The denial queue on a Tuesday morning runs a couple hundred fresh claims. The billing team works it the same way every week: high-dollar claims at the top, lower-dollar claims at the bottom. The $300 to $500 denials are aging toward appeal deadlines while the team focuses on what moves the AR. That triage math feels rational at the claim level, but that same triage mentality forfeits revenue at the portfolio level. A two-year litigation record and a congressional investigation both reveal why.
The courts offer the most direct evidence. In 2022, Cigna deployed an algorithm called PXDX that denied more than 300,000 claims over two months. Cigna physicians averaged 1.2 seconds of review per claim. One medical director denied 60,000 claims in a single month, according to the complaint in Kisting-Leung v. CIGNA Corp. The Eastern District of California allowed the class action to proceed, finding the allegation that Cigna violated its own plan terms sufficient to move to discovery. The complaint cited Kaiser Family Foundation (KFF) data showing only 0.2% of denied claims were appealed. The argument was direct: Cigna deployed PXDX knowing that low appeal rates would let most improper denials go unchallenged. Cigna disputes the characterizations. The complaint documents the core design principle: issue denials at scale, knowing the appeal rate is low enough to make most wrongful denials financially durable.
The UnitedHealth nH Predict case documents the same payer design in Medicare Advantage post-acute care. UHC's denial rate for post-acute care more than doubled, from 10.9% in 2020 to 22.7% in 2022. The timing tracks the rollout of nH Predict, a predictive algorithm built by an Optum subsidiary. The lawsuit alleges UnitedHealth pressured case managers to follow the algorithm's recommendations even when clinicians and families objected. Appeals overturned 90% of those denials. A federal magistrate ordered UnitedHealth as recently as March 2026 to produce internal documents on how their tool was built and deployed. Congress reached the same conclusion when the Senate Permanent Subcommittee on Investigations pulled more than 280,000 pages of internal documents from UnitedHealthcare, Humana, and CVS. Its October 2024 report found the three largest MA insurers were, in the subcommittee's words, intentionally targeting a costly area of medicine and substituting medical necessity judgment with a financial calculation.
The AMA's 2025 Prior Authorization Physician Survey, conducted in December 2025, confirms the system is working as built. Physicians complete 40 prior authorization requests per physician per week. The physicians and their teams spend 13 hours weekly managing the process. Nearly one in three reports their PA requests are often or always denied. The number that closes the argument: 59% of physicians do not appeal because past experience tells them it will not succeed.
That belief runs counter to what the overturn data shows. KFF's 2024 Medicare Advantage analysis puts the appeal overturn rate at 80.7%. That gap between physician belief and actual overturn rates did not develop by accident. It was built over years of administrative friction designed to make individual-claim appeal feel like a bad use of time.
For RCM directors and CFOs managing denial portfolios across a multi-site platform, understanding the design intent behind denial systems changes their required operational response. Staffing and workflow improvements matter. But workflow improvements alone will not close the revenue gap when algorithmic tools issue denials faster than any billing team can work them claim by claim.
UnitedHealthcare sets a 65-day commercial appeal window against the 180-day standard most other major payers use. That structural pressure compounds the triage problem. The platforms recovering the most from their denial portfolios address both causes: internal workflow that defaults to write-off rather than appeal, and external conditions built to make that default feel like the right call.