Prior Auth Is Now a Throughput Variable (Not Just a Billing Headache)

TL;DR: In 2023, Medicare Advantage (MA) plans made ~50 million prior-authorization determinations—about two per enrollee—and denied 3.2 million (6.4%) in whole or part. Only ~12% of denials were appealed, yet ~82% of those appeals were overturned. The real cost isn’t denial—it’s delay. With new CMS rules enforcing 7-day and 72-hour decision clocks, hospitals can now time and route discharges to minimize friction.

What the evidence actually shows

1. Volume is the signal.
   MA prior authorizations reached ~50M in 2023, up from ~42M in 2022—about two per enrollee. Nearly every discharge lane touches prior auth now, not just the complex ones. 1

2. For MA - Initial “no” often becomes “yes”, but delay becomes LOS.
   Only 11.7% of denials were appealed, yet 81.7% of appeals succeeded. That means most initial “no’s” would have become “yes” if pursued—just later. The lost time extends inpatient length of stay (LOS). 1

3. OIG confirmed inappropriate MA denials.
   A federal review found MA plans sometimes deny care that meets Traditional Medicare rules, due to plan-specific criteria or administrative errors. Fast, targeted re-submissions are often faster than appeals. 2

4. Post-acute prior auths face steep denial rates.
   In 2024, AMRPA reported 53% initial denial rates across 12,157 IRF/SNF PA requests—with >2-day waits common. Even overturned denials mean extra inpatient days. 3

5. The rules are shifting.
   CMS’s new Interoperability & Prior Authorization Final Rule (CMS-0057-F) requires MA plans to:

   • Deliver decisions in 7 days (standard) or 72 hours (expedited)
   • Publish PA metrics by 2026
   • Provide PA APIs for transparency and tracking 4

   Another rule, CMS-4201-F, narrows how plans can use PA and adds continuity-of-care protections. 5

Why care management teams are uniquely exposed

• MA is now 51% of all Medicare enrollment—it dominates the discharge lane.
• CM teams control order timing and placement sequencing, both of which influence whether a prior auth delay becomes a weekend hold.
• A “preferred SNF” doesn’t help if authorization lags. The AMRPA data explain why networks alone can’t fix flow; you need better sequencing. See building a preferred SNF network for strategies to handle this mismatch.

What to track (and publish) if you want this to improve

1. PA start timestamp.
   Capture when the first PA submission is sent, not just when the discharge order is written.

2. PA decision timestamp.
   Measure standard vs. expedited cycle times per payer. Public CMS metrics will help you benchmark by 2026.

3. Time-to-First-Accept and Order-to-Door.
   When these metrics worsen for MA discharges, it’s a signal that PA timing—not staffing—is driving LOS. Explore how bed delays shape cost curves.

What actually helps

• Start PA when criteria are met, not when discharge is ordered.
  Align submission timing with expected LOS and discharge target.

• Use the new CMS clocks strategically.
  With 7-day and 72-hour limits, you can stage PA initiation to complete near the discharge date rather than spill into weekends.

• Respond to the denial reason, not the entire chart.
  OIG data show most denials cite documentation gaps or plan-only criteria. Match your re-submission exactly to the cited reason.

What to tell executives

• “Initial denials aren’t denials—they’re delays.” 82% of appealed denials flip; time lost equals LOS.
• “We’ll use CMS decision clocks to our advantage.” 7-day standard, 72-hour expedited. We’ll stage PA earlier and push for expedited when delay risks harm.
• “Document to the denial reason, not the chart.” Matching plan criteria prevents multi-week appeal loops.
• “We’ll monitor payer PA metrics once public.” Hospitals can target high-friction payers and sequence accordingly.

Prior authorization isn’t a paperwork nuisance anymore—it’s a throughput variable. Treating it that way can free bed-days without adding staff.