The Risk of Automated EMR Coding

A frequent topic of conversation is how to improve the claims-coding process. If you are intrigued by the automated E&M coding offered by an EMR, I suggest caution. This feature has high marketing appeal—promising that office visits will be coded to the highest level of reimbursement possible. Unfortunately, however, EMR coding has led to severe financial and legal repercussions for practices, as reported in a recent study in the venerable Medical Economics journal.

An increase in average coding levels raises a red flag with payers, and EMR documentation does not stand up well in the resulting audits. According to the authors, who assisted several practices during Medicare audits, the danger is that EMRs automatically guide physicians to create records that document high levels of care, and the result can be a statistically significant (and noticeable) increase in the percentage of claims with level-4 and level-5 codes. The templated chart notes created by traditional (CCHIT-type) EMRs all tend to look the same and do not contain the information necessary to justify these higher levels of coding.

In a claims audit, typically between 20 and 100 charts per physician are reviewed, and the results are then extrapolated to the entire set of claims for that payer. For the practices discussed in this article, between 20% and 95% of the EMR-generated claims failed the audit, and the physicians were assessed penalties and subject to repayments to Medicare that ranged from $50,000 to $175,000 each.