August 21, 2014

Technology and ICD-10

By Mark Morsch, MS, AHIMA-approved ICD-10 trainer

Technology continues to play an important role in supporting facilities and physicians engaged in the transition to ICD-10. Many organizations have purchased technology such as computer-assisted coding (CAC), clinical documentation improvement systems, natural language processing (NLP) and/or coding workflow solutions. These technologies hold the promise of significant benefits to key functions such as clinical documentation, coding, and revenue cycle operations.

 As our team at Optum360 collaborates with many different healthcare organizations and physicians, we are advising them to use the extra time from the ICD-10 delay to move ahead with technology implementations. This is an opportunity to go both broader and deeper to more thoroughly test new systems that are being implemented alongside existing systems requiring remediation for ICD-10.

For computer-assisted coding, the focus on testing covers both existing ICD-9 workflows for today as well as dual coding and future ICD-10 coding workflows. Prior to the announcement of the most recent delay, Optum360 was collaborating with CAC users who were dual coding as part of an ICD-10 NLP early adopter program. This program provides comparative analytics to show how ICD-10 coding compares to ICD-9 benchmarks. 

With the announcement of the delay, we extended the timeline and expanded the scope of the program to more users. Organizations using CAC are interested in coding results of the Optum LifeCode NLP for ICD-10-CM and ICD-10-PCS, and how those results can be optimized in preparation for the Oct. 1, 2015 implementation date. Dual coding with CAC allows the system to measure coder edits compared to the results from the NLP. This provides data for coding benchmarks while also allowing the opportunity for review of coder decisions to identify areas of potential additional ICD-10 education offerings and code usage clarification. Participants in the ICD-10 NLP early adopter program also use the results to identify documentation gaps and improvement opportunities. Cases that lack sufficient information to code to full specificity can be marked and utilized as examples for physician education. This is particularly important for ICD-10 procedure codes, which often require significantly more details than the similar ICD-9 procedure codes.

Leaders of both hospital and physician organizations are concerned by the potential revenue cycle impact of ICD-10. There are risks for potential delays in reimbursement, higher levels of claim rejections, and the potential for reduced reimbursement because of incorrect or incomplete documentation. This is an important area, and placing the focus on complete implementations of technology that will provide ICD-10 value is a sound practice.

Along with the implementations, going deeper into the analytics as part of testing programs for information systems will help both quantify and mitigate the risks. Analytics from dual coding using CAC can be used to help measure shifts in DRGs. Financial impacts also can be estimated, helping to justify specific actions to reduce negative effects. The follow-up actions can range from physician training on documentation improvement so complete ICD-10 procedure codes can be supported, to coder education to ensure codes for complications and comorbidities are captured fully and accurately in ICD-10, to simple estimation of the potential increase in physician queries and related coding delays.

With the extra time, more data can be collected and analyzed to help make better operational decisions and create improved financial forecasts.

Looking more broadly, Optum360 also is conducting business partner testing in which clients utilize systems and interfaces just as they would after the ICD-10 transition date. This testing measures how the systems used by an organization meet the requirements of ICD-10 from one end of the coding process to the other. Specific functional deficiencies are identified through this process, pointing to the systems, processes, or interfaces that require further remediation or are not functioning as expected. Business partner testing also validates that transactions flowing between provider organizations, Optum, and third parties are accurate. The breadth and depth of this testing is also intended to reduce unpredicted and "potentially unfavorable" financial impacts by catching and addressing system errors that could result in processing delays or inaccurate coding. We view this as key to revenue cycle readiness for ICD-10 – and an excellent way to use the additional time.

All of this testing relies on sound and complete data putting a spotlight on the quality of a system’s reporting and analytics. Technologies such as CAC and NLP are collecting much more data on coding accuracy, coder productivity, case mix, documentation quality, and operational efficiency than was available with older processes. It is important to spend time with this data and understand the implications for your organization. This means correctly interpreting the data to draw the appropriate conclusions about the level of your performance relating to key quality, operational, and financial metrics.

Data is useful only when it is actionable, so establishing goals that can be measured will help reinforce and build upon the successful use of technologies for ICD-10.

Focus areas for Optum360 through 2015 include analytics and reporting to measure results across multiple dimensions (coding, clinical documentation, operations, financial, etc.) and identify opportunities for provider organization risk and improvement related to ICD-10. 

  • Technology and tools related to ICD-10 transition –
  • Technology initiatives important for ICD-10 –
    • Computer-assisted coding
    • Clinical documentation improvement
    • Documentation-related technologies
    • Workflow
  • Guidance (move ahead with technology implementations in these four areas):
    • Delay time is an opportunity to go “broader and deeper” in testing of technologies
    • Validation of functionality through focused testing programs
    • ICD-10 early adoption
    • Business process testing
    • Focus on related analytics and reporting
  • Optum360 ICD-10 NLP Early Adopter Testing
    • Coding results – CM and PCS of Optum LifeCode NLP
    • Dual coding analytics – measuring coder edits versus NLP
    • Benchmark analysis – comparing coding results to current ICD-9 benchmarks
    • Coding feedback – review of ICD-10 coding decisions to identify opportunities for  education
    • Documentation gaps and opportunities – documentation missing or lacking specificity
    • DRG shifts based on dual-coding results
  • Business process (partner) testing
    • Validation of operational readiness for ICD-10
    • Identify and mitigate issues and risks prior to the October 2015 deadline
    • Reduce unpredicted and “potentially unfavorable" financial impacts from inaccurate ICD-10 code selection
    • Validate that transaction processing between your organization, Optum, and third parties deemed necessary parties is accurate
    • Testing and validation of end-to end systems handling ICD-10 transactions
    •  Identify systems and processes that require further remediation or are not functioning as expected
    • Key to revenue cycle readiness
  • Focus on analytics and reporting
    • Accurate data is essential – spend time with data and results to understand the meaning for coding, clinical documentation, operations and finances
    • Actionable data is key
    • Focus area for Optum in 2014 include analytics and reporting to measure results across multiple dimensions (coding, clinical documentation, operations, financial) and the identification of opportunities for organization risk and improvement related to ICD-10.

About the Author

Mark Morsch, MS is the vice president of technology for Optum360.

Mark has 20 years of experience as a software developer, specializing in computer-assisted coding, data abstraction and data mining. Mark is co-inventor of the patented LifeCode NLP engine that is used by over 160 healthcare organizations in the United States.  He is also an AHIMA-approved ICD-10 trainer.

Contact the Author

Comment on this article