AI in Medical Billing and Coding: Reducing Errors and Alleviating Staff Burnout
With medical billing and coding, timeliness and precision are of paramount importance. AI can improve both.
Medical billers and coders must process claims as quickly and accurately as possible, explains Steven Carpenter, a billing and coding instructor at the University of Texas at San Antonio. “If they don’t code properly, then the claim can be denied or suspended, and that in turn affects the revenue cycle and profitability,” Carpenter says.
But instead of poring through the ICD-10-CM and medical records to identify the right codes for each and every diagnosis, coders can use AI to swiftly and correctly identify the appropriate, up-to-date codes.
“Manual billing and coding can be repetitive, error-prone and mentally taxing, especially when coders must handle high volumes of data under tight deadlines,” Carpenter says. “AI can help improve the speed and the accuracy.”
In addition to recommending codes, AI tools can analyze and validate claims, automate claim submissions, verify insurance eligibility, and get prior authorizations and other documents to payers — ultimately optimizing revenue cycle management.
“Revenue cycle management has a lot of moving parts, and on both the payer and provider side, there’s a lot of opportunity for automation,” says Aditya Bhasin, vice president of software design and development at Stanford Health Care. “There are huge benefits of leveraging this technology to remove friction from the system.”
EXPLORE: Overcome AI implementation hurdles in healthcare.
Reducing Errors and Alleviating Staff BurnoutMedical billing involves more than assigning the correct codes. Billing teams also must interact with patients to address continual queries about insurance coverage and medical charges. Stanford Health Care found that AI could augment the capabilities of its billing staff and, in the process, save time and avoid burnout.
“In healthcare, there has been a lot of focus on burnout among physicians and how AI can help clinically, but we also noticed that our billing folks have a lot of back and forth with patients and get really complex queries from them,” Bhasin says.
Every day, Stanford Health Care’s billing representatives respond to hundreds of questions from patients via the organization’s online portal. In the past, the billing reps responded to these written queries by consulting 25 templates, selecting the most appropriate one, and then personalizing it to respond to each patient query.
Recognizing that this time-consuming and repetitive activity could lead to staff burnout, Stanford Health Care this year piloted an AI tool that quickly generates a draft response for each query. The tool considers myriad variables, including the patient’s insurance policy, and then creates an accurate, well-formulated response that reflects Stanford Health Care’s brand and voice. “That work was done manually, and now all of it is generated by the AI tool,” Bhasin says.
After Stanford Health Care launched the pilot in January, 10 billing customer representatives used the AI model to process 1,000 patient billing messages. The tool saved the reps about one minute per message — adding up to about 17 hours over the two-month pilot. “We saw substantial time savings,” Bhasin says. “It’s about augmenting the people who have to do a lot of manual work and making their day-to-day lives easier.”
When Stanford Health Care surveyed its billing reps about the AI tool, “they were super excited about it,” Bhasin says. As of March, Stanford Health Care’s entire billing staff was using the AI tool.
“One of AI’s most valuable contributions is its ability to alleviate staff burnout,” Carpenter says.
healthtechmagazine
