate that this proposal retains a system in which providers are only required to bill once (to Medicare), and that the claim will be transferred to Medicaid or the Medicaid managed care plan to address the payment of Medicare cost-sharing.

To address the commenter's concern about when states indicate multiple denials on the state's remittance advice, we are clarifying our intent by finalizing § 438.3(t) with additional text specifying that the state's remittance advice must inform the provider that the claim was not denied by the state but was redirected to a managed care plan for adjudication. We regret that our intent was not clear in the proposed rule and believe this clarification will minimize provider confusion and reduce the risk of providers inadvertently perceiving forwarded claims as denied.

After consideration of the public comments and for the reasons articulated in the proposed rule and our responses to comments, we are finalizing § 438.3(t) as proposed with one modification to clarify that when a state elects not to require its managed care plans to enter into COBAs with Medicare, the remittance advice issued by the state must indicate that the state has not denied payment but that the claim has been sent to the MCO, PIHP, or PAHP for payment consideration. In addition, we are finalizing the regulation text with slight grammatical corrections to use the present tense consistently.

2. Actuarial Soundness Standards (§ 438.4)

3

3

In

Texas

v.

United States,

No. 7:15-cv-151-O, slip op. at 40, 62 (N.D. Tex. Mar. 5, 2018), appeal docketed, No. 18-10545 (5th Cir. May 7, 2018) (“Texas”), six states challenged the portion of the regulation previously codified at 42 CFR 438.6(c)(1)(i)(C) (2002) (now codified in portions of 42 CFR 438.2, 438.4), defining “actuarially sound capitation rates” as capitation rates “that . . . [h]ave been certified . . . by actuaries who meet the qualification standards established by the American Academy of Actuaries and follow the practice standards established by the Actuarial Standards Board,” on the basis that Actuarial Standard of Practice (“ASOP”) 49 defines “actuarially sound capitation rates” to mean rates that account for all fees and taxes, including the Health Insurance Provider Fee (“HIPF”). In a decision issued on March 5, 2018, the court declared the challenged portion of the regulation to be “set aside” under the Administrative Procedure Act, 5 U.S.C. 706(2)(B) through (C). Texas, slip op. at 62. In its decision, the court specifically allowed CMS to “continue to use ASOP 49 to make internal decisions whether capitation rates are `actuarially sound,' ” and only “cannot use ASOP 49 to . . . require Plaintiffs to pay the HIPF.” Texas, slip op. at 21. As of July 2019, the court had not issued a final judgment. The government has appealed the court's March 5, 2018 decision; during the pendency of the appeal, the government is complying with the court's order.

a. Option to Develop and Certify a Rate Range (§ 438.4(c))

Before the 2016 final rule was published, we considered any capitation rate paid to a managed care plan that fell anywhere within the certified rate range to be actuarially sound (81 FR 27567). However, to make the rate setting and the rate approval process more transparent, we changed that process in the 2016 final rule at § 438.4 to require that states develop and certify as actuarially sound each individual rate paid per rate cell to each MCO, PIHP, or PAHP with enough detail to understand the specific data, assumptions, and methodologies behind that rate (81 FR 27567). We noted in that 2016 final rule that states could continue to use rate ranges to gauge an appropriate range of payments on which to base negotiations with an MCO, PIHP, or PAHP, but would have to ultimately provide certification to us of a specific rate for each rate cell, rather than a rate range (81 FR 27567). We believed that this change would enhance the integrity of the Medicaid rate-setting process and align Medicaid policy more closely with actuarial practices used in setting rates for non-Medicaid plans (81 FR 27568).

Since publication of the 2016 final rule, we heard from stakeholders that the requirement to certify a capitation rate per rate cell, rather than to certify a rate range, has the potential to diminish states' ability to obtain the best rates when contracts are procured through competitive bidding. For example, we heard from one state that historically competitively bid the administrative component of the capitation rate that the requirement to certify a capitation rate per rate cell may prevent the state from realizing a lower rate that could have been available through the state's procurement process. States that negotiate dozens of managed care plans' rates annually have also cited the potential burden associated with losing the flexibility to certify rate ranges. States have claimed that the elimination of rate ranges could potentially increase administrative costs and burden to submit separate rate certifications and justifications for each capitation rate paid per rate cell.

To address states' concerns while ensuring that rates are actuarially sound and Federal resources are spent appropriately, we proposed to add § 438.4(c) to provide an option for states to develop and certify a rate range per rate cell within specified parameters. We designed our proposal to address our previously articulated concerns over the lack of transparency when large rate ranges were used by states to increase or decrease rates paid to the managed care plans without providing further notification to us or the public of the change. We noted that the rate range option at proposed paragraph (c) would allow states to certify a rate range per rate cell subject to specific limits and

would require the submission of a rate recertification if the state determines that changes are needed within the rate range during the rate year. Under our proposal, we noted that an actuary must certify the upper and lower bounds of the rate range as actuarially sound and would require states to demonstrate in their rate certifications how the upper and lower bounds of the rate range were actuarially sound.

Specifically in § 438.4(c)(1), we proposed the specific parameters for the use of rate ranges: (1) The rate certification identifies and justifies the assumptions, data, and methodologies specific to both the upper and lower bounds of the rate range; (2) the upper and lower bounds of the rate range are certified as actuarially sound consistent with the requirements of part 438; (3) the upper bound of the rate range does not exceed the lower bound of the rate range multiplied by 1.05; (4) the rate certification documents the state's criteria for paying MCOs, PIHPs, and PAHPs at different points within the rate range; and (5) compliance with specified limits on the state's ability to pay managed care plans at different points within the rate range. States using this option would be prohibited from paying MCOs, PIHPs, and PAHPs at different points within the certified rate range based on the willingness or agreement of the MCOs, PIHPs, or PAHPs to enter into, or adhere to, intergovernmental transfer (IGT) agreements, or the amount of funding the MCOs, PIHPs, or PAHPs provide through IGTs. We proposed these specific conditions and limitations on the use of rate ranges to address our concerns noted in this final rule; that is, that rates are actuarially sound and ensure appropriate stewardship of Federal resources, while also permitting limited state flexibility to use certified rate ranges. We stated in the proposed rule our belief that the proposed conditions and limitations on the use of rate ranges struck the appropriate balance between prudent fiscal and program integrity and state flexibility. We invited comment on these specific proposals and whether additional conditions should be considered to ensure that rates are actuarially sound.

Under proposed § 438.4(c)(2)(i), states certifying a rate range would be required to document the capitation rates payable to each managed care plan, prior to the start of the rating period for the applicable MCO, PIHP, and PAHP, at points within the certified rate range consistent with the state's criteria in proposed paragraph (c)(1)(iv). States electing to use a rate range would have to submit rate certifications to us prior to the start of the rating period and must comply with all other regulatory requirements including § 438.4, except § 438.4(b)(4) as specified. During the contract year, states using the rate range option in § 438.4(c)(1) would not be able to modify capitation rates within the +/− 1.5 percent range allowed under existing § 438.7(c)(3); we proposed to codify this as § 438.4(c)(2)(ii). We noted that this provision would enable us to give states the flexibility and administrative simplification to use certified rate ranges. We noted in the proposed rule that while the use of rate ranges is not standard practice in rate development, our proposal would align with standard rate development practices by requiring recertification when states elect to modify capitation rates within a rate range during the rating year. States wishing to modify the capitation rates within a rate range during the rating year would be required, in proposed § 438.4(c)(2)(iii), to provide a revised rate certification demonstrating that the criteria for initially setting the rate within the range, as described in the initial rate certification, were not applied accurately; that there was a material error in the data, assumptions, or methodologies used to develop the initial rate certification and that the modifications are necessary to correct the error; or that other adjustments are appropriate and reasonable to account for programmatic changes.

We acknowledged that our proposal had the potential to reintroduce some of the risks that were identified in the 2016 final rule related to the use of rate ranges in the Medicaid program. In the 2016 final rule, we generally prohibited the use of rate ranges, while finalizing § 438.7(c)(3) to allow

de minimis

changes of +/− 1.5 percent to provide some administrative relief to states for small changes in the capitation rates. This change was intended to provide some flexibility with rates while eliminating the ambiguity created by rate ranges in rate setting and to be consistent with our goal to make the rate setting and rate approval processes more transparent. We specifically noted in the 2016 final rule that states had used rate ranges to increase or decrease rates paid to the managed care plans without providing further notification to us or the public of the change or certification that the change was based on actual experience incurred by the MCOs, PIHPs, or PAHPs that differed in a material way from the actuarial assumptions and methodologies initially used to develop the capitation rates (81 FR 27567 through 27568).

We further noted in the 2016 final rule that the prohibition on rate ranges was meant to enhance the integrity and transparency of the rate setting process in the Medicaid program, and to align Medicaid policy more closely with the actuarial practices used in setting rates for non-Medicaid health plans. We noted that the use of rate ranges was unique to Medicaid managed care and that other health insurance products that were subject to rate review submit and justify a specific premium rate. We stated in the 2016 final rule our belief that once a managed care plan has entered into a contract with the state, any increase in funding for the contract should correspond with something of value in exchange for the increased capitation payments. We also provided additional context that our policy on rate ranges was based on the concern that some states have used rate ranges to increase capitation rates paid to managed care plans without changing any obligations within the contract or certifying that the increase was based on managed care plans' actual expenses during the contract period. In the 2016 final rule, we reiterated that the prohibition on rate ranges was consistent with the contracting process where managed care plans are agreeing to meet obligations under the contract for a fixed payment amount (81 FR 27567-27568).

We noted how the specific risks described in the proposed rule concerned us, and as such, were the reason for specific conditions and limitations on the use of rate ranges that we proposed. Our rate range proposal was intended to prevent states from using rate ranges to shift costs to the Federal Government. There are some states that currently make significant retroactive changes to the contracted rates at, or after, the end of the rating period. As we noted in the 2016 final rule, we do not believe that these changes are made to reflect changes in the underlying assumptions used to develop the rates (for example, the utilization of services, the prices of services, or the health status of the enrollee), but rather we are concerned that these changes are used to provide additional reimbursements to the plans or to some providers (81 FR 27834). Additionally, we noted that states would need to demonstrate that the entirety of rate ranges (that is, lower and upper bound) compliant with our proposal are actuarially sound. As noted in the 2016 final rule, 14 states used rate ranges with a width of 10 percent or smaller (that is, the low end and the high end of the range were within 5

percent of the midpoint of the range), but in some states, the ranges were as wide as 30 percent (81 FR 27834). We noted that we believed that our proposal would limit excessive ranges because proposed § 438.4(c)(1)(i) and (ii) would require the upper and lower bounds of the rate range to be certified as actuarially sound and that the rate certification would identify and justify the assumptions, data, and methodologies used to set the bounds. While we believed that our proposal struck the right balance between enabling state flexibility and our statutory responsibility to ensure that managed care capitation rates are actuarially sound, we noted that our approach may reintroduce undue risk in Medicaid rate-setting.

Therefore, we requested public comments on our proposal in general and on our approach. We requested public comment on the value of the additional state flexibility described in our proposal relative to the potential for the identified risks described in the proposed rule and in the 2016 final rule, including other unintended consequences that could arise from our proposal that we have not yet identified or described. We requested public comment on whether additional conditions or limitations on the use of rate ranges would be appropriate to help mitigate the risks we identified. We also requested public comment from states on the utility of state flexibility in this area—specifically, we requested that states provide specific comments about their policy needs and clear explanations describing how utilizing rate ranges effectively meets their needs or whether current regulatory requirements on rate ranges were sufficiently flexible to meet their needs. We also requested that states provide quantitative data to help us quantify the benefits and risks associated with the proposal. We also encouraged states and other stakeholders to comment on the needs, benefits, risks, and risk mitigations described in the proposed rule.

The following summarizes the public comments received on our proposal to add § 438.4(c) and our responses to those comments.

Comment:

Many commenters supported the proposed option to develop and certify a 5 percent rate range, stating it allows for increased flexibility in rate setting. Commenters noted that the rate range proposal will remove ambiguities in determining actuarial soundness and will put appropriate limits on unsustainable rates. Some commenters specifically noted support for the requirements to recertify rates when there are changes made within the approved rate range and for states to document the specific rates for each managed care plan. Commenters also noted support for the proposal that states cannot pay managed care plans at different rates within the range based on IGT agreements. Several commenters noted that the specific conditions proposed by CMS must be implemented and strictly enforced to ensure that actuarial soundness is achieved within the rate ranges. A few commenters urged CMS to adopt all of the conditions set forth in § 438.4(c) if rate ranges are finalized.

Response:

We continue to believe, particularly with the support of commenters, that the 5 percent, or +/−2.5 percent from the midpoint, rate range will permit increased flexibility in rate setting, while the specific conditions proposed will also ensure that the rates are actuarially sound. The proposed parameters and guardrails carefully strike a balance between state flexibility and program integrity and we are finalizing them, with some modifications as discussed in response to other comments.

Comment:

Several commenters opposed the proposal to allow states to certify rate ranges and urged CMS not to finalize it. Some of these commenters expressed concerns that rate ranges decrease transparency, do not ensure that rates are actuarially sound, and do not enable CMS to ensure the adequacy of state and Federal investments in patient care. Some commenters noted that our proposal represents diminished Federal oversight of the adequacy of payment rates to Medicaid managed care plans and that it would result in lower payments to managed care plans, which could limit patient access to care. One commenter specifically expressed concern that wider rate ranges may result in lower rates to managed care plans and in turn result in contracts being awarded to less qualified plans, which may lead to early contract terminations, plan turnover, and instability for beneficiaries; this commenter also noted that such plans may be unable to pay competitive market rates, which could reduce patient access to care. Commenters also stated that the rate range provision is unnecessary since the existing +/−1.5 percent adjustment under § 438.7(c)(3) is adequate to provide states with administrative flexibility. One commenter expressed concern that the rate range proposal would result in reduced services for enrollees and instability for managed care plans and providers.

Response:

We understand commenters' concerns related to the use of rate ranges in Medicaid managed care. We also acknowledge our own fiscal and program integrity concerns which were noted in the 2016 final rule, as well as in the 2018 proposed rule. However, we proposed this rate range provision because we heard from states that this was a critical flexibility to reduce administrative burden in state Medicaid programs. We developed our proposal to carefully strike a balance between state flexibility and program integrity. Balancing this flexibility with the fiscal and program integrity concerns was the driving reason for including comprehensive guardrails around the use of rate ranges in the proposal and this final rule. To ensure appropriate Federal oversight, we specifically proposed parameters to ensure that rate ranges: (1) Do not inappropriately use IGTs to draw down additional Federal dollars with no correlating benefit to the Federal Government or the Medicaid program; (2) are bounded at the upper and lower ends with rates that are actuarially sound consistent with the regulations at §§ 438.4 through 438.7; and (3) strike the appropriate balance between prudent fiscal and program integrity and state flexibility. With regard to this last point, we specifically proposed that states using rate ranges must document in the rate certification the criteria used to select the specific rate within the range for each managed care plan under contract with the state. The guardrails finalized in § 438.4(c) will enable CMS to review the establishment and use of rate ranges by states and ensure that all rates actually paid to managed care plans are actuarially sound. To address specific concerns about unsound capitation rates, this final rule requires both the upper and lower bounds of the rate range to be actuarially sound; therefore, actuarially unsound rates would not be consistent with § 438.4(c).

We agree with commenters that the existing regulation that permits a +/−1.5 percent adjustment to certified rates can help states appropriately address mid-year programmatic changes or mid-year rate adjustments. However, we also believe that the additional option to certify a rate range can be helpful to states, especially in circumstances where states are competitively bidding the capitation rates. We also agree with commenters that rate ranges can obfuscate payment rates, and that is why we included specific guardrails around the use of rate ranges in the proposed rule and are finalizing those requirements. For

example, § 438.4(c)(1)(i) requires that the state's rate certification identify and justify the assumptions, data, and methodologies used to develop the upper and lower bounds of the rate range. Also, § 438.4(c)(2)(i) requires that states document the capitation rates, prior to the start of the rating period, for the managed care plans at points within the rate range, consistent with the state's criteria for paying managed care plans at different points within the rate range. This means that the contract and rate certification must be submitted for CMS approval before the rating period begins. We specifically included this timing requirement to limit the obfuscation of rates. We believe that the guardrails we proposed and are finalizing, such as these two examples, provide a level of transparency on the use of rate ranges and provide a mechanism to avoid obfuscation, especially since this regulation requires the actuary to describe and justify the assumptions, data, and methodologies used to develop the rate range in the actuarial certification.

We understand that several commenters were concerned that rate ranges could be used to lower payments to managed care plans, thereby leading to reduced services for enrollees and instability for managed care plans and providers; however, we have incorporated safeguards to prevent such outcomes. Under § 438.4(c)(1)(ii), both the upper and lower bounds of the rate range must be certified as actuarially sound consistent with the requirements in part 438. Actuarially sound capitation rates must provide for all reasonable, appropriate, and attainable costs that are required under the terms of the contract and for the operation of the managed care plan for the time period and the population covered under the terms of the contract. Since the lower bounds of rate ranges must also be actuarially sound and developed in accordance with the regulations in part 438 governing actuarial soundness and rate development, we believe that rates within the range must all be actuarially sound. Under § 438.4(c) as finalized, states using rate ranges must also document the criteria for paying managed care plans at different points within the rate range and must document the capitation rates prior to the start of the rating period—this means that the criteria used to set managed care plans' capitation rates must be documented prior to the start of the rating period. We believe that these requirements will ensure that states are not arbitrarily reducing payments to managed care plans. We also want to reiterate that the regulations in 42 CFR part 438 contain other beneficiary protections meant to ensure that plans are not arbitrarily reducing services to managed care enrollees. For example, § 438.210 requires that the services covered under the managed care contract must be furnished in an amount, duration, and scope that is no less than the amount, duration, and scope for the same services furnished to beneficiaries under FFS Medicaid. We would also highlight the requirements in § 438.206 regarding the timely availability of services that states and managed care plans must ensure for all managed care enrollees.

Comment:

Several commenters supported the proposal to allow rate ranges but recommended that the range be expanded beyond 5 percent. Some commenters requested that CMS expand the rate range provision to 10 percent. Commenters also requested that CMS restore rate ranges to pre-2016 regulatory levels, noting their belief that limits on a rate range are not necessary if the requirement of paying actuarially sound rates remains in place. Several commenters also recommended a narrower rate range to ensure the actuarial soundness of the final rates and recommended that actuaries be required to consider specific factors in determining the width (or size) of the rate range, such as maturity of the program, credibility/quality of the base data, amount of statistical variability in the underlying claim distribution, and size of the population. Commenters suggested additional rate ranges with a width (or size) of +/−2 percent (total range of 4 percent) or +/−3 percent (total range of 6 percent) from the midpoint, or two times the risk margin reflected in the capitation rates as alternatives to our proposal. Some commenters considered the proposed 5 percent range to be overly broad and recommended smaller ranges for the rates to remain actuarially sound. Some commenters gave specific scenarios by which the proposed 5 percent rate range may be insufficient and recommended that CMS not finalize a prescriptive +/− rate range to permit additional state flexibility.

Response:

We are declining to adopt any of these specific recommendations, as some commenters requested wider permissible ranges, while other commenters requested narrower permissible ranges. Because of the mix of public comments on this topic, we believe that we struck the right balance in the proposed rule by permitting a rate range up to 5 percent, or +/−2.5 percent from the midpoint, between the lower and upper bound. We believe that 5 percent, or +/−2.5 percent from the midpoint, is a reasonable rate range to permit the administrative flexibility requested by states and also to ensure that all rates within the entire rate range are actuarially sound. We proposed, and are finalizing, regulatory requirements that the rate certification identify and justify the assumptions, data, and methodologies specific to both the upper and lower bounds of the rate range (paragraph (c)(1)(i)), and that both the upper and lower bounds of the rate range are certified as actuarially sound (paragraph (c)(1)(ii)). We believe that the 5 percent, or +/−2.5 percent from the midpoint, rate range is more appropriate to ensure that these requirements can be satisfied, rather than an unspecified limit, or a limit that is so wide that it would not be possible to find both the upper and lower bounds of the rate range to be actuarially sound.

Regarding comments about the factors used in determining a rate range, such as maturity of the program, credibility/quality of the base data, amount of statistical variability in the underlying claim distribution, and size of the population, we believe that such factors would be permissible for actuaries to consider as part of the assumptions, data, and methodologies specific to both the upper and lower bounds of the rate range in accordance with generally accepted actuarial principles and practices, and the requirements for rate setting in §§ 438.4, 438.5, 438.6, and 438.7. If actuaries use these factors in determining the rate range, it would be appropriate to document these factors in the rate certification as required under § 438.4(c)(1)(i) and (ii). However, we decline to require that actuaries must consider these factors when determining the width (or size) of the rate range, as such an approach is overly prescriptive. We believe that actuaries may consider other factors when identifying and justifying the assumptions, data, and methodologies specific to both the upper and lower bounds of the rate range.

Comment:

Some commenters submitted technical recommendations about the rate range option, including that the calculation of the rate range should exclude risk adjustments and pass-through payments, that states should be able to apply or adjust risk adjustment mechanisms outside of setting the certified rate range, that the calculation of rate ranges should not reflect incentive payments for managed care plans, and that state budget factors should not influence the calculation of rates within the rate range. Other commenters recommended that administrative expenses should not be

subject to rate range variances. A few commenters recommended that certain government-mandated costs be considered outside of the rate range, including the Health Insurance Provider Fee (HIPF). One commenter also recommended that rate ranges be limited to include only underlying benefit changes.

Response:

Under § 438.4(c)(1)(ii), both the upper and lower bounds of the rate range must be certified as actuarially sound consistent with the requirements of part 438. This means that the calculation of the rate range under § 438.4(c) must include all of the components of the capitation rate that are currently required to be included in the rate development and certification under §§ 438.4, 438.5, 438.6, and 438.7. This includes pass-through payments, administrative expenses, taxes, licensing and regulatory fees, government-mandated costs (including the HIPF and other taxes and fees), and underlying benefit costs, which are all required components of developing the capitation rates under our existing regulations.

4

We agree that it would be inappropriate to include the specific incentive payments for managed care plans made under § 438.6(b)(2) in the calculation of the rate range, as per longstanding policy since the 1990's, those incentive arrangements are provided in excess of the approved capitation rate and are already limited to 105 percent of the approved capitation payments attributable to the enrollees or services covered by the incentive arrangement. We also agree that it would be inappropriate for state budget factors to influence the calculation of rates within the rate range since such factors are not considered valid rate development standards (that is, state budget factors are not relevant to the costs required to be included in setting the capitation rates in accordance with §§ 438.4, 438.5, 438.6, and 438.7).

4

While proceedings in

Texas

v.

United States

, No. 7:15-cv-151-O, slip op. (N.D. Tex. Mar. 5, 2018), appeal docketed, No. 18-10545 (5th Cir. May 7, 2018) (“Texas”), are ongoing, CMS will not require that the HIPF be accounted for in capitation rates for the six plaintiff states in Texas (Indiana, Kansas, Louisiana, Nebraska, Texas, and Wisconsin) in order for such rates to be approved as actuarially sound under 42 CFR 438.2 & 438.4(b)(1).

Regarding comments about risk adjustment, we generally agree with commenters that risk adjustment mechanisms can be applied outside of setting the certified rate range, consistent with existing Federal regulations at § 438.7(b)(5). While the state's actuary is required to certify rate ranges and must describe the risk adjustment methodology in the certification and certify the methodology, the state's actuary is not required to certify

risk-adjusted rate ranges

(that is, the rate ranges with the risk adjustment methodologies applied to reflect the actual payments potentially available to the managed care plan). The Federal requirements for including risk adjustment mechanisms in the capitation rates are found in § 438.7(b)(5). As part of the 2016 final rule, we acknowledged that risk adjustment methodologies can be calculated and applied after the rates are certified (81 FR 27595); therefore, we finalized specific standards for retrospective risk adjustment methodologies at § 438.7(b)(5)(ii). Further, the regulation at § 438.7(b)(5)(iii), which we finalized in the 2016 final rule, provides that a new rate certification is not required when approved risk adjustment methodologies are applied to the final capitation rates because the approved risk adjustment methodology must be adequately described in the original rate certification; payment of rates as modified by that approved risk adjustment methodology would be within the scope of the rate certification that adequately describes the risk adjustment mechanism. We also clarified in the 2016 final rule, under the requirements in § 438.7(c)(3), that the application of a risk adjustment methodology that was approved in the rate certification under § 438.7(b)(5) did not require a revised rate certification for our review and approval (81 FR 27568). However, we noted that the payment term in the contract would have to be updated as required in § 438.7(b)(5)(iii). Requirements for risk adjustment and risk sharing mechanisms in § 438.6 must also be met. Therefore, as long as the Federal requirements are met for risk adjustment, we agree that such mechanisms can be appropriately applied outside of the certified rate range (meaning, applied to the rates after calculation of the rate range), consistent with existing Federal regulations and our analysis in the 2016 final rule.

Comment:

One commenter recommended that CMS describe the permitted rate range in terms of a percentage.

Response:

We confirm for this commenter that the permissible rate range is expressed as a percentage. Section 438.4(c)(1)(iii), as finalized in this rule, requires that the upper bound of the rate range does not exceed the lower bound of the rate range multiplied by 1.05. This means that the upper bound of the rate range cannot exceed the lower bound of the rate range by more than 5 percent, or +/−2.5 percent from the midpoint.

Comment:

A few commenters requested clarification on the use of the

de minimis

+/−1.5 percent range that is currently codified in § 438.7(c)(3). Commenters requested detail on whether the proposal to allow rate ranges adds new parameters on the use of the

de minimis

flexibility that is currently codified in § 438.7(c)(3). Commenters also requested clarity on how the new rate range provision and the +/−1.5 percent flexibility can be used together.

Response:

As proposed and finalized, § 438.4(c) does not add or require additional parameters on the use of the +/−1.5 percent adjustment as permitted under § 438.7(c)(3) for any state that does not use rate ranges. However, under § 438.4(c)(2)(ii), states that use rate ranges are not permitted to modify the capitation rates under § 438.7(c)(3). States are permitted to either use the rate range option under § 438.4(c)(1) or use the

de minimis

+/−1.5 percent range that is currently codified in § 438.7(c)(3), but states are not permitted to use both mechanisms in combination. As noted in the 2018 proposed rule, we believe that this prohibition on using rate ranges in combination with the

de minimis

revision permitted under § 438.7(c)(3) is necessary to ensure program integrity and guard against other fiscal risks. As finalized at § 438.4(c)(1)(i), the rate certification must identify and justify the assumptions, data, and methodologies specific to both the upper and lower bounds of the rate range. The rate range cannot be wider than 5 percent, or +/−2.5 percent from the midpoint; the

de minimis

revision permitted under § 438.7(c)(3) cannot be used in combination with this rate range. It is our belief that the upper and lower bounds of a 5 percent rate range can remain actuarially sound as long as all of the Federal requirements for rate development, including the requirements we are finalizing in § 438.4(c), are met. If states were permitted to use rate ranges in combination with the

de minimis

revision permitted under § 438.7(c)(3), this could result in final rates that are outside of the 5 percent range, and this is not permitted. As provided in this rule in a separate response, we continue to believe that 5 percent is a reasonable rate range to permit the administrative flexibility requested by states. We believe that the 5 percent rate range is appropriate to ensure that the rate development requirements in part 438

can be satisfied, rather than a wider rate range where it may not be possible to find both the upper and lower bounds of the rate range to be actuarially sound.

Comment:

Several commenters expressed concern about the requirement to recertify modified rates within the rate range, noting that recertification is too rigid and is burdensome for both states and the Federal Government. One commenter requested that additional documentation could be provided rather than a requirement to recertify rates. A few commenters expressed concern that states cannot modify capitation rate ranges using the

de minimis

flexibility in § 438.7(c)(3) and requested that CMS allow states to employ both approaches to increase flexibility and reduce the need for recertification when rates change because of minor programmatic changes. Some commenters requested that mid-year rate changes be permitted within the rate range during the rating year without the need to recertify the rates to reduce burden and actuarial costs for states. Some commenters also recommended that CMS permit

de minimis

modifications to rates used within the 5 percent rate range.

Response:

We disagree with commenters that states should be able to use the

de minimis

rule in § 438.7(c)(3) in combination with a rate range. We proposed and are finalizing a prohibition on such combinations in § 438.4(c)(2)(ii). States may use either the rate range option under § 438.4(c) or use the

de minimis

+/−1.5 percent range that is currently codified in § 438.7(c)(3), but states are not permitted to use both mechanisms in combination. As noted in the 2018 proposed rule, we proposed § 438.4(c)(2)(ii) to enable appropriate state flexibility and administrative simplification without compromising program integrity or other fiscal risks. It is our belief that the upper and lower bounds of a 5 percent, or +/−2.5 percent from the midpoint, rate range should be permissible only as long as all of the Federal requirements for rate development are met. If states were permitted to use rate ranges in combination with the

de minimis

revision permitted under § 438.7(c)(3), this could result in final rates that are outside of the 5 percent range and therefore could result in a rate that is not actuarially sound. We continue to believe that 5 percent is a reasonable rate range to permit the administrative flexibility requested by states, but also to ensure that each rate within the entire rate range is actuarially sound. We believe that the 5 percent rate range is appropriate to ensure that the rate development requirements in part 438 can be satisfied, rather than a wider rate range where it may not be possible to find both the upper and lower bounds of the rate range to be actuarially sound.

However, we are persuaded that our proposal at § 438.4(c)(2)(iii), which required states to recertify capitation rates for modifications of the capitation rates within the rate range, regardless of whether the modification was for minor programmatic changes or a material error, was too rigid and would likely add unnecessary administrative burden and costs for states. We reached this conclusion for minor changes within the rate range that would not result in scenarios where such changes resulted in capitation rates outside of the 5 percent, or +/−2.5 percent from the midpoint, range. Therefore, we are finalizing § 438.4(c)(2)(iii) as permitting changes (increases or decreases) to the capitation rates per rate cell within the rate range up to 1 percent during the rating period without submission of a new rate certification, provided that such changes are consistent with a modification of the contract as required in § 438.3(c) and are subject to the requirements at § 438.4(b)(1). Just as we do not permit rate ranges in combination with the

de minimis

revision permitted under § 438.7(c)(3), we will not permit any changes that could result in final rates that are outside of the 5 percent range or in rate ranges that have upper and lower bounds that are larger than 5 percent apart.

Any modification to the capitation rates within the rate range greater than the permissible +/−1 percent amount will require states to provide a revised rate certification for CMS approval that demonstrates compliance with the criteria proposed and finalized at § 438.4(c)(2)(iii)(A) through (C). We believe that this modification to what we proposed for this regulation will address commenters' concerns related to mid-year programmatic changes or mid-year rate adjustments. We note that the permissible +/−1 percent standard under § 438.4(c)(2)(iii) is slightly smaller than the

de minimis

standard (+/−1.5 percent) for changes that do not require a new rate certification under § 438.7(c)(3) when rate ranges are not used. We believe that it is appropriate to use the smaller amount under § 438.4(c)(2)(iii) when rate ranges are used because when states use rate ranges, they are already afforded additional flexibility, as rates are permissible within the upper and lower bounds of the rate range, than they are afforded when certifying the rates to a specific point. We believe that the ability to make a permissible +/−1 percent change provides states flexibility to make small changes while easing the administrative burden of rate review for both states and CMS. Further, permitting small changes facilitates CMS' review process of rate certifications in accordance with the requirements for actuarially sound capitation rates because we would not require revised rate certifications for minor programmatic changes that result in minor and potentially immaterial changes to the capitation rates; therefore, CMS' review of rate certifications can be more focused on substantial issues that impact the capitation rates.

Comment:

Commenters requested clarification on the acceptable criteria for paying managed care plans at different points within the rate range. Specifically, commenters requested if rates can vary based on state negotiations with managed care plans or a competitive bidding process.

Response:

We note that capitation rates, including permissible rate ranges under § 438.4(c), must comply with all rate setting requirements in §§ 438.4, 438.5, 438.6, and 438.7. This means, as finalized in § 438.4(b)(1), that capitation rates must have been developed in accordance with the standards specified in § 438.5 and generally accepted actuarial principles and practices. Under this final rule, § 438.4(b)(1) also requires that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations must be based on valid rate development standards that represent actual cost differences in providing covered services to the covered populations (see section I.B.2.b. of this final rule for a discussion of this provision in § 438.4(b)(1)). We clarify this here to ensure that commenters are aware that the standards for capitation rate development, including the development of rate ranges under § 438.4(c), do not change with the use of rate ranges under § 438.4(c). Regarding the acceptable criteria for paying managed care plans at different points within the rate range, which must be documented in the rate certification documents under § 438.4(c)(1)(iv), we confirm that such criteria could include state negotiations with managed care plans or a competitive bidding process, as long as states document in the rate certification how the negotiations or the competitive bidding process produced different points within the rate range. For example, if specific, documentable components of the capitation rates varied because of state negotiations or a competitive bidding process, the rate

certification must document those specific variations, as well as document how those variations produced different points within the rate range, to comply with § 438.4(c)(1)(iv) and (c)(2)(i). We understand that capitation rate development necessarily involves the use of actuarial judgment, such as adjustments to base data, trend projections, etc., and that could be impacted by specific managed care plan considerations (for example, one managed care plan's utilization management policies are more aggressive versus another managed care plan's narrow networks); under this final rule, states must document such criteria as part of the rate certification to comply with § 438.4(c)(1)(iv) and (c)(2)(i).

Comment:

Several commenters recommended that CMS add minimum transparency requirements on the use of rate ranges. A few commenters recommended that states be required to provide managed care plans with the CMS approved rate ranges and the data underlying those rate ranges prior to bidding, with sufficient time and opportunity for managed care plans' review and input. Some commenters recommended that states be required to provide a comment period for managed care plans to review the rate ranges. One commenter recommended that CMS engage with managed care plans through a technical expert panel to develop appropriate standards for rate ranges. Another commenter recommended that CMS hold a public comment period during which stakeholders can raise issues related to rate ranges before and during the bidding process each year. Commenters also recommended that CMS require a dispute resolution process when states and managed care plans do not agree on the rate ranges. One commenter recommended that CMS require states to conduct studies to ensure that the rate ranges are sufficient to facilitate patient access to care.

Response:

In the proposed rule, we specifically requested public comment on whether additional conditions or limitations on the use of rate ranges would be appropriate to help mitigate the risks we identified. Based on the comments we received, we understand that commenters have significant concerns about the lack of transparency inherent in the use of rate ranges. The lack of transparency in the use of rate ranges has also been a significant concern for us; when we finalized the 2016 final rule, we explained that elimination of rate ranges would make the rate setting and the rate approval process more transparent (81 FR 27567). Further, we explained how the requirement to develop and certify as actuarially sound each individual rate paid per rate cell to each managed care plan with enough detail to understand the specific data, assumptions, and methodologies behind that rate would enhance the integrity of the Medicaid rate setting process (81 FR 27567). We agree with commenters that a significant level of transparency is necessary, particularly if states are using rate ranges for competitive bidding purposes. We believe that managed care plans and other stakeholders should have access to the necessary information and data to ensure that rates are actuarially sound, and we believe that such transparency will also help to ensure that competitive bids are appropriately based on actual experience and appropriately fund the program, and that the bids are actuarially sound. Providing managed care plans with approved rate ranges prior to bidding, with sufficient time and opportunity for managed care plans' review and input, along with the data underlying those rate ranges ensures that there is transparency in the setting and use of rate ranges.

Therefore, we are finalizing a requirement at § 438.4(c)(2)(iv) to require states, when developing and certifying a range of capitation rates per rate cell as actuarially sound, to post specified information. States are required under § 438.10(c)(3) to operate a public website that provides certain information. As finalized, § 438.4(c)(2)(iv) requires that states must post on their websites specified information prior to executing a managed care contract or contract amendment that includes or modifies a rate range. We are including this standard to ensure that managed care plans and stakeholders have access to the information with sufficient time and opportunity for review and input, and to ensure that the information is available to meaningfully inform plans' execution of a managed care contract with the state.

At § 438.4(c)(2)(iv)(A) through (C), we are finalizing the list of information that must be posted on the state's website required by § 438.10(c)(3): (A) The upper and lower bounds of each rate cell; (B) a description of all assumptions that vary between the upper and lower bounds of each rate cell, including for the assumptions that vary, the specific assumptions used for the upper and lower bounds of each rate cell; and (C) a description of the data and methodologies that vary between the upper and lower bounds of each rate cell, including for the data and methodologies that vary, the specific data and methodologies used for the upper and lower bounds of each rate cell. We believe that these requirements ensure that managed care plans and stakeholders have access to a minimum and standard level of information, for reasons outlined in the public comments. We believe that these requirements are also appropriate and necessary to ensure a minimum level of transparency when states utilize rate ranges under § 438.4(c). We also believe that this level of information will help to ensure that capitation rates are appropriately based on actual experience and are actuarially sound since plans will have access to such information prior to executing a managed care contract.

Regarding the public comments recommending public comment periods, technical expert panels, dispute resolution processes, and specific studies on access to care, we decline to adopt these specific recommendations. While we believe that states should seek broad stakeholder feedback, we do not believe that it is necessary to create new and expansive Federal requirements to accomplish this goal. In our experience, states are already working with many stakeholder groups, including their managed care plans, and we believe that states should continue to have discretion in how they convene stakeholder groups and obtain stakeholder feedback to inform Medicaid managed care payment policy. If states want to utilize public comment periods, technical expert panels, or conduct specific studies on access to care to help inform their rate setting, including rate ranges, states are welcome to utilize such approaches. We also understand that commenters are interested in Federal dispute resolution processes; however, we do not believe that is an appropriate role for CMS in the Medicaid program. When plans and/or other stakeholders do not agree on rates, we would refer those groups to the state Medicaid agencies to appropriately address specific rate setting concerns. Since state Medicaid agencies are the direct administrators of the Medicaid program in their respective states, we believe that this approach is more appropriate.

Comment:

Some commenters expressed concern that requiring states to document the capitation rates at points within the rate range prior to the start of the rating period is too rigid and unrealistic. Commenters noted that the time and labor-intensive process of developing and certifying actuarially sound rates can, and often does, result in unexpected delays that push the

process into the rating period for which the rates are being developed. Commenters recommended extending flexibility to states around submission timing in a manner that maintains proper CMS oversight and is consistent with current CMS practice. One commenter further recommended that if the timing requirement is finalized, it should be delayed by 3 years.

Response:

We acknowledge that our proposed requirement in § 438.4(c)(2)(i) that states document the capitation rates at points within the rate range prior to the start of the rating period means, as a practical matter, that states electing to use rate ranges must submit contracts and rate certifications to us prior to the start of the rating period. We also note that section 1903(m)(2)(A)(iii) of the Act and § 438.806 require that the Secretary must provide prior approval for MCO contracts that meet certain value thresholds before states can claim FFP. This longstanding requirement is implemented in the regulation at § 438.806(c), which provides that FFP is not available for an MCO contract that does not have prior approval from us. This requirement is necessary and appropriate to ensure that rate ranges are not used to shift costs onto the Federal Government and to protect fiscal and program integrity. As we noted in the 2018 proposed rule, one of the goals of the guardrails we proposed, and are finalizing here, for use of rate ranges is to prevent states from using rate ranges to make significant retroactive changes to the contracted rates at or after the end of the rating period; this goal is served by the requirement that rate ranges and the specific rates per cell be documented and provided to CMS prior to the beginning of the rating period. While we are not prohibiting outright all retroactive rate changes, the limits on when rates can be changed under § 438.4(c)(2) will necessarily limit the types of retroactive changes that raise the most issues. As we noted in the 2016 final rule and the 2018 proposed rule, we do not believe that retroactive changes are made to reflect changes in the underlying assumptions used to develop the rates (for example, the utilization of services, the prices of services, or the health status of the enrollee), but rather we are concerned that these changes are used to provide additional reimbursements to the managed care plans or to some providers without adding corresponding new obligations under the contract. We do not believe that such changes are consistent with actuarially sound rates and represent cost-shifting to the Federal Government.

Because of these specific concerns, we decline to adopt commenters' recommendations about the timing guardrails included in § 438.4(c), including the recommendation that we delay this proposal by 3 years. We are finalizing the rule with the requirement in § 438.4(c)(2)(i) that states document the capitation rates (consistent with the requirements for developing and documenting capitation rates) at points within the rate range prior to the start of the rating period. However, since rate ranges were previously prohibited under the 2016 final rule (and before this final rule), we believe a transition period is appropriate to allow states that elect to utilize the rate range option at § 438.4(c) time to appropriately develop rate ranges and submit the rate certifications and contracts in advance of the start of a rating period. Therefore, we are delaying the effective date of this provision to rating periods starting on or after July 1, 2021.

Comment:

A few commenters expressed concern about the proposal to prohibit states from paying MCOs, PIHPs, and PAHPs at different points within the certified rate range based on the willingness or agreement of the MCOs, PIHPs, or PAHPs to enter into, or adhere to, intergovernmental transfer (IGT) agreements, or the amount of funding the MCOs, PIHPs, or PAHPs provide through IGTs. One commenter expressed concern that the proposal is too restrictive on states' ability to make use of non-Federal share sources and that our proposal constrains states' authority under sections 1902(a)(2) and 1903(w) of the Act to draw upon a variety of state and local sources to fund the non-Federal share of medical assistance costs, including in the managed care context. One commenter stated that the prohibition on varying payments within a certified rate range based on the existence of IGT arrangements is an expansive Federal restriction on the longstanding ability of states to make use of a variety of non-Federal share sources to improve reimbursement to safety-net providers in managed care. Commenters recommended that the regulation be amended to allow using IGT agreements in conjunction with other criteria for paying managed care plans at different points within the rate range.

Response:

We disagree that our proposal unnecessarily constrains states' authority under sections 1902(a)(2) and 1903(w) of the Act to draw upon a variety of state and local sources to fund the non-Federal share of medical assistance costs, as our proposal does not limit states from using permissible sources of the non-Federal share to fund costs under the managed care contract. Under § 438.4(c)(1)(v), the state is not permitted to use as a criterion for paying managed care plans at different points within the rate range either of the following: (1) The willingness or agreement of the managed care plans or their network providers to enter into, or adhere to, IGT agreements; or (2) the amount of funding the managed care plans or their network providers provide through IGT agreements. This prohibition is specific to states using amounts transferred pursuant to an IGT agreement to pay managed care plans at different points within the rate range under § 438.4(c) and is not a prohibition on states' authority to use permissible sources of the non-Federal share to fund costs under the managed care contract. Further, we explicitly clarify here that certain financing requirements in statute and regulation are applicable across the Medicaid program irrespective of the delivery system (for example, fee-for-service, managed care, and demonstration authorities), and are similarly applicable whether a state elects to use rate ranges or not. Such requirements include, but are not limited to, limitations on financing of the non-Federal share applicable to health care-related taxes and bona fide provider-related donations.

We are concerned that without these specific parameters in the regulation, states could try to use rate ranges to inappropriately use IGTs to draw down additional Federal dollars with no correlating benefit to the Federal Government or the Medicaid program. To address commenters' concerns related to increasing levels of provider reimbursement for safety-net providers, we note that states can use the authority for state directed payments under § 438.6(c) to direct specific payments to providers. However, we clarify here that certain financing requirements in statute and regulation are applicable across the Medicaid program irrespective of the delivery system (for example, fee-for-service, managed care, and demonstration authorities), and are similarly applicable whether a state elects to direct payments under § 438.6(c). Such requirements include, but are not limited to, limitations on financing of the non-Federal share applicable to health care-related taxes and bona fide provider-related donations. These financing requirements similarly apply when a state elects to direct payments under § 438.6(c). We understand that safety-

net providers play a critical role in serving underserved populations in states, including Medicaid managed care enrollees. We also understand that safety-net providers are critical to maintaining network adequacy and adequate access to care in many communities, including rural areas of the state, and we do not believe our proposal unnecessarily constrains states' authority under sections 1902(a)(2) and 1903(w) of the Act to draw upon a variety of state and local sources to fund the non-Federal share of medical assistance costs, as our proposal does not limit states from using permissible sources of the non-Federal share to fund costs under the managed care contract.

After consideration of the public comments and for the reasons articulated in the proposed rule and our responses to comments, we are finalizing § 438.4(c) as proposed with the following modifications:

• At § 438.4(c)(2)(iii), we are finalizing authority for a state to make changes to the capitation rates within the permissible rate range of up to 1 percent of each certified rate within the rate range without the need for the state to submit a revised rate certification. Under final § 438.4(c)(2)(iii), a state may increase or decrease the capitation rate per rate cell within the rate range up to 1 percent of each certified rate during the rating period provided that any changes of the capitation rate within the permissible +/−1 percent amount must be consistent with a modification of the contract as required in § 438.3(c) and are subject to the requirements at § 438.4(b)(1). Any modification to the capitation rates within the rate range greater than the permissible +/−1 percent amount will require states to provide a revised rate certification for CMS approval and to meet the requirements listed in paragraphs (c)(2)(iii)(A) through (C).

• At § 438.4(c)(2)(iv), we are finalizing a requirement that states, when developing and certifying a range of capitation rates per rate cell as actuarially sound, must post the following specified information on their public websites: (A) The upper and lower bounds of each rate cell; (B) a description of all assumptions that vary between the upper and lower bounds of each rate cell, including for the assumptions that vary, the specific assumptions used for the upper and lower bounds of each rate cell; and (C) a description of the data and methodologies that vary between the upper and lower bounds of each rate cell, including for the data and methodologies that vary, the specific data and methodologies used for the upper and lower bounds of each rate cell.

States certifying a rate range must document the capitation rates payable to each managed care plan prior to the start of the rating period for the applicable MCO, PIHP or PAHP under § 438.4(c)(2)(i). As noted previously in this final rule, this requirement means that states electing to use a rate range would have to submit rate certifications to us

prior to the start of the rating period

and must comply with all other regulatory requirements including § 438.4, except § 438.4(b)(4) as specified. In order to publish additional guidance needed to implement this requirement, we are delaying the effective date of this provision until the first contract rating period beginning on or after July 1, 2021. States that elect to adopt rate ranges must comply with § 438.4(c) as amended effective July 1, 2021 for Medicaid managed care rating periods starting on or after July 1, 2021.

b. Capitation Rate Development Practices That Increase Federal Costs and Vary With the Rate of Federal Financial Participation (FFP) (§ 438.4(b)(1) and (d))

In the 2016 final rule, at § 438.4(b), we set forth the standards that capitation rates must meet to be approved as actuarially sound capitation rates eligible for FFP under section 1903(m) of the Act. Section 438.4(b)(1) requires that capitation rates be developed in accordance with generally accepted actuarial principles and practices and meet the standards described in § 438.5 dedicated to rate development standards. In the 2016 final rule (81 FR 27566), we acknowledged that states may desire to establish minimum provider payment rates in the contract with the managed care plan. We also explained that because actuarially sound capitation rates must be based on the reasonable, appropriate, and attainable costs under the contract, minimum provider payment expectations included in the contract must necessarily be built into the relevant service components of the rate. We finalized in the regulation at § 438.4(b)(1) a prohibition on different capitation rates based on the FFP associated with a particular population as part of the standards for capitation rates to be actuarially sound. We explained in the 2015 proposed rule (80 FR 31120) and the 2016 final rule (81 FR 27566) that different capitation rates based on the FFP associated with a particular population represented cost-shifting from the state to the Federal Government and were not based on generally accepted actuarial principles and practices.

In the 2016 final rule (81 FR 27566), we adopted § 438.4(b)(1) largely as proposed and provided additional guidance and clarification in response to public comments. We stated that the practice intended to be prohibited in § 438.4(b)(1) was variance in capitation rates per rate cell that was due to the different rates of FFP associated with the covered populations. We also provided an example in the 2016 final rule, in which we explained that we have seen rate certifications that set minimum provider payment requirements or established risk margins for the managed care plans only for covered populations eligible for higher percentages of FFP. Under the 2016 final rule, such practices, when not supported by the application of valid rate development standards, were not permissible. We further explained that the regulation did not prohibit the state from having different capitation rates per rate cell based on differences in the projected risk of populations under the contract or based on different payment rates to providers that were required by Federal law (for example, section 1932(h) of the Act). In the 2016 final rule, we stated that, as finalized, § 438.4(b)(1) provided that any differences among capitation rates according to covered populations must be based on valid rate development standards and not on network provider reimbursement requirements that apply only to covered populations eligible for higher percentages of FFP (81 FR 27566).

Since publication of the 2016 final rule, we have continued to hear from stakeholders that more guidance is needed regarding the regulatory standards finalized in § 438.4(b)(1). At least one state has stated that if arrangements that vary provider reimbursement pre-date the differences in FFP for different covered populations, the regulation should not be read to prohibit the resulting capitation rates. We explained in the 2018 proposed rule that while we believe that the existing text of § 438.4(b)(1) is sufficiently clear, we also want to be responsive to the comments from stakeholders and to eliminate any potential loophole in the regulation. Therefore, we proposed to revise § 438.4(b)(1) and added a new paragraph § 438.4(d) to clearly specify our standards for actuarial soundness. We did not propose changes to the existing regulatory requirement in § 438.4(b)(1) that capitation rates must have been developed in accordance with the standards specified in § 438.5

and generally accepted actuarial principles and practices but proposed to revise the remainder of § 438.4(b)(1).

We proposed that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations must be based on valid rate development standards that represent actual cost differences in providing covered services to the covered populations. Further, we proposed that any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of FFP associated with the covered populations in a manner that increases Federal costs consistent with a new proposed paragraph (d). Our proposal was intended to eliminate any ambiguity in the regulation and clearly specify our intent that variation in the assumptions, methodologies, and factors used to develop rates must be tied to actual cost differences and not to any differences that increase Federal costs and vary with the rate of FFP. The proposed revisions to § 438.4(b)(1) used the phrase “assumptions, methodologies, and factors” to cover all methods and data used to develop the actuarially sound capitation rates.

In conjunction with our proposed revisions to § 438.4(b)(1), we also proposed a new paragraph (d) to provide specificity regarding the rate development practices that increase Federal costs and vary with the rate of FFP. We proposed in § 438.4(d) a regulatory requirement for an evaluation of any differences in the assumptions, methodologies, or factors used to develop capitation rates for MCOs, PIHPs, and PAHPs that increase Federal costs and vary with the rate of FFP associated with the covered populations. We explained that this evaluation would have to be conducted for the entire managed care program and include all managed care contracts for all covered populations. We proposed to require this evaluation across the entire managed care program and all managed care contracts for all covered populations to protect against state contracting practices in their Medicaid managed care programs that may cost-shift to the Federal Government. We noted that this would entail comparisons of each managed care contract to others in the state's managed care program to ensure that variation among contracts does not include rate setting methods or policies that would be prohibited under our proposal.

We also proposed at § 438.4(d)(1) to list specific rate development practices that increase Federal costs and would be prohibited under our proposal for § 438.4(b)(1) and (d): (1) A state may not use higher profit margin, operating margin, or risk margin when developing capitation rates for any covered population, or contract, than the profit margin, operating margin, or risk margin used to develop capitation rates for the covered population, or contract, with the lowest average rate of FFP; (2) a state may not factor into the development of capitation rates the additional cost of contractually required provider fee schedules, or minimum levels of provider reimbursement, above the cost of similar provider fee schedules, or minimum levels of provider reimbursement, used to develop capitation rates for the covered population, or contract, with the lowest average rate of FFP; and (3) a state may not use a lower remittance threshold for a medical loss ratio for any covered population, or contract, than the remittance threshold used for the covered population, or contract, with the lowest average rate of FFP. We proposed § 438.4(d)(1) to be explicit about certain rate development practices that increase Federal costs and vary with the rate of FFP. Our proposal was to explicitly prohibit the listed rate development practices under any and all scenarios; we also noted that the rate development practices under § 438.4(d)(1) were not intended to represent an exhaustive list of practices that increase Federal costs and vary with the rate of FFP, as we recognized that there may be additional capitation rate development practices that have the same effect and would also be prohibited under our proposed rule. In the 2018 proposed rule, we explained our goal of ensuring that the regulatory standards for actuarial soundness clearly prevent cost-shifting from the state to the Federal Government.

Finally, in § 438.4(d)(2), we proposed to specify that we may require a state to provide written documentation and justification, during our review of a state's capitation rates, that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations or contracts, not otherwise referenced in paragraph (d)(1), represent actual cost differences based on the characteristics and mix of the covered services or the covered populations. We noted that our proposal was consistent with proposed revisions to § 438.7(c)(3), to add regulatory text to specify that adjustments to capitation rates would be subject to the requirements at § 438.4(b)(1), and to require a state to provide documentation for adjustments permitted under § 438.7(c)(3) to ensure that modifications to a final certified capitation rate comply with our regulatory requirements. We requested public comments on our revisions to § 438.4(b)(1) and new § 438.4(d), including on whether these changes were sufficiently clear regarding the rate development practices that are prohibited in § 438.4(b)(1).

The following summarizes the public comments received on our proposal to revise § 438.4(b)(1) and add § 438.4(d) and our responses to those comments.

Comment:

Many commenters supported the proposal to prohibit certain rate development practices and to require a state to provide written documentation that rate variations are based on actual cost differences. Many commenters also opposed this proposal and noted that there are often legitimate and actuarially sound reasons for varying pricing assumptions between rate cells that are independent of differing levels of FFP. Commenters stated that there are valid actuarial reasons where varying rating components would be supported by actuarial experience and data. Commenters recommended that were CMS to finalize the proposed amendments to § 438.4(b)(1) and (d), we do it in a way that would allow states to continue to have differentials in margins, payment levels, and MLR remittance thresholds for higher FFP contracts when those differences are justified in data and actuarial experience.

Commenters stated that valid rate development practices would be prohibited under the proposal, including using a lower margin assumption for populations with more stable costs, varying MLR thresholds based on actual administrative cost differences, adjusting the underwriting gain used, and using higher reimbursement for highly specialized providers or services or in areas where it is difficult to recruit providers. Commenters stated that the proposal is too prescriptive and duplicative of current requirements and recommended that CMS allow states to use assumptions that reflect different levels of risk so that rate cells are appropriately funded. Commenters stated that restricting actuarial variables from being determined by certain program characteristics will result in rates that are not actuarially sound. A few commenters also believed that the proposal could unintentionally result in new cost-shifting to the Federal Government, such as requiring higher margin assumptions for certain populations or requiring higher levels of provider reimbursement in specific

programs. One commenter requested that the regulation differentiate situations where rate development assumptions are intended to increase Federal costs from those where such an outcome is incidental and that CMS should only prohibit the former.

Several commenters recommended that instead of prohibiting certain rate development practices, CMS should instead require documentation and justification that variations related to margin, provider reimbursement, or MLR are actuarially valid. One commenter recommended that CMS only require such documentation in circumstances when we believe that the variation is related to FFP. One commenter expressed concern that the requirement to provide documentation duplicates existing policy. One commenter requested clarification on whether the written justification is part of the rate certification process and supporting documents, the managed care contract review, or is an additional requirement.

Response:

Our goal in proposing these revisions to § 438.4(b)(1) and (d) was to clarify the standards that capitation rates must meet to be approved as actuarially sound capitation rates eligible for FFP under section 1903(m) of the Act. Our proposal was also intended to eliminate any ambiguity in the regulations and to clearly specify that variation in the assumptions, methodologies, and factors used to develop rates must be tied to actual cost differences and not to any differences that increase Federal costs and vary with the rate of FFP. We remain committed to these goals, and to our overarching goals of improving fiscal and program integrity within Medicaid managed care rate setting. However, we believe it is appropriate to finalize the proposal with changes to address the concerns of commenters that our proposal was too restrictive and overlooked scenarios by which our prohibited rate development practices under proposed § 438.4(d) may be actuarially appropriate in limited circumstances.

We reiterate that our overarching policy goal of prohibiting variation in capitation rates associated with the FFP for a particular population, which we explained in the 2015 proposed rule, 2016 final rule, and the 2018 proposed rule, has not changed and is not changing as part of this final rule. Specifically, we explained in the 2015 proposed rule (80 FR 31120) that different capitation rates based on the FFP associated with a particular population represented cost-shifting from the state to the Federal Government and were not based on generally accepted actuarial principles and practices. In the 2016 final rule (81 FR 27566), we finalized, at § 438.4(b)(1), a prohibition on different capitation rates based on the FFP associated with a particular population as part of the standards for capitation rates to be actuarially sound. Also in the 2016 final rule (81 FR 27566), we provided additional guidance and clarification in response to public comments that the practice intended to be prohibited in § 438.4(b)(1) was variance in capitation rates per rate cell that was due to the different rates of FFP associated with the covered populations; that discussion included an example where rate certifications set minimum provider payment requirements or established risk margins for the managed care plans only for covered populations eligible for higher percentages of FFP. We note that setting minimum provider payment requirements for covered populations under the managed care contract is permissible as long as such requirements apply broadly, are not selectively applied to only those covered populations eligible for higher percentages of FFP, are supported by valid rate development standards that represent actual cost differences in providing covered services to the covered populations, and do not shift costs to the Federal Government. In the 2016 final rule, we explained how § 438.4(b)(1), as adopted there, required that any differences among capitation rates according to covered populations must be based on valid rate development standards and not on network provider reimbursement requirements that apply only to covered populations eligible for higher percentages of FFP (81 FR 27566). In the 2018 proposed rule (83 FR 57268), we clarified our policy that § 438.4(b)(1) was intended to prohibit variances in capitation rates based on the rate of FFP, even if such variances in capitation rates were the result of variances in provider reimbursement that pre-date the differences in FFP for different covered populations. We explained that our current proposal would eliminate ambiguity on this point and eliminate any potential loophole in § 438.4(b)(1) by more clearly specifying the scope of the prohibition. We reiterate these published statements here as part of this final rule and remind commenters that CMS has not changed our position on this topic. As finalized with the amendments in this final rule, § 438.4(b)(1) prevents states from cost-shifting onto the Federal Government and prohibits any variances in capitation rates associated with the rate of FFP for different covered populations. Further, we explicitly clarify here that § 438.4(b)(1) is not premised on nor require a state's intention to shift costs to the Federal Government; we believe that an intent to cost shift is immaterial compared to the actual effect of cost shifting.

Therefore, as part of this final rule, we are finalizing amendments to § 438.4(b)(1) to codify this policy clearly. Section 438.4(b)(1), as amended, continues to require that capitation rates be developed in accordance with the standards specified in § 438.5 and generally accepted actuarial principles and practices. We are also finalizing the proposed new and revised regulation text that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations must be based on valid rate development standards that represent actual cost differences in providing covered services to the covered populations and that any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of FFP associated with the covered populations in a manner that increases Federal costs. We are not finalizing the text proposed in paragraph (d)(1) to address the concerns from commenters that proposed § 438.4(d)(1) was too restrictive and overlooked scenarios where the proposed list of prohibited rate development practices may be actuarially appropriate.

We will generally use the list of prohibited rate development practices in interpreting the prohibition finalized in paragraph (b)(1) and we will consider the state's documentation and justification in applying the prohibition. We originally proposed § 438.4(d)(1) in conjunction with our proposed revisions to § 438.4(b)(1) to provide specificity regarding the rate development practices that we believed increased Federal costs and varied with the rate of FFP; however, based on public comments, we agree with commenters that there could be legitimate and actuarially sound reasons for varying pricing assumptions between rate cells that are (and must be) independent of differing levels of FFP, and that there could be valid actuarial reasons for an actuary to vary rating components that would be supported by actuarial experience and data. Therefore, as part of this final rule, we are not finalizing the list of rate development practices that we proposed in § 438.4(d)(1). We agree with the commenters that we are unable to

predict every future scenario and there might be situations where one or more of the items on that list of rate development practices is actuarially appropriate. We remind commenters that it is still our view that these rate development practices generally increase Federal costs and vary with the rate of FFP. As such, in situations where one of those practices is not actuarially appropriate and where it increases Federal costs, we will apply § 438.4(b)(1) to deny rates that have been developed based on such practices. To fully evaluate scenarios where differences in the assumptions, methodologies, or factors used to develop capitation rates appear to vary with the rate of FFP, we believe that we will need additional information and explanation from the state. If states or actuaries intend to utilize these rate development practices, we need to be able to require written documentation and justification that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations or contracts represent actual cost differences based on the characteristics and mix of the covered services or the covered populations. We had originally proposed a requirement for submission of information and documentation for this purpose under § 438.4(d)(2). Since we are not finalizing § 438.4(d), we are finalizing this proposed standard as part of the new text in § 438.4(b)(1). To address commenters' request for clarity, we note that such written documentation and justification would be required as part of CMS' review of the rate certification.

We are finalizing the introduction in proposed paragraph (d) as part of the new text in § 438.4(b)(1) that the evaluation of compliance with § 438.4(b)(1) be on a program-wide basis, including all managed care contracts and covered populations. The final rule continues to prohibit any differences in the assumptions, methodologies, or factors used to develop capitation rates that vary with the rate of FFP associated with a covered population in a manner that increases Federal costs. To ensure that this requirement is met, the final rule requires an evaluation of any differences in the assumptions, methodologies, or factors used to develop capitation rates for MCOs, PIHPs, and PAHPs that increase Federal costs and vary with the rate of FFP associated with the covered populations. This evaluation must be conducted for the entire managed care program and include all managed care contracts for all covered populations. We are finalizing this requirement for an evaluation across the entire managed care program and all managed care contracts for all covered populations to protect against any potential loopholes where state managed care contracting practices may cost-shift to the Federal Government. Specifically, as noted in the proposed rule, this requirement would entail comparisons of each managed care contract to others in the state's managed care program to ensure that variation among contracts does not include rate setting methods or policies that would be prohibited under § 438.4(b)(1).

Comment:

A few commenters noted that risk margin differences can apply between TANF, ABD, and LTSS populations and expressed concern that the proposal would require inappropriate comparisons between populations that have legitimate cost differences. Other commenters provided that the proposed regulation may have the unintended effect of causing actuaries to increase margins on disabled or LTSS populations to maintain justifiable higher margin assumptions for non-LTSS populations, which could increase Federal and state costs for the Medicaid program. Commenters stated that from an actuarial standpoint, the percentage risk margin may appropriately vary by population characteristics due to insurance risk differences. Commenters also explained that populations may have very different PMPM costs and, in particular, that expansion populations may require higher risk margins to account for unknown risks associated with a population not previously covered by the Medicaid program. Commenters recommended that CMS monitor for inappropriate rate setting practices or require additional documentation if we believe that cost-shifting is occurring, but these commenters recommended that CMS not finalize the proposal prohibiting specific rate development practices. One commenter stated that existing CMS authority enables us to enforce appropriate rate setting and that the proposed revisions to § 438.4(b)(1) and (d) are unnecessary.

Response:

We understand the issues raised by commenters and reiterate that to the degree the pricing assumptions are based on actual cost differences in providing covered services to the covered populations under the contract, these varying pricing assumptions would be permissible under § 438.4(b)(1) as valid rate development factors. To the degree that varying pricing assumptions represent actual cost differences in providing covered services to the covered populations, we would find the assumptions to be consistent with valid rate development standards. If a population has documented higher costs, supported by actual experience, we would not find the pricing assumptions to be in violation of finalized § 438.4(b)(1), even if the higher cost population is also one with a higher FFP percentage. However, we emphasize that varying pricing assumptions must not include using a rate development practice that increases Federal costs and varies with the rate of FFP when not supported by valid rate development standards that represent actual cost differences in providing covered services to the covered populations. As finalized in this rule under § 438.4(b)(1), any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of FFP associated with the covered populations in a manner that increases Federal costs unless those variances represent actual cost differences in providing covered services to the covered population.

Under the regulations governing rate setting at §§ 438.4 through 438.7, including as revised in this final rule, states and actuaries can vary the pricing assumptions based on actual cost differences in providing covered services to the covered populations under the contract, but the prohibition on shifting costs to the Federal Government through use of such variances remains. We also believe that there are other tools that can be used to mitigate the issues that were raised by commenters without inappropriately shifting costs onto the Federal Government and running afoul of § 438.4(b)(1). Although we are not finalizing a list of specifically prohibited rate development practices (as proposed at § 438.4(d)), it is still our view that these rate development practices generally increase Federal costs and vary with the rate of FFP, and as such, are generally prohibited under § 438.4(b)(1) as finalized in this rule. If states or actuaries intend to utilize these rate development practices (for example, higher margin assumptions for non-LTSS populations), we will require written documentation and justification that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations or contracts represent actual cost differences based on the characteristics and mix of the covered services or the covered populations.

Comment:

Several commenters provided that our proposal to restrict capitation rate development practices that are based on minimum levels of

provider reimbursement would likely result in unintended consequences for states seeking to comply with the provisions under at least two scenarios: (1) States would be required to decrease provider reimbursement rates to the lowest common denominator of the lowest FFP contracts, which could diminish access to care for some Medicaid populations; or (2) States would be required to increase provider reimbursement rates to the highest common denominator of the higher FFP contracts for the lowest FFP contracts, which could increase Federal and state Medicaid expenditures. Commenters also provided that many states have contracts for a specific population, such as a population at the average FMAP rate, with state statute setting the rate structure at the state's FFS rates; in these circumstances, this proposal would make the state's FFS rate structure the standard by which other managed care contracts would be evaluated, which may not be actuarially appropriate. A few commenters also expressed concern that fee schedule variation is limited under the proposal and noted that there is often a need to increase the fee schedules for certain provider types to meet network adequacy and encourage provider participation. These commenters expressed concern that such limitations on provider fee schedules may unfairly burden managed care plans. Commenters urged CMS not to finalize this provision as part of the proposal.

Response:

We understand the issues raised by commenters and reiterate that to the degree the pricing assumptions are based on actual cost differences in providing covered services to the covered populations under the contract, these varying pricing assumptions would be permissible under § 438.4(b)(1) as valid rate development factors. To the degree that varying pricing assumptions represent actual cost differences in providing covered services to the covered populations, we would find the assumptions to be consistent with valid rate development standards. If a population has documented higher costs, supported by actual experience, we would not find the pricing assumptions to be in violation of finalized § 438.4(b)(1). However, in our experience in reviewing and approving capitation rates, we have seen rate certifications that set minimum provider payment requirements for the managed care plans for covered populations eligible for higher percentages of FFP. We note here that such practices, when they shift costs to the Federal Government and when not supported by the application of valid rate development standards, are not permissible. Any differences among capitation rates according to covered populations must not shift costs to the Federal Government and must be based on valid rate development standards rather than network provider reimbursement requirements that apply to covered populations eligible for higher percentages of FFP even in cases where provider reimbursement requirements for such populations are mandated by state statute. Furthermore, we reiterate that setting minimum provider payment requirements for covered populations under the managed care contract is not permissible if such requirements shift costs to the Federal Government, even if such differential provider payments are authorized under § 438.6(c). For example, we have seen one state use § 438.4(b)(1) to vary provider reimbursement for covered populations eligible for higher percentages of FFP, as mandated by state law and not on valid rate development standards, and this state has stated that when such arrangements pre-date differences in FFP, the regulation should not be read to prohibit the resulting capitation rates. Our proposal and the amendment to § 438.4(b)(1) we are finalizing here eliminates that particular argument as a potential loophole. Regardless of when the differential rates were started, § 438.4(b)(1) as amended in this rule requires that differential rates be based on valid rate development standards and that they not shift costs to the Federal Government; such non-compliant differential rates must be eliminated. As revised, § 438.4(b)(1) prevents states from cost-shifting onto the Federal Government and prohibits any variances in capitation rates based on the rate of FFP for different covered populations, regardless of whether arrangements that vary provider reimbursement are mandated by state statute and/or pre-date the differences in FFP for different covered populations. We note that this state also stated that the different rates were intended to better align Medicaid rates with commercial rates but did not demonstrate that differential provider payments for one covered population was a valid rate development factor. As noted above in a previous response to public comments in this section, setting minimum provider payment requirements for covered populations under the managed care contract is permissible as long as such requirements apply broadly, are not selectively applied to covered populations eligible for higher percentages of FFP, are supported by valid rate development standards that represent actual cost differences in providing covered services to the covered populations and do not shift costs to the Federal Government. We note that varying pricing assumptions based on provider payment requirements mandated by state legislation that shift costs to the Federal Government do not constitute actual cost differences in providing covered services.

To the extent that states need to enhance reimbursement for specific providers or specific services, we believe that states can utilize other means to accomplish that goal, such as enhancing fees for covered services across all of their programs rather than varying fee schedules only for higher FMAP populations. We also understand that some states may have legislatively mandated fee schedules; however, as long as such states comply with all applicable regulatory requirements and are not including additional costs or mandating higher levels of reimbursement for higher FMAP populations, states can comply with mandated fee schedules and this regulation without a conflict. Mandated fee schedules that comply with all applicable regulatory requirements and do not result in higher payment for higher FMAP populations may be used as the basis for rate setting for the managed care contracts. We emphasize that varying pricing assumptions must not include using a rate development practice that increases Federal costs and varies with the rate of FFP when not supported by valid rate development standards that represent actual cost differences in providing covered services to the covered populations regardless of whether such differences are mandated by state legislation. As finalized in this rule under § 438.4(b)(1), any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of FFP associated with the covered populations in a manner that increases Federal costs.

Although we are not finalizing a list of prohibited rate development practices (as proposed at § 438.4(d)), it is still our view that these rate development practices generally increase Federal costs and vary with the rate of FFP, and as such, are prohibited in most cases under § 438.4(b)(1) as finalized in this rule. If states or actuaries intend to utilize these rate development practices, we will require written documentation and justification that any differences in the assumptions, methodologies, or

factors used to develop capitation rates for covered populations or contracts represent actual cost differences based on the characteristics and mix of the covered services or the covered populations.

Comment:

A few commenters expressed concern with the proposal (at proposed § 438.4(d)(1)(iii)) that states may not use a lower MLR remittance threshold for expansion populations than the MLR remittance threshold used for TANF, ABD, and LTSS contracts. Commenters stated that it is impractical and not actuarially sound to use an average MLR remittance threshold without acknowledging the actual costs of each managed care program and covered population. One commenter noted that remittance thresholds vary as a function of the administrative load of a product and is unrelated to the FFP for the program. Some commenters expressed concern that the proposal will force states to reduce MLR remittance thresholds for all managed care contracts, which will increase Federal Medicaid costs. Some commenters also stated that there are valid actuarial reasons to establish a higher MLR remittance threshold for LTSS populations, and that states should not be prohibited from designing such reasonable approaches based on actuarially sound practices. Commenters provided that the administrative costs for an LTSS program as a percent of revenue is lower than an expansion program (managed care plan covering the Medicaid benefits for the expansion population). As such, if a minimum MLR threshold is developed with an equal likelihood of being triggered by each program, the LTSS MLR threshold would need to be higher for the LTSS program.

Response:

We agree with commenters and acknowledge that our proposed rule failed to account for varying MLR thresholds for high-cost populations, such as LTSS populations. We agree that if a minimum MLR threshold is developed with an equal likelihood of being triggered, the MLR may need to be higher for LTSS programs because the administrative costs, as a percent of revenue, may be lower. Under § 438.4(b)(1) as finalized here, we will require states to provide valid reasons for varying the MLR threshold component in contracts where the FFP percentages are different. For approval of rates that are developed using such different MLR thresholds, a state could demonstrate that it has used factors to develop rates based on valid rate development standards and not on differences that increase Federal costs and vary with the rate of FFP, and it has applied the same methodologies for developing the administrative costs within the capitation rate, and therefore, the corresponding MLR remittance threshold is based on those same underlying methodologies. In such a situation, we would not find this approach to be a violation of § 438.4(b)(1) despite the different MLR thresholds used in setting the rates for high and low FMAP populations.

We emphasize that varying pricing assumptions must not include using a rate development practice that increases Federal costs and varies with the rate of FFP when not supported by valid rate development standards that represent actual cost differences in providing covered services to the covered populations. We note that varying pricing assumptions based only on provider payment requirements mandated by state legislation do not constitute actual cost differences in providing covered services. As finalized in this rule under § 438.4(b)(1), any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of FFP associated with the covered populations in a manner that increases Federal costs. Although we are not finalizing a list of prohibited rate development practices (as proposed at § 438.4(d)(1)), it is still our view that these rate development practices generally increase Federal costs and vary with the rate of FFP, and as such, are prohibited in most cases under § 438.4(b)(1) as finalized in this rule. If states or actuaries intend to utilize these rate development practices, we will require written documentation and justification that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations or contracts represent actual cost differences based on the characteristics and mix of the covered services or the covered populations.

Comment:

One commenter expressed concern that the proposal does not account for recent statutory changes made by section 4001 of the Substance Use-Disorder Prevention that Promotes Opioid Recovery and Treatment (SUPPORT) for Patients and Communities Act (Pub. L. 115-271, enacted October 24, 2018), which allows states to retain a larger share of the remittances collected from managed care plans by remitting funds back to the Federal Government for expansion enrollees at the state's standard rate of FFP, provided that certain statutory conditions are met.

Response:

Section 4001 of the SUPPORT for Patients and Communities Act, enacted October 24, 2018, amended section 1903(m) of the Act to add a new paragraph (m)(9). Section 1903(m)(9) provides a time-limited opportunity (after fiscal year 2020 but before fiscal year 2024) for states that collect an MLR remittance from their Medicaid managed care plans for the eligibility group described in section 1902(a)(10(A)(i)(VIII) to apply the state's regular Federal medical assistance percentage (FMAP) match rate (calculated pursuant to section 1905(b) of the Act) to determine the Federal share of that remittance instead of the higher FMAP match rate specified under 1905(y) for use in connection with the Medicaid expansion group. Since this statutory provision is limited to requirements on the amounts paid to the Federal Government on certain MLR remittances within the specified parameters of the statute, and not related to varying the remittance thresholds for specific populations or contracts, section 4001 of the SUPPORT for Patients and Communities Act is outside the scope of this final rule. Specifically, we clarify for this commenter that this final rule does not implicate the requirements under section 4001 of the SUPPORT for Patients and Communities Act regarding the amounts paid to the Federal Government on certain MLR remittances.

Comment:

A few commenters requested clarification regarding whether the proposal would apply to CHIP programs.

Response:

We clarify here that our proposal under § 438.4(d) was never intended to and our amendment of § 438.4(b)(1) does not apply to CHIP programs. The CHIP requirements for rate development are found in § 457.1203, which does not incorporate or reference the Medicaid managed care regulations on actuarial soundness and rate setting.

Comment:

A few commenters expressed concern that CMS refers to a list of prohibited rate development practices that are “including but not limited to” certain practices. These commenters expressed concern that this non-exhaustive list requires additional clarification and recommended that specific rate development practices be identified through notice and comment rulemaking.

Response:

The list of specific rate development practices that we proposed to prohibit outright in proposed § 438.4(d) is not being finalized. However, should such practices be used and result in rates that violate the standard we proposed and are finalizing in the amendment of § 438.4(b)(1), the

resulting rates will not be approved. We confirm for commenters that should we find it necessary to prohibit specific rate development practices in the future, we would do so through notice and comment rulemaking. Here, however, we are limiting how rates must be developed to ensure that differences in the assumptions, methodologies, and factors used to develop rates are based on valid rate development factors that represent actual cost differences and do not vary with the rate of FFP in a manner that increases Federal costs.

After consideration of the public comments and for the reasons articulated in the proposed rule and our responses to comments, we are finalizing our proposed amendments to § 438.4(b)(1) with modifications and are not finalizing the proposed addition of § 438.4(d); specifically, we are finalizing amendments to § 438.4(b)(1) as follows:

• At § 438.4(b)(1), we are finalizing the proposal to add regulation text to provide that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations must be based on valid rate development standards that represent actual cost differences in providing covered services to the covered populations and that any differences in the assumptions, methodologies, or factors used to develop capitation rates must not vary with the rate of Federal financial participation (FFP) associated with the covered populations in a manner that increases Federal costs.

• At § 438.4(b)(1), we are finalizing that the evaluation of compliance with § 438.4(b)(1) be on a program-wide basis, including all managed care contracts and covered populations. The final rule will require an evaluation of any differences in the assumptions, methodologies, or factors used to develop capitation rates for MCOs, PIHPs, and PAHPs that increase Federal costs and vary with the rate of FFP associated with the covered populations. This evaluation must be conducted for the entire managed care program and include all managed care contracts for all covered populations. This provision was proposed as part of the introduction text to paragraph (d).

• At § 438.4(b)(1), we are also finalizing the authority for CMS to require a state to provide written documentation and justification that any differences in the assumptions, methodologies, or factors used to develop capitation rates for covered populations or contracts represent actual cost differences based on the characteristics and mix of the covered services or the covered populations. This provision was proposed as part of paragraph (d)(2).

• At § 438.4(b)(1), we are not finalizing any references to paragraph (d).

3. Rate Development Standards: Technical Correction (§ 438.5(c)(3)(ii))

In the 2016 final rule, we finalized at § 438.5(c)(3) an exception to the base data standard at § 438.5(c)(2) in recognition of circumstances where states may not be able to meet the standard at paragraph (c)(2) regarding base data. We explained in the 2016 final rule preamble (81 FR 27574) that states requesting the exception under § 438.5(c)(3) must submit a description of why the exception is needed and a corrective action plan detailing how the state will bring their base data into compliance no more than 2 years after the rating period in which the deficiency was discovered.

Regrettably, the regulation text regarding the corrective action timeline at § 438.5(c)(3)(ii) was not as consistent with the preamble or as clear as we intended. The regulation text finalized in 2016 provided that the state must adopt a corrective action plan to come into compliance “no later than 2 years from the rating period for which the deficiency was identified.” The preamble text described the required corrective action plan as detailing how the problems “would be resolved in no more than 2 years after the rating period in which the deficiency was discovered.” This discrepancy resulted in ambiguity that confused some stakeholders as to when the corrective action plan must be completed and when a state's base data must be in compliance. To remove this ambiguity, we proposed to replace the word “from” at § 438.5(c)(3)(ii) with the phrase “after the last day of.” The preamble of the 2016 final rule used the term “discovered”, while the regulatory text used the term “identified.” We proposed to retain the term “identified” in the regulatory text since we believed this term to be more appropriate in this context. We explained that our proposed change would clarify the corrective action plan timeline for states to achieve compliance with the base data standard; that is, states would have the rating year for which the corrective action period request was made, plus 2 years following that rating year to develop rates using the required base data. For example, if the state's rate development for calendar year (CY) 2018 did not comply with the base data requirements, the state would have 2 calendar years after the last day of the 2018 rating period to come into compliance. This means that the state's rate development for CY 2021 would need to use base data that is compliant with § 438.5(c)(2).

The following summarizes the public comments received on our proposal to revise § 438.5(c)(3)(ii) and our responses to those comments.

Comment:

A few commenters supported the proposed language change. One of these commenters supported the proposal to use the term “identified” in § 438.5(c)(3)(ii) instead of the word “discover,” which was used in the preamble of the 2016 final rule to describe the regulation. One of these commenters also urged CMS to ensure that the base data used by a state submitting a corrective action be improved to meet the standards in § 438.5 and recommended that CMS enforce these requirements. One of these commenters also requested that the base data be required to include all available and emerging experience, such as pharmacy utilization experience.

Response:

We agree with commenters that the term “identified” in the regulatory text is appropriate, and therefore, we used it in the proposed rule and this final rule. We also agree with commenters that states and actuaries should be utilizing base data that is compliant with the standards and requirements set forth in the 2016 final rule, and we assure commenters that CMS is enforcing those rules. While we also agree with commenters that our base data standards should include the use of appropriate available and emerging experience, we did not propose any changes to the standards governing base data and are not finalizing any changes to those standards in § 438.5(c)(1) and (2). We remind commenters that the general rule for base data at § 438.5(c)(2) already requires states and their actuaries to use the most appropriate data, with the basis of the data being no older than from the 3 most recent and complete years prior to the rating period, for setting capitation rates. Such base data must be derived from the Medicaid population, or, if data on the Medicaid population is not available, derived from a similar population and adjusted to make the utilization and price data comparable to data from the Medicaid population. Data must also be in accordance with actuarial standards for data quality.

After consideration of the public comments and for the reasons articulated in the proposed rule and our responses to comments, we are finalizing the amendment to § 438.5(c)(3)(ii) as proposed.

4. Special Contract Provisions Related to Payment (§ 438.6)

a. Risk-Sharing Mechanism Basic Requirements (§ 438.6(b))

In the “Medicaid and Children's Health Insurance Program (CHIP) Programs; Medicaid Managed Care, CHIP Delivered in Managed Care, Medicaid and CHIP Comprehensive Quality Strategies, and Revisions Related to Third Party Liability” proposed rule (the 2015 proposed rule) (80 FR 31098, June 1, 2015), we proposed to redesignate the basic requirements for risk contracts previously in § 438.6(c)(2) as § 438.6(b). In § 438.6(b)(1), we proposed a non-exhaustive list of risk-sharing mechanisms (for example, reinsurance, risk corridors, and stop-loss limits) and required that all such mechanisms be specified in the contract. In the preamble, we stated our intent to interpret and apply § 438.6(b)(1) to any mechanism or arrangement that has the effect of sharing risk between the MCO, PIHP, or PAHP, and the state (80 FR 31122). We did not receive comments on paragraph (b)(1) and finalized the paragraph as proposed in the 2016 final rule (81 FR 27578) with one modification.

In the 2016 final rule, we included the standard from the then-current rule (adopted in 2002 in the “Medicaid Program; Medicaid Managed Care: New Provisions” final rule (67 FR 40989, June 14, 2002) (hereinafter referred to as the “2002 final rule”)) that risk-sharing mechanisms must be computed on an actuarially sound basis. The 2015 proposed rule inadvertently omitted the requirement that risk-sharing mechanisms be computed on an actuarially sound basis but we finalized § 438.6(b)(1) with that standard included in the 2016 final rule (81 FR 27578). As managed care contracts are risk-based contracts, mechanisms that share or distribute risk between the state and the managed care plan are inherently part of the capitation rates paid to plans for bearing the risk. Therefore, the risk-sharing mechanisms should be developed in conjunction with the capitation rates and using the same actuarially sound principles and practices.

We explained in the 2018 proposed rule how we expect states to identify and apply risk-sharing requirements prior to the start of the rating period because they are intended to address the uncertainty inherent in setting capitation rates prospectively. Because we believed that the 2016 final rule was clear on the prospective nature of risk-sharing and our expectations around the use of risk-sharing mechanisms, we did not specifically prohibit retroactive adoption and use of risk-sharing mechanisms. However, since publication of the 2016 final rule, we have found that some states have applied new or modified risk-sharing mechanisms retrospectively; for example, some states have sought approval to change rates, or revise a medical loss ratio (MLR) requirement, after the claims experience for a rating period became known to the state and the managed care plan. As noted in the 2018 proposed rule, we acknowledge the challenges in setting prospective capitation rates and encourage the use of appropriate risk-sharing mechanisms; in selecting and designing risk-sharing mechanisms, states and their actuaries are required to only use permissible strategies, use appropriate utilization and price data, and establish reasonable risk-sharing assumptions.

We also acknowledged in the 2018 proposed rule how, despite a state's best efforts to set accurate and appropriate capitation rates, unexpected events can occur during a rating period that necessitate a retroactive adjustment to the previously paid rates. We explained that when this occurs, states should comply with § 438.7(c)(2), which provides the requirements for making a retroactive rate adjustment. Section 438.7(c)(2) clarifies that the retroactive adjustment must be supported by an appropriate rationale and that sufficient data, assumptions, and methodologies used in the development of the adjustment must be described in sufficient detail and submitted in a new rate certification along with the contract amendment.

To address the practice of adopting or amending risk-sharing mechanisms retroactively, we proposed to amend § 438.6(b)(1) to require that risk-sharing mechanisms be documented in the contract and rate certification documents prior to the start of the rating period. We also proposed to amend the regulation at § 438.6(b)(1) to explicitly prohibit retroactively adding or modifying risk-sharing mechanisms described in the contract or rate certification documents after the start of the rating period.

In the proposed rule, we acknowledged that our proposed requirement that risk-sharing mechanisms be documented in a state's contract and rate certification documents prior to the start of the rating period meant, as a practical matter, that states electing to use risk-sharing mechanisms would have to submit contracts and rate certifications to us prior to the start of the rating period. We noted that section 1903(m)(2)(A)(iii) of the Act, as well as implementing regulations at § 438.806, require that the Secretary must provide prior approval for MCO contracts that meet certain value thresholds before states can claim FFP. This longstanding requirement is implemented in the regulation at § 438.806(c), which provides that FFP is not available for an MCO contract that does not have prior approval from us. We have, since the early 1990s, interpreted and applied this requirement by not awarding FFP until the contract has been approved and permitting FFP back to the initial date of a contract approved after the start of the rating period if an approvable contract were in place between the state and the managed care plan. This practice is reflected in the State Medicaid Manual, section 2087.

The following summarizes the public comments received on our proposal to amend § 438.6(b)(1) and our responses to those comments.

Comment:

Several commenters supported the proposed amendment that risk-sharing mechanisms be documented in a state's contract and rate certification documents prior to the start of the rating period. Commenters noted that doing so would improve transparency and facilitate CMS' oversight of these risk-sharing mechanisms. One commenter noted the proposed amendment to § 438.6(b)(1) would promote a more reliable and predictable method for risk-adjusting payments to managed care plans. Commenters also stated that risk-sharing mechanisms should be documented in the contract prior to the start of the rating period to provide certainty to both states and their contracted managed care plans.

Response:

We appreciate commenters' support and agree that risk-sharing mechanisms should be documented in a state's contract(s) and rate certification(s) prior to the start of the rating period for all of the reasons commenters provided. As risk-sharing mechanisms are intended to address the uncertainty inherent in setting capitation rates prospectively, we believe that states should develop risk-sharing requirements prior to the start of the rating period and that risk-sharing mechanisms should be developed in accordance with actuarially sound principles and practices.

Comment:

Several commenters stated that retroactive adjustments should not be limited to adjustments to rates but should also apply to risk-sharing mechanisms. These commenters stated that states transitioning new populations or services into managed

care programs, such as LTSS, are more likely to need retroactive adjustments to payment structures due to the unknown risks in covering new populations in managed care for the first time.

Response:

We disagree with commenters on permitting retroactive adjustments to risk-sharing mechanisms. We do not believe that it is appropriate to modify risk-sharing mechanisms between states and plans after the claims experience for a rating period is known, because such retroactive changes undercut the need for states and plans to address uncertainty prospectively. We are not foreclosing retroactive adjustments to rates when appropriate. As provided by § 438.7(c)(2), if the state determines that a retroactive adjustment to the capitation rate is necessary, the retroactive adjustment must be supported by a rationale for the adjustment and the data, assumptions, and methodologies used to develop the magnitude of the adjustment must be adequately described with enough detail to allow CMS or an actuary to determine the reasonableness of the adjustment. These retroactive adjustments must be certified by an actuary in a revised rate certification and submitted as a contract amendment to be approved by CMS. These types of changes are distinct from application of a previously set risk-sharing mechanism that is retrospective. While CMS will not permit a retroactive change to the risk-sharing mechanism under this final rule, the state can pursue a retroactive change to the capitation rates if the requirements under § 438.7(c)(2) are satisfied.

Comment:

Commenters requested that CMS clarify the difference between risk adjustment and risk mitigation. One commenter requested that CMS create definitions for risk adjustment and risk mitigation. Commenters also requested that CMS clarify that the proposed change in this section does not apply to risk adjustments as permitted in § 438.7(b)(5). Another commenter noted that the new language explicitly prohibiting retroactively adding or modifying risk-sharing mechanisms may be seen as not allowing retroactive rate adjustments and requested that CMS add language to this section that clearly states retroactive rate adjustments under § 438.7(c)(2) are still permitted.

Response:

First, we clarify here that risk-sharing mechanisms, which can include a risk mitigation strategy, are a distinct and separate concept from risk adjustment. We note that “risk mitigation” is not a phrase used in part 438. Risk adjustment is defined at § 438.5(a) as a methodology to account for the health status of enrollees via relative risk factors when predicting or explaining costs of services covered under the contract for defined populations or for evaluating retrospectively the experience of MCOs, PIHPs, or PAHPs contracted with the state. The requirements regarding risk adjustment are found at §§ 438.5(g) and 438.7(b)(5). Risk-sharing mechanisms, on the other hand, are any means, mechanism, or arrangement that has the effect of sharing risk between the MCO, PIHP, or PAHP, and the state. A risk mitigation strategy is a means to protect the state, or the managed care plan, against the risk that assumptions (not only based on health status of enrollees) underlying the rate development will not match later actual experience. In other words, “risk-sharing” is about the aggregate actual experience, while “risk adjustment” is about paying based on the health status of enrollees at the individual level and how health status is assumed to result in higher costs.

Second, we explain how the regulations that address these concepts interact or do not interact. We confirm here that § 438.6(b)(1), including the proposed change that we are finalizing here, does not regulate and has no impact on risk adjustment as addressed in §§ 438.5(g) and 438.7(b)(5). We also confirm that our proposed change to § 438.6(b)(1) does not impact states' ability to revise or adjust capitation rates retroactively under § 438.7(c)(2) when unexpected events or programmatic changes occur during a rating period that necessitate a retroactive change or adjustment to the previously paid rates. Section 438.7(c)(2) clarifies that the retroactive adjustment (or change) to capitation rates must be supported by an appropriate rationale and that sufficient data, assumptions, and methodologies used in the development of the adjustment must be described in sufficient detail and submitted in a new rate certification along with the contract amendment. Changes to a risk-sharing mechanism are not changes to the capitation rates themselves; they are changes to an arrangement or mechanism that results in a separate payment from a state to a managed care plan or a remittance to a state from a managed care plan.

Section 438.6(b)(1) applies to any and all mechanisms or arrangements that have the effect of sharing risk between the MCO, PIHP, or PAHP, and the state on an aggregate level. We believe that this concept includes risk mitigation strategies and other arrangements that protect the state or the MCO, PIHP, or PAHP against the risk that the assumptions used in the initial development of capitation rates are different from actual experience. Common risk mitigation strategies include a medical loss ratio (MLR) with a remittance, a risk corridor, or a risk‐based reconciliation payment. Under § 438.6(b)(1), we included a non-exhaustive list of risk-sharing mechanisms, such as reinsurance, risk corridors, or stop-loss limits. We also defined risk corridor in § 438.6(a) as a risk-sharing mechanism in which states and MCOs, PIHPs, or PAHPs may share in profits and losses under the contract outside of a predetermined threshold amount. Because the regulations in part 438 do not use the term “risk mitigation strategy,” we do not believe it is necessary to define the term or add it to the regulations. Section 438.6(b)(1) is clear that all risk-sharing mechanisms are subject to its scope.

Comment:

One commenter requested CMS add risk pools to the list of risk-sharing arrangements in § 438.6(b)(1) to clarify that such arrangements are subject to actuarial soundness requirements and must be documented in the managed care contract prospectively.

Response:

If a risk pool is used as a mechanism to share risk between the MCO, PIHP, or PAHP, and the state, then we agree with commenters that a risk pool is subject to the requirements in § 438.6(b)(1). We reiterate that any mechanism, strategy, or arrangement that protects the state or the MCO, PIHP, or PAHP against the risk that the assumptions used in the initial development of capitation rates are different from actual experience is subject to the requirements in § 438.6(b)(1). We decline to add a specific mention of “risk pools” into the regulations because we believe that § 438.6(b)(1) adequately indicates that it applies to all risk-sharing mechanisms and only lists certain mechanisms as examples.

Comment:

One commenter requested clarification from CMS regarding whether restrictions on profits are to be considered a risk-sharing mechanism, including minimum MLR requirements and contractual profit caps. Another commenter requested that the proposed risk-sharing mechanism be open to modifications while CMS is reviewing the rates, so that if CMS does not accept the initially proposed risk-sharing mechanism, then the state can modify and propose to CMS an alternative, acceptable strategy.

Response:

We confirm that a minimum MLR requirement with a remittance would be considered a risk-sharing mechanism and subject to the requirements in § 438.6(b)(1). We also

confirm that additional restrictions on profits or contractual profit caps would also be considered risk-sharing mechanisms under this regulation. To the degree that arrangements (like the examples provided by the commenter or other arrangements) function to explicitly share risk between states and managed care plans, such arrangements would be risk-sharing mechanisms and subject to the requirements in § 438.6(b)(1). Regarding possible modifications to a risk-sharing mechanism while CMS is reviewing the rates, we confirm for commenters that such modifications would only be possible prior to the start of the rating period to comply with the final regulation text. The requirements in § 438.6(b)(1) to document the risk-sharing mechanism in the contract and rate certification documents prior to the start of the rating period, as well the prohibition on adding or modifying risk-sharing mechanisms after the start of the rating period, would apply to states, plans, and CMS. If states are seeking CMS review and approval prior to the start of the rating period, CMS and states can work toward modifications that would ensure that arrangements are reasonable, appropriate, and compliant with Federal requirements, as long as such modifications are in place and documented in the contract and rate certific

This text is long and has been trimmed here. Open the source document for the complete record.