Was Your Insurance Denial Made by AI? A Practical Checklist
The appeals process is not a formality. It is a genuine second review — and insurers are counting on most people won’t use it.
The actual humans working at insurance companies have been denying well-supported claims for decades. It's simply payor MO. But something shifted in the last few years: major insurers began deploying automated algorithms to review — and reject — claims at a scale no team of human reviewers could match. A Senate investigation found that UnitedHealthcare’s denial rate for post-acute rehab more than doubled after implementing automated review tools. Cigna’s system reportedly processed 300,000 denials in just two months.
Here’s what makes this worth paying attention to: the denials aren’t always substantively different from what a rushed human reviewer might produce. That’s what makes them hard to spot — and why the question “was this even reviewed by a person?” matters more than it might seem.
Less than 1% of denied claims are ever appealed — yet studies consistently show that over 80% of those appeals succeed in Medicare Advantage plans. The appeals process works. Most people simply don’t use it.
The checklist below won’t let you prove with certainty that AI made your decision. But it will help you identify the hallmarks of automated review — and give you the language to demand accountability in your appeal.
Signs Your Denial May Have Been Automated
Check each item that applies to your denial.
The more that apply, the stronger the case that your claim deserves a thorough human review.
☐ The denial came back unusually fast — within hours or even 1–2 days of submission.
Complex clinical reviews take time. A denial returned faster than a physician could realistically read the records is a red flag. I've worked on the provider side, submitting claims, and on the patient side, holding EOBs. Document the exact submission date and denial date — that timestamp alone can be significant.
☐ The denial letter is generic and doesn’t reference your specific clinical picture.
Compare the denial language to the actual records submitted. If the letter cites only general criteria — like InterQual or MCG/Milliman guidelines — without explaining how your specific findings failed to meet them, that’s meaningful. A human reviewer is expected to engage with the actual documentation. A checklist algorithm is not.
☐ Coverage was approved initially, then cut off at a suspiciously round number of days.
AI tools like UnitedHealthcare’s NaviHealth predict discharge dates based on population averages for a given procedure — not on the individual patient’s progress. Coverage terminating at a neat benchmark, despite clinical evidence of ongoing need, is a signature of algorithmic review.
☐ The denial doesn’t engage with your treating physician’s documented recommendations.
A human reviewer is expected to weigh the clinical judgment of the treating team. If the denial doesn’t address — or even acknowledge — what the attending physician documented, that’s a substantive gap worth challenging.
☐ The denial doesn’t account for the patient’s home situation or support circumstances.
Algorithms assessing discharge readiness rarely have data about what happens at home — whether someone lives alone, has stairs, lacks a caregiver. Stanford researchers studying this issue specifically noted that these tools almost never factor in a patient’s social supports. If those circumstances were documented and the denial ignores them, that’s worth raising.
☐ You cannot identify who reviewed the claim, or what specialty they hold.
Federal rules and many state laws require that clinical denials be reviewed by a physician with relevant specialty expertise. If the denial letter doesn’t identify the reviewer — or identifies someone whose specialty doesn’t match the clinical issue — you have grounds to ask. Request this information in writing.
☐ You’re on a Medicare Advantage plan and the denial involves post-acute or rehab care.
This is the highest-scrutiny area. A Senate Permanent Subcommittee on Investigations report confirmed that UnitedHealthcare, Humana, and CVS Health all increased denials for rehab and skilled nursing care after adopting automated review tools. Medicare Advantage post-acute denials are specifically where algorithmic review has been most documented — and most litigated.
What To Do Next
Identifying these signs doesn’t mean you can prove AI made the decision — insurers aren’t required to disclose this. What you should do is make the insurer defend their process.
A note on deadlines — don’t wait.
Appeal windows are strict and vary by plan. UnitedHealthcare commercial plans allow only 65 days from the denial date — significantly shorter than the 180 days allowed by Aetna, BCBS, and Cigna. Medicare Advantage appeals have their own separate timeline. If you’re unsure of your deadline, find out today.
Step 1: Request Confirmation of Physician Review
You have the right to know whether a licensed physician with relevant specialty expertise reviewed your denial.
Include this request in your appeal letter:
«The denial does not appear to reflect engagement with the specific clinical documentation submitted. The denial letter cites only general criteria without addressing the documented findings of the treating team. We respectfully request confirmation that this denial was reviewed by a licensed physician with relevant specialty expertise, and reserve the right to raise procedural objections if it was not.»
Step 2: Request the Specific Criteria Used
Ask for the exact clinical criteria — by name and version — that were applied to your denial.
Ask how your records were measured against them.
This forces specificity and creates a record if criteria were applied incorrectly or mechanically.
Step 3: Ask for a Peer-to-Peer Review
Most major insurers allow your treating physician to speak directly with the insurer's medical director.
Written appeals alone succeed about 67% of the time for UHC commercial plans.
Adding a peer-to-peer review pushes that rate up by 15–20 percentage points.
Ask your provider if they're willing before escalating further.
Why This Is Worth Fighting
The gap between how often appeals succeed and how often people actually try is one of the most important — and least discussed — facts in health insurance advocacy.
Less than 1% of denied claims from Medicare Advantage Plans are ever formally appealed. Yet 83% of appeals result in a full or partial overturn. For UnitedHealthcare commercial plans specifically, written appeals succeed roughly 67% of the time — and that rate climbs significantly when a peer-to-peer review is added.
The appeals process is not a formality. It is a genuine second review — and insurers are counting on most people won’t use it.
Sources: KFF analysis of CMS data; AMA 2024 Prior Authorization Survey; NAIC 2024 survey of large health insurers; U.S. Senate Permanent Subcommittee on Investigations, October 2024.
Bio:
Nadene Bradburn is the Founding President of Elder Care Advocates. The company helps families make clear, informed choices when elder care becomes complex, urgent, or emotionally charged. We offer structured guidance through situations where medical, legal, financial, and emotional factors collide — starting with a comprehensive assessment and strategy report that brings the full picture into view.
If you’re navigating a denial and aren’t sure where to start, a consultation can help you understand your options before deadlines pass. To learn more, visit my website at The Elder Care Advocate, which helps families make clear, informed choices when elder care becomes complex, urgent, or emotionally charged. We offer structured guidance through situations where medical, legal, financial, and emotional factors all collide--starting with a comprehensive assessment and strategy report that brings the full picture into view. To learn more, visit my website at https://theeldercareadvocate.com