The US' Centers for Medicare & Medicaid Services is testing ACCESS, an outcome-based payment model for AI-driven medical care, with 150 tech companies
Neil Batlivala has spent seven years building a healthcare company that most of the tech industry has never heard of and that serves …
Context & Ripple Effects
CMS has previously opened reimbursement pathways for narrowly defined AI uses, including diabetic eye-disease and stroke detection. More recently, it drew a boundary around AI in Medicare Advantage by barring insurers from using it to determine care or deny coverage.
The new test therefore matters as a shift in emphasis: CMS is exploring where AI can be paid for based on care outcomes while retaining limits on AI as an autonomous coverage-decision tool.
First-order effects
- The 150 participating technology companies gain a route to test whether their AI-enabled care products can qualify for outcome-linked CMS payment rather than relying solely on conventional software purchasing.
- CMS will evaluate AI-driven care in a payment framework tied to results, placing greater immediate importance on demonstrating measurable clinical or operational outcomes.
Second-order effects
- Hospital systems already buying commercial AI licenses may face stronger pressure to distinguish tools that can substantiate outcomes from those sold mainly on workflow or productivity claims.
- Insurers and care providers will need to separate AI used to support and improve care from AI used in coverage determinations, an area where CMS has already set explicit constraints for Medicare Advantage.
Third-order effects
- If outcome-based payment proves workable, healthcare AI competition could shift toward evidence, implementation quality, and accountability for results rather than access to models alone.
- The policy direction suggests a two-track market: public reimbursement experiments for AI that improves care, alongside tighter scrutiny of AI that influences eligibility or denials.
The trend: Healthcare AI is moving from pilot and license adoption toward reimbursement models that demand demonstrated outcomes and clearer boundaries on automated decision-making.