News|Articles|June 18, 2026

How MCOs Should Be Preparing for the OBBBA's Medicaid Cuts: David Brueggeman, MBA

Fact checked by: Giuliana Grossi
Listen
0:00 / 0:00

Key Takeaways

  • Regulatory ambiguity around capped state-directed payments can destabilize MCO–state contracts by triggering compliance-driven payment redesigns, mid-cycle rate rebaselining, and reopening risk corridors and risk-sharing terms.
  • Constraining state-directed payments undermines provider breakeven economics, prompting demands for higher FFS rates, expanded value-based arrangements, or increased commercial pricing to offset Medicaid shortfalls.
SHOW MORE

David Brueggeman, MBA, breaks down the underpriced financial risks, redetermination lessons, and what accountable AI actually requires from health plans.

The One Big Beautiful Bill Act (OBBBA) authorizes $1 trillion in federal Medicaid spending reductions over 10 years, with phased implementation beginning in 2026.1 Among its most consequential provisions: caps on state-directed payments at 100% to 110% of Medicare rates, semiannual redeterminations for expansion adults beginning December 2026, and CMS guidance still pending on undefined terms embedded in the statute. For managed care organizations (MCOs), the window to model contract exposure and renegotiate is closing—and industry forecasters project that states will respond to federal funding contraction by ratcheting down Medicaid managed care rates, driving plan exits, and concentrating sicker, higher-need members among remaining carriers.

David Brueggeman, MBA, managing director at BRG Healthcare, spoke with The American Journal of Managed Care® (AJMC®) about what MCOs are underpricing in their contracts, what plans got wrong during the post-COVID redetermination unwinding, and where the line falls between deploying artificial intelligence (AI) responsibly and deploying it recklessly.

This interview has been lightly edited for clarity.

AJMC: The OBBBA caps state-directed payments at 100% to 110% of Medicare rates, and CMS guidance on terms like "good faith effort" and "completed preprint" is still pending. What is the single most underappreciated financial risk that MCOs are not yet pricing into their contracts?

Brueggeman: There are 2 ways to interpret that question: MCO-to-state contracts and MCO-to-provider contracts. The answer is different for each.

For MCO-to-state contracts, the underappreciated risk is what uncertainty in regulations does to contract stability. If the details and audit expectations can change, plans and/or the state’s actuaries can underprice that uncertainty. States that need to stay compliant may change payment methods or evaluation requirements, and even small payment design changes can force mid-cycle rate rebaselining and/or reopen rate corridors and risk-sharing provisions.

For MCO-to-provider contracts, modifications to state-directed payments have a substantial impact on provider revenues and often represent the only pathway to breakeven levels of payment for applicable services. The removal or modification of payments will cause providers to seek additional revenue either within existing contracts via increased fee-for-service payments, additional value-based contract opportunities, or significant commercial rate increases to cross-subsidize.

AJMC: We went through the post-COVID redetermination unwinding and learned hard lessons about administrative attrition. Now we are heading into semiannual redeterminations starting December 2026. What did plans get wrong the first time, and what does "getting it right" look like this time?

Brueggeman: The unwinding period demonstrated that significant coverage losses can be driven by process failure rather than true ineligibility, with the Government Accountability Office documenting repeated state-level compliance issues that contributed to eligible people losing coverage.

Plans tended to treat that disruption as a one-time enrollment event instead of a recurring one, which matters because the OBBBA increases the frequency of eligibility redeterminations. Getting it right means plans need to be prepared for similar issues going forward. How does more frequent churn, partially driven by administrative friction, change actuarial assumptions? What operational process changes are required, and what cost is associated with those changes? How quickly can a plan enable re-enrollment to provide care continuity and prevent members from waiting until there is a service need before reapplying?

What some plans got wrong was the estimation of the impact of administrative friction on redetermination. It is important for plans to account for it moving forward because it changes cost assumptions every time an eligible member loses coverage for procedural reasons.

AJMC: AI adoption in managed care is accelerating, especially for redetermination processing and utilization management. CMS's Wasteful and Inappropriate Service Reduction (WISeR) Model is now live, and guardrail proposals are in the pipeline. Where is the line between deploying AI responsibly at speed and deploying it recklessly? What does accountable AI actually look like inside a health plan?

Brueggeman: A helpful way to think about this is to separate automation from accountability. The WISeR Model explicitly pairs AI with human clinical review, which signals what I believe is CMS's goal: speed is the target, but humans remain responsible for the decision—and the underlying technical process can be explained and audited.

The line is crossed when the model becomes the policy. If a plan cannot explain why an adverse decision occurred, cannot demonstrate consistency in outcomes, and cannot show a low-friction override and appeals pathway when the AI is wrong, that opens the plan to scrutiny. We recommend that plans think about governance around accountable AI, which means defined use cases and boundaries, documented decision logic, routine performance monitoring, and a defensible audit trail that can be produced for regulators, providers, and members.

It is also important to distinguish large language model–based AI methods from standard algorithms that are now being labeled as AI. Large language models, while improving, can still produce different answers to the same question, whereas static logic-based algorithms have more reproducible results. Deploying AI responsibly means it accelerates decisions while humans remain accountable for them. Deploying it recklessly means the algorithm becomes the decision-maker, and the plan cannot explain the result.

Reference

Steinzor, P. 5 things to know about the converging Medicaid funding crisis. AJMC. May 8, 2026. Accessed June 18, 2026. https://www.ajmc.com/view/5-things-to-know-about-the-converging-medicaid-funding-crisis