On October 26, 2018, the Centers for Medicare and Medicaid Services (“CMS”) released for viewing a proposed rule that includes significant changes for Medicare Advantage organizations (“MAOs”), Part D prescription drug plan sponsors (“PDPs”), Medicaid managed care organizations (“MCOs”), and the providers and suppliers that contract with them (the “Proposed Rule”). CMS anticipates that the Proposed Rule will be formally published in the Federal Register on November 1, 2018.

Many of the Proposed Rule’s changes would implement provisions of the Bipartisan Budget Act of 2018 (the “BBA”). Another set of proposed changes relate to program integrity initiatives including the preclusion lists that CMS established in the CY2019 MA and PDP final rule, as well as MA risk adjustment data validation (“RADV”) audits. Changes to improve program quality and accessibility are also proposed.

This is the first in a series of Sheppard Mullin Healthcare Law Blog posts on the Proposed Rule. The focus of this post is proposed changes relating to RADV audits.

Medicare Advantage Risk Adjustment Data Validation Provisions

RADV is a MAO audit program that is designed by CMS to verify that diagnosis codes submitted for payment by an MAO are supported by medical record documentation for an enrollee. RADV audits for payment year 2007 recouped $13.7 million based on enrollee-level adjustments. No RADV audits were conducted of payment years 2008, 2009 or 2010, as CMS was considering a RADV audit methodology that would provide for contract-level payment error audit findings based on extrapolation.

CMS issued the final RADV audit methodology in February 2012, which provides for sampling techniques and the statistical calculation to be used for extrapolation. The methodology also provides for a fee-for-service adjuster (“FFS Adjuster”) as an offset before finalizing any audit recovery. The FFS Adjuster was to account for any effect of erroneous diagnosis codes in the data from Medicare FFS that are used to calibrate the MA risk adjustment model. The FFS Adjuster would be calculated based on a RADV-like audit of Medicare FFS records. CMS has completed that review and, based on its findings, CMS is proposing not to include a FFS Adjuster in any final RADV payment error methodology. See subsection 5 infra. This is an unwelcomed surprise to MAOs, particularly those MAOs that underwent RADV audits for payment years 2011, 2012 and/or 2013 using the final methodology announced in 2012 and have been waiting for CMS’ determination of the FFS Adjuster so those audits could be finalized.

CMS proposed the following RADV audit-related changes:

​1. Extrapolation

CMS intends to recover overpayments based on extrapolated audit findings through the use of statistically valid random sampling techniques. CMS uses the Proposed Rule to provide additional notice and seek public input on the agency’s methodology for calculating a contract-level payment error in RADV audits, including the sample sizes used in these contract-level audits. CMS notes that it is not required to set forth the methodology for calculating an extrapolated payment error through regulatory provisions, but, in the interest of transparency, is “updating stakeholders on our plans to use various sampling and extrapolation methodologies in RADV audits, as CMS deems appropriate.”

CMS is considering other potential methodologies for sampling and extrapolation, which would calculate improper payments made on the audited MA contract for a particular sub-cohort or sub-cohorts. A sub-cohort could be the enrollees for whom a particular hierarchical condition category (“HCC”) or one of a related set of HCCs (such as the three diabetes HCCs) was reported. After choosing an MA contract and a sub-cohort or sub-cohorts to audit, CMS would select a statistically significant sample of enrollees for the sub-cohort or sub-cohorts. And, after reviewing the medical records of those enrollees, CMS would use statistical extrapolation to calculate and recoup the improper payments made to the audited MA contract for covering enrollees for the sub-cohort or sub-cohorts in that payment year. CMS believes that the sub-cohort-based audit methodology would allow the agency to focus on cohorts of enrollees that appear to raise programmatic concerns.

In a footnote, CMS indicates that it may begin to conduct RADV audits for payment years 2014 and 2015 before the Proposed Rule is finalized, “pursuant to [CMS’] longstanding authority to review the medical records of any MA enrollee and recoup any improper payments identified.” These audits would be designed so that that the individuals selected would form a statistically significant sample that would support an extrapolated recovery, but CMS would not seek to recover on an extrapolated basis until the rule is final. “At the very least, these audits would support enrollee-level recoveries.”

​2. Application to Payment Year 2011 and Subsequent Years

The Proposed Rule would apply the finalized RADV payment error methodology or methodologies to payment year 2011 and all subsequent years. CMS does not “expect” to use a sub-cohort based methodology, if finalized, for any payment year before 2014. CMS notes that Section 1871(e)(1)(A) of Social Security Act (42 U.S.C. § 1395hh(e)(1)(A)) authorizes retroactive application of rules where “(i) such retroactive application is necessary to comply with statutory requirements; or (ii) failure to apply the change would be contrary to the public interest.” CMS is considering whether application of the finalized methodology or methodologies to payment year 2011 and all subsequent years, would require the exercise of this statutory authority to engage in retroactive rulemaking.

CMS points out that the February 2012 methodology put MAOs on notice that CMS expected to calculate a contract-level payment error for payment year 2011 and beyond by extrapolating from its review of a statistically valid sample of enrollees, and that MAOs have never been entitled to receive or retain payments associated with HCCs that cannot be validated by medical records. “Application of the finalized RADV payment error methodology or methodologies to payment year 2011 and all subsequent years therefore would not upset any settled interest.”

​3. Implementation

CMS would recover improper payments based on an extrapolation of the payment error from RADV audit samples to the MAO’s specified populations. MAOs would be required to remit extrapolated recovery amounts through CMS’ payment system, the Medicare Advantage and Prescription Drug system or “MARx”. In the event the recovery amount exceeds an MAO’s payment from CMS in one month, the recovery would be spread across multiple months until the full amount is recovered. CMS may likewise require MAOs to remit such recovery amounts based upon audit findings by the Department of Health and Human Services Office of the Inspector General.

​4. Recoupment of Improper Payments in Part C

The Proposed Rule would amend the risk adjustment regulation at 42 C.F.R. § 422.310(e), to require MAOs to remit funds that CMS calculates as improper payments through the extrapolated RADV audit methodology in the manner specified by CMS. RADV audit results can be appealed by MAOs in accordance with § 422.311. CMS is considering changes to the RADV appeal regulations to expand MAOs’ appeal rights, including the right to appeal the RADV payment error methodology.

​5. FFS Adjuster

As a follow-up to the February 2012 RADV audit methodology, CMS conducted a study on the presence and impact of diagnosis error in Medicare FFS claims data. According to CMS, the study suggests that errors in FFS claims data do not have any systematic effect on the risk scores calculated by the CMS-HCC risk adjustment model, and therefore do not have any systematic effect on the payments made to MAOs. An executive summary of CMS’ study is available on the CMS website.

Interestingly, CMS indicates that, even if it had found that the diagnosis errors in FFS claims data led to systematic payment errors in the MA program, CMS no longer believes that a RADV-specific payment adjustment would be appropriate. CMS notes that RADV audits are used to recover payments based on diagnoses that are not supported by medical record documentation and should not have been reported to CMS. In such cases, the entire payment is in error and should be recovered in full because the payment standard has not been met. Since RADV audits do not address issues with the accuracy of payments based on diagnosis codes that are supported by medical record documentation, CMS believes that an adjustment to RADV recoveries to remedy payment accuracy concerns is inappropriate. CMS believes that it would not be appropriate to correct any systematic payment error in the MA program through a payment adjustment that was only applied to audited contracts. “Doing so would introduce inequities between audited and unaudited plans, by only correcting the payments made to audited plans.”

CMS acknowledged the recent district court decision that vacated CMS’ Part C and Part D overpayment rule. That decision was the topic of a blog post and webinar by Sheppard Mullin attorneys. CMS is reviewing the decision and “considering its response”, but CMS noted that the decision was made on the basis of the administrative record before the court, which did not include the results of CMS’ study. This may portend CMS proposing a modified version of the overpayment rule, which continues to define any overpayment as an diagnosis code that is not supported by medical record documentation, but defines when an overpayment has been “identified” consistent with the underlying statute.