Accused, Then Abandoned: The Empty Chair in Medicare Overpayment Appeals
Imagine a prosecutor who files charges, seizes the defendant’s assets before trial, and then never appears in court. No opening statement, no witnesses, no cross-examination.
The defendant hires counsel assembles the record, and argues to an empty table while the charging document sits in the file as though it proves itself. In any criminal court, that case would be dismissed.
Before a Medicare Administrative Law Judge, it is routine: the contractor that alleged the overpayment, and often has already begun recouping the money, need not show up to defend its own accusation.
The process usually starts with a RAC or UPIC audit. A Medicare contractor reviews a sample of claims, decides some were not properly supported, and often extrapolates that finding across a much larger universe of claims. A modest disagreement over documentation can become an overpayment demand worth hundreds of thousands of dollars or more. The provider must then work through redetermination and reconsideration, two levels of review run by the same contractor system that made the accusation, before it can reach a quasi-independent ALJ. READ MORE
The AI Healthcare Promise Meets the Policy Reality
AI has been widely touted as a way to make healthcare more efficient, but the technology may increase costs before it delivers savings. That’s the warning from the Centers for Medicare & Medicaid Services (CMS) Administrator, who said last week that the short-term impact of AI on healthcare spending is likely to be inflationary because it can “turbocharge” existing billing systems.
This concern is already playing out in medical coding. The BCBS Association recently estimated that about $1 billion in additional healthcare costs may have resulted from AI-powered coding tools that classify more patients as medically complex. And insurers argue that the higher costs are not being accompanied by an increase in treatment. Vendors serving healthcare providers, however, dispute the characterization, saying AI is improving the accuracy of medical bills, rather than encouraging upcoding.
The debate highlights an important distinction: AI can make healthcare operations more efficient without necessarily making healthcare less expensive. READ MORE
Inside the Coder’s Mind
If you are a frequent listener to Talk Ten Tuesdays, and we certainly hope you are, you know we always ask for questions in the chat and comments on what you would like to hear about. Two weeks ago, a listener noted they would like to hear about what coders should have in mind when looking at providers’ clinical documentation.
As coders, we are expected to know not only clinical information, but the coding guidelines and official guidance, in order to assign for these conditions correctly. If you are a coder, you know you are a lifelong learner. New conditions arise, coding instructional notes change, and of course, we get new codes, and existing codes can be revised or deleted (a timely reminder, as we are in the first week of October).
With that being said, we follow our guidelines to accurately assign codes based on the supporting documentation we review. For example, the coder may review a case in which a patient presents with urinary frequency and burning on urination. We can note where a urinalysis was performed and see greater than 100,000 CFU/mL of E. coli, with orders to start an antibiotic. In the coder’s mind, we are already thinking that this patient probably has a urinary tract infection, but without the provider’s diagnostic statement, we can’t automatically assign N39.0, with the additional code identifying E. coli. We can use this information as clinical indicators for a non-leading query, asking the provider to document any appropriate diagnosis. READ MORE