Melinda Tully
Angela Carmichael

At a Glance

Computer-assisted coding tools have the potential to drive improvements in seven areas:

  • Transparency of coding
  • Productivity (generally by 20 to 25 percent for inpatient claims)
  • Accuracy (by improving specificity of documentation)
  • Cost containment (by reducing overtime expenses, audit fees, and denials)
  • Compliance
  • Efficiency
  • Consistency

With the upcoming transition to ICD-10 by Oct. 1, 2014, many hospitals and health systems are considering the adoption of computer-assisted coding to offset the expected declines in productivity that could accompany ICD-10 implementation.

The sheer increase in volume of codes with ICD-10 is anticipated to cause coding productivity to decline initially by as much as 50 percent with the implementation of ICD-10, based on Canada’s experience with the transition to ICD-10. The number of diagnosis codes will increase from 14,567 to 69,833 codes, while inpatient procedure codes will jump from 4,000 to 71,918.

Knowing this outlook, healthcare finance leaders face difficult spending decisions: Hire more coders? Hire clinical documentation specialists? Adopt new technology to ease the transition—and if so, what types?

Training is another concern: Working with the new ICD-10 data will require coders to have more advanced clinical understanding and biomedical training, including a better grasp of medical terminology, anatomy and physiology, pathophysiology, and pharmacology. Should a hospital or health system outsource training for ICD-10, or invest in trainers in-house to meet the demands for staff education?

With hospitals and health systems already operating on razor-thin margins, a drop in productivity for even a few months can be a tough proposition. That is why, according to surveys by research firm KLAS, nearly half of healthcare providers plan to purchase a computer-assisted coding product within the next two years. The hope is that computer-assisted coding will mitigate at least a portion of the anticipated drop in productivity.

Healthcare organizations have used computer-assisted coding to the greatest extent on the outpatient side of health care, and they also have used the technology retrospectively. However, the concerns about lost productivity are prompting finance leaders to consider concurrent computer-assisted coding. In the inpatient setting, computer-assisted coding can be a helpful tool within the context of a larger coding and documentation process and workflow, but only when adopting this tool is not the organization’s first course of action.

Advantages of Computer-Assisted Coding

On the inpatient side, concurrent computer-assisted coding could play an increasing role in boosting and managing productivity, and indeed, it historically has been viewed as a tool for enhancing coder productivity. But, as noted previously, the focus to date has been on its use in outpatient settings and retrospectively.

The benefits of computer-assisted coding also can extend beyond improving productivity to helping improve data integrity, consistency, and payment accuracy—but only when it is considered in tandem with a hospital’s or health system’s clinical documentation program. When computer-assisted coding is coupled with a strong clinical documentation program on the inpatient side, it can help to improve:

  • Transparency (by helping to identify documentation that supports the assignment of a code, so that proper identification can more easily be verified)
  • Productivity (generally by 20 to 25 percent for inpatient claims)
  • Accuracy (with a corresponding decrease in audit findings)
  • Cost containment (by reducing overtime expenses, audit fees, and denials)
  • Compliance (by improving specificity of documentation)
  • Efficiency (by streamlining code assignment and reducing the need to review records electronically or on paper)
  • Consistency

It is important to note that the level of benefits that computer-assisted coding brings to the inpatient setting will not equal those achieved in ambulatory settings for a number of reasons:

  • Workflows differ between settings, given that on the inpatient side, information is pulled from a greater number of sources and in a variety of formats (e.g., text-based, scanned, handwritten).
  • The volume of information for review and vocabulary are much greater in the inpatient setting than in the ambulatory setting.
  • Coding guidelines are more extensive in the inpatient setting (e.g., guidelines for present on admission, sequencing, abstraction, and grouping).
  • Codes are more complex and numerous in the inpatient setting than in ambulatory settings.

According to the American Health Information Management Association (AHIMA), computer-assisted coding “works best where there is a limited number of source documents that must be analyzed for code selection” (Delving into Computer-Assisted Coding, AHIMA practice brief, 2004).

Clinical Support Critical to Success

No matter how sophisticated the tool, a computer-assisted coding program is only as good as the documentation in the patient record, because codes are assigned by these programs based on such documentation. This is a classic “garbage in, garbage out” scenario.

Especially in light of the transition to ICD-10, having a clinical documentation improvement program in place is essential to ensuring the integrity of coding. Computer-assisted coding tools require clinical oversight. If a code cannot be validated in real time and clinical support for determining the appropriate code is not available, claims will be at risk of being denied or audited.

Clinical documentation specialists can help to clarify documentation through discussions with physicians. They can ensure that clinical support exists for diagnoses and obtain details related to the severity of the illness, the presence of a condition upon admission, and more. Such discussions have the potential to enhance the credibility and validity of the final code assignment.

For example, imagine that a patient has been admitted and the computer-assisted coding tool identifies a diagnosis of sepsis based on the documentation in the patient record. At this point, the clinical documentation specialist will become a clinical investigator of the suggested diagnosis. The specialist must review the chart and decide: Is the diagnosis clinically supported? Is there other conflicting documentation? Is there overall consistency in the documentation?

Let’s say that the clinical documentation specialist discovers that the patient has an indwelling urinary catheter—a detail that was included in the patient’s chart but not detected by the computer-assisted coding tool. The documentation specialist now must query the physician to determine whether a complication from the indwelling catheter caused the sepsis infection; if it did, a claim for sepsis would be denied.

In this and other instances, the clinical documentation specialist becomes both a user and benefactor of the computer-assisted coding tool. This type of tool does not sequence information in the chart, so the documentation specialist must review the chart to determine the most appropriate principal diagnosis based on myriad complex coding guidelines. In effect, computer-assisted coding tools enable clinical documentation specialists to make higher-level decisions at a quicker pace.

Best Practices for Implementation

Hospitals and health systems should wait to implement computer-assisted coding tools until they have strong clinical documentation programs in place. Then, they should introduce the tools first in the outpatient setting and later in the inpatient setting, supported by concurrent coding and documentation.

Computer-assisted coding tools should not be used for retrospective coding, as doing so will only exacerbate problems with coding efficiency. Requests for clarification from coders will undoubtedly increase after the transition to ICD-10—particularly for procedure reporting.

Coding and documentation specialists should work with computer-assisted coding tools in tandem, with coders validating the codes assigned while clinical documentation specialists concurrently verify that the clinical support and specificity required are included. Coders understand the guidelines and payment methodology associated with a diagnosis, while documentation specialists understand the clinical indicators, risk factors, and impact on care and management that supports the final diagnosis. Computer-assisted coding tools allow both coding and documentation professionals to do what they do best and work collaboratively toward the assignment of diagnosis-related groups (DRGs). If there is a mismatch between the final DRG suggested by the coder and the DRG suggested by the documentation specialist, computer-assisted coding and queries to the patient’s physician can help the two specialists arrive at an agreed-upon code by the time the patient is discharged.

Computer-assisted coding tools also can help clinical documentation specialists perform their duties more efficiently by highlighting terms and sections in the chart that relate to the assigned code. When a documentation specialist revisits a chart days later and sees that the assigned code has changed, the specialist can more easily track the progression of events that led to that changed code assignment.

In the past, concurrent coding has failed when coders lacking the clinical expertise and physician relationships needed to conduct an effective review have been brought onto the floor and asked to delve deeper into patient charts.

In sum, the new computer-assisted coding model constitutes a collaboration between two disciplines, represented by the clinical documentation specialist and the coder, and supported by software designed to speed up code assignment and to show links to the clinical documentation required to support each code. The clinical documentation specialist works on the floor as part of the clinical team, making sure code assignment is validated by content, and the coder focuses on the guidelines for compliance and payment.

Staff Are Still Key to Accurate Documentation

One of the positive aspects of the delayed deadline for ICD-10 implementation is that organizations have more time to assess the computer-assisted coding tools and to build or reinforce their clinical documentation programs. It is important to find best-of-breed solutions for both computer-assisted coding and clinical documentation improvement, even if that means using a separate vendor for each. But even more important is ensuring that the staff who are using and applying these tools possess the knowledge and skills needed to gain the most value from them.

Computer-assisted coding will not replace clinically trained documentation specialists and coders who must review, validate, and oversee the tasks related to the final dropped bill. However, such tools could better support documentation specialists in applying their clinical, critical thinking skills to review and analyze data in patient charts—which ultimately has the potential to improve productivity, efficiency, and compliance while reducing costs and improving revenue.

Computer-assisted coding tools are only as good as the people using them and applying them to improve the end product. As Bill Gates said, “If you give people tools, and they use their natural abilities and their curiosity, they will develop things in ways that will surprise you very much beyond what you might have expected.”

Melinda Tully, MSN, CCDS, CDIP, is senior vice president, clinical services and education, J.A. Thomas and Associates, Atlanta, and a member of HFMA’s Georgia Chapter (

Angela Carmichael, RHIA, CDIP, CCS, CCS-P, was formerly director, health information management compliance, J.A. Thomas and Associates, Atlanta. 

Publication Date: Monday, October 01, 2012

Login Required

If you are an existing member, please log in below. Username and password are required.



Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:


   Become an HFMA member instead