Medicare Advantage RADV Rule May Prompt More Legal Action from Payers

Payers have not reacted lightly to the Medicare Advantage Risk Adjustment Data Validation (RADV) final rule. Major payer Humana made the strongest display of disapproval by filing a lawsuit against CMS that seeks to invalidate the regulation.

More payers are expected to follow Humana’s lead. In the meantime, Medicare Advantage organizations should focus on having complete documentation and leveraging artificial intelligence to ensure compliance, Michael Stearns, MD, specialized consulting director of medical informatics and health language at Wolters Kluwer, told HealthPayerIntelligence.

Humana’s lawsuit revolves around CMS’ decision to remove the fee-for-service (FFS) adjuster from the RADV audit process. The payer alleged that the regulation violates the Administrative Procedure Act and claimed that CMS did not reasonably justify the change.

The FFS adjuster established an acceptable margin of payment errors related to unsupported diagnosis codes and limited RADV audit recovery to payment errors above that level. CMS asserted that the adjuster did not impact risk scores or Medicare Advantage payments.

However, Humana maintained that removing the FFS adjuster will impact payments.

“It would be a significant hit on revenue and it would make even a single error here and there start to count against the payer, which is pretty tough,” Stearns said. “Even the RADV process has a margin of error so that part does seem rather strict to me.”

CMS has yet to respond to Humana’s lawsuit, but Stearns expects more payers to take legal action against the final rule.

“I would be surprised if none of the other major payers joined in; given the amount of money, it’s worth putting in the effort and the costs associated with filing the litigation for larger payers,” Stearns indicated.

“Humana has given them a bit of a roadmap, so they can emphasize Humana’s points or agree with them as they go forward. Humana has opened the door. The [payers] on the fence are probably strongly considering joining in.”

Whether or not Humana and other payers will succeed in their lawsuits depends on how CMS responds. Humana has stated that CMS is leading on legal and not factual arguments. CMS may fight the lawsuit based on the legal constructs Humana is relying on, but the agency may also draw from data and analyses to support its argument, according to Stearns.

The litigation could go on for years, depending on how many payers weigh in and what the arguments are, he mentioned. When asked for a comment, a CMS spokesperson said, as a matter of policy, the agency does not comment on pending litigation.

While Stearns acknowledged that the FFS adjuster does not make the most sense, he asserted that change is necessary.

“Some sort of calibration related to how differently doctors document in the fee-for-service model versus the risk adjustment model is needed,” he said.

As Humana’s litigation is ongoing and additional lawsuits are expected, payers still need to master a way to comply with the RADV final rule policies to avoid repayments.

CMS will use extrapolation to estimate error rates starting with diagnoses from 2018. Given this timeline, payers are facing significant pressure to eliminate claims that would be rejected diagnoses due to a lack of supporting documentation, Stearns explained.

However, recent developments in machine learning, artificial intelligence, and natural language processing have made it easier for payers to identify these issues.

“Rather than having an auditor look at something like five records an hour, a machine can look at 500 records an hour and, based on the calibration of the tools, can point out things you want to focus on,” Stearns said.

The Office of Inspector General (OIG) has identified nine high-risk diagnosis groups that have had high error rates over the past two years: acute stroke, acute heart attack, embolism, vascular claudication, major depressive disorder, lung cancer, breast cancer, colon cancer, and prostate cancer.

Some mistakes may be easier to fix than others.

For example, if a patient recently diagnosed with a stroke comes in for a follow-up visit one week after discharge, physicians may incorrectly code this visit as a stroke instead of an outpatient visit. Using a software to identify this mistake can help prevent inaccurate coding, Stearns shared.

In addition to coding errors, a lack of supporting documentation can lead to rejections, suggesting that clinical validation may be useful.

“Clinical validation is the process of determining whether a condition can be reported based on documented clinical information that meets clinical validation criteria,” Stearns wrote in a blog post.

“While clinical validation requires more clinical knowledge than coding, it may be performed by a wider range of professionals, including coding professionals, clinical documentation integrity (CDI) professionals, registered nurses, health information management (HIM) professionals, physicians, and others.”

While Medicare Advantage audits do not seem to be headed toward clinical validation, Stearns said it would make sense to help manage diagnoses.

However, Medicare Advantage plans can leverage automation tools and other software to focus on the diagnoses OIG typically targets in its audits. These tools can help identify inaccurate documentation by focusing on codes or medications that should or should not be included in a chart, given the diagnosis.

 “Payers can say, ‘If the OIG is going after these nine things, we better review the records as best we can.’ With the help of software tools, they might be able to do it effectively and efficiently and identify areas they need to correct,” Stearns concluded. “They may also identify where provider education is needed or additional guidance and policy changes.”