December 8, 2008

Look North — Canada’s Slant on Smooth ICD-10  Strategies
By Elizabeth S. Roop
For The Record
Vol. 20 No. 25 P. 20

Health information professionals looking to gain a sense of what lies ahead can learn a few lessons from how their Canadian counterparts approached the task
.
Meeting the Centers for Medicare & Medicaid Services’ (CMS) proposed plans to require full implementation of ICD-10 by 2011 will be neither cheap nor easy. According to a 2004 RAND report, prepared on behalf of Health and Human Services, the transition price tag will run anywhere from $425 million to $1.15 billion in one-time costs for system changes and training, plus an additional $5 million to $40 million per year in lost productivity.

But it is necessary for many reasons, including the fact that the United States is now the only industrialized nation that has not switched to an ICD-10–based classification system since the code set was first endorsed by the World Health Organization in 1990.

Canada, which completed its five-year transition in 2006, is often lauded for the proactive, highly strategic approach it took for the move to ICD-10-CA. As such, there may be a few pages the United States can take from its northern neighbor’s playbook to help ease its own transition.

Key Challenges in Canada
Canada began implementing both ICD-10-CA for morbidity coding and the Canadian Classification of Health Interventions (CCI) in 2001 following a development, product, and testing cycle that lasted approximately 12 months and a pilot program in New Brunswick.

In addition to significantly expanding code sets to achieve greater specificity, Canada also needed to create versions in English and French. But the greatest challenge did not come from the rapid deployment, code-set expansions, or even dual-language system development.

“The major change and challenge for Canada at the time was that, previously, coding books were books. They were paper,” says Mea Renahan, BScPT, MBA, CHE, manager of classification standards for the Canadian Institute for Health Information (CIHI). “These classifications were completely electronic, but not everyone had a computer on their desk with Windows capabilities. That was the first challenge … to computerize everyone and also to train people to go from DOS to a Windows environment.”

The second greatest challenge was educating coders on the rule changes created by the higher specificity level of ICD-10-CA and the concurrent introduction of the CCI. Exacerbating this particular challenge was the volume of codes HIM professionals would be dealing with, which expanded from 3,500 in the Canadian Classification of Diagnostic, Therapeutic, and Surgical Procedures to 20,000 in the CCI.

A third challenge was a shortage of professional coders—something with which the United States is very familiar—that made it difficult for some provinces to fill open positions. The resulting coding backlogs would only worsen as coders struggled to adjust to the new system.
 
The end result was an average learning curve of four to six months and an average of six months for productivity to return to pre–ICD-10-CA/CCI levels.

“One of the biggest problems was the resistance to change and the shortage of coders and health record professionals. Most of the coding previously was done by memory, which is true of the United States for many providers and facilities as well. It was difficult for the coders to unmemorize diagnosis codes,” says Deborah Grider, CPC-E/M, CPC-I, CPC-H, CPC-P, CCS-P, a healthcare consultant, an author for the American Medical Association, and president of the American Academy of Professional Coders National Advisory Board. “Many coders in Canada were familiar with abstracting software but were unfamiliar with Windows-based products. They were also unfamiliar with coding-lookup software, and many had not even used a mouse prior to implementation.”

Grider also notes that additional problems cropped up during implementation, including the need to manually maintain nontabular components such as front and back matter and tables, an SQL server incapable of efficiently handling coder demand, lost formatting, and delays throughout the iterative process of working with an external contractor.

Strategies for Success
To effectively manage what promised to be an exceptionally daunting task, the CIHI developed a comprehensive, four-phase plan for transitioning to ICD-10-CA and implementing the CCI.

It started with the preplanning phase, which was perhaps the most important element to the project’s success. This included the development of a comprehensive provincial plan that centered on the establishment of committees made up of key stakeholders in the process who served in either an advisory or task-based capacity. A provincial leader or project coordinator or team was tasked with overseeing the plan’s implementation, which started with an environmental assessment to identify needs and resource requirements, funding responsibilities and sources, and monitoring processes.

Success depended on gaining the early commitment to and engagement in the project from a broad range of industry stakeholders, including all levels of governments and their agencies, professional associations, colleges and universities, healthcare facilities, vendors, and the CIHI. Also critical was a transparent communication strategy for keeping everyone informed of progress, next steps, issues, and resolutions.

In addition to conducting a detailed computer-readiness evaluation focused on hardware, software, and computer literacy of end users, a robust education and training program component was designed to target the specific needs of each stakeholder group.
 
The CIHI took a three-pronged approach to education. First, it provided every facility and vendor with an implementation tool kit, a self-learning package, and a basic training workshop. The organization also trained individuals in each province to serve as trainers and resource personnel.

“CIHI trained every HIM professional in the country. … It was important that everyone understand what needed to happen in order to change from a numeric system to an alphanumeric system and also how to include the CCI,” says Renahan. “We tried to emphasize where the system had changed from a coding perspective and also to familiarize them with the search engine and Folio tools to get them comfortable with the search functionality of the product. We then had two days of hands-on, face-to-face [training] with our staff and provincial representatives.”

The CIHI also launched an online coding query service, where coders could post specific questions to the organization, review responses to other queries, and access resources to help them navigate the new system. The query service, which has since been expanded to encompass six databases and case-mix grouping methodologies, has in excess of 10,000 queries in its databank.

The second component of the CIHI’s plan was a testing phase, during which hardware and software were checked against the established deliverables and expected outputs, and a determination was made regarding the success of the education and training programs for end users.

Key activities in phase 2 included the establishment of procedures for measuring and reporting on the functionality and quality of the input and output and ensuring the availability of adequate in-house IT systems support, as well as technical and coding support for the coding staff. Also important were ongoing communications among the CIHI, vendors, and other stakeholders to ensure early detection and resolution of any problems and adequately prepare facilities for expected productivity dips.

The process entered phase 3 once the system was up and running and facilities were working to comply with provincial/territorial and CIHI submission deadlines. Again, communication was key—particularly among vendors, the CIHI, and end users—to continue identifying and resolving problems and share experiences with others undergoing the transition process.

The fourth and final phase—maintenance and upgrading—is ongoing. It is designed to encourage active participation in the process and ensure the integrity and value of ICD-10-CA and the CCI as tools to gather the information needed to guide future decisions regarding Canadians’ health and the country’s healthcare system as a whole.

Renahan notes that continuing education is a priority for the CIHI, which offers a range of e-learning programs, workshops, and annual case studies geared toward the concerns and challenges identified by coders and to emphasize the best use of classifications.
Although there were glitches and roadblocks, which are to be expected in a transition of this magnitude, the project was ultimately successful.

“Everyone was keen to do this. It was a challenge because it was a completely electronic environment, but once they got into it, no one would ever want to go back,” says Renahan. “From the coders’ perspective, they are thrilled to be able to code what they are seeing [in the chart] and to have much more in-depth information to work with.”

Differences Aside, Key Lessons for the United States
It is important to note that there are a number of critical differences between Canada and the United States that are likely to limit the lessons that can be culled from the former’s ICD-10 implementation experience.

Canada’s universal, single-payer healthcare system is funded, regulated, and managed by the federal government, unlike in the United States. As such, “They can make a centralized decision to change the system and make it much more easily,” says Robert Tennant, senior policy advisor with the Medical Group Management Association.

The Canadian government also footed the bill for the transition, including software and hardware upgrades and training HIM, coders, researchers, and clinicians. Implementation was also phased in over five years.

Meanwhile, the CMS is proposing that the system be transitioned by October 1, 2011, giving everyone less than three years to get up to speed. Further, the transition’s full costs will be funded by the private sector.

“But the critical difference is that Canada decided not to implement ICD-10-CA in the physician practice setting but rather focused strictly on the hospital setting. When asked about that decision, they said that it would have been too expensive and complicated and of only marginal value for public health data collection,” says Tennant.

Despite these differences, there are several key lessons U.S. healthcare leaders can learn from Canada’s experience. According to Grider, a few of the areas that went particularly well for Canada include the following:

• having a dedicated information systems staff to support the classification staff;

• the development of a custom JAVA application in Oracle developer that could be edited in XML in an unformatted text box or downloaded to a user-friendly XML editor;

• the use of Gantt charts and detailed work plans to help the transition go smoothly; and

• the development of a copy code edit tool to reduce key errors and search engines to enable high-speed searches.

“What did not go so well included that shadow files were not transferable with version 2002 and 2003,” she says. “Canada felt they needed to start sooner. There should have been a longer testing phase with more frequent meetings with expanded participation. The detailed work plan was too tight, and the timeline needed to be expanded.”

For Tennant, the pilot testing done by Canada was perhaps the most important step the country took that the United States should be but is not duplicating.
 
“No industry should ever go through a change of this magnitude without doing a pilot,” he says. “There are lots of positives that come from doing a pilot, not the least of which is the identification of problems and solutions.”

A pilot test could also answer key questions about the impact the transition will have on the different facets of the U.S. healthcare system, as well as determine whether there is real value in making the transition in both the inpatient and outpatient sectors at the same time.

“If Canada did their pilot and they did their analysis and found the value was only on the hospital side, that might be the most important lesson the U.S. can learn,” says Tennant. “I don’t want to come across as overly negative about ICD-10, but it’s one of those complex and costly changes where you are on a cliff and it’s foggy. Do you just jump and hope the ground isn’t that far away? Of course not.”

— Elizabeth S. Roop is a Tampa, Fla.-based freelance writer specializing in healthcare and HIT.