Every so often a datapoint lands that reframes a whole category of denials. The latest comes from federal oversight of Medicare Advantage: as reported by the AMA on July 29, Medicare Advantage organizations overturned 95 percent of prior authorization denials for skilled nursing facility admissions that were appealed. The same reports found one payer denied 80 percent of long-term acute care hospital admission requests - and that only about a third of patients appealed their denials at all.
Sit with that pair of numbers for a moment. A denial that gets appealed almost always flips. Two-thirds of denials are never contested. The gap between those facts is where post-acute revenue goes to die.
The reason the overturn rate is so high is worth being clear-eyed about: these denials are largely administrative screens, not clinical judgments. When nearly every appealed denial reverses, the initial denial is not expressing a considered coverage position - it is pricing the odds that nobody will push back. That is why the OIG findings matter for budgeting: the recoverable share of prior auth denials is knowable, large, and sitting unworked in queues that nobody owns.
Denials priced on non-participation
Payer adjudication is a business process, and businesses respond to incentives. When overturn rates run this high on appeal, the denial is not a coverage determination in any meaningful clinical sense - it is a filtering mechanism that works because most denials go unchallenged. The patients who appeal (or whose providers appeal for them) recover care or payment at extraordinary rates. The ones who do not, absorb the loss.
For providers, this changes the ethical and financial calculus of triage. A denial with a 95 percent historical overturn rate is not a maybe. It is near-certain recoverable revenue, and the only question is whether anyone does the work before the deadline.
Building an appeal-default operation
The operational gap is rarely willingness; it is capacity. Appeal work is document-intensive and deadline-bound, and it lands on the same staff who are already working the claim queue. Making appeal-by-default real requires taking three burdens off the humans:
- Denial capture. Every prior auth denial, from every portal, logged the day it appears - including the ones payers communicate only inside their own systems.
- Records assembly. The clinical documentation for the appeal pulled and organized automatically, because that is where the hours go.
- Deadline management. Appeal levels, windows, and escalation dates tracked per case, with nothing expiring quietly.
Budget discipline belongs in the same list. Appeals have a measurable recovery yield - in the OIG data, extraordinarily high - which means every unworked prior auth denial is budgetable lost revenue, not bad luck. Finance teams that credit recovered appeals against the cost of running the program usually find the program funds itself several times over. That framing also survives leadership turnover: "the appeals program pays for itself at a 10x margin" outlives any individual champion of it.
With those three in place, a trained appeal writer reviews and signs rather than assembles - which is the difference between an appeal program that exists and one that keeps up.
The compounding benefit is payer behavior. Denial patterns respond to resistance: payers that face systematic appeals on a service line adjust, because contested denials cost them review time too. Practices that appeal everything train their payers that denials on this population will be fought. Practices that appeal selectively train the opposite lesson. The queue you work today shapes the denials you receive next quarter.
The OIG findings reported by the AMA show MA organizations reversed 36 percent of appealed long-term acute care denials and 43 percent of appealed inpatient rehab denials - with SNF appeals overturning at 95 percent. The variance by service line is why every case needs working, not sampling.
The takeaway
The Oversight Committee's data confirms what post-acute billing teams have long suspected: these denials are largely arbitrary, and the appeal process is the real coverage determination. Two-thirds of denials going unappealed is a capacity failure, not a choice - and it is the most addressable revenue leak in post-acute billing. Automation is how appeal-by-default stops being a slogan: capture every denial, assemble every records package, track every deadline, and put a human decision behind all of it.