What You Should Know:
– A newly released report from the Peterson Health Technology Institute finds that while AI is reducing administrative burden for individual organizations, it is not lowering system-wide costs—and may, in fact, be increasing them.
– The reason: AI is accelerating activity across the system—more prior authorization submissions, more billing complexity, and more back-and-forth between providers and payers—without fixing the underlying structural inefficiencies.
The Prior Authorization “Bot War”
Prior authorization is a significant source of friction, with each submission cycle costing health plans $40–$50 and providers $20–$30. While AI is being deployed to automate these tasks, it has yet to reduce overall system-level costs.
- Increased Activity, Not Efficiency: AI allows providers to submit more requests and plans to process more denials, leading to “bot wars”—automated back-and-forth exchanges that increase communication volume without resolving the underlying clinical questions.
- The Scalability Gap: Emerging real-time adjudication models aim to approve requests during a patient visit, but these proofs of concept remain narrow and difficult to scale due to fragmented medical policies across different health plans.
- Standardization Limits: While new rules like CMS-0057-F mandate data standards for APIs, they do not standardize the actual medical necessity criteria, which vary wildly between insurers.
Medical Billing and the Risk of Inflation
In medical billing, AI-enabled ambient scribes are becoming standard, with all large health systems in a recent survey reporting their adoption. However, the primary result has been increased billing intensity.
- Coding Intensity: AI tools capture clinical complexity more thoroughly, leading to a higher proportion of visits being coded at higher levels. One system saw a 5% increase in Level 5 encounters after deploying AI scribes, raising revenue by over $1,000 per provider per month.
- Payer Retaliation: Health plans are responding with automated “downcoding” and reimbursement reductions, which may disproportionately harm smaller or rural providers who haven’t adopted AI tools.
- Cost Shift to Patients: This “AI arms race” threatens to increase patient cost-sharing and annual premiums as total medical expenditures rise.
For more inforamtion about the report, visit https://phti.org/administrative-ai-current-use-and-potential-impact/
