D'Arcy Gue

ICD-10 and Computer Assisted Coding

October 28, 2014

ICD-10 6 Minute Read

It appears that Computer Assisted Coding (CAC) applications are about to experience an influx of growth. According to a HIMSS Analytics report, profiling 25 support service applications, CAC shows to have the highest growth potential. *Note: the HIMSS Analytics report is only available to HIMSS Analytics members.

This growth is almost certainly in response to the projected productivity impact of ICD-10. As the deadline gets closer, many healthcare organizations are examining Computer Assisted Coding Technology (CAC). Given the increased interest, it’s important that hospitals truly understand how CAC works and set realistic expectations for how the technology will impact coding productivity. 

CAC uses computer software to generate a set of medical codes for review, validation, and use by the coder, based upon provider clinical documentation. The integration of CAC technology into the coding shop is a complex process that ultimately involves using CAC tools to review all of the electronic documentation in a patient’s chart before the chart is seen by a coder, and providing the coder with proposed codes to work with when finalizing the coding of the chart. The general idea is that the computer does the more “tedious” work of reviewing many pages of documentation, and that the coder is then free to spend more time making the careful judgments about which diagnoses are appropriate to code and in what sequence to accurately represent a clinical visit.

It should be made clear that, while CAC provides technological assistance in the uniform assignment of valid codes and descriptions, it does not generally replace the role of coders. The assistance frees the coder to focus on the review and validation of the CAC output, and other high-value tasks such as ensuring compliance with correct coding initiatives, CMS and other payer specific policies, and local coding rules prior to code acceptance and billing. Because of these benefits to both the facility and the coder, AHIMA has recommended broad adoption of CAC technology.

how  computer assisted coding worksHow It Works

CAC software applications process clinical information from electronic documents and generate codes for validation by medical professionals. Generally, this is done using some form of Natural Language Processing, which employs complex algorithms to recognize language patterns, generate codes, and enable querying electronic text. The logic to do so can be very complex. Imagine the following four clinical scenarios:

  • The patient has long-standing CHF, which does not appear to have changed in severity.
  • The patient has new-onset CHF.
  • The patient does not have any signs of CHF.
  • The patient will be referred to the CHF for follow-up.  (Where the CHF is the coronary health facility)

Each of these four sentences uses the term CHF (which in the first three cases is a reference to congestive heart failure), but the interpretation of the four options is significantly different.


  • CAC as an ICD-10 training tool — Most CAC products analyze the clinical language contained within the chart for both ICD-9 and ICD-10 codes, and the coder has the option to review those recommended codes side by side. This presents the coders with feedback about potential ICD-10 codes, even before the transition date.
  • Productivity — As I mentioned previously, CAC is being evaluated specifically as a tool to address coding productivity, and there are numerous case studies suggesting that productivity may increase from 10 to 15, or even 20 percent after the implementation of CAC tools. Of particular note is that the tool seems to most benefit slower coders.
  • Accuracy — Many facilities are considering CAC as a tool to help mitigate the risk for increased coding errors because of unfamiliarity with the new codes.
  • Case Mix — One thing that CAC does particularly well is identification of secondary diagnoses, some of which may be mentioned in only one place in the chart.    Because the CAC technology does so well at identifying complicating conditions and secondary diagnoses, many facilities find that they see an improvement in their case mix index after implementing CAC. University of Pittsburgh Medical Center implemented CAC software in three hospitals and found that their case mix rose 8 percent, a $22 million increase in revenue. Gwinnett Hospital System in Lawrenceville, GA., saw a 3.8 percent in case mix in the first year after they implemented CAC. 

Practical Considerations

New Technology

CAC applications are still an evolving technology. Although research and development in the technology has been going on for years, it’s only in the last three years, with a proliferation of EMRs in the US because of the Meaningful Use / MIPS incentive program, that the technology has been deployed in a broad manner. According to Klass research’s 2012 report, 21 percent of providers already use CAC applications, and another 15 percent plan to purchase in the next two years


CAC technology is not cheap. Hardware, Initial licensing fees, ongoing maintenance fees, and needed IT support can easily run over $500,000 a year for a smaller health system (1,000 beds), and even though smaller hospitals generally pay less in licensing fees for the software, the per-chart cost is often higher. Many hospitals find that the ROI on the projects is actually very reasonable, however, because of the expected increases in case mix and decreases in overtime, use of outsourced coders, and higher coder productivity. CAC vendors have a variety of pricing strategies available to help address the burden of the up-front costs,  so the forward-looking hospital should not see cost as an absolute barrier until analysis has been performed.

Electronic Record Availability

To reap maximum benefit from CAC, it is important that most of the clinical record be available electronically. Many facilities have not yet implemented EMRs in all of their facilities, or have only partial implementations (such as progress notes not being available). Facilities need to evaluate their particular EMR configuration in order to determine if they can fully realize the benefits of the technology.


The advantages of CAC technology and the realities of coder productivity under ICD-10 have caused a dramatic increase in hospitals wanting to implement the technology, and most large CAC vendors prefer a three to six month implementation cycle once staffing and hardware are available. As a result, most facilities who haven’t made the decision to implement CAC yet will probably not be able to do so before October 1, 2014.

Bottom Line

The decision to implement CAC technology is not a simple one. Facilities need to do a complete analysis of the potential cost savings and case mix benefits before committing to CAC as part of their coding strategy, and the strategic implications of the implementation must not be overlooked.

When performing analysis of this nature, it’s important to have experienced advisors involved. If you’re in the process of accessing your coding strategy and considering CAC, Phoenix can help. Our ICD-10 experts are happy to provide advisement. If you’re interested, check out our Services page or contact us.

Related Posts