While supplies last! Free 2022 Essentials of Interventional Radiology Coding book with every ICD10monitor webcast order. No code required. Order now >

Greetings from the world of ICD-10 testing. I wanted to share some testing insights and best practices for ICD-10 implementation, based on the methodology behind the two largest ICD-10 testing pilots currently underway: that of the Healthcare Information and Management Systems Society (HIMSS) and Workgroup for Electronic Data Interchange (WEDI), and that of the North Carolina Healthcare Information and Communications Alliance (NCHICA).

These pilots are based on the use of real medical records for testing the end-to-end lifecycle of claims. The critical items to consider in determining the correct starting point for test data generation include usefulness for all industry segments, coding accuracy, billing, contracts, remittance, benefit programs and medical policies.

Test data must be applicable to health information management (HIM) and practice management software (PMS) vendors, the provider, and hospitals, clearinghouses and health plans. One-sided testing artifacts provide very little long-term value and can’t be used for end-to-end testing. Entities must be careful in selecting the right kind of test data. We have seen some vendors using fake scenarios that are not real medical records, but cleverly disguised using the term “clinical scenario” – or, worse yet, they are general equivalency mappings (GEMs) based conversions with no medical record at all behind them to validate proper coding techniques in ICD-10.

Remember that real medical record data originates from the provider and mimics the exact production flow of actual healthcare transactions. Fabricated clinical scenarios are payer-generated and do not represent a complete, real medical record – and they will fall short in extracting true hospital coding patterns across medical specialties. The other troubling aspect of payer-generated scenarios is that payors want providers to code them to fit their own maps and expectations instead of adjusting their crosswalks to reflect what hospitals will actually code, which appears to be a case of putting the proverbial cart before the horse.

I think most people remember when NASA lost a $125 million Mars orbiter a few years ago because one engineering team used metric units while the other team used English units for a key spacecraft operation.

“Our inability to recognize and correct this simple error has had major implications,” NASA’s Jet Propulsion Laboratory (JPL) Director Edward Stone said at the time.

This bears a direct correlation to ICD-10 in that the same thing is happening in ICD-10 testing. Health plans are using GEMs or “clinical scenarios” for testing, yet hospitals and providers are using real medical records. They are not the same thing and they will produce different results. Interoperability only can be achieved when both sides collaborate and use the only testing artifact applicable to both sides – authentic medical records with mutually expected outcomes in both ICD-9 and ICD-10.

Why are we seeing such disparity in approaches? It is due in large part to organizations trying to use technical solutions, tools and maps, to solve a clinical problem. No one has seen real ICD-10 coding across thousands of encounters as of yet, but many make the assumption that a technical tool can accurately replicate human coding behaviors when ICD-10 goes live. Even the smartest and most highly qualified ICD-10 master trainers and coders do not have those answers yet, but most do repeat the same chorus, noting that only the medical record can determine future ICD-10 coding patterns. ICD-10 is a clinical issue requiring clinical solutions, period. As a matter of fact, all healthcare testing should be viewed with a clinical lens and should always have a clinical solution behind it.

Testing that contains traceability and interoperability is a critical component for any successful national testing effort, especially for those involving ICD-10. Medical records are the only traceable artifact throughout the entire end-to-end testing process.

Since providers must dual-code medical records, and considering that those records contain all the information needed to code and create claims for adjudication and remittance, why not use them for provider, clearinghouse and payer testing? In the aforementioned pilots, medical records are aligned and shared among all participants, and expected results can be verified at every step along the way. As such, payers get a more accurate depiction of true ICD-10 coding patterns before we go live, and hospitals and providers get to see real clinical test cases pass adjudication with correct remittance. This is why we created a national test bed repository of dual-coded medical records for industry participants to use, delivering a much more comprehensive solution with long-term reusability. The medical record encounter is a stable, peer-reviewed artifact that can be used for ICD-10, 6020, 7010, Operating Rules, EFTs, HIM and PMS vendor certification, and more.

As an industry we can chose to repeat the same non-collaborative testing processes that were used for 4010, NPI and 5010 and expect a different result – which could be clinically described as a blending of F22 with a little F28 mixed in, unless you mapped it, which would make it an F29!

Stay tuned for my next piece, which will cover issues regarding the 90 percent of hospitals and providers that will be left out of the end-to-end testing process because of outdated linear on-boarding methodologies.

About the Author

Mark Lott is CEO of the Lott QA Group and chairman of the HIMSS-WEDI ICD-10 Testing Workgroup. He also serves as co-chairman of the NCHICA ICD-10 Testing Pilot and is a  member of the ICD-10monitor editorial board.

Contact the Author


To comment on this article please go to editor@icd10monitor.com

State HIM Association (CHIA) Promotes ICD-10 Education


You May Also Like

Leave a Reply

Your Name(Required)
Your Email(Required)