Good news on the coding front: With more than half its coders already using ICD-10 diagnosis codes on a daily basis, McKesson Business Performance Services (McKesson) has found that the acute documentation and productivity problems many feared would overwhelm the ICD-10 transition have not materialized.
McKesson began shifting coders to the new code set in December 2013 via an in-house application that automatically converts ICD-10 codes to ICD-9 for claims submission. The software, coupled with extensive training, has allowed the company to get a running start on the Oct. 1, 2015, national adoption deadline.
Cindy Slocum, McKesson’s project manager for ICD-10 implementation, said about 300 coders working in a variety of medical specialties have collectively processed about three million patient encounters in ICD-10 since late 2013. She said the balance of McKesson coders should complete the transition by March 2015.
“The whole rationale behind our early adoption program was to gradually phase in ICD-10 in order to get our staff acclimated and proficient, and to be able to work with our customers to make sure their documentation supported the new codes,” Slocum said. “Taking this approach has allowed us to get ahead of the curve and learn a lot of valuable lessons.”
Default Code Eases Transition
One of the most important of those lessons, Slocum said, has been the revelation that the documentation deficiencies many feared would sabotage ICD-10-coded claims have so far not been an issue.
“Basically what we’ve discovered is that if you can code the clinical document in ICD-9, there is an extremely high probability that you’ll be able to code in ICD-10,” Slocum said. “Clean claims are going out with the appropriate diagnosis codes, and that’s a huge relief for a lot of people.”
The smooth transition may be in part due to concerted physician efforts aimed at providing the level of encounter detail required by ICD-10. But Slocum said it is more likely that the general absence of problems stems from the expansive ICD-10 code set, as well as the coding leeway provided by the Centers for Medicare & Medicaid Services (CMS.)
ICD-10 allows coders to use an unspecified code when documentation specifics are lacking. Slocum cited a provision from an April 2013 CMS guidance document: “When sufficient clinical information is not known or available about a particular health condition to assign a more specific code, it is acceptable to report the appropriate unspecified code.”1
For example, if the clinician fails to document which hand has been affected by an injury, the claim can nonetheless be pushed through with an “unspecified” code.
Slocum noted that while the clinical specificity at the heart of ICD-10 is essential to improved financial and clinical analysis, quality measures, disease surveillance and overall quality improvement, the unspecified code was established as an interim tool to make it easier for the physicians to make the transition to the new system.
“Over time I think that the use of unspecified codes will be phased out,” she said.
Despite the flexibility provided by the unspecified codes, Slocum cautioned that groups that currently have clinical documentation returned for additional information, or that experience high denial rates in ICD-9, will likely face significant issues when transitioning to ICD-10.
McKesson has developed a feedback mechanism that identifies documentation shortcomings as they emerge during the coding process, Slocum added, and the information is being shared with clinicians as well as coders. This continual improvement process will help ensure that physician documentation is up to speed by next October, she said.
Productivity Largely Unaffected
In addition to highlighting the absence of major documentation problems, McKesson’s early adoption has also revealed that the serious productivity declines many anticipated with the transition have not occurred. McKesson and the others in the healthcare industry last year warned that practices could see a drop in coder productivity of up to 60% for as long as six months after implementation.
“We have actually had coders that saw no decline in productivity whatsoever, and I think that’s a tribute both to their training and their own level of commitment,” Slocum explained. “There is a lot of variability in terms of coding for different specialties, so it’s difficult to generalize. But by and large, we just haven’t seen the sharp and sustained declines that we had anticipated.”
That said, Slocum said there have been significant productivity drops in a few specific areas. The conversion of evaluation and management (E&M) codes, for example, has been accompanied by productivity declines of up to 85%. Yet the fall-offs lasted only several weeks and productivity recovered as coders became proficient with the new codes.
Despite the positive results emerging from McKesson’s early adoption program, Slocum said practices shouldn’t become complacent.
“There is still a long way to go and a lot of work to do, so groups need to keep their eye on the ball to ensure a smooth transition,” she said. Slocum acknowledged that the decision earlier this year to delay implementation of ICD-10 by one year damaged momentum and undermined training for many smaller groups.
“It has been difficult for groups that invested time, training and IT resources in the original timeline,” she said. “But they’ve got to make sure that the training continues and that they take advantage of the additional time provided. If our early adoption has taught us anything, it’s that training and practice are essential to a smooth transition.”