The landscape of healthcare administration is undergoing a seismic shift, with artificial intelligence (AI) increasingly at the forefront of complex processes like prior authorization. While initially hailed as a potential panacea for reducing administrative burdens, the integration of AI into these systems is proving to be a double-edged sword, leading to both efficiency gains and novel challenges for healthcare providers, patients, and insurers alike. This evolution is creating a "battle of the bots," where automated systems on both sides of the healthcare equation are clashing, impacting the delivery of care and straining the provider-patient relationship.
The Unforeseen Consequences of Algorithmic Decision-Making
For otolaryngologists and other medical professionals, the advent of AI in prior authorization has introduced a new layer of complexity and frustration. Dr. Yolanda Troublefield, an attending surgeon and otolaryngologist at Southcoast Physicians Group in North Dartmouth, Massachusetts, and a member of the Physician Payment and Policy Workgroup of the American Academy of Otolaryngology–Head and Neck Surgery, recently encountered this firsthand. She was attempting to secure prior authorization for an eight-year-old child requiring an adenoidectomy with ear tube placement due to recurrent otitis media. The medical necessity was well-documented, supported by records from both primary care and otolaryngology, and corroborated by an audiogram. Even the patient’s parents, both physicians, understood the standard of care.
Despite meticulous documentation, the request was denied. Dr. Troublefield attributes this to the AI, specifically a large language model, that appears to analyze documentation selectively. In this instance, while the overall diagnosis was "recurrent otitis media," a specific part of the assessment-and-plan section noted "acute otitis media." This seemingly minor discrepancy, intended to indicate the need for an antibiotic at that particular visit, was misinterpreted by the AI as a reason to deny coverage for the surgical intervention. This required Dr. Troublefield to navigate further steps, potentially including a time-consuming peer-to-peer conversation with a reviewer, disrupting her schedule and delaying patient care.
"We’ve been doing tonsils and adenoids and ventilation tubes for the last 50 years—seems pretty straightforward," Dr. Troublefield remarked. "Every single ‘i’ has been dotted, ‘t’ has been crossed." Yet, the AI’s interpretation superseded established clinical practice and documentation standards.
Denials at Machine Speed: A New Era of Administrative Friction
The increasing deployment of AI by health insurance companies in prior authorization processes, and for the review of previously rendered services (clawbacks), has led to an acceleration of claim denials. Physicians report that these denials are often delivered with unprecedented speed and can be perplexing in their reasoning. This necessitates additional administrative work for providers to secure coverage for necessary patient care, leading to growing exasperation. The trend toward automated exchanges between provider and payer systems, using AI for both submissions and responses to denials, is expected to have significant, yet currently unknown, long-term effects on the healthcare system.
"What we’re seeing is faster denial of claims because of the use of AI," Dr. Troublefield stated. "You can set the dial in different ways. You can set it high, medium, or low. And they’re set high for denials." This suggests a deliberate calibration of AI systems to err on the side of denial, thereby potentially reducing insurer payouts.
Concerns about AI’s impact on denial rates are widespread among physicians. A survey conducted by the American Medical Association (AMA) last year revealed that 60% of physicians believe AI has already increased or will increase prior authorization denial rates. Furthermore, 55% of physicians reported that prior authorization delays access to necessary care either always or often, and a significant 79% indicated that prior authorization sometimes leads to patients abandoning treatment altogether. Alarmingly, 26% of physicians reported that prior authorization has resulted in a serious adverse event for a patient under their care.
Dr. Bruce Scott, a former president of the American Medical Association and an otolaryngologist at ENT Care Centers in Louisville, Kentucky, expressed initial optimism that AI would alleviate administrative burdens. "I think physicians were very hopeful that AI was going to be a boost, a solution if you will, to reduce the administrative burdens that we all face," he said. However, his experience suggests otherwise.
"Eventually, their bot is going to talk to my bot," Dr. Scott predicted. "Because my bot is going to tell me how to document so I get authorization, and then their AI is going to get even smarter and deny that, so I get a denial letter back. And in the meantime, the physicians and patients are stuck in the middle, and that’s the problem." This sentiment highlights a growing perception of an adversarial relationship being amplified by technology.
Regulatory Intervention: States Step In to Govern AI in Healthcare
In response to the burgeoning issues surrounding AI in healthcare claims processing, a wave of new state laws has emerged to regulate its use in claim denials. These legislative efforts aim to ensure that AI systems do not undermine established principles of patient care and provider autonomy.
For instance, an Alabama law mandates that insurers base determinations on a patient’s unique circumstances, rather than relying solely on aggregated group datasets. This seeks to prevent AI from making generalized decisions that may not apply to individual patient needs. Similarly, an Indiana law prohibits insurers from using AI as the sole basis for downcoding a claim, a practice that can reduce reimbursement rates. In Washington state, a law requires that claim determinations be made exclusively by licensed and qualified health professionals, ensuring human oversight in critical decision-making. These are but a few examples of the growing regulatory landscape aimed at curbing the potential misuse of AI in healthcare.
The federal government is also exploring the application of AI in health insurance, albeit with a different objective. The Centers for Medicare and Medicaid Services (CMS) has launched the WISeR (Wasteful and Inappropriate Service Reduction) program. This initiative aims to leverage AI, alongside human review, to "reduce clinically unsupported care by working with companies experienced in using enhanced technologies to expedite and improve the review process for a pre-selected set of services that are vulnerable to fraud, waste, and abuse." The program commenced in January in six states: New Jersey, Ohio, Oklahoma, Texas, Arizona, and Washington. While intended to combat fraud and waste, the deployment of AI by governmental entities raises similar questions about transparency and potential for error as its use by private insurers.
Fighting Automation with Automation: Providers Adapt to the AI Landscape
Faced with the increasing sophistication and ubiquity of AI in insurance operations, healthcare providers are finding it increasingly necessary to adopt similar technologies to streamline their own administrative processes. Dr. Bradford Bichey, a rhinologist at Indiana Sinus Centers, has developed an AI-powered product designed to enhance physician office operations, including the generation of documentation for prior authorization submissions.
Dr. Bichey observed a significant increase in clawback attempts several years ago, where insurers retrospectively questioned coverage for services rendered. He noted a surge from approximately five requests per time period to around fifty, often accompanied by unusual language suggesting early-stage AI experimentation by insurers. "The verbiage was slightly off," he recalled. "I got one letter that was almost hateful." While the tone of these communications has since become more professional, the underlying trend persisted.
In response, Dr. Bichey created "Blue," an AI product developed under his corporation, Nemedic. Blue generates not only clinical notes based on patient encounters but also prior authorization documents specifically tailored to the requirements of individual insurers. The software can, for example, recognize keywords within a patient encounter—such as the mention of a specific insurance company—and then automatically compile a document that meets that insurer’s documented criteria for prior authorization.
These submissions often require extensive detail beyond a standard clinical note. For a nasal endoscopy, for instance, the AI can generate a justification like, "routine anterior examination was insufficient to visualize the deeper nasal passage and assess for ongoing mucosal disease. Diagnostic rigid nasal endoscopy was medically necessary to direct visualization of the sinonasal mucosa," while also carefully noting exclusions such as "this procedure was not performed for routine screening, but to inform management."
"You have to know [the insurer’s] inclusion and exclusion," Dr. Bichey emphasized. "And it’s almost impossible for someone who does billing to constantly keep up with all these plans because we see hundreds of different types of insurance." The ability of AI to parse and apply these complex, ever-changing rules is seen as a critical advantage.
Dr. Bichey also noted that appeals for denied claims often become simpler with this approach. Frequently, denials stem from information that was present in the initial submission but was overlooked or misinterpreted. "When we do this, we just resubmit the same note and say, ‘You’re wrong—look at paragraph three,’" he explained. This improved upfront documentation has allowed him to avoid peer-to-peer calls for the past two years.
Governance, Not Just Algorithms: The Underlying Issue
While the challenges presented by AI in prior authorization might suggest a fundamental flaw in the technology itself, experts argue that the core problem lies in the governance surrounding its implementation. Dr. Matthew Crowson, assistant professor of otolaryngology–head and neck surgery at Harvard Medical School and director of clinical informatics and artificial intelligence at Massachusetts Eye and Ear, posits that AI is not inherently the issue.
"It’s not so much that AI is the problem," Dr. Crowson stated. "It’s more the governance around this stuff, or lack thereof." He contends that AI has not resolved the inherent tension between providers and payers; instead, it has intensified it. "It’s turning into a battle of the bots," he observed. "The application of AI by providers and payers is cutting both ways. It doesn’t resolve the fundamental tension between approving and denying based on medical necessity. It’s scaling up existing decision-making workflows faster."
However, Dr. Crowson also acknowledges instances where AI has demonstrably improved administrative efficiency in ways that benefit patients. He cites the example of a patient requiring a specific ear drop not on a health plan’s formulary. Historically, a nurse or medical assistant would have to manually compile supporting information and file an appeal. With AI, a contextualized appeal can be drafted rapidly, potentially expediting medication access for the patient. "What would take a human process maybe seven business days to do, now you can do it in one," he said. "I think administratively, it’s massively speeding up our ability to respond quickly."
Efficiency Gains or Amplified Activity? The Unclear Impact of AI
The overall impact of AI on the U.S. healthcare system remains a subject of ongoing analysis. A report by the Peterson Health Technology Institute, based on workshops involving senior leaders from across the healthcare spectrum, indicated an unclear picture regarding AI’s effects. While AI may reduce the cost of prior authorizations for individual organizations, it has not yet translated into systemic cost reductions.
Furthermore, the report suggests that the adoption of AI by providers might be increasing "billing intensity." This could involve identifying more severe diagnoses or more advanced treatments, which, in turn, leads to higher associated costs. Participants in the workshops expressed concern that the optimized use of AI by both insurers and providers could transform the prior authorization process into a more "activity-intensive" endeavor rather than a genuinely more efficient one. One healthcare provider, quoted in the report, lamented, "Bots don’t get tired of asking questions, so my review queue keeps growing."
Patients Caught in the Crossfire: The Human Cost of Algorithmic Delays
Beyond the administrative and operational challenges, the prior authorization process, amplified by AI, continues to exert a significant strain on the patient-physician relationship. Dr. Troublefield notes that discussing insurance coverage requirements and navigating hurdles has become a daily, time-consuming aspect of patient interactions.
"It takes more time—I don’t think any of us runs on schedule at all," she stated. "I would say five years ago, we probably could run on schedule. Not anymore. It never, ever, ever happens. We’re finding that we’re constantly having to apologize, apologize, apologize. And that’s not the way people want to be treated."
Dr. Scott recounted a distressing case where a patient with a maxillary sinus tumor, who had finally agreed to surgery, received a denial letter from her insurer. The denial was based on the preposterous assertion that the patient hadn’t yet been on an antibiotic, despite the clear indication for surgical intervention. The patient, confused and swayed by the insurer’s communication, questioned the necessity of surgery and asked if an antibiotic trial should be pursued. This situation forced Dr. Scott to re-establish trust with his patient, highlighting the erosion of the physician’s authority and the patient’s confidence due to insurance-driven delays and miscommunications.
Dr. Scott advocates for consistent appeals of denied claims, stating that his practice appeals every denial. However, he acknowledges the pervasive issue of physician burnout, which can leave practitioners feeling unable to engage in every necessary battle. He and the AMA are actively pushing for greater physician involvement in the development of regulations and processes surrounding AI in healthcare, aiming to ensure that these technologies serve the best interests of patient care.
A Fork in the Road: Navigating the Future of Prior Authorization
Dr. Crowson believes the field is at a critical juncture. One path forward involves both providers and payers utilizing AI to efficiently process routine paperwork and basic tasks, freeing up human resources to focus on complex cases. The alternative is an "escalating compliance and gaming war," where both sides continuously adapt their AI strategies to gain an advantage.
He expresses hope for a "pragmatic, middle-of-the-road approach," where AI is employed for its strengths: automating administrative monotony and handling simple tasks that can be safely and efficiently scaled.
Dr. Troublefield remains concerned that AI might introduce new hurdles, requiring physicians to use specific "buzzwords" in patient conversations to satisfy algorithmic requirements. However, she finds the concept of software that automatically generates insurer-compliant templates to be an attractive proposition, provided it is practical and user-friendly.
She expresses cautious optimism that technological advancements, coupled with "smart legislation" that regulates AI’s role in insurance decisions, could offer solutions. The inclusion of patient voices, and even public figures, in advocacy efforts could further bolster legislative initiatives. Ultimately, Dr. Troublefield believes physicians must adapt to the evolving system. "The system is changing, and either you have to be part of the change, or you’re going to get left behind," she concluded. "It’s the same thing as when the automobile happened, right? You’re not stopping it, so you have to figure out ways within the system to make sure that what you want, and what your goals are, are actually achieved."
The ongoing integration of AI into prior authorization processes represents a profound transformation in healthcare administration. While the promise of efficiency and reduced burden remains, the current reality is a complex interplay of automated systems, evolving regulations, and persistent challenges for physicians and patients. The path forward will likely require a delicate balance between technological innovation and robust human oversight, guided by a commitment to patient well-being and equitable access to care.
Thomas R. Collins is a freelance medical writer based in Florida.
