On 9th February 2019, the American Medical Association (AMA) published the revised Current Procedural Terminology evaluation and management (CPT E/M) coding guidelines. Over the past couple of decades, since its establishment in 1995, E/M codes have been the single source of truth for physician-patient encounters. These codes concisely capture the information of the interaction in numeric codes and are key to facilitate billing. Hence complete and compliant E&M coding is critical for physicians and clinics.
However, in the years since its inception, the E/M coding procedure has received mixed reactions from the physician community. The utility of the process is undeniable, but so is the growing burden it imposes on physicians.
Physicians face an increasing battery of measurements meant to drive quality, satisfaction, and efficiency. The combination of electronic health records (EHRs) and the documentation burden caused an overload of E/M documentation requirements. This in turn resulted in physicians spending time away from patient care as well as inaccuracy and “note bloat”.
The updated CPT E/M coding guidelines, coming into effect on 1st January 2021, were published with the primary objective of reducing administrative burden on physicians.
But the healthcare industry can realize ways to achieve this even before the new AMA guidelines come into effect. And one such way is the application of AI-powered computer assisted coding (CAC). The applicability of AI is clearly depicted in a report that projects thevalue of AI in medical coding automation to hit USD 14 billion by 2026.
The Need for AI-powered CAC in E/M Coding
The healthcare industry is currently at a crossroads. The widespread adoption of value-based care models in a healthcare consumerism ecosystem along with increasingly stringent regulations is acting as a fulcrum for a shift in industry practices. And accurate documentation has become the key driver of hospital revenue systems.
The implementation of ICD-10 in 2015 further emphasized this fact. Many hospitals across the country are faced with complex revenue cycle management. For outpatient visits, the revenue is a direct function of the accuracy of E/M coding. With legacy infrastructure falling short of driving revenues through improved E/M documentation, healthcare organizations need to embrace more effective coding tools.
Enter, computer assisted coding (CAC).
Computer-assisted coding or CAC is not a singular technology, rather the amalgamation of various facets of AI and NLP. NLP-powered, CAC software is capable of scrutinizing and interpreting unstructured physician notes. Using dedicated linguistic algorithms, the software can extract the clinical facts and subsequently assign the appropriate E/M code. Physicians, therefore, no longer need to undertake the rigorous task of identifying and extracting data from the documents and feeding it into the system.
Voice-based NLP improves the process further. With dictation being one of the largest sources of unstructured medical data, CAC software can leverage voice-based NLP’s speech recognition capability to extrapolate the right E/M codes. This significantly reduces the coding burden on physicians while simultaneously enhancing the accuracy of the process. This allows for better matches both E/M coding guidelines as well as payer reporting systems, thereby significantly improving the billing process and decreasing denials and audit discrepancies. Moreover, AI-powered CAC provides real-time feedback to physicians, and the healthcare system, regarding the precision of E/M codes of patient evaluations during outpatient visits.
The ezDI Advantage
The needs and benefits of computer-assisted coding are quite clear. However, simply going for CAC software does not yield the desired results. Take the instance of Richmond University Medical Center (RUMC) in Staten Island, New York that faced a situation common across the country’s hospital community. Their existing encoder capabilities were not able to provide adequate workflow functionality. RUMC needed a comprehensive CAC framework that would not only allow them to improve productivity and accuracy of the coding process but also ensure regulatory and payer compliance.
Partnering with ezDI to work towards this objective, RUMC noticed significant results. Complex denials dropped by 13% while coder productivity increased by 33%.
Combining the capabilities of NLP and machine learning (ML) with E/M logic algorithm allows ezDI’s CAC solution (ezCAC) to automate E/M code assignment as well as suggest case history code. The automation capabilities can be further extended to assign examination types as well as complexity levels of outpatient evaluation. Moreover, the accompanying E/M calculator helps coders assign the correct E/M code level.
Partnering with ezDI does not just mean having technology by your side to improve your E/M coding process. Unlike other providers, we are with you every step of the way. We take full, end-to-own ownership of the process and can implement a solution that works for you in just 4-6 weeks.
To learn more about ezDI’s AI-based mid-revenue cycle management solutions visit www.ezdi.com and to see the live product demo of our Clinical Documentation, Coding, Compliance/Auditing, Quality Measures, Encoder, and Enterprise Analytics request a live demo.