Health insurers have been doubling down on Medicare Advantage (MA) programs in recent years to capitalize on the fast-growing number of retirement-age seniors. However, increased scrutiny by federal regulators, including allegations of outright fraud, is highlighting the very real challenges and risks payers face in producing clean and accurate patient data. 

CMS has become more aggressive in examining risk adjustment data from MA plans as the program grows in popularity and profitability. In a report examining MA in 2020, the Kaiser Family Foundation found that enrollment in the public-private program has doubled over the past decade to 24.1 million members, or nearly four of every 10 Medicare enrollees. Nearly four out of five of them are in plans that have received four-star ratings or higher and related bonus payments from CMS.

At the same time, a report from the Department of Health and Human Services’ Office of Inspector General (OIG) identified $2.6 billion in risk-adjusted payments to MA plans related to diagnosis codes that don’t correlate with clinical services actually rendered. OIG said the findings raise concerns about data integrity, among other issues. 

As such, there has been a flurry of recent activity related to MA claims data from payers: 

  • CVS Health, in a recent regulatory filing, acknowledged that HHS-OIG is auditing the MA plans sold by its payer subsidiary Aetna, which cover more than 2.9 million enrollees.
  • In July, the Department of Justice joined six whistleblower lawsuits accusing Kaiser Permanente of falsifying diagnosis codes to gain $1 billion in fraudulent reimbursement.
  • Earlier this year, an audit determined Humana had overcharged CMS by nearly $200 million with inaccurate MA data. 
  • An earlier lawsuit by the DOJ against Anthem alleges the insurer received $100 million in overpayments because it failed to correct inaccurate diagnosis codes. 
  • The DOJ has also sued Cigna alleging falsification of member health data to the tune of $1.4 billion in overpayments. 

This puts enormous pressure on payers, who may be required to refund premium payments if an audit finds that risk adjustment payments aren’t matched by their clinical data. CMS pays plans on a per-member basis and then adjusts payments upward for more significant health conditions, as reflected by diagnosis codes and other clinical data. For its audits, CMS uses a small, random sample of just 200 medical records, underlining the importance of having clean and accurate data on MA enrollees.

Unfortunately, while clinical data is readily available from electronic health records, it’s often incomplete, redundant, filled with errors, stored in separate records, or inconsistently coded. Information from providers in EHRs doesn’t always translate due to the wide variety of HL7 formats, making it difficult for data analysts to decode and translate the information that can be used for other purposes like MA claims. 

Payers need a way to identify gaps in diagnosis codes to help improve the accuracy of risk adjustment scores and ensure reimbursement payments are appropriate.

How Verinovum can help

At Verinovum, we take data from various sources, then aggregate and standardize it using our advanced curation and enrichment platform. We also employ our data science algorithms and semantic harmonization tools across coding systems including ICD9, ICD10, LOINC, and RxNORM. Lastly, we use algorithm-based, cross-message inference logic to fill in data gaps including medications, diagnosis, and prescribing providers. 

The result is a much cleaner, more accurate, and near real-time data set for payers. We can help MA payers be more proactive in working with members and providers to schedule appointments, validate and verify diagnoses, and ensure that members get the most timely and effective care.