Acute Myocardial Infarctions (AMIs) have emerged as a high-risk category for overreported HCC values in recent OIG audits. These findings confirm what many compliance teams have suspected: errors in this area are not isolated to a single coding mistake but stem from a broader misunderstanding of ICD-10-CM guidelines specific to AMIs.
Let’s explore some of the most common issues identified in the audits and how to prevent them.
1. Coding “Old” MIs as Acute Events
One frequent error was the reporting of an acute myocardial infarction (AMI) when the condition was actually resolved or classified as an “old MI.” According to ICD-10-CM Official Guidelines for Coding and Reporting, FY 2025, Section I.C.9.e.1:
“For encounters occurring while the myocardial infarction is equal to, or less than, four weeks old… codes from category I21 may continue to be reported… For encounters after the 4-week time frame… the appropriate aftercare code should be assigned… For old or healed myocardial infarctions not requiring further care, code I25.2, Old myocardial infarction, may be assigned.”
This distinction is often missed when providers select diagnoses from EMR search tools. These tools rarely display the full guideline context, and many include misleading or outdated code descriptions. Providers search by keyword, then pick the closest match—often without realizing that code I21 is no longer valid once the 4-week window has passed.
2. Reporting an AMI Without Active Treatment
Another error involved the documentation of an acute MI without any supporting treatment plan—particularly when a subsequent MI was also reported. Per Section I.C.9.e.4 of the coding guidelines:
“A code from category I22, Subsequent ST elevation and non-ST elevation myocardial infarction, is to be used when a patient who has suffered a type 1 or unspecified AMI has a new MI within the 4-week time frame of the initial AMI.”
An AMI—especially a second one—should come with clear evidence of an active treatment plan. However, audit findings frequently revealed documentation that lacked any such plan, raising red flags about the accuracy and appropriateness of the codes used.
3. Documented Angina Misreported as AMI
Auditors also found cases where documentation supported a diagnosis of angina, not an AMI, yet an AMI code was billed. This resulted in inflated payments unsupported by clinical documentation. While the OIG did not specify the circumstances leading to this error, one possibility is that an AMI was initially suspected but ruled out in later testing. In such cases, the correct code might be a form of angina or chest pain, not I21.
4. Lack of Hospitalization or Active Management
Some cases involved reporting an acute MI without any hospitalization within the prior four weeks or any documentation of ongoing management. This again points to a breakdown in documentation and coding processes—especially in outpatient settings where historic diagnoses may be re-reported without verifying whether they are still active or relevant.
What’s Driving These Errors?
These issues aren’t about carelessness; they’re about workflow and system design. Most EMRs don’t give providers access to the full coding guidance they need to make informed decisions. Providers are making the best selections they can based on EMR pick-lists, which are often filled with vague or inconsistent terms.
A Better Approach: Team-Based Workflow Design
To reduce RADV exposure and improve accuracy in high-risk HCC categories like AMIs:
- Establish coder review queues for any encounter where AMI, angina, or chest pain is assigned.
- Empower coders to verify documentation against official guidelines and to query providers when inconsistencies arise.
- Provide education and audit feedback to both coding and clinical teams, especially in high-error categories.
- Leverage risk adjustment expertise to refine EMR templates and search logic to support compliance.
Need help designing coding workflows that prevent audit risk and ensure accurate HCC capture? Contact us at info@bcarev.com to get the conversation started.