The implementation of the new ICD-10 codes to replace the ancient ICD-9 codes for procedures and diagnosis is required by Oct. 1, 2013. That means that all reimbursements on Sept. 30 have to be in ICD-9 coding and at midnight a big switch has to occur to change the way business is performed.
As John Glaser, CEO of Siemens, mentioned in his presentation at the HIMSS session on new technologies, the best, and maybe only way to create disruptive innovation in the United States healthcare system is by changing the payment method. For other countries where the government is directly involved with the delivery of healthcare, it might be different, that is probably why Canada was able to complete the switch ten years ago in 2001. The good news is that most vendors who serve the Canadian market are actually US based, so one might expect that at least on the vendor side, the impact might be minimal. Nevertheless, there will be major changes needed: additional drop-down menus and selections have to be available when coding the procedures because of the increased granularity that is inherent to the new code set.
Based on the Canadian experience, there seems to be three different approaches to meeting the deadline. The first one is the transformational approach, which means taking the opportunity to make changes in the workflow, and reengineer the way that information is captured. This not only applies to implementing a new code system, but also when implementing any major change, such as an EHR for example.
There is no question that such large-scale change has is a major impact that can temporarily reduce productivity. Some institutions report that even one year after EHR implementation, they still are not at the pre-EHR level of productivity. I would argue that there are two reasons for this. The first is selecting the wrong solution for your needs, and the second is failing to reengineer the workflow. I was discussing EHR implementation with a physician who complained that he had to turn his back to his patient to enter information from a keyboard in a text-based questionnaire while taking basic information. This is a failure to select the right solution for the task. A much better solution might have been to replace the paper “clipboard” that he had been using, with a tablet computer that has a touch screen interface and allows for easy drop-downs and data entry. There are vendors on the market that are able to meet these requirements.
In addition to the transformational approach to meet a given deadline, there is the tactical approach. This basically means doing as little as possible, while just trying to “keep the lights on.” While doing this may allow continued operation, albeit with reduced effectiveness, it will prevent taking advantage of the major benefits of a new technology.
The approach that is probably best is the blended approach, which is based on cherry picking a couple of opportunities that may be easiest to implement while having the greatest impact. For example, one of the meaningful use requirements is the implementation of Computerized Physician Order Entry or CPOE. One might take the opportunity to reengineer this radically to take advantage of the new technology and the new coding.
Implementing a new technology requires a readiness analysis. This consists of several components. First is an assessment of whether the systems ready, which means in the case of a new coding system, making sure that your applications are capable of entering this information to the degree and granularity needed. Second, the technology infrastructure has to be ready. For example, reporting systems have to be modified, and libraries have to be uploaded. The Canadian experience suggests that converting and/or migrating the old data to the new coding can create more problems than necessary. It is a lot of work and the new implementation will take so much effort by itself that it is not worth spending any time on that. As an example converting old data to the new coding, resulted in very disjointed reporting that obscures trends because comparison data will be based on different information, i.e. from before and after the switch.
Education is also a major factor in creating smooth transitions with major change. The Canadian experience showed that a five-day computer based self-study coupled with two-day face-to-face training was insufficient for coders to do their jobs effectively. One also should not underestimate the amount of training needed for physicians, nurses and technologists. For example a cardiologist may know by heart the major codes used as of today, but he or she has to start from scratch learning the new ones. Experience has also shown that knowing ICD-9 actually has proven to be a hindrance to learning the new codes as new graduates, who never knew any ICD-9, performed much better than those who had a command of the old system. It appears to be easier to start with a clean slate than to carry any old luggage.
In conclusion, the experience from Canada provides valuable lessons that can be used to apply to the upcoming changes in the US coding systems. Most of these lessons can be generally applied to any changes in technology. The most important take-away is not to panic, and, if needed, take a tactical approach and do the bare minimum. However, to fully utilize a new technology, it is recommended to take a blended approach and implement workflow changes while re-engineering the department. A recurring theme is that you cannot train your staff enough and that the time it takes to thoroughly train them, is almost always under estimated.
Herman Oosterwijk, VP OTech Media, live from HIMSS2011