ICD-10 Conversion: Coming to your EHR

The way we have all become accustomed to encoding diagnoses in health care will be changing quite dramatically in 2013. Since 1977, healthcare has captured structured diagnosis documentation using the ICD-9 system – in the U.S. the National Center for Health Statistics (NCHS) expanded on the ICD-9 code set (developed by the WHO) to create the ICD-9-CM, which we have become accustomed to using. Both the NCHS and the Center for Medicare and Medicaid Services (CMS) are responsible for overseeing the edits and modifications to ICD-9-CM, which is updated annually.

ICD-9 codes have been used mainly for billing, historically. Every clinical encounter that gets billed to an insurance payor (including CMS) includes diagnosis designations, encoded as ICD-9 codes. Additionally, codes for specific items of service are also submitted in bills – these are encoded as CPT codes, which have Resource-Based Relative Values Units (RBRVUs) associated with each one, and translate into payable amounts for each service line-item.

As physicians move from paper recordkeeping to Electronic Health Records (EHRs), generating and maintaining diagnoses as structured data (ICD-9 codes) also becomes something that clinicians keep on their end. Of course, generating bills from clinical encounters make passing of the coded diagnosis data from the EHR onto bills sent to insurance payors much easier.

Keeping diagnosis lists in structured formats is actually one of the HHS Certification criteria for EHRs that will qualify to allow clinicians’ access to Meaningful Use incentive money beginning in 2011. From this stem all sorts of other capabilities that an EHR can (or, given Certification, is supposed to) offer – reports of patients with specific conditions, clinical quality measures for patients with various diagnoses, public health surveillance reporting, etc.

Changing the fundamental method of encoding diagnoses to a whole new system will have a profound impact. The rationale for making such a change (given the disruption that will occur) is that the ICD-10 code set is more detailed and extensible, allowing for more than 155,000 different codes, and permits the tracking of many new diagnoses and procedures (a significant expansion on the 17,000 codes available in ICD-9). Developed by the WHO and released in 1992, the ICD-10 system was adopted relatively swiftly in most of the world.

CMS has made resources available for the conversion to ICD-10. Most of the focus of its impact has been on billing. The Health Insurance Portability and Accountability Act (HIPAA) – which we generally think of in terms of privacy and security – also defines the standard kinds of files that are used when billing systems submit claims to insurers, and when electronic remittance advice files (the electronic version of EOBs) are sent back to the billers from the insurers. The file version (defined by HIPAA) used to transmit claims information – not just to CMS, but also to every commercial insurer covered by HIPAA – has been Version 4010/4010A1. Part of the conversion to ICD-10 means that a new standard file format needs to be adopted by billing and accounting systems – the new Version 5010 standards for electronic health care transactions (which accommodates ICD-10 codes) need to be adopted by January 1, 2012, and need to be universally in place prior to the 2013 mandate for using ICD-10.

But how does this impact EHRs? Given the much larger code set of ICD-10 compared to ICD-9 (155,000 codes rather than 17,000), there is not a one-to-one conversion. There are mapping tools that create a “best guess” of translating ICD-9 codes to ICD-10, but it is not that straightforward.

From a technology standpoint, incorporating the ICD-10 code set in a pick-list for diagnosis entry (rather than the ICD-9 code set) is fairly easy. In fact, Practice Fusion has both code sets in place already, and “throwing the switch” to ICD-10 is not that difficult.

There is the task of re-figuring the queries when creating lists, reports, clinical quality measures, and the like – but, again, it is simply plugging in the ICD-10 code ranges rather than the ICD-9 ones.

Much more impact is in store for billing systems. Not only do sending (and receiving, on the insurer end) systems need to be able to handle ICD-10 codes, but also the data transmittal standards need to be migrated from the 4010 version to the 5010 version – something behind-the-scenes that adds burden. Billing systems, and especially claims processing systems at the insurer end, are often massive legacy enterprise solutions that will require some major (i.e. expensive) re-tooling.

Learning about ICD-10 is something that will impact more than just coders for billing. Yes, billers will receive (appropriately so) much focus of attention, training, and dialog in order to understand how to code things the “ICD-10 way.” Clinicians will also need to learn about the new coding system, and will need to re-code their diagnoses lists in their EHRs with appropriate new codes. EHR vendors may develop tools to help re-code the diagnoses that are already in place, as well as code new diagnoses as they are created. It will be a bit like learning a new language – or like learning the metric system (after all, the rest of the world uses metric measurements, and also ICD-10).

The deadline for conversion to ICD-10 is October 13, 2013 (it was already postponed once). Conversion to ICD-10 from the Practice Fusion EHR standpoint is straightforward (and already in place for users outside the U.S.). It is a bigger technical challenge for billing systems. It is even a bigger challenge for insurance payor accounting systems. And the largest impact for clinicians will mainly be a learning curve. We will all need to use a broader, more systematic, more flexible and extensible system – in short we all will need to learn a new language.

Robert Rowley, MD
Chief Medical Officer
Practice Fusion EMR