IT teams can accomplish great things when they must meet or beat a deadline, like Y2K. The ICD-10 transition has been delayed, but CMS can still act to keep momentum building.
Easy-to-Mock ICD-10 Diagnosis Codes
(Click image for larger view and slideshow.)
When InformationWeek asked me to write about my journey towards compliance with the ICD-10 mandate, President Obama had just signed a law delaying the mandate for compliance until October 1, 2015 (at the earliest). Implementation of this diagnosis code standard had been delayed before, but until Congress got involved, the word from the Centers for Medicare & Medicaid Services (CMS) was that October 1, 2014, was a firm deadline.
When the House and Senate were voting, the buzz around my IT department varied from "Of course it will pass" to "No, they'd never extend it; that would be a waste of the time, effort, and money already expended." The House passed the bill with little debate, and the Senate did the same a few days later. Still, the buzz on the floor was that the President might not sign it into law. We simply could not fathom that a deadline we had been told would absolutely not move was indeed going to be moving.
Part of a larger bill on physician reimbursement for Medicaid, the section imposing the delay reads, "The Secretary of Health and Human Services may not, prior to October 1, 2015, adopt ICD-10 code sets as the standard…" Our speculation turned to what CMS would have to say about the date, and what the options would mean for our work.
For the uninitiated, ICD-10 is the tenth revision of the International Classification of Diseases, created by the World Health Organization in 1989 and widely used for decades outside the United States. ICD-10 increases the detail of the diagnosis codes used on claim forms, which also increases complexity. The plan developed by CMS would have increased ICD-9's 14,000 codes to 69,000 ICD-10 codes.
Currently, for example, a provider can report a diagnosis of diabetes as 250.00 in ICD-9. The ICD-10 list has specific codes for possible complications, such as Type 2 diabetes with chronic kidney disease (E11.610) or with kidney complications not elsewhere classified (E11.29), unless the diabetes with chronic kidney disease was due to drug or chemicals (E09.22). These are just three examples of the vast number of diagnosis codes available for reporting a diagnosis of diabetes.
ICD-10 also includes more exotic codes. One that caught my imagination is V96.03XA: balloon collision injuring occupant, initial encounter. This code cannot be the primary diagnosis (it is not an injury, but a causal code). A primary diagnosis to use could be J67.2: bird fancier's lung. Used together, the two codes spin a story of a hot air balloon passenger who inhaled bird feathers after the collision and developed a respiratory ailment. In ICD-9, this would have been 495.2, with no causal code. (Let me know if you have used some interesting combinations of codes in your testing.)
ICD-10 does not deal only with diagnosis codes, however. Hospital claims are more involved, and these are changing with the use of the new procedure codes along with the diagnosis codes. I deal with the professional (physician) side of billing, where the choice of coding systems has real financial consequences.
Y2K vs. X12 version 4010 to 5010 Let's compare what's happening with ICD-10 to the Y2K deadline. The Y2K deadline obviously could not move -- not even Congress could change it. Come midnight on January 1, 2000, either the world would end when all the computers on the planet thought it was 1900 again, or nothing dire would happen. Vast resources were called in as companies geared up
Deborah Graham is a senior programmer/analyst at a large hospital system in Massachusetts working in the IT department on the provider practice side of the organization. She has more than 14 years of healthcare IT experience and over 25 years of programming experience in a ... View Full Bio
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?