February 15, 2016

ICD-10 Coding Culprits: Paracentesis and Arterial Lines

By Scot Nemchik, CCS and Sabrina Yousfi, MBA, RHIA, CDIP, CCS, AHIMA-approved ICD-10 CM/PCS Trainer

EDITOR’S NOTE: In this new series, Scot Nemchik and Sabrina Yousfi will identify ICD-10 “coding culprits” and offer practical advice and insights for addressing issues as they arise. 

ICD-10 implementation was a hugely important event, but not catastrophic, as some anticipated. For the most part, we in the healthcare industry have been successful. This is undoubtedly due to the incredible levels of preparation; while we could prepare for the codes themselves, it was very difficult to prepare for the interaction of those codes and the effect on DRG assignment.

For example, a few unexpected DRG behaviors have been observed in our day-to-day coding: culprit codes that no one could have predicted. If not addressed promptly, these anomalies in the code set and its associated DRG system could pose a major threat to coding quality, productivity, hospital statistics, and proper reimbursement. 

This first article in the series targets two issues that have come to our attention, paracentesis and arterial lines, along with suggestions for mitigating problems going forward. 

One character today, thousands of dollars tomorrow 

Firstly, we’re seeing some unpredicted and significant shifts in DRG assignment based on very small differences within procedure codes. These codes are seven characters in length, and each character represents an aspect of the procedure.

With certain procedure codes, if you shift a single character from one value to another, it takes the DRG on a “wild ride” – resulting in a huge payment differential. This speaks to the delicate nature of ICD-10-PCS, wherein thousands of dollars of payment can hinge upon a single character in a single code. Following are two examples based on recent feedback from the field.

Paracentesis is a procedure frequently performed to remove fluid that has accumulated in the abdominal cavity. For example, a patient presenting with ascites may require this procedure to both alleviate abdominal distention and potentially diagnose the underlying cause.

The last character of the code denotes the intent of the procedure – whether it is therapeutic or diagnostic. Unfortunately, providers don’t always clearly state the intent. The truth is, they might not even know that it’s important to state the intent. Ideally, the procedure would be documented clearly as “diagnostic,” “therapeutic,” or using the example provided earlier, “diagnostic and therapeutic.” Absent this detailed documentation, a coder may be tempted to inferintent. 

In the aforementioned example, coding paracentesis as “therapeutic” has no effect on DRG assignment. On the other hand, coding it as “diagnostic” shifts the DRG to a surgical DRG, nearly doubling the relative weight and payment. Assigning codes for both a “therapeutic” and “diagnostic” paracentesis has the same net effect.

Many dollars can ride on one wrong character. As this example shows, a single character can shape-shift an entire PCS code, the corresponding DRG, and revenue received.

Arterial line insertion is another common procedure performed in various critical care settings, often to monitor arterial blood pressure for acutely ill patients. In ICD-9, the coding of this ancillary procedure had no influence on DRG assignment for any conceivable scenario.

In ICD-10, adding the arterial line insertion code can shift an innocuous medical DRG to a high-risk surgical DRG. For example, consider a patient presenting with a stroke who subsequently undergoes an arterial line insertion. Omit the procedure code and you’ll stay firmly planted in a traditional stroke DRG. Add the procedure code and you’ll land in one of three dreaded DRGs describing an “extensive OR procedure unrelated to principal diagnosis.” In ICD-9, these DRGs were usually the result of a bad coding decision. In ICD-10, they appear attributable to bad logic in the coding structure. 

With early coding culprits identified, what can organizations do to make sure they’re documenting and coding properly?

Addressing issues as they arise

Ensuring accurate and complete coding under ICD-10-PCS is a complex proposition at best. Organizations experiencing new coding issues are becoming increasingly concerned about denials, audits, and compromised revenue. Additional coding guidelines are suggested to address issues as they arise and avoid an avalanche of problems in the long term. 

The first step is to make the organization aware of the situation. The organization then can develop internal policies and procedures that clearly direct coders to consistent practices related to the procedures in question. For coding paracentesis, we recommend the following four steps: 

  • Notify clinical documentation improvement (CDI) teams to take a close look at paracentesis documentation to ensure that a therapeutic versus diagnostic designation is clear and physician queries are kept to a minimum. To avoid query fatigue, the forms committee could handle this by simply changing the documentation form to include a checkbox for the physician. 
  • Adjust organizational-specific coding guidelines to accommodate unique issues and idiosyncrasies—particularly those for which unintended code logic (such as that occasionally seen with arterial lines) may be impacting DRG assignment drastically. While remaining compliant with official coding guidelines, coders and organizations alike certainly will benefit from specific coding policies for each anomaly. 
  • Meet with physician department (radiology) heads to discuss challenges in documentation that affect DRG assignment. A little discussion can go a long way, and most are willing to help support the goals of the organization.
  • Contact the American Hospital Association Central Office to get questions answered when there is ambiguity. We have to rely on our professional resources to provide accepted guidance when needed.

Any internal coding guidelines should clarify unusual circumstances, support consistency, and yield accurate reimbursement. Seek clarification and guidance to address the problem through a reliable body of resources, shared stories, revised policies and procedures, and additional training. And ask yourself: how can the coding issue be rectified to eliminate risk?

Throughout 2016, we’ll continue to target common culprits, provide practical examples, and offer recommendations for avoiding negative impacts on revenue, quality reporting, and data integrity. Early detection is important; prevention is paramount.

We also invite you to join the conversation – we’d like to hear your stories, insights, and suggestions. If you’re experiencing common culprits of your own, here are questions to consider, along with tips from the field: 

  • How will the DRG shift if the case is coded incorrectly?
  • What are the implications for proper billing/reimbursement if the case is coded incorrectly?
  • What specific follow-up training could occur? 
  • Should a specific coding guideline be developed? And if so, what would it entail?
  • Given your experience with this issue so far, what do you think is the best solution?

Quick tips from the field

  • Keep in mind that coding “diagnostic” in ICD-10 can potentially change the DRG and affect payment, which was not the case in ICD-9.
  • Consider a documentation checkbox, giving providers the option to choose diagnostic, therapeutic, or both.
  • Remember that a single wrong character in ICD-10 can completely shift a PCS code, the DRG, and revenue received.
  • Make the organization aware of ICD-10 anomalies as they arise, and conduct training to ensure consistency.
  • Submit issues to the Coding Clinic and work with your CDI team to address culprits until the Centers for Medicare & Medicaid Services (CMS) publishes further guidance.

- Patricia Stingl, Inpatient Coding Manager, Froedtert Health

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.

Related Stories