In partnership with
Medicare Advantage plans denied 12 percent of standard prior authorization requests in 2025, and then they reversed themselves on 67 percent of the denials patients bothered to appeal. Read that number twice. The plan got its first answer wrong two times out of three whenever somebody pushed back, which means the no was never a clinical judgment, it was an opening offer. On October 15, Medicare open enrollment opens. A 72 year old woman will sit at her kitchen table with a plan comparison in front of her, she will see a premium that dropped, and she will see nothing at all about whether that plan intends to approve her hip.
THE DATA
CMS published its 2027 landscape release on September 28 and built the headline around price. The weighted average monthly Medicare Advantage premium falls from $14.37 in 2026 to $12.00 in 2027, a decline of 16.5 percent. Enrollment keeps climbing, with CMS estimating 34 million people in Medicare Advantage in 2027, roughly 47.4 percent of everyone on Medicare. Part D moves the other way. CMS projects the total average monthly stand alone Part D premium rising from $35.09 to $36, and the agency plans to discontinue the Part D Premium Stabilization Demonstration in 2027. Stable, by the measure CMS chose to report.
Now put the denial data next to it. KFF published an analysis on August 13, 2026 covering 2025 plan reported data from six Medicare Advantage insurers representing 25 million enrollees, the first year CMS required plans to report these numbers publicly. Those plans denied 12 percent of standard prior authorization requests and 10 percent of expedited ones. They reversed 67 percent of the denials that reached appeal. The pattern holds across other markets and runs worse. ACA Marketplace plans denied 18 percent of standard requests and reversed 43 percent on appeal. Medicaid managed care plans denied 14 percent and reversed 47 percent.
Then there is the report that should have ended the argument. HHS Office of Inspector General, report OEI-09-24-00331, issued June 8, 2026, looked at June 2024 skilled nursing facility admission requests across 19 Medicare Advantage organizations. Those organizations denied 12 percent of SNF admission requests, with individual plan rates running from 23 percent down to 0.4 percent for the same benefit under the same program. When enrollees appealed, the plans reversed themselves 95 percent of the time. Enrollees and providers appealed only 18 percent of denials. OIG notes no confidence intervals on the report page, so take the figures as the census they are rather than an estimate with error bars. One more line from that report deserves its own sentence. naviHealth processed half of all SNF admission requests, denied 14 percent overall, denied requests from people already living in nursing homes 40 percent of the time, and watched 97 percent of its denials get reversed.
WHERE THIS BREAKS
Start with the structure, because the structure explains more than anyone's intent does. A plan that denies first and reverses later collects the full value of every denial nobody appeals. OIG measured that gap precisely: 95 percent reversal, 18 percent appeal rate. The unappealed 82 percent carries the economics. Nobody needs a bad actor for that math to work, they just need a process that makes appealing hard and denying cheap. Layer the vendor on top and the accountability disappears entirely. naviHealth sat between the plan brand and the patient, took the hardest call in the system, and applied it hardest to nursing home residents, the people least able to file anything. CMS, for its part, measured what it can average. Premium dollars and plan counts fit in a press release. A denial that sends somebody home without rehab does not.
Speak naturally. Send without fixing.
Wispr Flow turns your voice into clean, professional text you can send the moment you stop talking. No rough transcription you have to clean up. Actual polished text, ready for email, Slack, or any app.
Speak the way you think. Go on tangents. Change your mind mid-sentence. Flow strips the filler, fixes the grammar, and gives you text that reads like you spent five minutes writing it.
Now the part that stings, and I am including my own field in it. CMS has required plans to publish prior authorization and appeal outcome data since March 31, 2026. That data sat public for six months. Open enrollment opens in ten days. Patient advocacy organizations and pharma advocacy teams will run hundreds of enrollment education sessions between now and December 7, and almost all of them will cover premiums, networks, and formularies. Almost none will hand a patient the plan level denial rate and appeal reversal rate. I have watched the field demand transparency in enough comment letters to know the sentence by heart, and the first year we actually got it, nobody built the comparison tool. The data came. The advocacy infrastructure did not.
That gap has a familiar shape. Archō has fielded ELAVAY annually since 2015 across more than 20 disease states, asking patient advocacy organizations, professional societies, and community based organizations how biopharmaceutical companies behave as partners. The finding that repeats, year over year, looks like this one: companies grade their own advocacy performance well, and the organizations they name as partners describe something different.
WHAT THE GOOD ONES DO
The organizations that get this right changed their counseling inputs, not their mission statement. They pull the plan level prior authorization metrics CMS now requires and build them directly into enrollment counseling, so a patient sees a denial rate and an appeal reversal rate sitting next to the premium, in the same table, at the same decision. They treat the appeal rate as a program metric and staff for it, because an 18 percent appeal rate tells you the problem lives in the filing, not in the patient's understanding. More education will not move that number. Somebody filing on the patient's behalf will.
The sharpest ones go one layer deeper and name the utilization management vendor, not just the health plan, in their policy comments and their patient materials. The vendor contract holds the denial logic. A comment letter that only names the plan leaves the actual decision engine out of the record, and the median organization writes exactly that letter every cycle, then wonders why the rule comes back unchanged.
WHAT TO DO THIS QUARTER
Three moves, each one assignable to a name.
First, give one person the job of downloading the 2025 prior authorization metrics for the five plans your patient population uses most, and publish a one page comparison before October 15.
Second, add one question to your intake or hotline script this quarter, was this service denied and did you appeal, then report the appeal rate monthly against the 18 percent OIG baseline and beat it.
Third, have your policy lead file comments on drug prior authorization rule making in the CMS-0062-P line that ask for denial and overturn reporting broken out by utilization management contractor, not only by plan.
If you want to know whether your advocacy function sits close enough to the decision to matter, take the Advocacy Influence Diagnostic. Free, anonymous, five minutes, and it returns an Advocacy Visibility Index and a Strategic Risk Score. The 2025/2026 ELAVAY Report is available now, and fielding for 2026/2027 opens this month. Reach me directly at [email protected].
