At this point in your ICD-10 preparations, you should have a clear view of your expected coding under ICD-10, and how that coding translates to DRGs. Once you’ve answered these three questions:

1. Is my current documentation sufficient to derive revenue neutral coding?
2. Is my coding sufficient to produce revenue neutral DRGs?
3. Does the DRG assignment of my visits change under ICD-10?

The final step is to translate the results into proposed dollars.  Here’s how:

For Medicare and any contracts you have based on the Medicare fee schedule,  follow the procedure below (per payer):

• First, for the areas where you’ve determined that DRGs will differ, match your ICD-9 and ICD-10 derived DRGs against the most recently available Medicare Fee schedule for your facility.
• Second, use historical data to determine the frequency of the DRGs using ICD-9 coding.
• Third, use the data produced in the previous steps to estimate the new frequency of DRGs.
• Finally, use the most recent fee schedule to calculate the dollar impact.

An example of the analysis for one DRG triad should make this clear:

The same procedure can be used for contracts with a DRG-based proprietary fee schedule.

A similar approach can be taken with procedure code based (and percent of fee) contracts, using the procedure codes created in the previous steps.

Limitations of the Analysis

The financial model we’ve presented in these posts goes a long way toward providing hospitals with a view into their reimbursement in ICD-10. However, the model has a few key limitations that you must be aware of:

First, CMS (and other payers) have stated very clearly that their intent is that the ICD-10 transition be revenue neutral. The greatest challenge to this is that not even CMS has an accurate database of ICD-10 coded claims to create a detailed reimbursement model. The analysis that CMS has relied on in developing the ICD-10 groupers and associated logic is based on a well thought out map of ICD-10 expected claims frequency based on ICD-9 data.  Once provider data is available in the first few quarters after implementation, all payers may have to adjust their reimbursement models. 

Secondly, the model assumes that providers have adequate documentation and coding skills to produce ICD-10 claims.  If documentation is lacking, or if coders cannot accurately extract the relevant codes, reimbursements will suffer.

Finally, please take note that documentation and coding issues affect all patient visits, not just those inpatient visits paid directly by DRG. Outpatient and other procedure-based fee schedules must be analyzed separately, and medical necessity rules based on procedure and diagnosis code combinations will affect reimbursement from all payers.

In the next post,  we will wrap up this series, and discuss how to use the data you’ve developed in this analysis to your advantage.