Last week, the U.S. Supreme Court issued an odd decision in the case of Douglas v. Independent Living Center. Douglas is the consolidation of three suits challenging cuts in California's Medicaid (Medi-Cal) reimbursement for a wide range of health care services. The Ninth Circuit affirmed lower court decisions halting the cuts because they were found to violate a provision of the Medicaid Act that requires rates be sufficient to ensure equal access to quality care. This provision, 42 USC 1396a(a)(30(A), is commonly known as the "Equal Access" or "30A" Requirement. The Supreme Court did not take up the issue of whether the cuts actually violated this requirement.

The Court only granted cert to decide whether patients and providers had a right to challenge such cuts in federal court using the Supremacy Clause, a Constitutional provision that essentially says federal law preempts conflicting state law. The problem is that the Court never answers this question. In a five-to-four decision, the Court reframed the question and remanded it back to the Ninth Circuit for briefing by the parties because during litigation the rates were approved by CMS (Centers for Medicare and Medicaid Services), the agency charged with administering Medicaid. The revised question is whether private citizens can use the Supremacy Clause to challenge state cuts after CMS has approved them.

As Professor Tim Jost has already commented, the practical effect of delaying the decision is a win for plaintiffs because it preserves their right to sue, for now. The majority declined to follow the dissent's approach, which would have held that the Supremacy Clause cannot be used regardless of CMS approval. Despite this apparent win for plaintiffs, some view the majority's opinion as foreshadowing a more favorable rule for states. According to KPCC news, for example, Governor Brown characterized the decision as giving states leeway in rate-setting and said the court gave the Ninth Circuit "clear instructions that are very favorable to the state's position." And in Professor Kevin Outterson's recent blog, he predicts that on remand the Ninth Circuit will defer to CMS and find no violation of the Supremacy Clause - he says "it won't even be a close decision."

While I agree that language in the majority's opinion provides some evidence of a more favorable approach for states, I disagree that the Douglas decision clearly favors the states or demands unqualified deference to CMS in the future. Rather than clarifying questions about future payment challenges, Douglas raises a new and more challenging question about the level of review required and how courts decide when deference to federal approval is warranted. In fact, three recent decisions from a California district court highlight this question, and are probably a better predictor of what the Ninth Circuit will do on remand. In each case, the court temporarily enjoined Medi-Cal cuts despite CMS approval. Far from resulting in a slam dunk win for states, the court refused to defer to CMS's approval because of serious defects in the state's rate-setting process that made it look arbitrary and capricious, and because CMS's own inconsistency and approval of such a defective process did not warrant the kind of deference typically required for agency action.

The Importance of Deference and the APA

In their predictions about what Douglas means for future payment suits, Governor Brown and Professor Outterson are probably relying on language in the opinion that speculates about two effects of CMS approval - both of which are favorable to states. First, the Court suggests that CMS approval "may change the answer" to the "underlying substantive question" about whether the rates violate federal law. Second, the Court notes that CMS approval changes the procedural posture and, thus, may change the level of scrutiny applied by the courts because it may require the plaintiffs to seek review of the agency determination under the Administrative Procedure Act (APA) rather than in an action against California under the Supremacy Clause. This is relevant, says the Court, because the APA generally provides for judicial review that is deferential to agency action, making the "Supremacy Clause challenge at best redundant." At worst, the Court worried, allowing both claims might "threaten[] potential inconsistency or confusion," if the Supremacy Clause action allows courts to avoid the kind of deference typically demanded under the APA.

In short, the Court is suggesting that federal courts, in fact, may be less likely to invalidate cuts that have been approved by CMS, and that legally courts may have less discretion to do so under this more deferential APA standard. While this certainly sounds good for the states, people should not rush to assume that this will mean a slam dunk for states. For now, the Court can only speculate about the effect of CMS approval because the parties have not briefed these issues; this is why it is remanding the issue back to the Ninth Circuit. Through this process, I expect the Court to learn much more about the level of review that federal courts have been applying to states' rate-setting processes and CMS's decisions. What it will discover is that Supremacy Clause claims have not been used to avoid the typical deference applied to agency action under the APA; federal courts do consider the APA's basic principles of deference in these cases, but deference is not absolute. A look at Medicaid payment suits shows that courts often have good reasons for finding that deference to agency decisions is not warranted, even under the APA.

Limits on Deference: Why States Can Still Lose

The APA does create a deferential standard for reviewing agency action, but the Supreme Court has made clear that a reviewing court can set aside agency action if it is found to be "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." The Court has gone further to explain that a decision is arbitrary and capricious "if the agency relies on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise." When federal courts have invalidated state Medicaid rates, even in the face of CMS approval, it is because they found that the agency action did not meet these minimal standards.

The most common and compelling example of this is where states cut rates in clear violation of federal law, and the federal regulator does nothing about it. California has one of the worst histories of repeatedly disregarding federal rate-setting requirements. It has tried several times to cut rates solely in response to its budgetary needs, and without any consideration for how it would impact access for Medicaid patients -- an act that federal courts have long held to be a violation of federal law. CMS (and its predecessors) have an equally troubling history. Historically, federal regulators have not done much, if anything, to enforce the Equal Access Requirement. They are more likely to actively review and reject state plans when proposed rates look too high, but not when concerns are raised that rates are too low to ensure equal access. Reviews of state plans have been cursory, at best, and are often approved by default. It is in this context that courts have consistently refused to defer to agency action, holding that even traditional APA deference does not require courts to "rubber stamp" federal approval of state plans that are clearly arbitrary and a violation of federal law.

Admittedly, the question about whether deference is warranted is much harder to answer where the state undertakes some process to comply with federal law, and where CMS actively reviews and approves that process. This is the scenario presented by recent developments in California, which is likely why the Supreme Court remanded the case. In the latest round of cuts, California conducted an access review and concluded that the rates complied with federal law. CMS was also engaged in the process; before approving California's rate cuts, CMS requested additional information from state officials concerning the impact on access, and was confident in the state's findings. In fact, under the Obama Administration, CMS generally has been more active in reviewing state plans to determine compliance with the federal Equal Access Requirement, and it has recently proposed regulations that give states guidance about the kind of process they should undertake in order to satisfy this requirement. This certainly does not look like the kind of clearly arbitrary or capricious behavior described above.

Nonetheless, the APA's deferential standard permits courts to take a "hard look" at the agency decision-making process to determine whether an agency has engaged in reasoned decision making or is acting in an arbitrary and capricious manner. Not all courts look closely agency actions, and some defer to agency action despite evidence of serious flaws in the process. But other federal courts have rejected state rates despite federal approval where they find that the state failed to undertake a "bona fide process" that would ensure compliance with federal law. And the Ninth Circuit, in Orthopaedic v. Belshe, has gone the furthest in interpreting 30A as specifically requiring states to consider provider cost as part of the rate-setting process. According to the Ninth Circuit, ignoring provider cost is inconsistent with the federal law, which is one of the grounds for setting aside agency action under the APA.

Recent Injunctions Illustrate the Next Challenge for the Court

Perhaps the best predictor of what the Ninth Circuit will do on remand, and the legal questions likely to reach the Supreme Court in the next round, is the result of three recent California district court opinions enjoining Medi-Cal cuts despite CMS approval. Applying Ninth Circuit precedent, the court found a strong likelihood that the state's rate-setting process, and specifically its access review, was so defective as to be arbitrary and capricious, and it found that deference to CMS approval was not warranted. I have described these cases in greater detail in an earlier blog, but I'd like to give just two examples of the kind of defects that led the court to its conclusion.

First, the court held that state officials' failure to consider provider cost in its rate-setting process violated 30A based on the Orthopaedic case described above. CMS argued for a different interpretation of 30A - one that would not require states to consider provider cost - and it insisted that its interpretation should receive deference. But the court rejected CMS's interpretation and argument for deference, in part, because of CMS's own inconsistency. It was particularly troubled by the fact that CMS was treating provider cost as irrelevant for determining whether the rates are too low to ensure equal access, but that in 2004 CMS argued the opposite position when it denied a state plan amendment by Alaska that would increase Medicaid rates. In that case, CMS denied the plan, in part, because of Alaska's failure to consider provider cost, and, ironically, CMS quoted language from the Ninth Circuit's Orthopaedic decision stating that "the requirements of 30A are ... not so flexible as to allow the [State] to ignore the costs of providing services." To take such apparently inconsistent positions without a rational explanation looks like precisely the kind of arbitrary action that does not warrant APA deference.

The court was also troubled by the state's failure to consider the kind of data necessary to accurately measure access. State officials measured access by counting the numbers of Medicaid providers that accepted at least one claim per year, but did not inquire into how many Medicaid patients these providers were able or willing to see, what kinds of services they could provide, or how this compared to the need of beneficiaries in a specific geographic area. Even CMS (through informal guidance and recent proposed rules) has made clear that this kind of analysis is inadequate, and that states should measure access based on patients' needs and a realistic picture of whether they can access services in a timely manner. Yet CMS approved the rate cuts, without acknowledging this failure or demanding further information.

These recent cases illustrate that some courts continue to be troubled by what they view as arbitrary, capricious and inconsistent decision making by states and CMS, and that they will continue to take a hard look at agency decisions in order to determine whether deference is warranted. If I had to make a prediction, my guess would be that on remand the Ninth Circuit will be equally troubled by the actions of state officials and CMS, and that they will be just as vigilant as the district court in reviewing the agency's process before deciding whether deference is really warranted under the APA. This is why I think Douglas raises more questions than it answers: Douglas' suggestion that CMS approval will trigger deference by courts is not a new concept nor does it mandate a particular outcome because deference is not absolute.

In the last few months, a California district court has issued temporary injunctions to prevent Medi-Cal state cuts challenged in three separate lawsuits: California Hospital Association v. Douglas; California Medical Association v. Douglas; and Managed Pharmacy Care v. Sebelius. The cuts were part of Assembly Bill 97 ("AB 97"), which enacted significant payment reductions for many kinds of services, including skilled nursing, physician, clinic, dental, emergency medical transportation, durable medical equipment and supply, and pharmaceutical services. Lawmakers are using AB 97 to try to solve California's fiscal crisis, but providers and beneficiaries are fighting the cuts. They claim that cuts to provider reimbursement will exacerbate existing barriers to health care for Medi-Cal beneficiaries who already have trouble finding providers willing to serve them, and that the state's rate-setting process violates federal law.

Previously I have blogged about this problem and about the lawsuits brought by Medi-Cal providers and beneficiaries to prevent these kinds of cuts from going into effect. States must comply with procedural and substantive guarantees found at 42 U.S.C. Section 1396a(a)(30)(A), commonly referred to as the "Equal Access Provision" or "30A" requirement. Under this provision, states must "assure that provider payments are consistent with efficiency, economy, and quality of care and are sufficient to enlist enough providers so that care and services are available to the general population in the geographic area." State officials cannot honestly and reasonably make such assurances without undertaking some kind of analysis to assess the impact that rates will likely have on access, and federal courts have consistently held that states cannot cut provider rates solely in response to fiscal concerns. Yet this is precisely what California lawmakers have tried to do a number of times before, until federal courts were asked to intervene and prevented the cuts.

Federal regulators are supposed to oversee the state rate-setting process to ensure 30A compliance. CMS (the Centers for Medicare and Medicaid Services), is the division of the U.S. Department of Health and Human Services that administers Medicaid, and the Medicaid Act requires states to submit proposals to change rates to CMS for approval. Until recently, however, federal regulators have been largely absent in this arena: HHS has failed to promulgate regulations providing guidance to states about how to comply with 30A, though the Obama administration has recently proposed such rules; and HHS has failed to use its enforcement power to reject rate cuts, even where state violations of federal law have been clear and egregious. It is because of this regulatory void that federal courts historically have played such a critical role in Medicaid payment suits.

Significance of the AB 97 Rate Cut Cases

The most recent decisions by a California district court enjoining the AB 97 rate cuts are noteworthy for a couple of reasons. First, both federal regulators and state officials seemed to take their 30A obligations seriously. When California submitted its rate cuts for approval, CMS did not simply rubber stamp them, it requested additional information concerning the implications of the cuts on health care access. California performed an access review, ultimately concluding that the cuts would not adversely impact quality or access. Based on this, CMS approved the cuts.

These cases are also important because of the current threat to providers' and beneficiaries' right to challenge illegal rate cuts in federal court. As I describe in greater detail in a prior blog, the Supreme Court will be deciding this issue soon, and HHS has sided with the states in advocating for the elimination of this judicial last resort. HHS's argument against allowing suits to challenge payment cuts is based, in part, on the fact that HHS has been explicitly charged with Medicaid oversight in rate-setting. Implicitly, HHS seems to be claiming that it can be trusted to exercise its oversight effectively. Certainly, the Obama administration has done more than past administrations to try to earn this trust by stepping up its oversight activity with respect to access and proposing rules to give states guidance for complying with 30A.

There are still many unanswered questions about HHS's stated commitment to enforcing 30A obligations, and these recent cases provide a troubling glimpse of this "commitment" in practice. In all three cases, the court's opinion reveals fundamental flaws in the state's rate-setting process -- flaws that reflect a continuing disregard of federal law, and specifically access and quality protections. In approving these latest cuts, CMS seems to ignore these flaws, California's history of violating federal law, and even California's disregard of CMS's own suggested criteria for determining 30A compliance. Far from instilling trust in CMS's regulatory promises, these most recent cases reinforce the critical role that federal courts will need to continue to play in protecting Medicaid access in the future.

A Closer Look at California's Process: Determining 30A Compliance

Typically, courts are very deferential to action by administrative agencies: they affirm an agency's action unless it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." Federal courts have held that a decision is arbitrary and capricious if the agency relies on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. In this case, however, the California district court did not defer to the state's findings or CMS's approval. It found that the state's process did not comply with 30A, and thus CMS's approval was also inconsistent with federal law. Links to the full opinions are provided above, but I want to briefly highlight three important defects in the state's process that led the court to invalidate the cuts.

The Role of Provider Cost

One of the flaws in the state's process was its failure to consider provider cost as a factor in setting rates. Since 1997, the Ninth Circuit has interpreted 30A to require states to do cost studies and consider these costs in rate-setting. In fact, Medi-Cal cuts have been invalidated in the past on this basis, so lawmakers are aware of this requirement. Their failure to comply is simply the latest in a pattern of disregarding federal rate-setting requirements. The important link between the sufficiency of payment rates, provider cost, and provider participation has been recognized either explicitly or implicitly since Medicaid's enactment in 1965. Whether rates are sufficient to cover the "reasonable cost" of providing care is relevant for determining if rates encourage economical and efficient care (in other words, not too high), as well as for determining if they are consistent with access and quality (in other words, not too low).

CMS's position about the role of provider cost in determining 30A compliance has been, at best, inconsistent. In 1990, Medi-Cal beneficiaries challenged the extremely low payment rates and access for dental care in the case Clark v. Kizer. In the course of litigation, HHS filed an amicus brief explaining the informal criteria it used for measuring 30A compliance, affirming the relevance of provider cost in the process. As recently as 2005, in Alaska Department of Health and Social Services v. CMS, CMS denied Alaska's proposed rate changes out of concern that the rates were too high, citing as one reason for its denial Alaska's failure to consider provider cost. Currently, however, CMS is taking the opposite position: In its recent proposed regulations and in the AB 97 cases, it insists that 30A does not require cost studies, and it seems to devalue the importance of the link between rates, provider cost, and access. The court was disturbed by this inconsistency. CMS cannot assert that cost is relevant in one circumstance (when concerned that rates are too high), but claim that it is irrelevant in another (when the concern is that rates are too low or after a change in administration) without some rational explanation. This is precisely the kind of arbitrary and irrational agency action which is not entitled to judicial deference.

Measuring Beneficiary Access

State officials and CMS seemed to focus their review on access concerns, and, specifically, whether rates would continue to guarantee equal access for Medi-Cal beneficiaries as to those in the general population. However, the state failed to consider the kind of data necessary to effectively measure access. A lack of clear statutory or regulatory requirements for 30A compliance means that states have had a great deal of discretion to establish their own process for access review. However, HHS has offered a framework and suggested criteria for measuring access, through informal guidance, and, more recently, in its proposed rules. This framework includes obvious criteria, such as the level of physician participation, beneficiary need, and patient utilization. But HHS and the California district court in the Clark case have also made clear that these criteria must be assessed meaningfully and accurately; that is, states must try to assess the reality of access on the ground and from the perspective of Medi-Cal beneficiaries.

In the case of the AB 97 rate cuts, the court found that the state failed to do this; instead the state used a process and data that likely created a distorted picture of access. In the case of physician services, for example, the state determined the number of participating Medicaid providers by counting the number of physicians who had submitted at least one claim per year. It did not, however, gather information about how many of these physicians actually treated Medi-Cal beneficiaries regularly, and what kind of services they provided in light of beneficiaries' needs. The state's access analysis also failed to take into account the proximity of providers to beneficiaries, beneficiaries' level of need, and patterns of utilization (what kind of care they received and where). The lack of meaningful inquiry into each of these factors meant that the state made a number of unrealistic and overly optimistic assumptions about the level or kind of services that certain providers would be able or willing to provide.

Quality of Care

Many of the potential access concerns raised by plaintiffs implicate quality of care concerns as well. Plaintiffs alleged the record demonstrated "'no consideration' at all by DHCS or CMS of the impact of the rate reduction on quality of care." In response, state officials argued that they designed a monitoring plan to monitor both access and quality of care. They also claimed that they could rely on existing health and safety laws that require a minimal standard of quality care to ensure that Medi-Cal beneficiaries would get adequate care. The court rejected both responses. Section 30A requires states assess the possible impact of rate cuts on access and quality prior to rate implementation, and to use this information to either adjust the rates or implement other safeguards to prevent unnecessary harm. The state's monitoring plan was inadequate because it took a wait-and-see approach, creating a potential response only after a problem had been identified. Moreover, a number of courts have held that states can not rely on external quality protections as a way to satisfy 30A requirements. States must do their own quality impact assessment based on actual delivery trends and patient needs. States cannot use the fact that other laws regulate quality as a way to avoid their obligation to assure "that the payments themselves be consistent with quality care."

So what does this mean for the future of Medicaid access?

In a prior blog I posed a similar question about whether the Medicaid expansion would really improve health care access in light of existing barriers, including low reimbursement. At that time, I suggested that the answer would depend, in part, on the willingness of CMS to meaningfully review state Medicaid plans and to use its oversight power to enforce federal access protections. This case gives us the first glimpse of what we can expect under the Obama administration, and what it reveals is not very hopeful.