By deploying their own artificial intelligence tools, healthcare providers are fighting back against the prior authorization systems that insurers use to control costs and sometimes deny patient care. This battle of algorithms aims to streamline the burdensome approval process, which has been a major source of physician burnout and a cause of treatment delays for patients.
How insurers use AI to manage prior authorizations
Major insurers like UnitedHealth Group, Cigna, and Humana have invested heavily in AI to automate their administrative functions, including prior authorization requests.
- Automating reviews: Algorithms extract and analyze data from provider requests to check for compliance with clinical criteria and determine if a service is medically necessary.
- Denying claims: Insurers have faced lawsuits alleging their AI tools systematically deny claims with little to no human oversight, sometimes at a much higher rate than typical.
- Expanding oversight: Automated systems allow insurers to apply prior authorization rules to a broader range of lower-cost procedures, a practice called “review creep,” that was previously too expensive to manage manually.
The push to automate prior authorizations has reshaped how payers evaluate medical necessity, with implications for speed, consistency, and accountability.
How providers are leveraging AI in response
A new wave of AI startups and major health IT vendors are equipping providers with tools to automate their side of the prior authorization process. These solutions are designed to mirror and counter the automated processes of insurers.
- Automated submission: AI systems can automatically extract patient clinical data from electronic health records (EHRs) and use it to auto-populate the specific authorization forms required by different payers.
- Policy monitoring: AI algorithms track constant changes in payer policies and automatically update providers on shifting documentation requirements, ensuring submissions are compliant and accurate.
- Denial prediction and appeals: Predictive analytics can flag requests that are likely to be denied, allowing staff to adjust the submission beforehand. If a claim is denied, AI can help draft and automate the appeal letter.
- Real-time status tracking: Instead of staff chasing down approvals manually, AI tools can track the status of a request from submission through to approval, providing real-time updates and notifications.
The consequences of this AI escalation
The battle of the bots could potentially improve efficiency but also raises serious concerns.
- Potential for harm: Some physicians fear that unregulated and unsupervised AI could override sound medical judgment, increasing denial rates and leading to delayed or forgone care.
- Exacerbating inequality: The “digital divide” could widen as smaller or rural practices lack the resources to adopt sophisticated AI tools, putting them at a disadvantage compared to larger, well-funded health systems.
- A push for human oversight: In response to concerns, some jurisdictions are taking action. California passed a landmark law in 2024 to require that a licensed physician, not just an AI algorithm, must make decisions regarding prior authorization denials.
Key takeaways
AI is accelerating both sides of the prior authorization process: insurers use AI to streamline reviews and expand oversight, while providers adopt AI to automate submissions, monitor policy changes, and predict denials.
The net effect could be faster approvals and reduced burnout, but it also risks unintended harm without appropriate oversight and transparency.
Policy and regulation are catching up. Laws mandating physician involvement in denials, like California’s 2024 measure, are signaling a shift toward preserving clinician judgment in AI-augmented workflows.
Further reading and resources
https://www.example.com/ai-prior-auth-news
https://www.example.com/ca-ai-denials
https://www.example.com/ama-prior-auth
