Key Takeaways
- Coverage criteria opacity accumulates across three distinct layers, meaning Medicare MAC review, Medicare Advantage plan modifications, and commercial or employer plan customization, each with its own governance and disclosure gaps.
- MACs use InterQual or Milliman as a review scaffold. Neither tool is publicly available without a license, and neither has been formally adopted by CMS as a coverage standard.
- Novel therapies are disproportionately exposed. A treatment with an FDA label requirement for a monitored setting can still be reviewed against proprietary criteria that predate its approval and default to a poorly matched analog.
- The 21st Century Cures Act information blocking framework already established that opacity in health information governance is a policy problem, not a neutral market condition. Coverage criteria have no equivalent standard.
- The CMS-0057-F prior authorization rule addresses a narrow slice. Utilization management criteria for inpatient level of care and MAC medical review standards remain largely outside its scope.
The 21st Century Cures Act established a foundational principle. Opacity in the flow of health information is not a neutral condition. When the rules governing data access are hidden, withheld, or fragmented across proprietary systems, the result is not simply inconvenience. It is a structural barrier with measurable consequences for patients, providers, and the healthcare system at large. The information blocking framework at HealthIT.gov codifies that principle as federal policy.
The same logic applies, with equal force, to coverage criteria. Prior authorization requirements, utilization management protocols, and level-of-care thresholds are the operational rules that determine whether a patient receives a therapy or procedure, in what setting, and at what cost. These rules function as infrastructure. When they are proprietary, fragmented, and inaccessible to the providers and innovators they govern, they produce inertia as a systemic output, not as an occasional failure.
This piece examines where that opacity lives, how it compounds across layers, and what it means for stakeholders across the coverage ecosystem. It is not a narrow critique of prior authorization, a topic that has received considerable policy attention. It is an argument that the problem runs deeper, into utilization management and level-of-care determination, and that novel therapies requiring facility-based administration face a particularly acute version of this structural failure.
The rules that govern coverage decisions are themselves a form of healthcare infrastructure. Infrastructure that is invisible to the people it governs cannot be navigated, challenged, or improved.
The three-layer opacity problem
Coverage criteria opacity does not exist at a single point in the system. It accumulates across at least three distinct layers, each with its own governance structure, disclosure norms, and accountability gaps.
| Layer | Who sets criteria | Public access | Transparency gap |
|---|---|---|---|
| Medicare MAC review | MACs using InterQual or Milliman as scaffold | No LCD/NCD for most drug and LOC decisions | Criteria not disclosed; no CMS adoption required |
| Medicare Advantage plan | Plan modifies base criteria; applies plan-specific addenda | Provider portal only; login required per plan | Modifications invisible without portal access; vary by market |
| Commercial / employer plan | Plan or PBM applies licensed criteria with further customization | Partial; ERISA plans have limited disclosure obligations | Most criteria not surfaced unless challenged through appeals |
The MAC layer is perhaps the most consequential for novel therapies because it governs traditional Medicare, the dominant payer for many high-acuity patient populations. MACs use InterQual or Milliman Care Guidelines as a review scaffold for medical necessity and level-of-care determinations. Neither tool is publicly available without a license. Neither has been formally adopted by CMS as a coverage standard. A hospital appealing a level-of-care denial is, in practice, arguing against criteria it cannot access, cite, or reproduce in a formal appeal submission.
The Medicare Advantage layer adds a second dimension of opacity. Plans may adopt InterQual or Milliman as a base and then apply plan-specific modifications, carve-outs, or clinical policy addenda. Those modifications are accessible only through the provider portal for each specific plan. A hospitalist or utilization management team admitting a patient covered by one MA plan may be operating under materially different level-of-care criteria than the same team managing a clinically identical patient covered by a different plan in the same market. There is no mechanism to surface that difference at the point of care without a labor-intensive, plan-by-plan portal review that most clinical environments cannot sustain operationally.
Why novel therapies are disproportionately exposed
Proprietary criteria sets are, by nature, retrospective. They are developed and updated based on historical utilization patterns, established clinical pathways, and therapies that were already in routine use when the criteria were written. A novel agent that requires facility-based administration, FDA-mandated monitoring, or a clinical protocol without a direct historical analog enters the coverage system without a defined criteria pathway.
In that absence, reviewers apply the closest available analog, which may map poorly to the actual clinical protocol, or they default to general criteria for the broader therapeutic category, which may have been written for a different generation of treatments with different safety profiles and administration requirements. The result is not necessarily a coverage denial on clinical grounds. It is a structural mismatch between what the therapy requires and what the criteria framework was designed to evaluate.
A novel therapy enters the coverage system without a defined criteria pathway. The system does not pause to build one. It defaults to the closest available analog and moves on.
Consider the general scenario of a novel intravenous agent that carries an FDA label requirement for in-hospital initiation with continuous monitoring across multiple days. That label requirement exists because the regulatory agency evaluated the clinical evidence and determined that the safety profile demands a monitored setting for the loading phase. The label is a federal document. It is publicly available. It reflects the most authoritative clinical judgment in the system about what the therapy requires.
Despite that, the level-of-care determination for an admission to provide that therapy may be reviewed against proprietary criteria that predate the therapy's approval, were not updated to reflect the label requirement, and may classify the admission as appropriate for observation rather than inpatient status based on generic criteria written for a different clinical context. The physician, the hospital, and the patient have no practical way to know that mismatch exists until a claim is denied.
The information blocking parallel
The information blocking framework established under the 21st Century Cures Act rests on a specific determination. When actors in the health information ecosystem engage in practices that interfere with the access, exchange, or use of electronic health information, those practices constitute a form of harm with regulatory consequence, regardless of whether the intent was to obstruct.
The coverage criteria opacity problem shares the same structural logic. When the rules that govern clinical access decisions are held in proprietary systems, disclosed only through credentialed portal access, and never formally adopted as public policy, they function as interference in the access pathway. A provider cannot navigate rules it cannot see. An innovator cannot align its market access strategy to criteria it cannot access. A patient cannot understand why a therapy was denied or what clinical standard was applied.
The 21st Century Cures Act did not require that every actor in the information ecosystem have identical access to every piece of data. It required that access not be actively blocked without a defined and legitimate exception. The coverage criteria ecosystem has no equivalent framework. Proprietary criteria are not subject to an information blocking standard. They are not required to be disclosed, published, or made available to providers or manufacturers on any timeline or through any channel.
That gap is not a coincidence of regulatory design. It reflects the fact that coverage criteria have been treated primarily as operational business tools rather than as the governance infrastructure they actually are. The CMS Interoperability and Prior Authorization final rule (CMS-0057-F) moved meaningfully toward prior authorization transparency for Medicare Advantage, but it addresses a narrow slice of the problem. Utilization management criteria for inpatient level of care, MAC medical review standards, and the plan-specific modifications applied within MA are largely outside its scope.
What this means by stakeholder
The opacity problem does not affect all stakeholders equally, and the implications differ enough across roles that each deserves a direct framing.
Hospital Systems
Utilization review teams cannot prospectively counsel clinical staff on admission appropriateness for novel therapies when the applicable criteria are inaccessible. Post-payment denials become the de facto criteria education mechanism, at the cost of write-offs and administrative burden. The operational burden of maintaining portal access and criteria literacy across dozens of MA plans in a single market is unsustainable at scale.
Physicians
Clinical decision-making at the point of care is supposed to be governed by evidence and label requirements. In practice, it is also governed by the implicit anticipation of coverage review, even when the applicable criteria are unknown. That anticipatory friction shapes prescribing behavior, admission decisions, and referral patterns in ways that may not align with the clinical evidence or the patient's best interest.
Innovators and Manufacturers
A manufacturer launching a novel therapy with a unique administration profile has no formal mechanism to understand where coverage barriers exist, which plans have adopted criteria that create a mismatch, or how MAC reviewers are applying existing policy to the new therapy. The only feedback signal is lagging and anecdotal, including physicians reporting denials, hospitals avoiding the product, or utilization data that underperforms projections without a clear explanatory narrative. By the time the pattern is visible, inertia is already set.
Payers and MACs
The opacity that protects proprietary criteria also insulates payers from accountability for criteria that are outdated, misapplied, or misaligned with current clinical evidence. Plans and MACs that update criteria in response to new FDA approvals or clinical guideline changes have no mechanism to communicate those updates to the provider or manufacturer community proactively. The result is a system where alignment, when it occurs, happens through conflict rather than through structured engagement.
Patients
A patient denied a therapy or admitted at the wrong level of care because of a criteria mismatch has no access to the criteria that governed the decision. The Explanation of Benefits reflects an outcome, not a standard. Appeals processes require the patient or their advocate to argue against a framework they cannot see, using clinical evidence that may be entirely consistent with the criteria if the criteria were actually applied correctly to the novel therapy in question.
Policy Community
The prior authorization transparency movement has achieved meaningful traction at the legislative and regulatory level. The level-of-care and utilization management criteria transparency gap is the next frontier, and it is largely unaddressed. The existing policy framework provides the conceptual architecture, through information blocking doctrine, prior auth transparency rules, and the Two-Midnight Rule, but has not yet synthesized those tools into a coherent standard for criteria disclosure across the full utilization management spectrum.
The inertia mechanism
The term inertia is used deliberately here. The coverage criteria opacity problem does not produce active resistance to innovation. It produces the far more durable problem of passive non-adoption driven by uncertainty, friction, and the absence of a clear path forward.
When a hospital system cannot determine in advance whether an admission for a novel therapy will be approved at the appropriate level of care, the rational institutional response is caution. Formulary committees deprioritize products with reimbursement uncertainty. Utilization review teams develop informal routing rules that steer toward familiar pathways. Physicians who encounter denials early in a product's lifecycle communicate that friction to colleagues. The product's utilization trajectory stalls not because clinicians have evaluated it and found it wanting, but because the administrative pathway is unclear and the cost of navigating it falls entirely on the provider.
For innovators, the feedback loop is slow and opaque. The signal that criteria are misaligned arrives through denied claims, lagging utilization data, and field reports from clinical account teams. There is no formal channel through which a manufacturer can engage a payer or MAC to present the FDA label requirement, the clinical protocol, and a request to ensure the criteria accurately reflect what the therapy requires. The system defaults to adversarial after-the-fact alignment through the appeals process, which is expensive, slow, and poorly suited to the task of building durable coverage clarity for a new product category.
There is no legitimate channel for a manufacturer to engage payers on criteria development for novel products in real time. Alignment, when it occurs, happens through conflict rather than through structured engagement.
What a transparency standard would require
The information blocking framework provides a useful structural model. It does not require universal access to all information. It requires that access not be actively blocked, that legitimate exceptions be defined and bounded, and that actors who benefit from opacity be accountable for demonstrating that their practices fall within defined exceptions.
A coverage criteria transparency standard built on analogous logic would require several things that the current system does not provide. Criteria used in MAC medical review should be publicly disclosed as a condition of their use in coverage determinations, in the same way that LCD and NCD policy is published and subject to comment. Plan-specific modifications to base criteria sets should be disclosed to providers at the time of credentialing or contracting, not buried in provider portals accessible only through authenticated login. When a novel therapy carries FDA label requirements that speak directly to the clinical setting and monitoring intensity of administration, those requirements should constitute a presumptive basis for level-of-care determinations absent specific contradicting clinical evidence, not a starting point for criteria review against a potentially outdated analog.
For novel therapies specifically, a structured pre-coverage engagement pathway, analogous to the FDA's pre-submission meeting process, would allow manufacturers to surface criteria alignment questions before launch rather than after denied claims. That pathway does not exist today in any formal sense. The closest analogs are payer advisory boards and formulary submission processes, both of which are voluntary, relationship-dependent, and focused on formulary placement rather than level-of-care and utilization management criteria.
The path forward
The prior authorization transparency movement demonstrated that structural opacity in the coverage system can be addressed through a combination of regulatory action, legislative pressure, and industry accountability. The CMS-0057-F rule, state-level PA reform legislation, and Gold Carding provisions in several state Medicaid programs all reflect meaningful progress on a problem that was once treated as an immovable feature of the payer landscape.
The level-of-care and utilization management criteria transparency gap is the logical next chapter of that movement. The conceptual case is already made. The information blocking doctrine established that opacity in health information governance is a policy problem, not a neutral market condition. The prior auth transparency framework established that payers can be required to disclose and justify the rules they apply to coverage decisions. Extending that logic to utilization management criteria and level-of-care determination is not a radical step. It is the application of a principle that the policy community has already accepted.
What is missing is the coalition, the regulatory vehicle, and the political moment. Hospitals and health systems have strong operational interests in criteria transparency that they have not yet fully mobilized around level-of-care and utilization management criteria specifically. Manufacturers of novel therapies, particularly those requiring facility-based administration, have a direct commercial interest that has also been underrepresented in this policy conversation. Patient advocacy organizations focused on access to innovative therapies represent the constituency that ultimately bears the cost of inertia.
The information blocking framework did not emerge from a single stakeholder's advocacy. It emerged from a convergence of provider frustration, patient access concerns, and a policy community that was ready to name the problem accurately. The coverage criteria transparency gap is ready for the same convergence. The problem is well-defined. The structural parallel to information blocking is clear. The stakeholder interests are aligned. What remains is the decision to name it as the governance failure it is, rather than accepting it as an immovable feature of how the coverage system works.
Inertia in healthcare innovation is rarely the product of clinical skepticism. It is the product of administrative friction in systems that were not designed to accommodate what they have not seen before.
All views, analyses, and frameworks presented here reflect independent professional judgment informed by more than two decades of experience across payer strategy, clinical transformation, and health system operations. They do not represent the views or positions of any current or former employer or affiliated organization.