On October 17, 2024, the U.S. Senate’s Permanent Subcommittee on Investigations (the Committee) chaired by Senator Richard Blumenthal released a report authored by the majority entitled, “Refusal of Recovery: How Medicare Advantage Insurers Have Denied Patients Access to Post-Acute Care,” (the Report). The fifty-four page Report details the strategies that Medicare Advantage plans use to avoid paying for sometimes lengthy and costly post-acute care for their members who require such care. As a result of the behavior detailed in the Report, hospitals are burdened with patients that they cannot discharge.
The Report focuses on the practices of three Medicare Advantage insurers – UnitedHealthcare (UHC), Humana, and CVS, which collectively cover nearly 60 percent of Medicare Advantage enrollees. In preparing the Report, the Committee gathered over 280,000 pages of documents from these three companies, which revealed the scope of the practices.
Specifically, the Report details how these three plans intentionally use prior authorization policies to target and deny post-acute care requests, which forces enrollees to choose between remaining in the hospital setting at their own expense because acute care is no longer medically necessary or being discharged home against the advice of their treating physicians and despite their need for post-acute care.
According to the Report, requests for authorization for post-acute care were denied at far higher rates than requests for other types of care. UHC and CVS denied these requests at three times the rate that they denied requests for other types of care, while Humana denied post-acute requests at sixteen times the rate at which it denied other requests. Between 2019 and 2022, UHC’s denial rate for requests for skilled nursing facilities increased ninefold.
The Committee also found that as part of its effort to reduce the time necessary to review and deny post-acute authorization requests, all three companies used strategies to automate or streamline the denial process. Specifically, an internal UHC utilization management program committee voted to approve the use of what it called “Machine Assisted Prior Authorization” in the company’s utilization management review process, which would reduce the average time necessary to review each request by six to ten minutes. During that same year (2021), UHC tested a “HCE [Healthcare Economics] Auto Authorization Model” which UHC found led to “faster handle times” and “an increase in adverse determination rate.” This model was tentatively approved after these findings were presented to the internal UHC committee. In 2022, UHC explored using AI and machine learning to predict which denials would be appealed and which of those appeals would be overturned.
The Committee found that while CVS’ denial rate remained relatively stable, the number of cases denied increased significantly due to a growth in the number of service requests subject to CVS’ prior authorization requirements. Like UHC, CVS tested a predictive model and artificial intelligence to reduce the amount of money spent on skilled nursing facility services and estimated that expanding the use of this initiative would save the company $77 million over three years. Documents reviewed by the Committee also revealed that CVS was aware of a consistent correlation between increasing prior authorization requirements and increased savings for the company, and that it deprioritized a plan to reduce prior authorizations because of the anticipated loss in savings.
According to the report, Humana’s denial rate increased by fifty-four percent for long-term acute care hospitals between 2020 and 2022. Humana modified its reviewer templates used to communicate authorization decisions in 2019, with one medical director noting that the new templates for post-acute facilities were “important for denial purposes” and would enhance the ability to “uphold a denial on appeal.” In training presentations given to the company’s reviewers, Humana included strategies for explaining denials to providers, including urging them to pose “surprise questions” to the recommending providers.
Based on its investigation and the findings set forth in the Report, the Committee recommended that CMS: (1) begin collecting prior authorization information broken down by service category; (2) conduct targeted audits of prior authorization data to reveal increases in adverse determination rates; and (3) expand regulation of insurer utilization committees to ensure that predictive technologies do not unduly influence human reviewers.
The full Report can be found here.
