AI Tools Boost Efficiency, Patient Experience

Integrating AI-powered tools into medical coding brings valuable opportunities that offset the increasing pressure on health systems to optimize operational efficiencies.

But accurately and efficiently capturing a patient’s care is often a complex and time-intensive task for medical coders.

"As medical coding evolves, there are increasing opportunities to automate routine tasks that don't require human input, allowing coders to focus more on those tasks that truly benefit from human judgment. This helps us clearly distinguish between what can be automated and what needs a personal touch," says Nicholas Judd, MBA, RHIA, Cleveland Clinic Senior Director, Revenue Cycle Management and Health Information Management.

Medical coding encompasses a wide range of activities essential for accurate billing, reimbursement, reporting and maintaining patient records. Among this workflow is mid-revenue cycle coding — the step that ensures the care a patient receives is accurately documented and translated into standardized codes. These codes are vital for communicating the complexity and scope of treatment, delivering high-quality care across providers and settings, which supports continuity of care and informed clinical decision-making.

“Correct coding facilitates access to necessary services. It helps ensure that prior authorizations are approved and that patients don’t face unnecessary barriers to receiving care. When coding is inaccurate or incomplete, it can result in miscommunication or delays that negatively affect the patient’s experience,” explains Gina DeFranza, Cleveland Clinic Director, Coding and Reimbursement.

Judd adds, “Accurate and properly coded documentation helps prevent errors that could lead to inaccurate patient outcomes, delays in insurance processing, and unexpected charges for patients or denied claims. This contributes to providing safer, more effective care, a smoother financial experience, and reduces administrative burdens for patients, coders and providers.”

Choosing the right partner

At Cleveland Clinic, revenue cycle teams typically review over 100 clinical documents for each case — such as progress notes, discharge summaries and pathology reports — before selecting codes from a pool of more than 140,000 options. This process can take up to an hour for a single patient encounter.

“Finding the right AI tool evolved from a comprehensive evaluation process with internal input from the Strategy Office, Digital and revenue cycle management,” says Judd.

That’s when Cleveland Clinic partnered with AKASA to implement generative AI tools that assist medical coding practices. This collaboration aims to enhance efficiency, accuracy and the overall quality of patient care through advanced automation.

Together, the organizations are addressing the complexities of coding and documentation that occur between patient care and billing.

Navigating future potential

“Healthcare revenue cycle management is a complex, highly regulated domain,” says Judd. “Sophisticated tools like generative AI are essential to keep pace with the rapid changes. The challenge is managing expectations—balancing the hype with the reality of AI capabilities.”

"Early results show improved speed and accuracy compared to our legacy workflow,” says DeFranza. “By automating routine tasks, we’re freeing up caregivers to focus on more complex work that leans on their clinical expertise and critical thinking."

Judd adds, “Accurate coding is essential for continuity of care, patient safety and quality outcomes. Streamlining revenue cycle tasks reduces friction in processes such as prior authorizations and payer denials. Ultimately, this ensures that care delivery remains the central focus, improving the overall patient experience.”

Looking ahead, both Judd and DeFranza agree that shared learning and demonstrated success will drive further innovations and advancements in the revenue cycle industry.