The New ICD-10 Deadline

By Scott Mace for HealthLeaders Media

When the Center for Medicare & Medicaid Services announced in April that the required go-live rollout of ICD-10 coding would be delayed one year, to October 1, 2014, healthcare providers reacted with sighs of relief, jubilation, disappointment, or stoic determination to stay the course—and occasionally, combinations of the four.

Aside from the American Medical Association, which clings to diminishing hopes that ICD-10 coding will just go away, the U.S. healthcare industry regards adoption of ICD-10 as inevitable, albeit protracted.

When CMS first announced that there would be a delay, but not now long it would be, planners everywhere scratched their heads. Too long a delay and it would be necessary to postpone training programs, lest the recently-trained coders forget the new codes in the interim before the official date.

Some healthcare systems have done enough work so far that they can afford to postpone some ICD-10-related investments into 2013 and attend to more pressing needs now.

“We’re not going to totally stop, but we’re not under the gun as much,” says Gary Barnes, CHCIO, CIO at Medical Center Health System in Odessa, Texas, which includes the 402-licensed-bed Medical Center Hospital.

“We were on target to have everything ready to go” in September 2013, Barnes says. He attributes CMS’ one-year delay to the struggles many providers have had implementing changes in version 5010 of HIPAA transactions. But payers weren’t ready to deal with the 5010 changes, which increased our days in accounts receivable by about three and a half days,” Barnes says. It took a while for this to get sorted out, he adds.

To prepare for ICD-10, MCHS reviewed all of its software applications and found that the new codes would affect eight different computer systems. So, Barnes says, “We were making sure all those systems were updated with the ICD-10-compliant version.”

ICD-10 upgrades were necessary for billing software, as well as clinical systems including computerized physician order entry, progress notes, and repository systems, Barnes says.

The Friday before the final one-year delay was announced, a committee at MCHS decided by consensus to postpone the beginning of ICD-10 training in light of the delay.

“If we train people too early, it’s going to be a waste,” Barnes says.

In the wake of the 5010 holdups, Barnes also questions whether the provider’s payers will be ready when MCHS finally deploys ICD-10.

“Every time we’ve done any of these projects it always seems like the vendors can be ready, but the payers are always the ones behind, and normally it’s the government,” particularly Medicare and Medicaid, Barnes says.

A year’s delay also means MCHS can continue to streamline its physicians’ workflows, ranging from CPOE to progress notes, he adds.

A task force continues to scrutinize existing workflows and find ways to make them more efficient.

The $500,000 budgeted for training physicians and coding staff on ICD-10 will now come out of the 2013 budget, Barnes says. The amount originally budgeted for 2012 will now go toward streamlining workflow efforts, he adds.

“It’s kind of surprising that we keep delaying the efforts in the United States to do that when ICD-10 codes have been used by other countries for years,” Barnes says. “What I’m hoping will happen is that vendors will create applications to help physicians determine the correct ICD-10 codes at CPOE and progress notes times.”

Physicians practicing at MCHS are either independently employed or employees of the Texas Tech Medical School. The hospital is providing them with electronic medical record software for their offices at a subsidized rate, Barnes says.

The ICD-10 training will probably be offered to all physicians because if CPOE and progress notes enter the MCHS already properly coded, then the hospital itself has to do less coding, and can begin to analyze the data in richer ways than before, he adds.

“We’re hoping to get to concurrent coding with ICD-10. This will help us do better predictive modeling on geometric length of stay,” Barnes says.

Currently, MCHS uses SNOMED coding. Barnes says his team is still evaluating whether some sort of transition will take place, or if it will come down to rip-and-replace with ICD-10. The solution will come from a mix of vendor input and outside consulting, he says. The one-year delay “gives us a little more time to evaluate that.”

At Seattle Children’s, the one-year delay did not change the “damn the torpedoes, full steam ahead” attitude toward ICD-10, says Drexel DeFord, senior vice president and CIO of Seattle Children’s.

“We were going to continue with all the work we were doing until we got a real date,” DeFord says. “We hadn’t really let up on anything. Now that the date’s been released, I think a one-year delay certainly feels appropriate. I think it releases a little bit of pressure on the folks who had gotten a late start or who had not quite yet started on ICD-10.”

The one-year delay lets Seattle Children’s better integrate its ICD-10 projects into the multitude of other projects under way throughout the organization, including an expansion from 254 licensed beds to 329 licensed beds by April 2013, DeFord adds.

Seattle Children’s transition from ICD-9 to ICD-10 coding began in mid- to late-2010 with the hiring of a program manager. “There was a lot of work really scoping what we needed to do,” DeFord says.

IT projects included making sure all software was updated to be ICD-10 capable. In many cases, software vendors weren’t ready yet, and even now, Seattle Children’s hasn’t been able to complete some of those transitions. But the October 1, 2014, date will still work, DeFord says.

ICD-10 coding touches 34 systems from 10 different vendors at Seattle Children’s, as well as internally developed research databases at the Seattle Children’s Research Institute.

“It’s certainly not everything we run—we have 197 applications at Seattle Children’s—but it’s definitely some of the more major systems that we run,” DeFord says.

After making an inventory of any systems using ICD-9 codes, DeFord’s team talked with vendors to understand their update timelines and how long Seattle Children’s would have to test these updates before going live with ICD-10 coding. In some cases, DeFord’s team pushed to get software in their hands “as far forward as possible,” he says.

Seattle Children’s is currently in the process of selecting a computer-assisted coding vendor that will work well with the Cerner electronic medical record software already in place, DeFord says. “There’s a lot of due diligence going on,” he adds.

“We know that we need to make a very careful decision here, but we think it’ll be one that will really pay off and will continue to drive our ongoing adoption of electronic medical records and especially physician documentation in the electronic medical record,” DeFord says.

At the executive level, DeFord’s efforts are being led by the chief financial officer and the chief medical officer. “We’ve really tried to make sure that it’s supported from the top by the three major executives that have the biggest effect on their areas, that impact hospitalwide,” he says.

DeFord, who also serves as chair of the College of Healthcare Information Management Executives, likes to use a football analogy when talking about ICD-10 preparation. “You get the ball on the 2 [yard line],” he says. “For those folks who are going to sit there and try to throw bombs and score a touchdown some time in 2013, I don’t think they’re ultimately going to be very successful.

“We’re going to gain a few yards every month until we’re in scoring position, or maybe have already scored the touchdown before we get to the go-live date.”

Seattle Children’s is already in the early stages of a clinical documentation improvement project, DeFord says. “Coding-savvy nurses are available on the floors looking at the documentation as it’s being created, giving feedback nearly in real time to physicians to make sure that they’re doing as complete a documentation as we can possibly get, which is really the key behind being able to code at an ICD-10 level,” he adds.

Paired with these efforts is “a lot of education for our physicians on documenting more completely, why it’s important, what ICD-10 is, and the level of coding and documentation that’s expected,” DeFord says.

DeFord’s estimate of total cost of ICD-10 conversion at Seattle Children’s: $2 million in capital expenses and $8 million in operating expenses, spread across several years.

For a while, Seattle Children’s will code in both ICD-9 and ICD-10. “There’s a period of time certainly around the compliance date where patients will have been admitted before the compliance date and will be discharged after the compliance date,” DeFord says. “The coding that will be in both ICD-9 and ICD-10 will really just be done to facilitate the claims around that time period.”

Almost all doctors and clinicians at Seattle Children’s are on the faculty of the University of Washington. “They will get the training for ICD-10,” DeFord says. “They might be one of the most important pieces of the care team.

“For freestanding independent small-business men and women or small group practices who have hung out a shingle and they’re doing their own thing, this is going to be a bit of a challenge,” DeFord says.

“They are doctors first. They want to practice medicine. The business part of healthcare for them—especially as we go through a lot of these transitions, whether it’s meaningful use or ICD-10—is a challenge,” DeFord says.

Back in the land of AMA members, at least one physician practice organization is taking a different path to ICD-10 readiness.

Northwestern Memorial Physicians Group, a Chicago-based multisite practice of about 100 physicians who are on the medical staff at Northwestern Memorial Hospital and faculty members of Northwestern University’s Feinberg School of Medicine, utilizes codes provided in software from Intelligent Medical Objects of Northbrook, Ill.

Today, those codes map to ICD-9, but when ICD-10 is ready, IMO codes will map to ICD-10 instead, says Lyle Berkowitz, MD, medical director of IT and innovation at NMPG.

“IMO codes are much more meaningful. I think they’re even much better than SNOMED,” Berkowitz says. “Are they perfect? Nothing is. But they’ll be good enough from our perspective.

“Since IMO terms are more precise and user-friendly than ICD-9, it makes sense for us to use IMO now,” Berkowitz says. “That way, we don’t have to wait for ICD-10 to use better terms in our problem list, and the IMO terms can all be easily mapped to both ICD-9 and ICD-10 for billing purposes when needed.”

For example, Berkowitz says, “I can use an IMO term to specify sidedness, such as the term ‘Osteoarthritis of the left knee.’ The related ICD-9 term is simply ‘Osteoarthrosis, localized, primary, lower leg’ versus the more specific ‘Unilateral primary osteoarthritis, left knee’ from ICD-10. Interestingly, there are some IMO terms that are more specific than either ICD-9 or ICD-10 nomenclatures. For example, the IMO term ‘bicuspid aortic valve’ maps to the more generic term of ‘congenital insufficiency of aortic valve’ in both ICD-9 and ICD-10. Looks like we’ll be ready for ICD-11, too,” Berkowitz says.