Key Takeaway: While encoders and computer-assisted coding (CAC) systems both offer gains in efficiency compared to manual medical coding, it is clear that neither technology compares to autonomous medical coding when it comes to the improvements in coding quality, speed, scalability, and cost for healthcare organizations.
This article provides an overview of the different medical coding technologies available today, diving into their unique capabilities, advantages, and shortcomings.
Traditional encoders, which were first developed in the early-to-mid 1990s, have long been a staple in medical coding workflows. These software tools provide medical coders with access to extensive code libraries and resources, using a terminology tree to help medical coders assign the most appropriate codes to a patient encounter.
As the oldest medical coding software currently available, encoders are a familiar and accessible tool for many of today's coders. While they can improve coder productivity by assisting with code selection, encoders still require coders to manually navigate through the terminology tree, validate their coding decisions, and assign the final code, ultimately resulting in a time-consuming process.
Computer-assisted coding (CAC) systems started to emerge in the early 2000s, and up until very recently, they were the most sophisticated medical coding tools available. CAC systems clinical documentation to suggest potential codes based on natural language processing (NLP) and coding rules. These types of systems have evolved since their initial development, and today, many CAC systems include work queue and flow management capabilities in addition to coding suggestions.
CAC systems were a significant improvement upon encoders and brought the medical coding process one step closer to full automation. According to a study by the American Health Information Management Association (AHIMA), CAC systems can increase coder productivity by up to 20 percent compared to coders who are not leveraging CAC.
Despite the apparent productivity gains, CAC systems don’t always result in improved efficiency and are notorious for falling short when it comes to accuracy and scalability. As noted in a 2019 literature review, “...productivity impacts [of CAC systems] vary widely, depending on the specific deployment. Some studies reported a drop in productivity when medical coders were forced to validate, and frequently eliminate, a large number of suggested codes.”
Additionally, CAC systems are limited in terms of their scalability. Because they require humans to validate every code suggestion before a patient encounter is sent to billing, CAC systems do little to solve the estimated 30 percent shortage of medical coders. Revenue cycle leaders must find solutions that do not depend so heavily on humans so that their teams are in a position to seamlessly support organization expansions and growing patient volumes.
Thoughts from Nym's Coding Compliance Officer, Paul Wojnar, MPA, CPC, CPCO, CRC, CEMC.
“While CAC systems guide decisions, they can highlight text that may influence inexperienced coders to select charges or diagnoses that are not fully supported by documentation or coding guidelines. In this way, the benefits of CAC rely on unified coding understanding as well as unified documentation practices. This is especially evident in large, centralized coding departments where turnover and other issues create learning/experience gaps in coding understanding.”
The latest advancement in medical coding technology is autonomous medical coding. Powered by multiple subfields of artificial intelligence (AI) such as natural language processing, machine learning, and deep learning, the purpose of autonomous coding solutions is to accurately assign codes to patient encounters and send them to billing, all without any human intervention.
Autonomous medical coding has been very well-received by health systems, hospitals, and other healthcare organizations, and interest continues to increase year after year. A recent survey revealed that around 60 percent of healthcare organizations either use autonomous coding or plan to, with half of the respondents who plan to incorporate the technology intending to adopt a solution within six to 12 months. By truly automating medical coding with an autonomous coding solution, healthcare organizations can expect to see significant improvements in coding speed, quality, cost, and operational efficiency
However, like any emerging technology in healthcare, autonomous coding technology has room for improvement. Depending on the solution or vendor, the implementation of autonomous coding solutions can be a lengthy process. Additionally, autonomous coding solutions are only capable of automating a certain percentage of coding volume (typically between 40 to 90 percent depending on the specialty area), but it should be noted that this coverage percentage increases the longer a solution is up and running.
Thoughts from Nym’s Coding Compliance Officer, Paul Wojnar:
“I think it’s important to call out that the future of autonomous medical coding is not just about coding. What I mean by that is that clinical documentation improvement, audit, and education processes need to be developed, or adjusted, with autonomous solutions in mind. Things like provider templates, orders, flowsheets, and other elements should be standardized and clarified in preparation for this new technology.”
At the end of the day, autonomous coding solutions are the first and only medical coding solutions to enable full medical coding automation. While encoders and CAC will continue to have their use cases in certain care settings, all signs point to autonomous coding being the gold standard in medical coding technology moving forward.
Interested in learning more about autonomous medical coding with Nym? Check out these helpful resources!
Learn more about Nym's Autonomous Medical Coding Engine
Get in touch with a member of our team
Schedule a demo and see Nym's engine in action