A new study of cancer drugs denied by Medicare Part D insurance plans found that more than one-fifth of denials were overturned upon review by a contracted reviewer. This suggests there are a considerable number of inappropriate denials.
The study, led by Youngmin Kwon, PhD, a postdoctoral research fellow in the Department of Health Policy at Vanderbilt Health, implemented natural language processing to analyze the text of second-level appeals of cancer drugs denied under Medicare Part D, the supplemental plans that cover prescription drugs. The study was published in JAMA Network Open.
“Cancer drugs are expensive, and their use is closely monitored and controlled by insurance plans. They also belong to a protected drug class in the Part D program, meaning that plans must cover all or nearly all of these drugs. While plans cannot broadly exclude coverage, they can deny coverage of drugs that are not deemed medically necessary,” Kwon said.
The most common reason for initial denials was “non-medically acceptable” indications, or drugs prescribed “off-label,” which generally means using a drug outside its formally approved indication by the Food and Drug Administration.
When treating cancer, prescription drugs are often combined or used for cancer types or populations that aren’t always specifically included in the FDA-approved labeled indication. To govern coverage of off-label use, Medicare relies on clinical compendia that summarize supporting clinical evidence or other peer-reviewed studies of medications prescribed off-label.
“Off-label prescribing is exceptionally common in oncology, especially for treating rare or treatment-resistant cancers. Because it does not follow the ‘standard’ treatment protocol, insurers may have a greater justification for heavily scrutinizing off-label use and refuse coverage,” Kwon said.
However, the study found that 88% of successful appeals for off-label use were for medically acceptable indications already supported by a Medicare-approved clinical compendium or peer-reviewed scientific literature, suggesting that many requests for acceptable off-label use may have been denied.
Further, appeals for off-label use were often denied due to clerical issues (72%), such as failure to include adequate documentation and supporting citations. This suggests that many coverage decisions may have been made purely on administrative grounds, rather than through a formal clinical review, Kwon said.
Among drugs prescribed on-label but that were initially denied for failing to follow the preapproval process, 60% were ultimately approved by reviewers, a higher rate of favorable review.
Kwon said these findings echo concerns that utilization management tools used by insurers, such as prior authorization, are imposing barriers to necessary and evidence-based care.
“Patients and their providers may face considerable administrative burden in the appeals process for coverage of drugs that could be lifesaving. A simple mishap, such as forgetting to attach a lab result or a supporting citation, could mean weeks of delay. This creates stress for patients and their care team and could even lead to worse health outcomes if care is substantially delayed.”
Kwon said the new findings can be helpful in ensuring that Medicare beneficiaries undergoing cancer treatment receive timely, high-quality medical care they need.
“Given the rapid adoption of AI into medical review processes, it will be important for Centers for Medicare & Medicaid Services and insurers offering Part D plans to continue to improve the coverage and appeals process for Medicare beneficiaries. For instance, reducing documentation burden for on-label therapies, while establishing clearer guidelines for determining effective off-label use by insurers, may go a long way in improving patients’ access to prescribed cancer drugs.”
The research was conducted in collaboration with researchers from the American Cancer Society, Emory University, the University of Pittsburgh and University of Hawai’i at Mānoa, and Wake Forest University.
