Last month I watched in amazement as NASA’s Perseverance Rover touched down softly on Mars. After traveling six and a half months and over 100 million miles, Perseverance aimed for the Jezero Crater, a 5-by-4-mile area featuring dangerous pits, cliffs and boulders. As I watched, I marveled at the precision of each calculated step that led to touchdown, and I celebrated (not always knowing why) whenever the NASA team cheered. NASA prepared for years and applied innovative, automated technology that mapped and analyzed rough terrain to find a precise spot for a flawless landing.
Being a technology geek as well as someone who spends his time thinking about how to streamline revenue cycle workflow and processes, naturally I started to think about denial management. OK, it’s not the Mars landing, but it is a complex topic, made mission critical by the unprecedented strain on cash flow at healthcare entities across the nation.
The new technologies that allowed NASA to hit its target to perfection were not available for the early efforts at landing rovers or stationary devices, a few of which crash-landed. In healthcare revenue cycle today, our applications and processes are closer to NASA’s older technology. We need to advance denial management into “Mars-landing denial science.”
Hitting the wrong spot
Traditional denial management depends on data from remittance advice and other sources, and the denials are typically mapped to categories to identify trends. With this data, and in the spirit of “collaboration throughout the revenue cycle,” denials are assignedto the department that caused them, such as health information management, utilization review, clinical service areas and registration. These departments are responsible for investigating the root cause of the denial and implementing improvements to ensure future claims won’t meet the same fate.
What if Perseverance landed on Mars, but missed the Jezero Crater? NASA selected this spot following five years’ research because Jezero offered the most promise for uncovering whether Mars held life billions of years ago. Missing it by a small margin would have ruined the mission. The problem with traditional denial management is that it frequently lands in the wrong spot – in areas that did not cause the denial and/or in the wrong follow-up work queue. The fault lies in problematic data.
The data problem
Denial data is inconsistent and potentially misleading. Under HIPAA, the government established national standards for electronic transactions, including codes that explain denials. However, between codes changing periodically and payers interpreting them differently, we’ve found that the categorization is not simple and cannot be static. In a 90-day period for one of our clients, a reason for denial based on “non-covered service” included eligibility (50%), coding or coverage (45%) and insurance and other issues (5%).
Based upon our client data, payer inconsistency could affect from 10% to 30% of denials that healthcare providers receive. NASA would not be satisfied with anything like that kind of failure rate. When data are faulty, denial management teams make inaccurate assignments and conclusions, including holding the wrong departments accountable.
Denial science
The industry is ready for denial scientists and next-generation technology embedded in the business office.
Denial scientists apply scientific method to identify and correct data anomalies. Scientific method consists of making observations, formulating hypotheses, testing hypotheses, drawing conclusions and refining hypotheses. It implies that there is potential to continuously evolve as new hypotheses lead to new conclusions.
As a simple example, consider seeing a large volume of denials related to revenue codes. In health systems, there is typically a manager responsible for the charge description master (CDM) to map revenue codes to each service provided. In traditional denial management, the revenue code denials would be automatically routed to the CDM manager to fix issues. However, in this case the denial scientists identify that only one insurance is sending this denial. It is rare for revenue codes to be mapped differently for each payer, so the hypothesis is that this was not a CDM manager issue. Investigation reveals this was a false denial from the insurance company.
In a much broader context, an approach utilizing denial scientists to investigate denial abnormalities ensures accurate data and enables departments to review clearly defined problems, potentially saving their work on 10% to 30% of accounts. The time saved translates to more time to focus on improving processes or, more importantly, on patients.
Artificial intelligence will likely be the future of denial management. I have noted previously that right now AI is more of a dream than a reality in revenue cycle, however much the term is bandied about. In the meantime, we need real-world solutions that:
Enhance the accuracy of reporting denials and denial trends
Accurately assign denials to responsible departments so they can identify and correct the root causes
Simplify/optimize denial workflow and training
A focus on data integrity through denial science can ensure these goals are achieved and denial improvement objectives land precisely where they should, like Perseverance.
In today’s healthcare revenue reality, such an outcome, however far from the headlines, would be something to cheer.