The AI Revolution in Medical Coding: Friend or Foe?

An interesting article from Bertalan Meskó, MD, PhD about an emerging technology that may impact medical/clinical coding.

Medical coding - the process of translating medical records into standardized codes - is a crucial but laborious task in healthcare. With coder shortages looming and errors common with manual coding, many view AI automation as the logical next step. But will AI truly improve medical coding or replace human jobs?

The Promise of AI Coding

AI promises huge potential benefits for medical coding through natural language processing. Companies like Nym, Fathom and CodaMetrix claim AI can autonomously code charts in seconds with up to 96% accuracy, reducing manual labor by 70%. This could greatly increase efficiency and cost savings for healthcare organizations struggling with backlogs and coder shortages. AI can take on the tedious, repetitive coding tasks, freeing up human coders to focus on more complex cases.

Concerns About Replacing Human Coders

However, some worry widespread AI adoption could replace human coders entirely. Given estimates of a 30% coder shortage in the US, is it ethical to replace these professionals with machines? Proponents counter that AI will instead augment and enhance coders' abilities. Just as calculators didn't replace mathematicians, AI will allow coders to operate at the top of their licenses as auditors and quality controllers.

The Need for Human-AI Collaboration

The key is striking the right balance between human expertise and AI productivity. Coders will need to become "more of an auditor in this job, more than just a coder." The humans must oversee AI training data for accuracy and provide ongoing quality checks. In turn, AI can handle large volumes rapidly, with people resolving exceptions. Together, human coders and AI may form an unstoppable revenue cycle optimization team.

Preparing for an AI Future

AI automation will likely be an inevitable next chapter in medical coding. For a successful integration, healthcare organizations need quality data to effectively train AI tools. Coders should receive training on using and auditing AI systems. With the right preparation and perspective, the humans and machines can work side by side to take medical coding to the next level.

Follow me on LinkedIn