Technological breakthroughs have led to advances in diagnosis and treatments.  And technology is an asset to the business side of healthcare too. Software makes it more efficient to manage clinical information and run the business of medicine.

So it’s curious that one of the problems identified during the ICD-10 National Pilot Program is that medical coders relied too much on computer-assisted coding (CAC) applications.

CAC was supposed to make medical coding easier and boost medical coding productivity. If a healthcare provider has already started using a CAC system, it can raise productivity to a level that compensates for the drop caused by ICD-10 implementation.

CAC can do a lot for healthcare providers:

  • Increase medical coding productivity and efficiency
  • Increase medical coding consistency
  • Create a medical coding audit trail
  • Create data queries
  • Allow more comprehensive medical code assignment
  • Improve medical coding accuracy
  • Decrease medical coding costs
  • Use free text for recording documentation
  • Improve systems through feedback

But the National Pilot program suggests that CAC doesn’t do so much for medical coding accuracy and productivity. CAC can be an asset in the ICD-10 transition if:

  • Templates and interfaces are built to do the job properly.
  • Electronic health records (EHR) templates are customized to maximize clinical documentation improvement (CDI) initiative.

The National Pilot Program probably was infected by GIGO (Garbage in. Garbage out.)

Paul Weygandt (Nuance Services) noted the problem with using CAC systems in ICD-10 implementation strategies. “Those systems don’t work when physicians don’t enter the correct information to generate ICD-10 codes. There are some structurally obvious coding queries which could be generated by computer assisted coding,” he wrote. “But the breakdown in the system will occur if the physicians are simply unaware of the content they need to provide for accurate coding.”

It’s also worth noting that the Cleveland Clinic tested medical coders who used CAC to help assign medical codes to cases and compared their work against unassisted medical coders. The study found:

  • CAC helps medical coders process claims more quickly.
  • CAC did not increase errors in medical claims that were more quickly processed by medical coders.
  • If CAC were only used to process medical claims with medical coders, errors increased.
  • As CAC systems “learned,” accuracy improved.

The Cleveland Clinic study reinforces the lesson that medical coders cannot rely on CAC systems. They need to verify the suggested codes even when clinical documentation is specific enough to support ICD-10 diagnoses.
Here at, we have always stressed that no matter the marvel of technology, coders still have to verify diagnoses.

If you are interested in learning the ICD-10 Coding Process, visit us at

This article first appeared in ICD-10