Medicare Advantage insurers are using prior authorization requirements to systematically deny seniors access to post-acute care facilities, according to a recent U.S. Senate investigative report. Dated October 17, 2024, the Majority staff report of the Senate Permanent Subcommittee on Investigations states that “Medicare Advantage insurers are intentionally using prior authorizations to boost profits by targeting costly yet critical stays in post-acute care facilities.” This finding emerged from the Subcommittee’s broader investigation into barriers that Medicare Advantage beneficiaries face when seeking healthcare services.
PRIOR AUTHORIZATION BACKGROUND
While both Traditional Medicare and Medicare Advantage programs share the requirement to cover services that are “reasonable and necessary for the diagnosis and treatment of or treatment of an illness or injury,” they differ in their approach to care delivery. A key distinction lies in the timing of care. Under Traditional Medicare, patients generally receive care first. However, as the report notes, “Medicare Advantage insurers, by contrast, frequently require patients and providers to obtain prior authorization before receiving care.” This means that before certain treatments or services are provided within a Medicare Advantage plan, approval must be obtained from the insurer.
The process of prior authorization in Medicare Advantage involves more than just medical necessity determinations. The report notes that “Along with medical necessity determinations, which are supposed to be the purview of doctors and nurses, Medicare Advantage insurers also make ‘administrative’ determinations, which involve assessing the terms of a member’s plan and can be made by non-clinical employees.” This indicates that decisions about prior authorization can be influenced by factors beyond clinical judgment, such as the specific details of a patient’s plan coverage.
The use of prior authorization in Medicare Advantage plans has become increasingly prevalent. The report notes this practice “has particularly increased in the last five years.” This trend is reflected in data from the Kaiser Family Foundation, which, as the report states, found that by 2023, “99 percent of Medicare Advantage enrollees were in a plan requiring prior authorization for some services.” This near-universal adoption of prior authorization within Medicare Advantage highlights its significant role in shaping access to care within these plans
THE ROLE OF ARTIFICIAL INTELLIGENCE
The expansion of prior authorization in healthcare has been enabled by advances in information technology, according to the Subcommittee report. Specifically, the report notes that “the widespread use of prior authorization became practical to implement only with computer technology and the ‘proliferation of information resources, assessment tools, and organizations that [made] case-by-case review of proposed services feasible on a large scale.’
On the provider side, the report notes the potential of technology to correspond to such widespread use by insurers, but that this is not yet a reality because the “structure of automation technologies varies considerably both across the healthcare industry and within the companies that employ them.”
FINDINGS
UnitedHealthcare
The Subcommittee report reveals a correlation between UnitedHealthcare’s efforts to automate prior authorization and an increase in denials for post-acute care. It states that “UnitedHealthcare’s denial rate for prior authorization for post-acute care significantly increased at the same time the company was launching initiatives to automate the process.” This suggests a potential link between automation and a higher rate of denials.
Further evidence of this connection is found in the report’s description of an internal decision-making process within UnitedHealthcare. According to the report, a UnitedHealthcare committee approved an “auto authorization model” after learning that it resulted in “faster review times and increased denials.” This indicates that the company was aware of and accepted the trade-off between speed and denial rates when implementing automated prior authorization.
The report also highlights the role of naviHealth, the company behind nH Predict, an algorithm linked in media reports to AI-powered denials of care, and its impact on prior authorization denials. The report notes that “Prior authorization denials for skilled nursing facilities accelerated significantly once naviHealth began managing post-acute care for Medicare Advantage beneficiaries.” This points to a potential connection between the involvement of naviHealth and a rise in denials, particularly for skilled nursing facilities.
CVS
The Subcommittee report indicates that CVS was aware of the financial benefits of prior authorization denials and actively expanded the use of this process for post-acute care. It states that “CVS knew prior authorization denials generated huge savings and subjected more post-acute care requests to the process.” This suggests a deliberate strategy to increase denials as a cost-saving measure.
The report further emphasizes the financial impact of denials compared to approvals, stating that “Savings from prior authorization denials vastly exceeded savings from automated approvals.” This highlights the prioritization of denials as a cost-containment strategy over other approaches, such as automated approvals.
CVS is also said to have employed a targeted approach to prior authorization, focusing on requests likely to be denied. The report notes that “CVS developed a data-driven strategy of focusing on prior authorization requests with ‘a significant probability to be denied.'” This suggests the use of data analysis to identify and target specific types of requests for denial.
Additionally, the report connects CVS’s use of artificial intelligence to cost-cutting measures within its Medicare Advantage division. It states that “CVS began to use artificial intelligence to reduce spending at post-acute facilities amid pressure to reduce costs in its Medicare Advantage division.” This suggests that AI was implemented as a tool to further reduce costs in this area.
Humana
The Subcommittee report reveals a likely correlation between Humana’s training practices and an increase in prior authorization denials at long-term acute care hospitals. The report states that “Humana’s denial rate at long-term acute care hospitals jumped significantly after prior authorization training sessions emphasized denials.” This suggests that the training itself contributed to the higher denial rate.
The report also highlights Humana’s efforts to reinforce denials during the appeals process. It notes that “Humana crafted templates to respond to post-acute prior authorization requests that enabled them to ‘uphold a denial on appeal.'” This is strongly suggestive of a strategy to maintain denials even when challenged.
Further evidence of Humana’s focus on denials is found in the content of its training sessions. The report states that “Humana training sessions for requests for long-term acute care hospitals emphasized cost and provided strategies for handling denials.” This reinforces the notion that cost containment and denial management were key priorities in these training sessions.
The report also reveals concerns among Humana staff regarding the suggestion of hospice care as an alternative to long-term acute care and raised questions about the level of human oversight in Humana’s decision-making processes. It states that “Internal Humana policies appear to give naviHealth [which was used by Humana in addition to UnitedHealthcare] and other contractors greater latitude to exclude humans from decision making.” This suggests that these contractors may have significant autonomy in making prior authorization decisions.
RECOMMENDATIONS
The Subcommittee report provided the following three recommendations:
- CMS should begin collecting prior authorization information broken down by service category;
- CMS should conduct targeted audits if insurer prior authorization data reveal notable increases in adverse determination rates; and
- CMS should expand regulations for utilization management committees to prevent predictive technologies from unduly influencing human reviewers.
Categories: AI, DATA ACCESS & INTEROPERABILITY
Leave a Reply