April 17, 2012

Do Not Rely on GEMs to Calculate the Financial Impact of ICD-10

By

As healthcare payers and providers prepare for ICD-10, a key area of focus involves navigating the switch from ICD-9 diagnosis and procedure codes to the new code set. Converting from one code set to the other is challenging because the sets are so dramatically different. There are few cases in which there is a direct mapping of codes between ICD-9 and ICD-10. In most cases there are no one-to-one matches, while in othercases there is no translation at all. ICD-10 contains about five times as many codes, reflecting advances in medical technology and allowing for greater specificity when documenting medical encounters.

To bridge the two code sets, the General Equivalence Mappings (GEMs) were created as a general reference tool. If it is not already clear, payers and providers need to exercise care in considering their strategy and use of the reference data for anticipated benefits. Such use should occur only by involving coding experts in making mapping decisions, and the purpose for doing so should be clearly defined. While GEMs have their purpose, they do not represent a true crosswalk between the code sets. Therefore, any organization intending to use GEMs or some form of them should invoke the knowledge of ICD code experts. Remember, the GEMs should never be used as a substitute for native coding. Payers and providers should be cautious about how they use GEMs mapping and not rely on it for direct translation or to determine how revenue will be impacted. There are too many other variables to consider, such as the impact of clinical documentation improvement programs and modifications to contractual reimbursement terms. Those who have evaluated the details of GEMs find that mapping is not always accurate and assumptions made in developing results are not always appropriate for their environment. Organizations can obtain great value from the work done to create GEMs, but again, they should use it as a starting point only. It can help payers and providers recognize where there are differences between coding sets and offer a shortcut in the process of formulating a strategy that makes sense for their practices. The old adage of “check your assumptions” is extremely valuable in this case, because many assumptions have gone into the creation of GEMs.

Where the Differences Lie

To understand the issues, it’s important to take a step back and review how GEMs were developed. They were created as a temporary measure to enable the industry to transition to new systems, processes and the ICD-10 code set. For the MS-DRG project, the specific purpose was “to replicate the ICD-9 MS-DRGs to minimize payment variances.”

The starting point involved selecting the best ICD-9 code to map to each ICD-10 code based on Medicare data as well as data from the American Hospital Association (AHA) and American Health Information Management Association (AHIMA). Today, while using GEMs works in the CMS environment, with results showing little change in payment, there may be greater variances when testing with other entities. That’s because a different patient population will result in a different outcome. Plus, in using GEMs to evaluate impact to reimbursement in an ICD-10 environment, one needs to assume that everything remains stable and there are no changes to items such as case mix or reimbursement terms.

CMS created the mappings in part to prove revenue neutrality. Considering the focus on improvement in ICD-10, with anticipated changes in the code sets used to report care, the code sets themselves are not the only variables that can change. If everything else were constant, it would work, but there is a great deal of change taking place with the advancement of EHRs, focus on structured data organized via computer assisted coding, and an overall emphasis on revenue enhancement opportunities in the provider community.

One Size Does Not Fit All

Some organizations, as part of their ICD-10 implementation processes, have established centralized mapping repositories or other standards as a way of utilizing GEMs. In other words, they are using GEMs as a starting point for their own applied mappings. Those that created their own mappings most likely based expected results on their own claims and patient populations. Similar to how CMS operated in creating the GEMs, this was done in part to accomplish a specific objective within their own organization. Creating baseline mappings help organizations build a process to meet their needs, whether for a particular application or purpose. Using such a strategy may make sense for members of the payer community as they look at eligibility, benefit plans and payment schedules. The provider community also may find a need for, and a benefit of, implementing a mapping or crosswalk solution. However, such a solution needs to be specific to the activities and events that create the need to use such a tool. For example, if a particular application is not going to be upgraded to ICD-10, then a provider may choose to implement a crosswalk or mapping solution to process data throughout the system. In other cases, a research grant may require certain clinical results or diagnosis trending. So a provider may want to utilize some form of crosswalk or mapping standard to trend or view  data between the two code sets.

GEMs do have value, but one needs to consider carefully any strategies that offer shortcuts in the process of anything other than native coding.

 


 

Mapping a Strategy

As payers and providers transition to the new code set and begin make mapping decisions, they should involve code experts to ensure that results are correct, consistent and understood. Since there is no direct translation, there is no substitution for looking at the actual record and selecting the right code. Failure to do so and relying on automation without a quality check increases compliance risk and can induce audit activity.

Using GEMs for internal data analysis can provide valuable insight to translate the meaning of data from one code set to another. However, to produce the most accurate translation, it is important to understand the purpose of the mappings. Customize GEMs to suit the purpose and apply different maps for different applications or processes, such as payment methodologies and clinical care management. If you need to determine financial impact to your organization, take a deeper dive into factors that impact financial results (such as case mix changes, coder knowledge, coding compliance, computer-assisted coding and documentation and payment terms). Tracking these micro-processes will give you the results and security of financial accuracy desired by both the payer and provider alike. Taking a teamed approach to claims processing and evaluating translation and results with a period of dual coding and testing can help organizations identify any ICD-10 translation issues prior to activation. The right ICD-10 management partner can help connect payers and providers work synergistically to test how claims might be changing and to understand the source of variances. This will help both payers and providers ensure financial neutrality as ICD-10 is implemented.

As payers and providers change over to ICD-10, they must acknowledge the aforementioned fact that the code sets are dramatically different. Organizations should prepare to go through the process of testing using real scenarios to identify variances in revenue or documentation, and understand how such changes will impact claims adjudication. Only then, after recognizing any and all variables, can you have confidence of knowing your financial impact of ICD-10 and whether you have adopted the new code sets successfully.

About the Author

Veronica Hoy, MBA, is vice president of SOURCECORP HealthSERVE Consulting, Inc., one of the largest pure-play business process outsourcing and consulting companies in the markets served.  SOURCEHOV serves more than 50 percent of the FORTUNE 100® companies and has a global workforce.

Contact the Author

To comment on this article please go to

Disclaimer: Every reasonable effort was made to ensure the accuracy of this information at the time it was published. However, due to the nature of industry changes over time we cannot guarantee its validity after the year it was published.