Daily insight on EHR and healthcare IT topics

Home

HHS Asks for a Delay to the Start of ICD-10

by Robert Rowley, MD on February 21, 2012

Last week, the Department of Health and Human Services (HHS) announced its intent to delay the date of mandatory compliance with the ICD-10 coding system. The coding system, which has already been delayed in the past, is scheduled to go into effect October 1, 2013. In preparation for this switch-over, the underlying billing-software message format, which is what billing software uses to send bills to payers (notably, CMS), switched from the old 4010 format to the new 5010 format the beginning of this year – the new 5010 format can accommodate both ICD-10 and ICD-9 diagnosis codes.

The whole effort to replace the ICD-9 system of classifying diseases, which was released in the 1970s, with a newer system – ICD-10 – which was released in 1992, has been to modernize nomenclature and capture diseases in a more logical and systematic way than can be done with the older system. It is not a one-to-one mapping, however: the ICD-10 system has some 68,000 codes, in addition to a supplementary list of procedure codes (ICD-10 PCS) used only in the U.S. containing an additional 76,000 codes.

So… what’s the problem?
The coding of diagnoses is central to pretty much every healthcare transaction. So a switch from ICD-9 to ICD-10 impacts everyone. In particular, it impacts very large sectors of the healthcare ecosystem that are heavily invested in older, entrenched enterprise systems which would all need upgrading – health plan billing payment systems, every business and provider that does quality measures (based on ICD-9) codes, pharmacy systems (both retail pharmacies as well as pharmacy benefits managers), hospital and academic institution systems. And, of course, the emerging use of Electronic Health Records (EHR) by community physicians. All of this would need upgrading.

The sheer cost and administrative burden to upgrade all the software systems that utilize diagnosis-based data in the entire spectrum of healthcare is onerous. It’s more than simply changing a learning curve for physicians, getting familiar with new codes for common diseases (uncomplicated diabetes is no longer 250.00, but instead E119). It involves cost, sometimes large, for the software upgrades needed in large institutional settings.

The American Medical Association (AMA) has been one of the louder voices pushing for a postponement of the October 1, 2013 deadline, noting that physicians are already strained with financial and administrative burdens in implementing EHRs for Meaningful Use (though that can certainly be minimized by adopting free, web-based EHRs), and that the timing of an ICD-10 conversion simply adds to the load. The benefit to physicians of making a change in their long-familiar diagnosis-coding system is not sufficiently compelling to be worth the burden.

On the other hand, the American Health Information Management Association (AHIMA) has been a voice calling for no delay in ICD-10 implementation. The plans to move health care into new payment paradigms, away from simple fee-for-service and toward a more global-payment method, is tied to the more specific information gathered by ICD-10 codes. As some have stated, “pushing the deadline back one year means that we can’t complete all aspects of healthcare reform.”

Should we leapfrog ICD-10 and move to ICD-11?
If things get delayed enough, some have speculated that we should simply bypass ICD-10 altogether, though the likelihood of that is quite low. After all, the ICD-10 system was released in the 1990s, based on 1980s medicine, before the Internet and the semantic web. The next-generation of medical diagnosis coding (ICD-11) is just being elaborated by the WHO, and will be able to address things like genomic data, SNOMED and other elements of health data that did not exist when the ICD-10 system was developed.

But ICD-11 is nowhere on the HHS agenda. They are responding to loud industry pushback just going from ICD-9 to ICD-10, and have accepted the argument of “excessive administrative burden” enough to postpone the mandatory implementation date. It is not clear by how long they will want to delay the implementation – some speculate a year, others longer.

What about docs using web-based EHRs?
The least-impacted segment of health care will be the smaller clinical practices using web-based EHRs. For our own system, the transition to ICD-10 is fairly straightforward from a technical standpoint. CMS has developed General Equivalency Mapping tables that make good cross-walks from ICD-9 to ICD-10. That way, all the legacy data already in a clinician’s charts, coded in the old ICD-9 framework, can be retro-converted to their most-likely ICD-10. All the Clinical Quality Measures would need to be reworked to query off of the new codes. And, going-forward, the diagnosis pick-lists simply use the new tables.

Such changes, though not trivial, can be done fairly easily. And, with web-based deployment, everyone using the system will have the upgrade at once. Those clinicians who have invested in locally-installed enterprise systems will, however, need to upgrade to a new version. This (depending on the software vendor) may take time, and will likely take money.

But, as we have seen, the most difficult places for such a change is in the realm of the large enterprise installations – the systems in place in academia, in hospitals, in pharmacies, and in health insurance payers. This is where the resistance is loudest, and this voice is well represented in Washington policy circles.

We will see, over the upcoming months, how long HHS will ask for a delay in moving to ICD-10. Its delay will delay elements of health reform. But its implementation will be a burden on big legacy “dinosaur” systems. As far as the more nimble, web-based technologies being rapidly taken up by smaller ambulatory clinical practices, the transition – though a learning curve about a new coding system will be needed – will be fairly straightforward.