Fraud, and especially healthcare fraud, has been a frequent topic in the news lately. If you are active on LinkedIn, I encourage you to follow the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) if you aren’t already doing so.
Last week they posted about a new audit, this time involving the Medicare Advantage (MA) program. The headline read, “CMS Potentially Overpaid Medicare Advantage Organizations $462 Million Based on Certain Unsupported Acute Stroke Diagnosis Codes.”
As a super-broad overview, the Centers for Medicare & Medicaid Services (CMS) uses a prospective risk-adjustment system to determine and adjust payments for enrollees based on health status and demographics. This Hierarchical Condition Category (HCC) model allows CMS to map diagnosis codes based on similarity, severity, and cost.
The plan enrollees’ diagnosis codes for one calendar year (which is the service year) are used to determine the HCC and are used to calculate their risk score for the following calendar year (which is the payment year). The audit report notes that in 2024, CMS paid around 760 MA organizations a total of $494 billion. This amount accounts for 44 percent of all Medicare payments for 2024.
The OIG noted that previous MA audits identified acute stroke diagnosis codes as an overpayment high-risk area. This may occur when physician records assign acute stroke ICD-10-CM codes during the same service year without an acute stroke code assigned on an inpatient or outpatient hospital record. This audit focused on service year 2020, when 773,999 MA organizations submitted an ICD-10-CM acute stroke code. For the purposes of the audit, a random sample of 100 enrollees was used.
After review, it was determined that three of the enrollees were not included in the sample because they were not enrolled in the appropriate identified MA plan. Of the remaining 97 in the sample, all of the acute stroke diagnosis codes reported were not supported by the submitted medical record documentation.
In 68 of the cases, documentation noted that the patient had a history of stroke. History of stroke, code Z86.73 is not an HCC condition, whereas acute stroke is. In 22 of the cases, documentation did not support an acute stroke diagnosis.
One of the cases did not meet CMS requirements for an acceptable data source, and one was not legible. Finally, one case had an acute stroke code assigned, rather than the documented hemiparesis following a stroke. This actually resulted in an underpayment to the MA organization.
Overall, the audit found that there was $187,122 in potential overpayments for sample of 97 enrollees. Based on this, the OIG estimated that CMS made $461,958,186 in potential overpayments.
As coders, we are aware that it is a CMS requirement that documentation-supported codes should be submitted according to the Official Guidelines. The audit report noted that it is the responsibility of MA organizations to ensure that the data they submit is accurate.
The majority of the organizations that submitted samples for this audit indicated that they had some form of procedure or technique to ensure correct coding of acute stroke. Twelve of the organizations noted that they are using the OIG Toolkit focusing on acute stroke diagnosis coding.
I believe this current focus on governmental fraud, especially in healthcare, will continue in the years and administrations to come. Whether you code for a large urban or small rural hospital, a physician practice group or individual practitioner, the focus on complete documentation and accurate coding has never been more important. Make sure your coding team is aware of ongoing OIG activity, and use the audit focus items as opportunities to refresh your staff on appropriate coding guidance.
