Implementation of ICD-10: Experiences and Lessons Learned from a Canadian Hospital

Kerry Johnson, CCHRA(C)

Introduction

When individual Canadian hospitals first started the journey towards The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canada (ICD-10-CA) and The Canadian Classification of Health Interventions (CCI) (from now on referred to jointly as "ICD-10"), there was little experiential information upon which to base implementation strategies. Humber River Regional Hospital (HRRH) had to prepare for implementation and understand the effects such an implementation would have on the hospital, particularly on information systems and data collection (coding) staff.

Up until that time, there were only a couple of provinces in Canada that had implemented ICD-10. Those provinces (Nova Scotia and British Columbia) were located at opposite ends of the country from HRRH's location. Also, as HRRH was preparing for implementation, other provincial implementations were still in their infancy. Although those provinces were ready and willing to offer any experience they had, it was still not enough to fully plan the implementation of a change of such magnitude. Any other global experience (such as that from Australia) would have been significantly different from what the Canadian experience would be, due to the "Canadianization" of the products and electronic nature of the database in which ICD-10 was published in Canada.

Background

HRRH is a multisite (three sites) general hospital located in the northwest corner of Toronto, in the province of Ontario. At the time of implementation, there were 605 beds in the hospital (including 18 general short-term rehabilitation beds). The hospital was a regional dialysis and pediatric center. Annual patient volumes applicable for coding were 30,000 inpatient visits (including 5,000 births), 102,000 emergency room visits, and 35,000 day surgery visits.

At the time, the Clinical Data Collection (coding) section of the hospital reported to the Director of Financial Planning and Decision Support Service in the Finance portfolio, and not the traditional reporting model to the Director of Health Records in Health Records/Information Services. At the time, there were 16 data collection staff members (coders) that needed to be trained on ICD-10, as well as three Clinical Information and Utilization Analysts, and the manager of the area.

In 1995, there was an agreement for Canada to adopt WHO ICD-10. Work then began in 1999 to develop the Canadian version of ICD-10-CA as well as the Canadian Classification of Health Interventions (CCI) with the European informatics standards. Both of these Canadian-tailored classifications were adopted for use commencing in April 2001.

In Canada, healthcare is a provincial jurisdiction. Although the federal government and agencies, such as The Canadian Institute for Health Information (CIHI), set the overall guiding principles and standards for universal healthcare delivery and information reporting within the country, how that service and data collection is administered is the decision of the province. Therefore, each province had the opportunity to say when was the "right" time for them to move to data collection within the ICD-10 framework. As a result, the provinces are making the transition to ICD-10 at different intervals starting in 2001 and staggered to April 2005.

Ontario hospitals were fortunate in that the provinces of Nova Scotia on the east coast, and British Columbia on the west coast of the country had already implemented ICD-10. Thus, Ontario was able to learn from those provinces and to some extent the experience that took place in Australia, with its healthcare delivery system that is similar to the Canadian one.

In Ontario, transition to ICD-10 was imminent. However, when that transition would happen was the outstanding question. There was a lot of uncertainty and confusion surrounding this question. Originally, an implementation date of April 2001was set for Ontario. Then it was moved to 2002 and then possibly to 2003. A point to note is that although these dates were somewhat fluid, they were fluid only a few months ahead of the proposed implementation and not years, as is the current state of ICD-10 implementation in the US.

This constant flux of timelines was extremely frustrating and anxiety inducing for those at HRRH. There were several within the organization who had experience with health information systems implementations and educational backgrounds that realized what the effects of an implementation of such magnitude could be. When the 2001 implementation date did not happen and there was some wavering on the 2002 date, HRRH knew it needed to proceed with preparation for the implementation anyway.

Methods for Implementation

Education

HRRH investigated education for key individuals within the staff that would be working on the implementation of the new classification system, which included the testing and evaluation of software. These staff members would also take a lead in helping guide the remaining staff at the facility through the transition from ICD-9-CM to ICD-10. There were some specific challenges presented in terms of finding the in-depth type of education the hospital was looking for: (1) Canada has a fundamentally different coding philosophy than many other jurisdictions with definitions of "codable" diagnoses and interventions; (2) the health information management market is not large in Canada, and as a result the number of educational materials available with a Canadian-based viewpoint are few and far between.

However, HIM professionals in Canada were fortunate that Joy Fletcher of Douglas College in New Westminster, British Columbia, had worked very closely with CIHI to develop materials for use in the educational system for HIM professionals. Because British Columbia had transitioned to ICD-10 in April 2001 and had been training their students for at least a year prior to that, there were already educational materials and formal distance education courses available for the key individuals in the transition to ICD-10, which HRRH started in late spring of 2001. This was advantageous in that all of the students in the country could potentially be trained using the same text that is based upon the coding standards set by CIHI.

Required Changes in Business Processes

The main preliminary work that needed to be done prior to any implementation of ICD-10 was the removal of the backlogs in data collection along with educational activities. Canadian coding is generally not part of a reimbursement-based system; as a result, Canadian timelines have typically been more relaxed than other jurisdictions, such as the US. Therefore, there were a number of weeks of work to eliminate. The hospital authorized overtime hours to catch up on the backlogs.

Some of the main considerations in the implementation were the changes in business processes that would need to take place, which are outlined in Table 1.

Table 1. Comparison of Canadian ICD-10 Products

Software Testing and Evaluation

Once initial key individual training was completed, preliminary testing of the abstracting software vendor (Meditech) with ICD-10 was initiated. Because Meditech already had ICD-10 clients, particularly in Nova Scotia, the company already had a version of the software that was Canadian ICD-10 compliant. Testing commenced in October 2001, and was completed without the benefit of a mandate from the government. However, there were rumours of a government mandate for April 2002.

As the testing process progressed, it became more apparent that there were going to be some issues with the Canadian ICD-10 version of the Meditech product that was provided, particularly with the use of the separate Folio InfoBase product. It also became apparent that extensive work to develop data quality strategies with the new cumbersome codes in the ICD-10 classification system was required.

Therefore, the department found itself in more of a climate of change than it had originally anticipated as outlined below:

• Hospital is preparing to implement ICD-10-CA/CCI.
• There was a definite increased need for data quality activity with no budget for increased staffing.
• It was apparent there was an increased need to enhanced IT functionality with software.
• Internal pressure for increased reporting that was timely and accurate.
• New Ministry of Health reporting requirements for this multisite hospital that was only required to report as "one facility" was now also faced with a mandate to report all three sites as separate institutions.

Key stakeholders from the Decision Support/Clinical Data Collection and Information Technology functions met to strategize, and they developed the following checklist of requirements for a software vendor to meet :

• Point-of-entry data quality edit checks
• Enhanced reporting capabilities for a "new" solution (if accepted) were required
• Solution had to get the information back to original hospital-wide integrated system abstracting module
• Windows based
• "Proven" technology, (that is, already in existence at another similar facility)
• Solution must go live with HRRH and ICD-10 (now it's March 2001!)

The selection for a new abstracting system (3M) that would interface (via Meditrain expertise) with existing hospital-wide integrated system (Meditech) was made. An implementation schedule of six weeks was developed and completed

Between the actual installation of the software and the "go live," there were several activities that took place. Reference tables compatible with those in the existing hospital system had to be created within the new software. Major functionality changes for the new abstracting software were identified by asking the question: would it stop us from coding and reporting what is required? The changes were then incorporated for implementation in the new system. A temporary manual process for the download of the required demographic information from the hospital-wide system to the new system was developed for the initial implementation date, and an automated download process was developed within three weeks after implementation.

One major decision that had to be made was whether or not any coded historical data would be converted into the new abstracting system. It was decided not to convert any of the historical data. This decision was based on the fact that the department was now working with a whole new database where there were new diagnostic and intervention codes as well as a new abstracts from what previously existed.

Staff Involvement

From planning to implementation, staff was involved in the process. A select few of the staff were involved in the site visit portion of the selection process. All staff was involved in an on-site demonstration of the software as recommended by their peers. They had been advised if they could not work with the proposed system, then the hospital would not go through with the change of system, but would continue with the implementation of ICD-10 with the existing software vendor.

After the installation of the hardware and software and training sessions were completed, the staff were involved in testing the software and the new methods of looking up codes. One of the major components of the new software system was a decision tree encoder, which was new to all of the staff. All staff was first trained on the Folio product, which is essentially an electronic book, through the CIHI training. They then were comparing their encoder results with what they knew from their coding training on the product delivered from CIHI. At the same time staff input was gathered for software modifications to suit their business processes.

All members of the staff were given the opportunity to take extra training (the Douglas College distance learning course) outside of the mandatory CIHI training (five-day Self-Learning Package [SLP] and two-day in-person training), but only a few did. All staff participated in Windows training and the mandatory CIHI training. Due to the existing union environment within the hospital, it was difficult to mandate any additional training other than what the government and CIHI had deemed as necessary.

Role of Information Systems Department

The role of the Information Systems Department (IS) was to ensure that servers and workstations adequately handled the new software. Specific changes required within the hospital were to ensure that workstation personal computers (PCs) would be able to run the software and do so in an efficient manner. From the users' perspective, monitors needed to be upgraded from 15 inches to 17 inches. Many of the staff found it difficult to view the lengthy codes on the 15-inch monitors. It was also necessary to ensure that there were sufficient computers with the new software test environment installed to perform the initial required testing and future testing while the software was in production. Server specifications were also reviewed to ensure that current technology was appropriate for the new systems.

Overall, the system would be managed by IS in conjunction with Clinical Data Collection and Decision Support Service. One of the most crucial roles of IS was to ensure that the interfaces between the software systems were working adequately. IS ensured that processes were in place for required reference tables to be set up initially on both systems and that a process for the continuing accurate and timely maintenance of those tables was in place. IS also ensured that communication between the various software vendors and Clinical Data Collection/Decision Support was maintained and productive. IS coordinated customization of software, installation of new software and upgrades, provided remote access of the software vendors to the hospital's servers, and coordinated "go live" activities.

Initial Findings

Staff Productivity

When implementing a change as big as the change of moving from ICD-9-CM to ICD-10 with the associated abstract changes, initial decrease in productivity is to be expected. Although there was anecdotal information received from British Columbia and Nova Scotia, the information was rather sketchy as to what effect the implementation had on staff productivity. Initial figures indicated that at the start a decrease of 50 percent productivity could be expected. Because HRRH's implementation planning was so close to the other provinces' implementations, it was difficult to assess at what point productivity would be back to normal. But it was realized that in all probability, tertiary hospitals and hospitals with a varied case load would have a greater time lag getting back to normal productivity than those hospitals whose case-mix range is relatively small and well defined as the staff are introduced to the wide variety of codes.

This lack of information made it extremely difficult to plan for any overtime or extra staffing costs that might be incurred to maintain turnaround times in the department.

Coding productivity numbers from HRRH are as listed in Table 2.

Table 2. Coding Productivity (Charts Completed per Hour) Pre- and Post-ICD-10 Implementation

It should be noted that the hospital was in the midst of training a number of individuals in day surgery coding and especially emergency room coding. This probably had some effect on the rebound of the numbers, but it is clear to see that after almost a year of using ICD-10, this particular hospital is not back to the productivity levels prior to ICD-10 implementation. It is noted that some of this may be accounted for by the transition in the coding tools, (that is, books to encoder and a software change). But what that effect would be is difficult to quantify, since the experience in productivity of this hospital is at least on par with many of its peer hospitals in terms of recovery from the implementation.

It is noted that it was at least three to six months post-implementation before there was any appreciable improvement in the decreased productivity and almost a year before productivity levels approached pre-ICD-10 levels. Again, there was a lack of shared information in to adequately quantify this hospital's experience compared to other hospitals. Anecdotally, this hospital's experience is not that much different than that of its peer hospitals, and it was noted that by mid-2003 productivity levels had peaked at below pre-ICD-10 levels. This may be due in part to the fact that the codes are now more detailed, requiring a more rigorous search in the record for the detailed information. This is particularly true in the cases with more intervention coding where the rebound has not been as pronounced as with the other coding.

Cost of Implementation

It is extremely difficult to provide a breakdown of the actual costs of implementation due to many reasons. Some of the most significant would be the amount of education that the hospital chooses offer and the amount of information systems infrastructure (hardware and software) that would need to be acquired for the implementation. However, key areas of expense can be identified for HRRH that could be modified and tailored to meet individual hospital needs that are contemplating the move to ICD-10 in the future.

1. Education

Pre-implementation education for all staff and key users was critical to the implementation. For HRRH, this included the mandatory five-day self-learning package provided by CIHI and the two-day in-person session. There was also work time that was spent prior to the implementation where staff was given the opportunity to practice using the new information system.

Key users that would be providing system testing, education, and guidance for the frontline staff completed a six-month distance-learning course from Douglas College in British Columbia. Tuition and text for the course was reimbursed by the hospital. The study was done on the individual's own time, however, examinations were written on hospital time.

2. Pre-implementation Readiness

To minimize the initial effect of reduced productivity that would result from the change from ICD-9-CM to ICD-10, the decision was made to put resources toward the removal of any coding backlogs. Because of the difference in Canadian clinical database use from those particularly in the US, it would not be unreasonable to find a coding backlog of 60 to 90 days. Each facility needed to make the determination of how much it was willing to invest in removal of backlogs. The decision was made at HRRH to get to a 21-day turnaround time from about 60 days. This was done over the course of November 2001 to April 2002 (a reduction of about one week per month) through overtime the existing staff was able to contribute. This was done at a cost of about 10 percent of the overall salary budget.

3. Post-implementation Production Volumes

Post-implementation, an additional 20 percent of the departmental budget was spent to remove the accumulation of charts due to the reduction in productivity. The hospital finally reached the 30-day turnaround target by January 2004. At that time, it was also known that there was a mandate coming from the Ontario Ministry of Health for mandatory quarterly deadlines that would require an internal 30-day turnaround standard.

4. Software and Hardware

The hospital needed to make a determination as to what new software and hardware requirements were needed in order to proceed with the implementation. Even if the hospital had stayed with the same vendor that was initially in place, there would have been a one-time conversion fee imposed by the vendor due to the increased costs incurred by the vendor for the transition. An inventory would need to be taken of software systems to determine the effect of converting to ICD-10. In some instances, particularly in the US, this may mean a more far-reaching effect (that is, financial systems) than are currently found in jurisdictions that use the coded data only as a clinical database.

HRRH needed to upgrade its software and hardware to some extent, so this cost was born and would be hospital-specific, depending on what would have to be done.

5. Other

Other associated costs will vary depending on the hospital and what is required or implementation. Some of the considerations faced by HRRH were: new supplies for start-up and the change in the existing business processes; time to test the new software; and time for extra data quality activity.

Post-Implementation Issues

Operational Issues

As with the conversion of any system, there is going to be all the unanticipated hurdles that need to be overcome as the implementation unfolds. All the "best laid plans of mice and men" seem to meet some form of unforeseen obstacle as one moves along in the implementation.

Clinical Data Collection continued to review and modify the functionality of the data elements in the software. This proved to be an ongoing process as more data quality issues arose post-implementation. Suggestions for enhancements came from front-line staff coders, analysts, and management. Suggestions involved data quality edits, flow of collecting the data on the abstract, and the effort to reduce the number of keystrokes necessary when completing an abstract.

One of the more notable problems with the transition to ICD-10 is that many times the documentation on the clinical record does not exist to support the detail of the classification system. To compound the problem, in many instances within the classification system (particularly the interventions), there is no accommodation for the instance when the detail is missing as with ICD-9 (that is, NOS option). As a result "next best" codes are selected that would be inaccurate.

Not specific to the implementation of ICD-10, but nonetheless a concern when changing systems, is the continuing challenge to ensure that the two interfaced systems reconcile in terms of the number of records discharged and coded each month. Many times there were discrepancies between the systems, which could be attributed to the patient registration process.

Initially, the vendor needs to be ready to meet facility requirements. However, some larger issues have come post-implementation because the vendors were not ready to support the ongoing changes to the classification system that occurred. There have been new ICD-10 diagnostic codes and intervention codes, especially as technology involved in the interventions changes. There are the subsequent deletions these new codes would create. This created difficulties with the vendors, particularly in relation to the timelines within which these changes were released by CIHI and the vendors' ability to incorporate the changes within the timelines they were given.

Informational Issues

There were some significant ICD-9/ICD-10 comparability and grouping methodology concerns in the Canadian experience, which are well documented by CIHI and are available through documents on its Web site. These concerns are briefly listed here:

• Many of the codes are not comparable from ICD-9 to ICD-10.
• There are different data collection standards and underlying principles between the systems.
• Because some of the differences are so fundamental to the coding schemes, crosswalks are not possible or somewhat unreliable at best.
• Because of the detailed nature of the codes, there has been a demonstrated shift in some case-mix groupings, that is, cases that were previously reported in one grouping under ICD-9 are now reported under a different grouping in ICD-10, creating a problem with comparability at the case-mix grouping level as well as the code level.

There will be time lags with such information as case weights and length of stay national databases, until there is a sufficient amount of information in the national database collected in ICD-10. This is somewhat due to the fact that the provinces were able to implement ICD-10 at different times. This resulted in incomplete national databases as far as collecting under the same classification and the ability to have the complementary financial information from those same provinces at the same time (for example, Alberta and Ontario). There will now be the time lag to analyze the information. A new grouping methodology with the associated case weights and length of stay databases is not expected until 2006.

With the change in the coding, all of the reports that previously existed needed to be examined and a large number of them rewritten to accommodate the new classification system. After the implementation, the hospital began its report writer training on the new software system. The hospital also began the process of converting all of its reports, not only from one vendor software to the ICD-10 solution, but from ICD-9-CM based reports to ICD-10 based reports. This meant there were many reports that accumulated over the years that had to be scrutinized for any traces of ICD-9-CM and old data collection abstract elements. In many cases, even the field sizes were changed, which required a change in the reports that had been written. The exercise provided the hospital with an opportunity to review carefully the number and quality of reports that it had within its reporting systems.

There was a time lag on reported data as the systems were converted. Old systems needed to be closed off before the new ones were implemented. Also, learning curves compounded the issue of turnaround times.

Discussion

Although there was a great amount of time, effort, and money put toward the implementation of ICD-10 at Humber River Regional Hospital, there were several aspects that contributed to what the hospital feels was a tremendous success in its implementation of ICD-10.

It is imperative that the appropriate resources be committed to the transition from ICD-9-CM to ICD-10. HRRH's Clinical Data Collection was fortunate that the hospital was willing to commit the appropriate resources to complete the project. Without those resources committed, this important implementation would not have been as successful and smooth a transition as it was.

Preparation is a must, and this will go hand in hand with providing the appropriate resources. It was necessary to m ake sure that there was enough time to adequately prepare for the implementation. HRRH started almost a year ahead of time to ensure that everything was ready for the implementation. One of the main tools used to prepare for and keep the implementation on track was the use of Microsoft Project software. It is also necessary to ensure that there is an individual who is responsible for the coordination of the project. It would have been more desirable to have more official lead-time for planning to eliminate some of the tight timelines that were realized by HRRH.

Another key to the success of the implementation was teamwork. Without teamwork, there is no way that any project of this magnitude will succeed. The key players for this particular implementation were the Health Information Management team, Information Services/Technology Department, software vendors, and Hospital Administration. It is imperative that Administration be kept informed of all that is going on with such an implementation. This process was started at HRRH with Administration as soon as it started education of the key users, a year ahead of the actual implementation.

Patience is also necessary for an implementation of this magnitude. As it stands, this particular implementation lasted for approximately two years from initial education and planning to the time the staff reached their pre-ICD-10 productivity levels. With the desire to have more planning time, conceivably, this could have been a three-year project from start to finish.

Strategies for Subsequent Implementations of ICD-10

From HRRH's implementation, there are a number of lessons learned that can translate into strategies to employ. If used, these strategies should ensure that any facility implementing ICD-10 would have as smooth a transition as possible.

Project Leader

It is critical to have a project leader within the organization to champion the implementation. This individual would take the lead in the coordination first of the planning of the implementation and then the project management throughout the course of the implementation. It is important that this individual have a background in classification systems, and preferably in HIM (experienced with ICD-9 or ICD-10 coding). This individual should take a lead in all of the activities related to the conversion. This would include the training or coaching of individuals who are being trained in the new classification system. He or she would ideally have some training in the new system prior to any others in the organization participating in the training. This individual would also be involved in the selection and evaluation of systems and their requirements as they relate to ICD-10.

It is also advisable to have a multidisciplinary task force to oversee the implementation to ensure all aspects are considered. This group might include, the project leader, IS representative, end-users of the data, and a sponsor for the project. As the scope of the implementation is determined, the group may grow.

Education

Education is critical to the success of the implementation. This cannot be stressed enough. There are a number of significant differences between ICD-9-CM and ICD-10. These need to be understood by all who are going to use the system.

A real benefit of education throughout the implementation is that those employing the new system understand the structure of the codes and how they are derived. In some respects,   this will help with identifying whether or not correct codes are being selected during the coding process. Also, it is important that all of the underlying principles and assumptions with the new coding system are well understood.

It is not sufficient to take a minimalist approach when it comes to this training. This is the approach that was taken by the Ministries of Health in conjunction with CIHI, and the result was less than ideal. Consequently, many find the mandatory training was not enough, and have now returned after the fact to get further training from the colleges, either through their organization or on their own.

It is important to make sure that the organization knows that the migration to ICD-10 is taking place and when it is taking place. The organization needs to be aware there are changes in how the data is collected, how it will be reported, requirements for documentation, lag times in reporting while the conversion is taking place, and so on.

Finally, as part of the education process, the "fear of change" factor needs to be minimized. Some of our greatest obstacles are with resistance to change. Employing change management strategies will greatly benefit the implementation of ICD-10.

There will ultimately be a personal investment of time and energy on the part of each one involved in this conversion. This is particularly true for those individuals who are intimately involved with the classification system on a day-to-day basis. Their success will depend upon the amount of personal commitment they are willing to put into this. Education is the key word!   Experience has demonstrated that those employees who gave of their personal time and energy to educate themselves on the new classification system were the most successful in the conversion.

Education is required for the following reasons: the new code structure of ICD-10 compared to ICD-9; ICD-10 requires a more detailed knowledge and understanding of anatomy, physiology, and interventions; and there are new coding standards and assumptions that are inherent in the new system

It must be realized that there will be a learning curve associated with this new system. The amount of usage an individual has with the system will dictate the amount of time it takes to get through the learning curve. Experience has shown that a period of about six months using the system on a daily basis in an acute general hospital allowed the coding staff to become proficient enough in using the new system that they almost returned to their previous productivity levels. Again, this depends on the amount of education that was invested at the front end of the system conversion.

Technical Considerations

The Information Systems department needs to be intimately involved in this project. They should be involved in all of the planning for this conversion, since it is largely an information systems conversion. One of the major factors, particularly in Canada is that the "code books" are now electronic.

The Information Systems department will need to:

• Inventory all software that carries the coding system
• Map interfaces, tables, and information flow
• Contact the vendors for affirmation on the upgrades
• Determine their staffing requirements
• Examine hardware/software requirements

Software vendor readiness is a huge factor. This is something that healthcare in Canada struggled with over several years, even after the initial implementation of the first two provinces. Although a number of vendors (abstracting) have moved to the ICD-10 products, there are still concerns with regard to their ability to continually upgrade every two years (as proposed) as new codes are created and obsolete ones are deleted.

Some specific software considerations are whether or not the current software vendor will be able to meet an organization's needs, or will an alternate vendor need to be considered?   There are differences in the code structure and field size--does this create any problems for existing systems that need to be corrected? Are there any interfaces that need to be modified to accommodate the new code structure? Regarding hardware, are there any changes that need to be made to accommodate the new system, such as larger monitors, more powerful hard drive to accommodate all work done with Folio product or to accommodate any other software changes that need to be made? Or, if the computers are slow now, how will that affect productivity if all the work (that is, coding and processing of an electronic patient record) is now done on the computer? Is a new server required?

Conversion of Business Processes and Systems

It is important to identify and remove any obstacles there may be that would stand in the way of the conversion of your systems. These barriers could be human or technological.

One way to assist in this is to ensure that there is some thought given to the switchover. This would include doing a project plan. There are many templates (Team Charter and Work Plans) and tools (MS Project) that exist to assist in identifying the tasks that need to be completed and help track the progress of your project. HRRH used these for their implementation and found them invaluable in assisting with a relatively smooth transition from ICD-9 to ICD-10.

Part of the plan would be to remove any backlogs that exist and get as current in workload as possible. It is a given that there will be a reduction in productivity levels in working with the new coding system. With that reduction will come the inherent backlogs. If the new backlogs are added to an existing backlog, then the result may be a situation from which a facility would never recover.

Another consideration is what will be done with the previous data sets?   A certain amount of storage capabilities of both systems will be required in most instances. But depending on the nature of the organization, the time frame and, therefore, the storage mechanism may differ. For example, research-oriented organizations, such as university hospitals, cancer facilities, and children's hospitals, will require more planning and adjustment than a 50-bed rural primary care facility.

Implementation Costs

One downfall of any project is the failure to consider the cost to the organization. Training costs would include the cost of tuition/registration fees for courses and includes the time lost from work while on the training courses. Overtime or agency costs may be incurred to remove backlogs prior to and post-implementation while the regular work continues. Some training may result in overtime to keep up with the regular day-to-day business while the staff is out at training. Hardware and software costs associated with the conversion need to be considered. New software may need to be purchased, or there may be costs associated with the upgrade of the current systems. There may be hardware costs incurred to accommodate the new software. Again, there may training costs associated with the new software or hardware.

Conclusion

Although there are a great number of challenges presented with the implementation of ICD-10, there are also many opportunities. There is the possibility for more enhanced reporting with the more detailed codes. There is also the opportunity to review business processes and information systems as they relate to data collection and reporting. ICD-10 provides a good classification scheme once the system has all of the technical data concerns addressed as previously mentioned. The system needs time to mature in Canada, especially the new CCI classification system for interventions and the grouping and case-weight methodologies.

In summary, there are four points to remember as individuals begin to implement and work with ICD-10:

• Preparation and planning is key to the success of the implementation.
• Sufficient education is a must.
• Understand the limitations of working with the systems (ICD-9 and ICD-10) in tandem.
• Learn from the experience of others.

References

Canadian Institute for Health Information. (1999). CCICD-10 Canada-Implementation Tool Kit. Ottawa, Ontario: Author.

Canadian Institute for Health Information. (2001). Coding with ICD-10-CA and CCI . Ottawa, Ontario: Author.

Canadian Institute for Health Information. (2001). Introduction to ICD-10-CA/CCI-Self Learning Program. Ottawa, Ontario: Author.

Canadian Institute for Health Information. (2003). Coping with the implementation of ICD-10 CA and CCI: Impact of new classification systems on the assignment of case mix groups/day procedure groups using fiscal 2001-2002 data. Retrieved May 13, 2004, from http://secure.cihi.ca/cihiweb/dispPage.jsp?cw_page=casemix_ICDimpact_e

Canadian Institute for Health Information. (2003). Coping with the implementation of ICD-10 CA and CCI: Impact of new classification systems on the assignment of case mix groups/day procedure groups using fiscal 2002-2003 data. Retrieved May 13, 2004, from http://secure.cihi.ca/cihiweb/dispPage.jsp?cw_page=casemix_ICDimpact03_e

Fletcher, J. (2000) ICD-10-CA/CCI Classification Primer . New Westminster, British Columbia: Douglas College.

Grant, T., Powell, S, and Steinbeck, B. (2002). Preparing for ICD-10-PCS. For The Record, 14 (25).