What is Artificial Intelligence?
Artificial Intelligence (AI) is defined as computers imitating “intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience” with the ability to achieve goals. AI in healthcare is known to “help manage and analyze data, make decisions, and conduct conversations, so it is destined to drastically change clinicians’ roles and everyday practices.”
How Can AI Help Manage Claim Denials?
A study conducted by Change Healthcare shows that $262 billion of providers’ medical claims revenue is denied out of a total of nearly $3 trillion. In other words, hospitals and independent physician practices need to rework about 9%, on average, of their claims.
Although providers can recover, on average, 63% of denied claims, it takes an extensive amount of resources. Providers can spend as much as $8.6 billion in administrative costs nationwide on the recovery of denied claims. The high cost of recovery causes about 65% of denied claims to not even be evaluated and resubmitted.
The problem of medical claim denials is complex, requiring a more intricate solution than simply analyzing claims through a rules-based engine. To enhance the denial management segment of the revenue cycle, it is a necessity to understand the complex relationship between different clusters of ICD 10 codes, external codes, and state-specific eligibility rules of different payers. The majority of providers rely on human intelligence to identify and comprehend these clusters and make changes to the claims accordingly. Artificial intelligence (AI) is one solution to automating the claims correction process.
AI models have the ability to handle large, complex sets of data to learn established rules, recognize patterns, and make predictions on whether a claim will be denied or accepted. This allows claim adjudicators to see real-time results and gives them a chance to correct any possible denials before sending them to the payers, saving some of the administrative costs associated with claim denials.
Other Ways AI Can Help with Revenue Cycle
Prior authorizations are one of providers’ biggest challenges when ensuring smooth revenue cycle management. In a study conducted by the American Medical Association (AMA), one participant indicated dissatisfaction with prior authorizations by explaining that “…prior authorizations kill us. Ugh, the amount of time we spend on prior authorizations.”
The AMA surveyed 1000 physicians to learn the impact prior authorization has had on their practices. 86% of physicians described the burden of prior authorizations within their practice as high or extremely high, with an increase over the past 5 years. Those surveyed also reported an average of about 2 business days per week spent dealing with prior authorizations.
An estimated 80% of claim denials are attributed to the complexity of prior authorization. The automation of prior authorization processes can help reduce the burden, yet the adoption of electronic solutions for prior authorization remains low. The Council for Affordable Quality Healthcare, Inc. (CAQH) reported in their 2019 CAQH Index® that certain fears such as vendor response and lack of standards prevent the adoption.
AI technology can be used to identify procedures that need prior authorization, submit requests to payers, and keep providers updated, in real-time, on the status of the requests, thus substantially increasing patient satisfaction and patient care.