The light at the end of the tunnel is not an oncoming train but ICD-10, according to one of our health information management leaders.

My organization has 64 computer systems to upgrade to the new coding structure by Oct. 1, 2013, along with testing about 125 interfaces to our payers and provider partners. Many believe that this will be a larger effort than Y2K; beyond just the technical hurdles, others believe that these computer upgrades pale in comparison to the work that will undergird the associated education, clinical documentation improvement and overall work flow changes.

All that's at stake is being paid for our services after the deadline and being paid based on complete and accurate ICD-10 codes. So there's plenty of reason for worry.

We began studying and planning our transition from ICD-9 to ICD-10 about two years ago. We formed our project governance, with executive sponsorship shared by one of our hospital CFOs and myself as chief information officer. Five work groups were formed to address applications, payers, provider affiliates, education and finance. We inventoried our affected applications and queried all partners for their plans. We developed a detailed project plan to manage the transition from software delivery in spring 2012 through four rounds of integration testing, leading to the October 2013 live date. To begin the necessary education, 16 of our most senior coders completed a comprehensive, Web-based ICD-10 curriculum.

An important element in any large project is identifying the risks—events that could jeopardize the project—and devising options should those risk events occur. Timely receipt of well-tested software upgrades is an obvious risk. We identified those applications most critical to billing and tagged them for the closest scrutiny.

As we progressed through the analysis, however, it became apparent that even larger risks surround attaining complete and accurate clinical documentation as the basis for coding. How do we educate our physicians on ICD-10 requirements and incentivize them to document in greater detail than may have been past practice? (In the words of some, first we made them clerks in adopting computerized physician order entry, and now we are making them coders!) That realization spawned an entirely new initiative around clinical documentation improvement.

The Advisory Board and other thought leaders have illustrated model work flows, starting at physician documentation and proceeding to coding, supported by automation. Tools such as speech recognition, controlled medical vocabulary, natural language processing and computer-assisted coding, all at varying levels of maturity, have been described and are at varying levels of maturity in leading hospitals around the country. These tools require integration with the electronic health-record system, where physician documentation occurs, on the front end and the grouper and billing system at the back end. They promise to significantly reduce the work required of physicians to complete their documentation.

Our team reviewed the leading emerging vendor products and visited hospitals using some or all the applications. The team observed that the tools, blended with solid supporting work flows, were successfully deployed in these model institutions. Besides demonstrating that the tools largely worked as intended, hospitals reported accurate billing, with no decline in net revenue. We concluded that these experiences can be extrapolated to ICD-10.

We are now in the final stages of determining documentation and product strategy. Clinical documentation improvement will entail engaging a broad constituency of physician leaders and supporting the effort aggressively in adopting the associated tool sets. Our goal will be to provide options for physicians to document in ways most suitable to their work requirements and personal preferences. A hybrid of structured and free text documentation is the likely result.

While this expansion in the scope of this project will be a huge hurdle for clinicians and IT alike, the benefits in terms of improved documentation, physician satisfaction and billing accuracy will be a significant payoff. Pulling it all off by Oct. 1, 2013? Well, we'll try to enjoy the challenges we'll be facing along the way.

Bill Spooner

Senior vice president and CIO

Sharp HealthCare

San Diego