Introduction
On August 13, 2021, a federal appeals court issued a ruling in a case brought by UnitedHealthcare, which challenged how the Centers for Medicare & Medicaid Services (CMS) goes about recovering overpayments made by Medicare Advantage (MA) plans. The ruling has several important implications for the Risk Adjustment Data Validation (RADV) process and MA payments. Foremost, the ruling, at least temporarily, reinstates the “Overpayment Rule,” which requires insurers to send a refund to CMS within 60 days from when they learn of an overpayment. Below, we provide more detail on what this and other impacts from the ruling means to MA plans. First, we describe how much money may be at stake in overpayments, as well as how the Overpayment Rule works. We also previously published a legal review of the recent court case.
The Extent of Overpayments
MA plans contract with CMS to offer Medicare beneficiaries a private plan alternative to the original (i.e., “traditional”) program and are paid a predetermined monthly amount by CMS for each enrolled beneficiary. These payments are risk adjusted to reflect differences in each enrolled beneficiary’s health status and projected spending for Medicare-covered services. As described in a prior blog, the amount that a payment is risk adjusted is largely dependent on the number and type of diagnoses documented on medical and billing records; larger adjustments and, consequently, higher payments, are generally associated with more documented diagnoses.
CMS conducts RADV audits of MA to help ensure the integrity and accuracy of risk adjusted payments. These audits can impact how much money MA plans are paid, as the RADV process can lead to recovery of improper payments from MA plans that submitted beneficiary diagnoses for risk adjustment purposes that were unsupported by medical records. Government oversightagencies have estimated that it improperly pays out billions of dollars annually to MA plans, primarily because of these unsupported diagnoses.
Overpayment Rule
The Overpayment Rule, which was codified in the Affordable Care Act (ACA), requires MA plans to “report and return any overpayment it received no later than 60 days after the date on which it identified it received an overpayment” and defines an overpayment as “identified” when the MA plan “has determined, or should have determined through the exercise of reasonable diligence, that the MA organization has received an overpayment.”
MA plans may rely on providers or consultants to submit diagnosis or claims to CMS for risk adjustment purposes. It is possible that an MA plan may not know that a diagnosis is unsupported when the information is submitted to CMS, but discovers that fact later on. This could happen when plans conduct retrospective chart reviews, not even looking for a particular overpayment issue. However, once the MA plan discovers that a diagnosis code is unsupported, it must “delete” that code. Under the Overpayment Rule, failing to report and return a known overpayment within 60 days of discovery is a violation of the False Claims Act (FCA) and can result in significant financial penalties. The FCA allows citizen “whistleblowers” to bring claims that the government can join. In the past, these whistleblower cases have often included former employees and consultants of MA plans.
Legal Challenges to the Overpayment Rule
UnitedHealthcare (UHC) challenged the overpayment rule in January 2016 using several arguments that we summarize below.
Failure of the Actuarial Equivalence Requirement. Its central challenge was that the rule violated the statutory requirement of actuarial equivalence between payments to MA plans and payments from traditional Medicare. The actuarial equivalence requirement is meant to ensure that CMS pays an MA plan the same, on average, as it would have paid if all of its enrollees were enrolled in traditional Medicare.
According to UHC, the Overpayment Rule violated the actuarial equivalence principle by requiring MA plans to refund payments related to unsupported diagnoses, but not requiring payments under the traditional Medicare program to undergo similar auditing scrutiny. That is, CMS was relying on both supported and unsupported diagnosis codes to calculate payments associated with each diagnosis in traditional Medicare, but only supported codes backed by medical records for payment in the MA program.
UHC further purported that this difference between the programs would result in different payments for similar enrollees in MA and traditional Medicare. Moreover, the payments to MA plans would be systemically lower. This argument was based on the notion that the value of each diagnosis for MA risk adjustment payment purposes is based on payments associated with each diagnosis in the traditional Medicare program. If there are more of each diagnosis in the traditional Medicare program (due to there being both supported and unsupported diagnoses), then the payments would be spread over a higher number of diagnoses leading to a lower payment per diagnosis. Hence, the diagnoses for MA risk adjustment payment purposes would be devalued.
Failure of the Same Methodology Requirement. In a related challenge, UHC argued that CMS was not following the “same methodology” requirement in the Medicare law, which requires CMS to establish risk factors for MA enrollees using the same methodology as would be expected to be applied in making payments under traditional Medicare.
Arbitrary and Capricious Implementation. In another methodological challenge, UHC noted that CMS was being “arbitrary and capricious” in implementation of the Overpayment Rule. More specifically, UHC claimed that CMS departed from its intent to apply a Fee-For-Service (FFS) Adjuster to the results of RADV audits; specifically when calculating the amount of any overpayment refund. The FFS adjuster was essentially an adjustment factor that would be applied to any recovered payments to account for the difference described above in the diagnosis documentation standards between the MA and traditional Medicare programs. However, CMS modified its position in the November 2018 proposed RADV rule that it no longer would apply a FFS Adjuster, citing research that showed that diagnosis coding errors in the traditional Medicare program had no real impact on the payment rates to MA plans.
Standard for Negligence. UHC also challenged the extent of negligence needed to invoke False Claims liability by the Overpayment Rule. In general, the FCA imposes liability for false claims that are intentionally submitted to the government. However, UHC alleged that the Overpayment Rule was being used to invoke FCA liability when an MA plan submitted a payment it “should” have known was false, even if it did not have actual knowledge.
In fact, CMS did adopt the more expansive interpretation of how negligence would be identified in the Overpayment Rule, stating: “A person has identified an overpayment when the person has or should have, through the exercise of reasonable diligence, determined that the person has received an overpayment and quantified the amount of the overpayment.” Importantly, CMS describes “reasonable diligence,” as requiring good faith and timely investigations in response to obtaining credible information of a potential overpayment, as well as proactive compliance activities to potential overpayments. A key notion with this definition is the expectation that MA plans would essentially be actively self-auditing themselves.
Court Rulings
There have been two significant court rulings on the UHC challenges to the Overpayment Rule. The two rulings were largely 180 degrees different from each other.
2018 Ruling. The first was in September 2018, when a federal district court judge essentially agreed with all of the UHC arguments, effectively invalidating the Overpayment Rule.
More specifically, the judge concluded that by basically imposing a 100% accuracy requirement on the data that the MA program uses for risk adjustment, the Overpayment Rule violated the statutory mandate of “actuarial equivalence” between CMS payments for healthcare coverage under traditional Medicare and MA. The judge went as far as to say that the data used for MA risk adjustment is “flawed” because it is “built on unaudited data about traditional, fee-for-service Medicare beneficiaries, which must contain errors.” The judge also found that the Overpayment Rule “fails to recognize a crucial data mismatch” in the data between MA and the traditional Medicare program by not including an adjustment (such as the FFS adjuster), and fails to use the “same methodology” to calculate traditional Medicare payments as it applies to MA payments.
The judge also agreed with UHC that the Rule was “arbitrary and capricious” because CMS abandoned its prior policy to include the FFS Adjuster, which recognized the differences between traditional Medicare and MA data. The judge also found that the interpretation of the negligence standard for failure to report and return overpayments extended beyond CMS’ statutory authority, as well as what was allowed by the FCA.
As a result of the 2018 decision, CMS paused enforcement of the Overpayment Rule. Thus, MA plans were able to at least temporarily operate under the assumption that the FCA did not require them to enact comprehensive monitoring and oversight processes to reduce the risk of any miscoded diagnoses.
2021 Ruling. CMS appealed a number of the decisions made in the 2018 ruling and in August 2021, a federal appeals court ended up reversing most of the 2018 rulings. Consequently, the Overpayment Rule is once again allowed to be enforced.
In regards to the actuarial equivalence requirement, the court decided that it did not apply to the Overpayment Rule. The main reason cited is that the law implementing the actuarial equivalence requirement does not reference either the statutory provision governing refunds of overpayments or the rule implementing that provision. In essence the Court’s opinion is that the actuarial equivalence requirement directs CMS on how to perform its risk adjustment modeling and focuses on the initial payments rather than the refunds that come later. Moreover, there is nothing in the actuarial equivalent requirement to prohibit CMS from recouping known overpayments from specific plans, after the initial payments have been shown to be actuarily equivalent.
The Court also concluded that UHC failed to show that the Overpayment Rule would systematically result in underpayments to MA plans. It conceded that miscoding of diagnoses included in the traditional Medicare data might lead to such a result. However, the Court pointed to evidence that providers often also under report diagnoses, as traditional Medicare providers have less financial incentive to comprehensively code diagnoses since they are largely paid on a procedure basis. The under reporting of diagnosis codes could offset the effect of not removing all of the incorrectly coded diagnoses. The Court also noted that UHC never challenged the values for risk adjustment that CMS assigns to each diagnosis that are used to determine the level of capitated payments to MA plans; which it sees as undermining its bid to challenge the same values that are used to calculate overpayment refunds.
The Court also decided that the “same methodology” requirement was completely unrelated to the process of MA risk adjustment calculations or the determination of the refund of overpayments. Thus, it dismissed the “same methodology” challenge out-of-hand.
CMS did not challenge the 2018 ruling regarding the standard used to define negligence for FCA applicability. Hence, the Overpayment Rule remains limited to cases where insurers have actual knowledge or act with reckless disregard or deliberate ignorance of an overpayment. That is, MA plans cannot be held liable for overpayments that they “should have determined through the exercise of reasonable diligence.”
Impact to RADV and Overpayment Recoveries
While the recent 2021 ruling puts the Overpayment Rule back in play, significant questions remain to be addressed. There will likely be other court rulings as well as the development of additional federal regulations that will factor in to the future of the Overpayment Rule.
One issue that is still open is the application of the FFS Adjuster. As discussed above, CMS proposed to not apply the FFS Adjuster when calculating overpayment refunds. CMS has until November 2021 to finalize the proposal.
Also still outstanding is a CMS proposal to recover overpayments based on auditing a sample of medical records to identify unsupported diagnosis. Using the audit results, CMS would recalculate MA payments for the sample. The audited payment amounts would then be compared to the initial payments to determine if any refund was needed. The findings would then be extrapolated across the entire population of enrollees under the same MA plan contract. In essence, this approach would remove all unsupported codes from an MA organization’s data. In the 2021 ruling, the Court left open the potential for additional thought on these issues when a Medicare Advantage insurer’s data will consist of only supported [diagnosis] codes,” as would theoretically be the case in RADV audits with extrapolated, contract-wide recoveries.
If CMS finalizes this approach of extrapolating recoveries and not applying a FFS Adjuster, any litigation challenging the audit methodology is likely to revisit these empirical disputes in greater depth.
Conclusion
The impact from these decisions will be made clearer as CMS moves forward with future proposed rulemaking and additional court cases are decided. However, for the time being, this ruling strengthens the position of CMS to collect overpayments, which will likely lead to larger recoveries from MA plans during risk adjustment data validation processes and an increased number of reviews. There will also likely be an increase in whistleblower cases, even though the courts raised the standard for when an MA plan can be accused of negligence subject to FCA penalties. MA plans should continue to remain vigilant in monitoring for inaccurate diagnoses, incomplete medical documentation, and other coding inaccuracies.