How Cigna Saves Millions by Having Its Doctors Reject Claims Without Reading Them

ProPublica's investigation into Cigna's PxDx system revealed something that many physician practices suspected but could not prove: a major health insurer was using an algorithm to deny claims at industrial scale without individual medical review. The reporting found that Cigna doctors were spending an average of 1.2 seconds per claim review, batch-processing denials for over 300,000 claims in just two months.

For physician practices on the receiving end of these denials, the investigation confirmed a frustrating reality — that the denial notices arriving in their inboxes were not the product of careful clinical evaluation, but of an automated system designed to reject claims based on diagnosis-procedure code combinations, regardless of the patient's individual clinical circumstances.

How PxDx Worked

The PxDx system flagged claims based on whether a particular diagnosis code was typically associated with a particular procedure code. If the combination fell outside the algorithm's expected patterns, the claim was routed for denial. Cigna's medical directors could then approve or deny these flagged claims — but the volume made meaningful individual review practically impossible, hence the 1.2-second average review time.

This is not clinical review. It is rubber-stamping algorithmic output. And it raises fundamental questions about the integrity of the denial process that every physician practice should be asking.

What This Means for Your Practice

If your practice has experienced unexplained denials from Cigna — particularly denials that seemed disconnected from the clinical reality of the patient encounter — this investigation provides important context. The denials may not have been based on a clinical judgment about your patient's care. They may have been based on a statistical model that decided the diagnosis-procedure combination was unlikely to be correct, without anyone examining the actual medical record.

This has direct implications for how practices approach appeals:

• Denials generated by algorithmic systems can often be overturned by providing the individual clinical evidence the algorithm ignored — the patient's specific history, examination findings, and the clinical reasoning supporting the treatment.
• Appeal language should address the specific clinical circumstances that make the denied service medically necessary for this particular patient, regardless of whether the diagnosis-procedure combination fits a statistical norm.
• Practices should track denial patterns by payer to identify when denials appear to be systematically generated rather than individually reviewed.

The Systemic Problem

The ProPublica investigation exposed a specific instance of a broader trend: the use of automated systems to manage claim adjudication at scale. When a payer processes millions of claims, the economic incentive to automate denials is significant. Each denied claim that goes uncontested represents retained revenue for the payer and lost revenue for the practice.

The math works in the payer's favor because most practices do not appeal most denials. Industry data consistently shows that fewer than half of denied claims are appealed, and in some settings, the appeal rate is below 10 percent. An algorithmic denial system that generates even a modest percentage of denials that go uncontested can save the payer substantial sums.

Documentation as Defense

For physician practices, the response to algorithmic denials is not resignation — it is documentation. The clinical record is the strongest counter-argument to a system that denies claims without reading them. When your documentation clearly establishes medical necessity with specific clinical findings, relevant history, and guideline-based reasoning, you have a foundation for appeal that can withstand scrutiny at every level of review.

The practices that fare best against automated denial systems are those that treat documentation not as a billing afterthought but as a clinical and financial asset — because when the payer's algorithm says no without looking, it is the medical record that says yes.

Regulatory and Legal Implications

The ProPublica investigation triggered congressional inquiries, class action litigation, and increased regulatory scrutiny of algorithmic claim processing. These developments are worth watching, but physician practices should not wait for regulation to solve the problem. The tools for contesting algorithmic denials are available now: thorough documentation, systematic appeal processes, and data-driven tracking of denial patterns that reveal when payer behavior crosses the line from clinical judgment to automated rejection.