The financial repercussions of the COVID-19 pandemic are considered secondary to the health crisis, but there’s more overlap between them than you may think. A viewpoint published in the Journal of the American Medical Association cites the challenge of running a financially solvent hospital during staffing shortages, overwhelming demand, and a pause in the outpatient and elective procedures that often make up the bulk of hospital revenue.
As the U.S. emerges from the pandemic, the financial impact to healthcare institutions of all sizes is significant. Delayed preventative treatments led to worsening population health, resulting in longer hospital stays. Economic conditions are fragile and exacerbated by the great resignation and inflation, which are putting more financial pressures on healthcare institutions. Many hospitals are laying off staff to meet budget needs. All the while, regulatory audits have become even more scrutinous. All of this contributes to the need for revenue cycle optimization.
One way to improve efficiency is to rely on technologies (such as artificial intelligence ) to automate tedious manual processes. Using technology to improve the speed and accuracy of coding allows people to shift their focus and perform more complicated tasks, such as gathering clinical documentation.
Clinical documentation has always been a challenge. Providers often document patient issues differently and use shorthand only they understand. Documentation clarity is critical for proper reimbursement. 56% of healthcare professionals believe clinical workflow optimization is obstructed because clinical documentation is unstructured data.
Most clinical documentation is stored in an electronic health record. Organizations evaluate the documentation accompanying each patient encounter. Payers add pressure by denying reimbursement to providers who fail to prove the necessity of services. New payment models containing risk scores have also made it critical for providers to document underlying issues during visits, creating additional complexity and opportunities for process improvements.
The AI approach
Machine learning, natural language processing, computer vision, and, to a degree, robotic process automation all fall under the umbrella of artificial intelligence. These subcategories require an implementation road map, including reference content to teach an algorithm how to process raw data and turn it into useful information. NLP, for example, helps computers read and process written information, such as that contained in a clinical knowledge graph.
An NLP solution can accurately determine the relationship between diagnosis codes, charge codes, prescription drug names, body parts, and syntax. Once the baseline capabilities have been established, the content can be updated with abbreviations, reference language, and nuances unique to a facility’s EHR template or a physician’s notation preferences.
AI possesses all the raw capabilities mentioned above, but further refinement and integration are critical. There’s a big difference between being able to read something and understanding it. Clinical NLP powered by knowledge graphs is capable of scanning charts for specific words and identifying them. However, it cannot understand the meaning of those words in context as a physician, nurse, or coder could.
Adding additional layers of analysis could be the answer. We’re within a few years of AI being able to evaluate