June 15, 2015

Coder Auditing in ICD-10: Five Nuances for Managers to Recognize Now

By Paul Strafer, RHIA, CCS

It’s important for coding managers to assess coding accuracy now — while there’s still time to remediate and improve. Defaulting to an unspecified code simply won’t be an option under ICD-10, as many insurers and auditors will be looking for codes that capture the greater specificity inherent in the new coding set. Although ICD-10 won’t go live until Oct. 1, there are definitely ways in which managers can simulate a real production environment to assess coding compliance sooner rather than later.

Last month I addressed why ongoing coder assessments are important heading into ICD-10. The June 2015 article identified three hurdles experienced by managers and suggested practical coding audit strategies to implement now, in advance of the implementation date: 

  • Use real medical records versus hypothetical scenarios.
  • Deliver immediate coder feedback.
  • Provide dynamic and targeted education.

This article, the second in a two-part series, addresses the nuances of measuring and auditing coders in the months ahead.

Quality Will Trump Productivity in the Early Days of ICD-10

Quality assurance will be the name of the game in ICD-10. Although coders will have to continue to meet strict productivity standards, they must place equal value on coding quality. According to most recent CMS ICD-10 end-to-end testing results, only 2 percent of test claims were “rejected due to invalid submission of ICD-10 diagnosis or procedure code(s).”

Despite these positive results, many finance experts predict an avalanche of claim denials during the months immediately following ICD-10 implementation. At HFMA’s Dixie Chapter meeting in February 2015, speakers suggested that over half of claims may not pass claims scrubbers, and the organization highly recommended that providers build cross-functional denials management committees now, including for health information management (HIM). More revenue cycle predictions for ICD-10 are listed online as presented during a recent HFMA webinar.

To mitigate this risk, there are five specific steps HIM professionals can take now. The list below includes important dos and don’ts of measuring coder accuracy and coding quality in ICD-10. 

Five Dos and Don’ts of ICD-10 Code Auditing

Consider the following to simulate as close to a real coding environment in ICD-10 as possible and evaluate coders accordingly:

1.  Type of documentation used for assessments

Don’t: Require coders to practice using hypothetical scenarios. Unlike real documentation, these scenarios tend to be contrived, even when they are well-written.

Do: Use your facility’s own medical records. As mentioned in June, the use of actual documentation and cases is critical. This is the only way to paint an accurate picture of coding quality. Where do documentation gaps occur? What cases take the longest to code in ICD-10? What conditions and procedures do coders query most frequently

2.  DRG shift data to track and evaluate

Don’t: Assume DRG shifts are due to ICD-10 coding errors. DRGs can shift for a variety of reasons, including incorrect assignment of the principal diagnosis or accidental omission or inclusion of a CC or MCC. There also may be uncontrollable shifts due to differences in ICD-10 guidelines compared to ICD-9.

Do: Track the reason for DRG shifts. This allows managers to understand whether the coder is struggling with assignment of an ICD-10 code or simply application of a particular coding guideline. For example, did the coder report a CC or MCC that wasn’t clinically validated?

Also track coding accuracy, DRG accuracy, and trends in coding errors by code category, subcategory, and individual code.Then ask: is there a departmental trend that must be addressed?

 3.   New quality standards

Don’t: Use the same standard established for ICD-9. Instead, assume that quality will continue to be a challenge as coders enhance their ICD-10 knowledge through live production. The ICD-10 National Pilot Program conducted by HIMSS and WEDI, for example, found that coders achieved an accuracy rate of only 63 percent. Productivity standards also must be reassessed and reset. The pilot program found that productivity decreased by 50 percent.

Do: Set an ICD-10 quality standard to which coders are held accountable while practicing. The new accuracy benchmarks will get coders accustomed to coding under the pressures of a daily workload. An 89-percent quality rate is not unreasonable, particularly when coders use a dynamic assessment tool that automatically scores test cases and provides remedial information.

 4.   Reliance on the encoder

Don’t: Value quantity and speed over quality. Relying entirely on the encoder for ease of coding may prove detrimental in the long run.

Do: Encourage coders to double-check the codes suggested by the encoder. Coders should reference the ICD-10 coding book to strengthen their knowledge, develop new skills, and ensure that the encoder software has been updated correctly. An extra minute or two per case is well worth the investment in terms of mitigating denials and avoiding future audits.

5.   ICD-10 templates and queries

Don’t: Wait until Oct. 1 to update electronic health record (EHR) documentation templates and physician queries. 

Do: Go live with ICD-10 templates and queries now. This will get coders and physicians accustomed to using them, and as a by-product, it will improve the documentation with which coders are practicing. As documentation becomes more granular and specific, coding accuracy and productivity will likely improve as well. 

Conclusion

Coder assessments are a complicated yet necessary part of ICD-10 readiness. Managers can make their lives a whole lot easier by automating this process using a dynamic assessment tool. This type of tool can analyze results and provide feedback to coders immediately. Performing assessments manually could take days. Time is of the essence as we all head toward the ICD-10 deadline. Time is also one of a coding manager’s most valuable commodities. Consider an assessment tool to yield comprehensive and efficient results. 

About the Author

Paul Strafer is the Coding and Education Manager for H.I.M. On Call.

His experience includes ROI, DNFB management, and medical coding. Paul obtained both his RHIT and CCS certifications from the American Health Information Management Association (AHIMA) in 2013 and has recently obtained his RHIA credential. He also earned the 2014 Kathleen A. Frawley Memorial Scholarship through NJHIMA. Prior to joining H.I.M ON CALL, Mr. Strafer worked with several universities, including New York College of Technology, in developing curriculums and implementing full education programs for inpatient and outpatient coding, as well as AHIMA’s CCS certification prep.

Comment on this Article

Related Stories

  • ICD-11 is Coming – Take Time to Adjust
    The new classification is designed as a database and has up to 13 dimensions. The World Health Organization (WHO) will be releasing the 11th Revision to the International Classification of Diseases, or ICD-11, this May. The WHO and many of…
  • “Assumptive” Coding for Heart Disease – A Coder’s Perspective
    Official guidance on ICD-10-CM coding raises questions regarding how to document cardiac care. The first step in choosing the proper ICD-10-CM code is reading the medical documentation to identify the diagnosis the provider has documented and confirmed. If there is…
  • How 2018 Impacts ICD-10-CM Cardiology Diagnosis Coding
    There are now five types of myocardial infarction (MI) code categories, and the author describes these new options and how they will impact reimbursement. For cardiology, the focus of ICD-10 is generally on increased specificity and documenting the downstream effects…