A Special Report by Peter Borans and Dr. Stacy Borans
Let’s get this out of the way: Auto-adjudication does save money. It allows TPAs to do more with fewer people. Auto-adjudication processes plow through a large number of claims cheaply and swiftly. The industry likely could not exist without it. But it’s not a perfect substitute for human intervention, not by a long shot. It seems ironic that a tool that is designed to save money ends up costing payers – but it does, every day. Complexity is the watchword here. Even the most sophisticated auto-adjudication procedures are not designed to handle complex cases.
Bad things can happen when nobody is watching.
The Bad Things
Auto-adjudication attempts to find patterns and flag claims for review often based on incomplete or inconsistent data. It is a substitute for qualified claim review. It mimics a qualified set of human eyes, but not nearly as well.
Take high-cost claims for complicated health issues, like inpatient stays. Auto-adjudication, even the best set ones, assumes that all costs are A) Necessary, B) Truthful, and C) Cost about the right amount. In the majority of cases, this makes sense. But in a payor chain with ample opportunity for human error, such as you find in catastrophic or complex claims, it doesn’t make a lot of sense to leave reimbursement decisions up to an algorithm.
For example, no patient is ever immediately discharged from the ICU, their level of care is appropriately downgraded to less intensive and less costly levels as their health improves. Yet claims are consistently filed that do not accommodate this level of care, requesting reimbursement for an ICU stay until the moment of discharge, and many auto-adjudication suites miss this because they are just not set up to catch that simple nuance. Billing codes and diagnosis codes match, and preauthorization may have been obtained, so the claim passes auto-adjudication, even though it is unlikely to have occurred as-billed.
A human eye would see it; many auto-adjudication setups miss it.
This is only the most egregious sort of auto-adjudication miss. What auto-adjudication missing is the ability to actually check and tie the clinical and financial pieces of a claim. Check and Balance things like medical necessity, cost, FDA approval. It’s just not adequate compared to a good claim examiner or a third party. And with a fiduciary responsibility to clients (and carriers looking over their shoulders), “not adequate” isn’t a good place to be.
A Rock and a Hard Place
TPAs are in a tight spot. Administrators must walk the line of efficiency and meeting mandated claim turnaround while at the same time performing rigorous review to effect cost containment. Auto-adjudication can and should be sufficient for 80%-85% of all claims coming through (though that number may be shrinking). It gets a little less sufficient when it comes to complex and catastrophic claims: large hospital bills, chemotherapy, transplants. Systems are built to look for specific codes and value settings, not to examine nuance or notice patterns.
This is a problem when the complexity of claims is increasing. On the one hand, you have increasing complexity coming from rapid advances in medicine. On the other, you have increasing costs across the entire spectrum of healthcare. In the middle you have auto-adjudication technology that was not built for a geometric growth in complexity and must be constantly updated in order to function as a cost-controller.
But updating auto-adjudication processes will only flag them for independent review. What happens when a larger percentage of claims starts becoming complex, or when rapid coding and billing changes further outpace TPAs? When does auto-adjudication stop being an asset and start turning into a liability?
The answer is almost certainly outside experts will need to step in to guide TPAs and other payors to make the best decisions possible.
What’s The True Cost of Auto-adjudication?
The true cost of auto-adjudication would be difficult to measure. Industry estimates place overpayment on specialty drug claims alone at somewhere between between 3% and 10%, and inappropriate payment on critical care to be almost 23%. None of this includes money spent on recovery, manual examination of systemic auto-adjudication problems, provider services costs, and the myriad other expenses that cascade from commonplace auto-adjudication inadequacy. Similarly it is difficult to measure money lost from a failure to notice patterns like multiple claims with high cost and low discounts, or whatever might be happening on the less-scrutinized in-network side of things.
These problems will only be exacerbated as healthcare costs increase, medical technology advances, and healthcare reform strains claim processing systems. Fast, reliable, expert manual review by human beings will necessarily grow to be the determining factor, and the need for payers to be able to accurately predict costs and program auto-adjudication processes so that discrepancies and problematic patterns can be noticed and corrected will be key.
It’s time for us to realize that auto adjudication isn’t the great savings future we always hoped it would be. Humans are always necessary when claims are complex. The sooner we come to terms with that, the stronger our industry will become.
Advanced Medical Strategies provides physician lead claim auditing, medical review, and diagnosis cost prediction services to the healthcare payor industry. Contact us to learn more about our tools to help your claims processing team spot and eliminate overpayment.