Implementation of the ICD-10-CM/PCS code sets creates a large degree of financial uncertainty and risk for payers. How reimbursement parameters will shift for each individual payer is only a matter of conjecture until claims start to be processed with ICD-10 codes. However, the delay of final implementation until 2014 gives payers an opportunity to conduct more thorough testing to better understand the impact on their organizations.
Payer risks associated with ICD-10 implementation are varied. To start, it is possible that increased granularity of ICD-10 will result in higher claims payments. Whether claims are paid correctly--either underpaid or overpaid--is another issue that will need to be closely monitored once ICD-10 is implemented. A potential increase in the number of "pending" claims that require manual intervention could place greater strain on staff time. These are all legitimate concerns. Evaluating them now can help payers predict potential disruptions to processes and make decisions to minimize those disruptions.
Up to this point, most payers have been focused on the technical aspects of transitioning to the 5010 electronic transaction standards. Ensuring the ability to receive, send and process ICD-10 codes was a critical first step, but it is now time to move beyond the technical specifics. Content should be tested to evaluate the use of ICD-10 codes to update policies, benefit plans and claims rules to take advantage of the increased granularity of ICD-10.
The best time to start testing the effect of ICD-10 implementation is now. To test the impact of ICD-10 claim content, payers must gather the data necessary to make the right decisions for updates. Luckily, between 5010 and shifts to new reimbursement models, the past few years have provided the industry with great lessons learned around claims testing--helping to identify key strategies payers can use to mitigate financial risks, including:
• Convert actual claims. Simulate the financial, operational and technological impact of ICD-10 implementation by converting actual claims from ICD-9-CM to ICD-10/PCS. Because there is no way to predict the specific codes that will be used by providers, the best way to gauge potential claims is to base your testing on the actual claims you are receiving now.
• Look at all claims. The sheer volume of claims data to analyze can be overwhelming, but it is important not to limit testing to diagnoses or procedures that are currently high-volume or high-risk. Once ICD-10 is fully implemented, the nuances of the more detailed code sets may change the risk levels associated with the different diagnosis and procedure codes.
• Evaluate the effect on different plans. ICD-10 implementation may affect different benefit plans in different ways based on each plan's patient population. Paying attention to these differences enables a payer to more accurately update policies related to specific plans. For example, a private plan may have a younger beneficiary population that results in more maternity claims while a Medicare plan likely will have a higher concentration of claims related to care for chronic conditions.
• Prepare for the detection of potential fraud. With the increased number of diagnosis codes for ICD-10-CM--roughly 68,000 compared to the current 14,000 ICD-9 codes--payers should expect a learning curve for detecting the potentially fraudulent use of codes. However, the ability to detect fraud can be improved with thorough testing prior to implementation. By using tools that run simulations for a wide variety of claims for a number of different plans, payers can collect data that could help spot anomalies when ICD-10 is fully operational.
• Partner with hospitals to obtain real ICD-10 data. Some hospitals that are early adopters have begun dual coding in both the ICD-9 and ICD-10 code sets as they prepare for full implementation. Partnering with these organizations to use real-world, real-time claims on a testing basis gives both organizations a chance to test the impact of ICD-10.
It is a mistake to evaluate only the high-risk or high-volume codes being received today; this strategy runs the risk of making assumptions based on ICD-9 codes that may not be valid under ICD-10. Instead, the closer a payer's ICD-10 testing mimics actual claims submission, the smoother the actual ICD-10 implementation. The problem, of course, is that mimicking real life requires cooperation.
The good news is that because ICD-10 is such a monumental challenge for all payers and providers, there is incentive on both sides to work together. A few larger hospital organizations already are starting to reach out to payers to discuss how ICD-10 will affect their billing processes and contracts. Some large payer organizations have begun to offer ICD-10 readiness testing to some of their providers.
There is no crystal ball to accurately predict the effects of ICD-10 implementation for all payers. However, wise use of the time between now and 2014 to gather data needed to update policies, processes and claims rules can help to reduce the financial risk of the transition.
Brian Levy, M.D., is the senior vice president and chief medical officer for Health Language, Inc. (HLI) and is also a practicing board-certified internist.