Tuesday, July 8, 2025

Prior Authorization Demo in Fee for Service Medicare

CMS has announced a regional, pilot program to test Prior Authorization of targeted services in Fee for Service Medicare.  

While none of these are genomic tests or PET imaging, diagnostic providers should still track this initiative for future implications.

  • ...Third party entities [will be] leveraging enhanced technologies, that would be paid under a novel payment approach where the model participants are compensated based on a share of averted expenditures. Further, the WISeR model would test: the speed and accuracy of new technology-assisted decision-making.   90FR28751.

Wow.  "Compensated based on a share of averted expenditures" or in plain English, "paid for claims denied."

  • See press release here - June 27, 2025.
  • See two-page fact sheet here.
  • See home page for the project, WISeR, at Center for Innovation, here.
  • See 35-page project description and Request for Applications here.
  • See FAQ here.
  • See one-pager here.
  • See Fed Reg here - 90 FR 28749, 5pp, July 1, 2025.
  • See article at "Kiplinger," here.
  • See a Medicare Advantage portal for genomics pre-authorization here.
The Fed Reg names each NCD or LCD that will be implemented via the new program.

Some Services are Odd

Some of these are real odd; the Parkinson/deep brain stimulation is based on NCD 160.24, written in 2003, and based on even older Medcac and TA materials (here).  See CR2553.

Similarly, "treatments for impotence" fall under NCD 230.4, which merely makes the statement, "payment made be made for the diagnosis and treatment of impotence," with an "effective date" of... 1966.  But, "if the service is provided to patients with a mental condition," you must first "apply additional limitations" to be found at a useless link.  In examples like this, it's impossible to see what pre-existing CMS rules or guidance the PA will deploy in its review.

Incontinence control devices are covered per conditions in 1996...see NCD 230.10.

• Electrical Nerve Stimulators (NCD 160.7)
• Sacral Nerve Stimulation for Urinary Incontinence (NCD 230.18)
• Phrenic Nerve Stimulator (NCD 160.19)
• Deep Brain Stimulation for Essential Tremor and Parkinson’s Disease (NCD 160.24) 
• Vagus Nerve Stimulation (NCD 160.18)
• Induced Lesions of Nerve Tracts (NCD
160.1) • Epidural Steroid Injections for Pain Management excluding facet joint injections (L39015, L33906, L39036, L39240, L39242, L36920, L38994, L39054)
• Percutaneous Vertebral Augmentation (PVA) for Vertebral Compression
Fracture (VCF) (L33569, L34106, L34228, L38201, L34976, L35130,
L38737, L38213)
• Cervical Fusion (L39741, L39799, L39770, L39758, L39762, L39793,
L39773, L39788)
• Arthroscopic Lavage and Arthroscopic Debridement for the Osteoarthritic Knee (NCD 150.9)
• Hypoglossal Nerve Stimulation for Obstructive Sleep Apnea (L38276,
L38307, L38398, L38387, L38310,
L38312, L38385, L38528)
• Incontinence Control Devices (NCD 230.10)
• Diagnosis and Treatment of Impotence (NCD 230.4)
• Percutaneous Image-Guided Lumbar Decompression for Spinal Stenosis
(NCD 150.13)
• Skin and Tissue Substitutes (LCDs below)—only applicable to MAC
jurisdictions and states that have  n
active LCD in place  
++ Application of Bioengineered Skin Substitutes to Lower Extremity
Chronic Non-Healing Wounds (L35041)
++ Wound Application of Cellular and or Tissue Based Products (CTPs, L36690)

I'm not sure what the "++" sign means.

###
The Fed Reg notes that services were picked due to evidence of fraud/waste, CERT error rates, prior OIG reports, and other factors.  Service providers will obtain prior auth from technology providers under WISER.  While pre-submission is optional, failing to pre-submit will simply trigger a post-service review by the same technology providers, plus, potentially, records requests.

As I understand it, "Model Participants" will be required to coordinate with MACs, including Uniform Tracking Numbers (UTNs).

Participants in the model will be "third party entities" and "subject to HIPAA."  MACs will "function as HIPAA business associates" (Fed Reg.)

If a service is provided after an initial "non-authorization, non-affirmed" expect it to be MAC-denied and trigger the normal appeals routes after that.

###
AI CORNER
###

WISeR Model: A Health Policy Overview of Third-Party “Model Participants” Coordinating with MACs

[AI generated article]

The Wasteful and Inappropriate Service Reduction (WISeR) model, launched by the CMS Innovation Center, represents a bold new approach to prior authorization in Medicare Fee-for-Service (FFS). Running from 2026 to 2031, the WISeR model is structured as a 6-year, two-phase demonstration project designed to test the use of enhanced technologies such as artificial intelligence (AI) and machine learning (ML) to improve oversight and reduce unnecessary or inappropriate services in Original Medicare.


❖ Purpose and Scope of WISeR

CMS estimates that up to 25% of U.S. health care spending is wasteful, including billions in Medicare expenditures. Services targeted in WISeR include those with little or no clinical benefit, as well as services vulnerable to fraud, waste, and abuse (FWA). Examples include:

  • Skin and tissue substitutes

  • Electrical nerve stimulators

  • Knee arthroscopy for osteoarthritis

  • Spinal injections, vertebroplasty, and hypoglossal nerve stimulation

Crucially, Medicare Advantage (MA) plans have demonstrated success using enhanced prior authorization tools; WISeR now tests the transferability of such strategies into the Original Medicare space.


❖ Who Are the “Model Participants”?

In contrast to many CMMI models that engage providers or health systems, WISeR’s only participants are third-party technology companies—not health care providers or insurers.

These model participants are:

  • Private-sector companies with prior experience in streamlining prior authorization for Medicare Advantage or commercial payers;

  • Firms that use enhanced technologies, including AI/ML algorithms, predictive analytics, and electronic portals;

  • Required to include appropriately licensed human clinicians to make final non-affirmation decisions, consistent with Medicare's requirement that adverse decisions are clinically reviewed.

These entities coordinate directly with Medicare Administrative Contractors (MACs) to process prior authorization requests and associated claims reviews.

CMS refers to these participants as technology innovators—companies that will serve as delegated prior authorization reviewers in designated states. 

Participants are compensated not via fee-for-service, but through a shared savings model based on averted Medicare expenditures attributed to their interventions.


❖ Coordination with MACs

WISeR introduces a triangular workflow:

  1. Providers and suppliers submit prior authorization (PA) requests:

    • Either to the MAC, which forwards it to the model participant,

    • Or directly to the model participant, who conducts the review.

  2. Model participants use their AI-powered platforms to evaluate documentation, determine coverage, and issue provisional affirmations (or not).

  3. The MACs then:

    • Receive a unique tracking number (UTN) from the model participant to process claims accordingly,

    • Conduct post-payment actions for any claims not submitted through prior authorization.

Claims without a prior authorization will be subject to pre-payment medical review by the model participant, again coordinated via the MAC.

This system is designed to reduce administrative burden, especially for compliant providers who may qualify for "gold card" exemptions, and to standardize the interface between CMS systems and the private sector.


❖ Oversight and Quality Metrics

Model participants will be evaluated on:

  • Volume of requests processed

  • Timeliness of decisions

  • Appeal overturn rates

  • Beneficiary experience

  • Clarity of documentation and communication

Final authority over non-affirmations lies with licensed clinicians, not machines—ensuring ethical oversight.


❖ Geographic Coverage and Implementation

WISeR will be piloted in six states across four MAC jurisdictions:

  • JL: New Jersey

  • J15: Ohio

  • JH: Oklahoma, Texas

  • JF: Arizona, Washington

These jurisdictions were selected for their claim volume, LCD infrastructure, and evaluability across control and test states.


❖ Conclusion

WISeR represents a major structural innovation in Medicare policy: contracting directly with private technology firms to manage utilization review, guided by AI but grounded in traditional clinical standards. This shift challenges CMS’s historic reliance on MACs and provider-led models by introducing a novel, delegated review system focused on efficiency, accuracy, and savings.

As the first model where CMS pays third-party vendors a cut of averted expenses, WISeR’s outcome will likely influence future efforts to modernize Medicare FFS oversight and incorporate private-sector technology innovations into federal healthcare administration.