What's Happening?
Health insurance companies, including Cigna and UnitedHealthcare, are facing federal litigation and increased scrutiny over their use of algorithms to deny patient claims. These automated systems, such as Cigna's PXDX and UnitedHealthcare's nH Predict,
process hundreds of thousands of claims in seconds, often with minimal human review. For instance, Cigna's PXDX system reportedly denied over 300,000 claims in two months in 2022, with doctors signing off on batches of denials in as little as 1.2 seconds per claim. UnitedHealthcare's nH Predict tool estimates the length of post-hospital care, and a proposed class action alleges it overrides physician recommendations. A significant finding is that over 90% of denials tied to nH Predict are reversed upon appeal, yet only about 0.2% of affected Medicare Advantage enrollees actually file an appeal. This indicates a system where errors are frequently overturned when challenged, but most patients do not contest the initial denial.
Why It's Important?
The widespread use of algorithms by health insurers to deny claims has profound implications for patient care and the healthcare system. The high reversal rate on appealed denials, coupled with a very low appeal rate, suggests that many patients are being unfairly denied necessary medical services. This places a significant burden on patients and their families, who are often in vulnerable situations mid-treatment or during a health crisis, making it difficult for them to navigate the appeals process. The practice can lead to premature discharge from facilities or force families to deplete savings to cover care that physicians deem necessary. Furthermore, the financial incentives for private vendors in Medicare's new AI-driven prior authorization model, where they receive a share of savings from denied claims, raise concerns about potential conflicts of interest and the accuracy of denial decisions. This shift from human medical professionals to algorithms in critical coverage decisions impacts patient access to care and trust in the insurance system.
What's Next?
Federal litigation against Cigna and UnitedHealthcare is ongoing, with judges allowing breach-of-fiduciary-duty and breach-of-contract claims to proceed. UnitedHealthcare has been ordered to provide nearly a decade of internal records related to its AI use in claims adjudication. Concurrently, the Centers for Medicare & Medicaid Services (CMS) launched the WISeR initiative in January 2026, testing AI and machine learning for prior authorization in traditional Medicare in six states. This model, which pays private vendors a share of savings from denied claims, is facing legislative challenges, including a proposed 'Ban AI Denials in Medicare Act.' Health law attorneys are advising patients to explicitly ask if an algorithm was involved in a denial, request specific coverage criteria, and appeal promptly. Regulators are also implementing new rules, such as CMS-0057-F, requiring standardized electronic prior authorization and public reporting of denial rates for various health plans, aiming for greater transparency and faster response times.
Beyond the Headlines
The increasing reliance on artificial intelligence in health insurance claims processing raises fundamental ethical and legal questions about accountability and patient autonomy. While AI can streamline administrative tasks and flag potential fraud, its application in making critical medical necessity judgments shifts decision-making power from medical professionals to automated systems. The documented pattern of high reversal rates on appeal highlights a systemic issue where algorithms, designed for cost containment, may be generating erroneous denials that disproportionately affect vulnerable populations. This trend could erode patient trust in healthcare providers and insurers, fostering a perception that financial motives outweigh patient well-being. The ongoing legal battles and legislative efforts reflect a broader societal struggle to define the appropriate role of AI in sensitive areas like healthcare, ensuring that technological advancements serve humanity rather than creating new barriers to essential services. The debate underscores the need for robust oversight, transparency, and patient advocacy in an increasingly algorithm-driven healthcare landscape.
