It’s certainly not breaking news that artificial intelligence (AI) is making inroads in healthcare. Nearly every new article on the topic seems to describe how AI is doing the work that heretofore was being accomplished by humans: nurses, administrators, lab technicians, and coders.
But just how effective is AI when it comes selecting evaluation and management (E&M) codes? A recent assessment conducted by Mazars USA LLP, an independent auditing firm, validated that the AI coding engine in the Calm Waters system recently achieved a 97-percent accuracy rate – two points better than the industry benchmark of 95 percent for human coders.
“This is powerful validation for our company and our technology solution,” Calm Waters AI President, Joe Ferro, said in a recent press release. “We are thrilled that a globally respected third party like Mazars has assessed that Calm Waters AI is as accurate as we have always believed and delivers the bottom-line results for provider groups that we have promised.”
Ferro went on to say that of the more than 1,000 providers Calm Waters serves and have audited, the company has found an average of a 63 – 65 percent accuracy rate across organizations of different sizes and different specialties.
“In seconds, Calm Waters AI delivers the bottom-line results for providers that we’ve promised,” said Ferro. “As a result, practice groups are optimizing chart accuracy and reducing audit risk, while gaining up to an hour per provider each day that would otherwise have been spent on documentation.”
Such high rates of accuracy would lend assurance in the potential for AI coding, particularly as it pertains to mitigating audit risk. For coders, it should generate a greater sense of confidence in the quality of the output, according to Angela Jordan, executive vice president of ChartPal, the E&M documentation service of Calm Waters AI, Inc.
Jordan also noted that coders’ responsibilities are likely to evolve when using the technology.
“Instead of spending their time pulling out components of the medical decision-making (MDM), they will simply validate the E&M level and then focus on other coding activities,” Jordan said in a statement to ICD10monitor. “I am thrilled that our AI is setting new standards in E&M accuracy. The results from this review underscore our commitment to being seriously precise. Calm Waters AI is not just advancing technology, we are making a significant, positive impact on the business of healthcare.”
