by Kristin Rowan, Editor

WISeR Model

Dangerous and Troubling

WISeR model is not as wise as CMS had hoped. For many months, we have been reporting on the waste, fraud, and abuse initiatives from HHS and CMS. From stricter oversight to fewer new agencies to broad investigations, the crackdown on wasteful spending of the Medicare Trust Fund has increased steadily. One initiative, the Wasteful and Inappropriate Service Reduction (WISeR) model, launched in Arizona, Ohio, Oklahome, Texas, New Jersey, and Washington. As the name implies, the model aimed at reducing unnecessary services by using an AI powered prior authorization algorith.

Sounds like Medicare Advantage

The WISeR model includes a limited number of procedures that require prior authorization. Even before the model launched, some were concerned that the preauthorization requirement would delay or deny necessary care, much like the prior authorization requirements in Medicare Advantage plans. The Center for Medicare Advocacy testified in opposition to the WISeR model. The spokesperson for CMA, David Lipschutz, supported three bills discussed during that hearing, including one that would stop the WISeR model altogether and prohibit using or testing any payment models for prior authorization in traditional Medicare.

Foresight was 20/20

Three months before the WISeR model launched, CMS founder Judy Stein said the program would create barriers between physician orders and approved procedures. She cautioned:

“Adding prior authorization requirements to traditional Medicare will create costly problems and barriers to necessary care. Instead, to truly address waste and abuse in Medicare, CMS should look to the dramatic overpayments and unreasonable denials in Medicare Advantage.”

Judy Stein

Founder & Senior Advisor, Center for Medicare Advocacy

Early Reports

Physician struggles

Just three months into the WISeR model demonstration, early reports from providers and patients are discouraging. The Washington Post analysis of the model didn’t include a lot of positive feedback. Physicians report challenges with the approval process. Patients wait in pain until needed care is approved. Issues with the technology include problems with online portals as well as struggles with coordination and communication with tech firms and claims processors. Physicians also report denials for care that is within coverages guidelines and decisions that take longer than federal guidelines allow. 

Unintended consequences

Part of the model intended to reduce unnecessary spending is a built-in incentive program for cost savings. The tech companies handling the AI algorithms are paid partly on the savings realized after denying medical services. This creates an inherent goal of adjusting the algorithm to deny more services, even if those services are necessary. The model supposedly balances this with pay adjustments for quality measures that include accuracy in decision-making. These tech companies are not staffed by medical professionals and it is unclear what they use as the basis for their decision-making algorithm.

Exemptions

Officials from Medicare say they intend to offer exemptions to physicians with a high authorization rate. These physicians would no longer be subject to the AI prior authorization if, over time, most of their submitted services are authorized. Taking this to its logical conclusion:

  • Physicians may dial back on recommended procedures until they get the exemption
  • Once exempt, physicians can circle back and get the procedures for their patients without prior approval
  • Tech companies may increase denials to ensure continued work
  • AI oversight of federal health spending could increase if the pilot program saves money, regardless of patient outcomes

Writing was on the Wall

Long before the WISeR model existed, AI algorithms for care authorizations had denied needed care. Insurance companies started using AI to process prior authorizations in 2020. Both UnitedHealth Group and Humana used nH Predict AI for care authorizations for Medicare Advantage. The family of an elderly couple sued UnitedHealth after their care was denied and the couple died. A class action suit against Humana claimed nH Predict AI Model predictions are highly inaccurate and are not based on patients’ medical needs. In February of 2024, CMS clarified the use of AI cautioning that “compliance is required with all of the rules at § 422.101(c) for making a determination of medical necessity, including that the MA organization base the decision on the individual patient’s circumstances.”

Unclear Future

After only a few months, officials have not indicated whether the WISeR model could expand into additional states or to cover additional procedures in the future. Beneficiaries say they stayed with the traditional Medicare plans specifically to avoid the prior authorization hurdles inherent in Medicare Advantage plans. CMS may have additional information by July, when the program has more data.

# # #

Kristin Rowan Editor The Rowan Report
Kristin Rowan Editor The Rowan Report

Kristin Rowan is the owner and Editor-in-chief of The Rowan Report, the industry’s most trusted source for care at home news. She is also a sought-after speaker on Artificial Intelligence, Technology Adoption and Lone Worker Safety. She is available to speak at state and national conferences as well as software user-group meetings.

Kristin also runs Girard Marketing Group, a multi-faceted boutique marketing firm specializing in content creation, social media management, and event marketing. She works with care at home software providers to create dynamic content that increases conversions for direct e-mail, social media, and websites.  Connect with Kristin directly at kristin@girardmarketinggroup.com or www.girardmarketinggroup.com

©2026 by The Rowan Report, Peoria, AZ. All rights reserved. This article originally appeared in The Rowan Report. One copy may be printed for personal use: further reproduction by permission only. editor@therowanreport.com