Much has been made about the challenges associated with the ICD-10 transition in October, including expected problems with medical billing coder productivity. But little has been said about another challenge likely to affect the industry: fraud detection.
There are several types of fraud prevention measures currently in place, and some are not affected by the coding system (e.g., anonymous reporting outlets for Medicare fraud). But fraud detection algorithms used by payers are based on machine “learning” related to the code set. A white paper from Jvion discusses the potential impact on the industry.
“ICD-10 will at least temporarily handicap the capabilities of these automated detection algorithms by completely changing the coding language. Most of the patterns that the machines learned over time will be wiped away, leaving open opportunities for costly mistakes,” write the authors. They estimate a period of 18 to 24 months for machines to regain those capabilities, during which unethical organizations may be able to commit fraud without detection.
The issue for providers is that this will almost certainly slow down the claims approval process. “Payors will take longer to process ICD-10 claims as they adapt to new processes based on the increased specificity within the code set. Medical claim edits and rules will undergo a period of adjustment and fluctuation as algorithms and systems adapt. Denials will increase as providers and payors adjust to new coding requirements,” say the report authors.
The paper calls for new solutions that are less dependent on the code set. It remains to be seen how many payers will develop such tools. In the meantime, ICD10Watch advises providers to mitigate their risk of being falsely accused of fraud by:
- Developing a high level of ICD-10 coding skills
- Preparing excellent documentation to support diagnoses and claims
- Hiring an independent auditor to examine recent claims for issues that can be corrected
- Coding claims in ICD-10 as soon as possible to establish norms that can support diagnoses and claims after the changeover
Last Updated on December 28, 2013