Reverse Engineering Payer Behavior to Improve Denied Claims Management

Denials are one of the most persistent problems in the revenue cycle and its about to get worse – a lot worse with the rollout of ICD-10.

The reason it is bad actually has little to do with why its about to get worse, but denials and ICD-10 are united in a common solution – something called machine intelligence.

Approximately 5% of net revenue is denied or rejected. While stunningly high, it does provide a massive data foundation to evaluate the problem from.

The labyrinth of payer rules has proven an ineffective guide to providers. Their application is highly uneven – resulting in pockets of tribal knowledge of what claims are likely to be denied, despite rules that explicitly state otherwise.  

This is because most revenue cycle management systems are dependent on a query based approach to solving this problem. What if the code looked like this…or what if the sequence of codes was that?

The challenge here is the concept of combinatorial complexity. Because there are so many different variables at play (patient, procedure, location, doctor, sequencing, payer, adjudicator, rules to name a fraction) the possibility of finding the right answer are few and far between.

That is why the problem still exists.  

The answer lies in a learning system that is designed to find the answers by detecting all of the relationships associated with the data.

Ayasdi’s Allison Gilmore delivers a superb take on the problem, its underlying complexity and how to tackle it using a combination of machine intelligence, process and people.  It is must watch video if you are anywhere in the revenue cycle.