February 5, 2018
Source: Health Affairs
Photo Source: Unsplash,
Experimenting On The Health Of The Poor: Inside Stewart v. Azar
Two Medicaid work demonstrations have now been approved: Kentucky’s on January 12 and Indiana’s, two days ago.
In a theoretical world, the recent Washington Post story, which focuses on whether Medicaid work demonstrations actually trigger more ACA Medicaid expansions, might suggest the broader value of allowing states to experiment with work requirements. Were the policy to be used this way, it would cut against the Administration’s deeper intent: Since taking office, the Administration has been vocal in its opposition to Medicaid for the poorest adults, labeling it a “clear departure from the core, historical mission of the program [that will] deprioritize the most vulnerable populations.” It would be ironic indeed were several more states to launch an expansion under these terms.
But neither Kentucky’s nor Indiana’s poor live in a theoretical world. The Kentucky HEALTH experiment shows the consequences of using work, reporting requirements, and lock-outs not to temper the reach of an expansion but strip benefits away. Indiana appears to do the same. Simply put, whatever vague and totally pragmatic offsetting gains might come from the Administration’s January 11 Medicaid work experiment solicitation, it is clear that in an Indiana and Kentucky context, the experiment simply is a significant threat to health and health care.
The Significance Of 1115 As An Experimental Statute
Section 1115 is not an open-ended invitation to the Executive Branch to unilaterally redesign social programs, which itself could raise constitutional separation-of-powers protections. To the contrary, 1115 rests on the concept of sound, ethical, and lawful experiments. Indeed, it is paradoxical that only last week the Administration allowed final regulations protecting human research subjects to take effect. A 1983 decision by the Reagan Administration to exempt 1115 demonstrations from human subject rules was accompanied by promises of careful attention to the risks raised by social welfare experimentation. Even a cursory consideration of the terms under which the Administration will permit work demonstrations underscores that this promise has not been kept.
As the complaint in Stewart v. Hargan (now Stewart v. Azar), filed on January 24 in federal district court in Washington D.C. makes abundantly clear, both the work experiment solicitation and the approval of Kentucky’s 1115 waiver in its wake are remarkable for the clarity of their intent to use punishment—in the form of denial or loss of health insurance coverage—to promote work. It doesn’t take a wealth of empirical research to conclude that people who are healthy probably are more likely to hold jobs and work more. Furthermore, there is actual empirical research that suggests that the very act of providing access to health care under the ACA expansion has had a positive effect on health and work. There simply is no evidence to support any reasonable hypothesis that health can be promoted by threatening the future and stability of health care for people whose health is worse to begin with and who—owing to the concentrated nature of poverty in the U.S.—tend to live in communities with higher unemployment and reduced job prospects.
To be sure, in its effort to induce states to adopt the ACA Medicaid expansion after the United States Supreme Court effectively made it optional, the Obama Administration agreed to 1115 demonstrations that contained a number of critical dimensions that carried health risks for the expansion population, including imposition of premiums that normally are barred and lock-out penalties for certain expansion adults for non-payment. But in this regard, the Obama-era experiments essentially followed a long history of Medicaid eligibility expansion demonstrations pursued by Democratic and Republican Presidents alike, all of which tend to mix risks and gains. The Clinton Administration permission for Oregon to combine a Medicaid adult expansion with a now-famous rationing approach to coverage for mothers and children was so provocative that 60 Minutes devoted a full segment to the experiment.
In the same vein, President Clinton approved the TennCare demonstration, which, like Oregon’s demonstration, combined an expansion with other spending curbs, in TennCare’s case mandatory managed care enrollment for hundreds of thousands of Medicaid beneficiaries. The George W. Bush Administration allowed Utah to couple limited benefits for a new group of extremely impoverished working-age adults with benefit reductions for traditional populations.
But the Trump Administration has now proposed to blow past the guardrails drawn by its predecessors. The Administration is pursuing work requirements and make-work reporting requirements designed to reach both traditional and expansion populations alike. The Administration raises “considerations” couched as optional cautionary notes, not as enforceable limits on state design options. Indeed, as the complaint points out, the Kentucky approval gives the state broad powers to define
who must work,
what qualifies as work,
who will be exempt from the requirements,
how beneficiaries will be able to earn back key (and ironically essential to work) vision and dental benefits,
how its complex new health savings account plans will be designed and operationalized,
the premiums people will have to play,
which populations can be penalized for non-payment, and
what barriers people locked out of insurance will have to overcome in order to earn back their coverage.
Every dimension of the Kentucky demonstration is in play, with no hypothesis other than the assumption that people are healthier when they work. The evidence regarding how Medicaid actually promotes work is not even acknowledged.
Stewart is the latest in a line of cases stretching back to the early 1970s that have challenged the Secretary’s use of 1115 authority to conduct certain types of experiments. Many of these cases have involved cash welfare experiments, while others have focused on Medicaid. See Aguayo v. Richardson, 473 F 2d 1090 (2d Cir., 1973), Crane v. Mathews, 417 F. Supp. 532 (N.D. GA, 1977), C.K. v. New Jersey Department of Health and Human Services, 92 F. 3d 171 (3d Cir., 1996), Beno v. Shalala, 30 F. 3d 1057 (9th Cir., 1994); Newton-Nations v. Betlach, 660 F. 3d 370 (9th Cir., 2011), G v. Hawaii 676 F. Supp. 2d 1006 (D.HI 2009).
As Nick Bagley points out in a recent editorial, few of these cases have succeeded. But none of the previous Medicaid 1115 cases comes even close to the extreme example posed by Stewart, not only in the sweep of the experiment but in the clear evidence of intent on the part of the Administration to use demonstration powers to take benefits away from people whom it, on the record, considers undeserving. Indeed, even a rapid reading of the complaint—what the Secretary permits, what the state intends to do, and the threats the experiment poses to the individual plaintiffs—underscores the ways in which the Administration’s new guidance and Kentucky’s demonstration appear to violate the law and the scope of the harm they could cause.
Inside the Stewart Complaint
As with all complex, fact-driven litigation, the Stewart complaint is designed to lay out both facts and legal claims as simply and clearly as possible. The complaint begins with an explanation of Medicaid’s basic purpose and structure. It then considers the scope of HHS authority under 1115, turns to the situations confronting the individual plaintiffs in the case (given the magnitude of the impact of the demonstration, the litigants not surprisingly seek class certification), and explains how the demonstration violates federal law.
Medicaid's Core Purpose And Essential Structure
The complaint describes Medicaid’s basic purpose, codified directly in the statute: “to furnish medical assistance [to people] whose income and resources are insufficient to meet the cost of necessary medical services” and to provide “rehabilitation and other services to help such families and individuals attain or retain capability for independence and self care.”
In furtherance of this purpose, the complaint notes, the statute offers states a generous financial deal aimed at assuring their acceptance of the conditions that come with the funding. These conditions include coverage for all who need assistance and qualify for it, and a full embrace of people who present health and social risks. Medicaid’s design also bars discrimination against members of assistance groups covered under a state Medicaid plan, and explicitly bars the addition of eligibility factors not contained in the statute itself.
The complaint lays out other core features of the statute designed to promote Medicaid’s ability to function not simply as an ordinary insurer but as a health care safety net program:
retroactive eligibility, which protects against the denial of catastrophic, high cost care;
a best-interest standard in measuring state program administration;
continuous coverage without interruption through a process of automatic renewal;
the opportunity to quickly gain reinstatement if coverage lapses as a result of delays in required reporting; and
strict limits on premiums and cost sharing that contain explicit statutory prohibitions, drafted directly into the Medicaid statute itself (and nearly always overlooked), that virtually bar what otherwise might be permissible 1115 demonstrations aimed at elevating the out-of-pocket cost burden. (This direct constraint on the discretion of the HHS Secretary was added to Medicaid in 1982 by a Republican Senate and Democratic House, alarmed at planned Medicaid cost-sharing experiments planned by the Reagan Administration).
The Limits Of 1115
Medicaid’s core features, essential structure, and express prohibitions on certain types of experiments essentially put the brakes on how far 1115 experiments can go. Furthermore, as the complaint notes, 1115 as an experimental statute has its own brakes. Its agency powers can be invoked only for experiments that further program objectives. In order to proceed, the HHS Secretary must build a record showing the value and relevance of the experiment and must proceed on an experimental basis. Certain procedural requirements apply.
Above all, perhaps, only certain parts of the Medicaid statute can be waived; 1115 does not provide carte blanche to waive the Medicaid act in its entirety, and it offers no basis for waiving other federal laws that continue to apply, ranging from federal civil rights laws to the Fair Labor Standards Act. Ultimately 1115 is a relatively narrow grant of agency authority to test certain program changes of potential value given the program’s objectives, not to simply invent a new program that has political appeal.
Kentucky's Aim: Selectively Culling The Eligible Population
Given the fact that federal law bars states from selecting winners and losers among the eligibility groups they cover, what is so striking about the Kentucky experiment is the lengths to which the state and its federal partner have gone to come up with a strategy, under the guise of experimentation, for selectively culling the Medicaid population. This selective culling of current and future beneficiaries is set to unfold against perhaps the nation’s most successful ACA Medicaid expansion, a fact illustrated by extensive evidence. As the complaint notes, some 428,000 people have gained coverage; hundreds of thousands have received essential medical care, with improved access to office visits, preventive dental care, tests to identify and manage chronic conditions, and treatment for serious conditions such as substance use disorders. The complaint also cites other research evidence pointing to the positive impact of the expansion: increased use of preventive care, decreased reliance on emergency departments, and reduced out-of-pocket spending for prescribed drugs essential to treating serious health conditions.
What clearly emerges against this evidentiary backdrop is the extent to which the experiment is likely to, and in fact designed to, reverse these figures. The state’s proposal is framed as an ultimatum, according to the complaint: the “terms under which the Commonwealth will continue Medicaid expansion.” Its purpose is to thin the ranks in ways not permitted under federal law to save money. Short of repealing entire categories of optional coverage groups, a rank-thinning experimental design thus is required to do this.
This design combines work requirements with new program features that, according to the state, reflect the far less forgiving world of commercial health insurance. In this version of Medicaid, its core objective becomes the inverse of its statutory aim: to limit access as much as possible so that people will stop relying on it. In this regard, the long-term deterrent potential of the proposed design elements probably is as important as their near-term impact—communities will quickly learn about the hoops they will need to jump through to get and keep Medicaid and enrollment will fall.
These hoops are multiple: Work requirements with few, poorly defined exemptions; lengthy loss of coverage for failure to follow additional reporting requirements; stripping away essential benefits with little hope of regaining them; using Medicaid to expose people to the limits of commercial insurance through escalating premiums, delayed enrollment, and elevated cost-sharing; and selectively withholding benefits from people considered undeserving.
The Plaintiffs' Legal Arguments
The plaintiffs make several basic legal arguments. The first argument is that the HHS Secretary violated APA procedures by not issuing the January 11 work experiment solicitation as a proposed rule. Issuing the document as a proposed rule would have given affected persons the ability to comment on its general terms, the basis of its justification, or its application to specific cases such as those, like Kentucky, where the demonstration has been intentionally deigned to have a downward and punitive impact on enrollment.
The second argument is that, in several respects, the Secretary has exceeded the outer bounds of his 1115 authority by approving work requirements. Conditioning Medicaid on work, according to plaintiffs, adds a condition that has never been recognized as falling within the bounds of Medicaid eligibility. (Indeed, when TANF was first created in the 1996 welfare reform, Congress deliberately amended federal Medicaid law to preserve Medicaid for those who had received it because of their entitlement to aid under the predecessor AFDC program and who now risked the loss of cash welfare because of work requirements and other restrictions built into TANF.) Furthermore, creating a work requirement implicates other federal laws related to the civil rights of people with disabilities and federal labor laws, neither of which are covered by 1115.
Finally, plaintiffs claim, a work experiment that punishes people who cannot fulfill its requirements undermines rather than promoting Medicaid’s basic objective. What is so striking in this case is that although Nick Bagley is correct that courts are hesitant to infer motive in reviewing the lawfulness of agency action, in this case the Administration could not have been clearer about its desire to block Medicaid for poor working-age adults.
The third and fourth arguments focus on the fact that the Medicaid statute itself blocks 1115 demonstrations that impose premium and cost-sharing requirements, because, in the absence of highly specific standards not present in this case, the statute bars experiments aimed at exposing the poor to higher out-of-pocket costs. These special standards (42 U.S.C. § 1396o) apply a 5-factor test for Medicaid cost sharing experiments:
a demonstration that will test “unique and previously untested use of copayments”;
an experimental time period not to exceed two years;
an experiment that “provide[s] benefits to recipients of medical assistance which can reasonably be expected to be equivalent to the risks”;
an experiment that is “based on a reasonable hypothesis which the demonstration is designed to test in a methodologically sound manner, including the use of control groups of similar recipients of medical assistance in the area; and
an experiment that “is voluntary, or makes provision for assumption of liability for preventable damage to the health of recipients of medical assistance resulting from involuntary participation.” This last factor is a direct reflection of the human subject protection standards that, until their revocation by the Reagan Administration, had been held by at least one court to apply to social welfare experiments under 1115.
In their fifth, sixth and seventh claims, plaintiffs argue that lock-out penalties, elimination of retroactive eligibility, and elimination of non-emergency medical transportation lie outside the authority of the Secretary to permit under 1115 because these alterations carry no legitimate experimental purpose that is likely to promote Medicaid objectives.
Plaintiffs’ eighth claim goes directly to the Kentucky approval in its entirety, arguing that in approving the HEALTH experiment, the Secretary “relied on factors which Congress has not intended him to consider, entirely failed to consider several important aspects of the problem,” and failed to make the kind of findings regarding the experiment’s relationship to Medicaid’s core objectives that brings an experiment within the limits of 1115. Indeed, the findings contained in the Kentucky approval letter go virtually entirely to the value of work rather than the value of Medicaid.
Plaintiffs’ final argument is a constitutional one. This claim focuses on two separate provisions of the United States Constitution: Article 1 § 1, which reserves the power to write laws to Congress, and Article 1 § 3, which requires the President to “take Care that the Laws be faithfully executed.” In essence, plaintiffs argue that in using 1115 to rewrite Medicaid in ways that reflect a political agenda rather than careful experiments aimed at promoting Medicaid’s objectives, the Administration has overstepped Presidential bounds.
The Plaintiffs
It is worth considering the facts that the 15 named plaintiffs present the court about themselves and the threatened impact of the experiment. This portion of the complaint, which covers nearly 30 pages, paints a remarkable portrait of the essential role played by access to Medicaid. A few of the cases:
Ronnie Maurice Stewart is a 62-year-old man who worked most of his life, has lost his job, and experienced homelessness. He works, but owing to his poor health can no longer do so for long periods. He is impoverished.
Glassie Mae Kasey, 56 years old, has spent her life in physical, low-paid work; her health precludes such work now. She subsists on unemployment benefits of $300 every two weeks.
Lakin Branham is 20. She lives in poverty, does not drive, lives in a county with no public transportation, and takes care of multiple family members.
The Ballingers are a young family. The husband works full-time at Hobby Lobby for poverty-level wages; his hours fluctuate constantly (a common aspect of many of the working plaintiffs’ stories), making continual reporting of changes in income exceptionally difficult. His wife has a series of medical conditions. Their monthly expenses already exceed monthly income. Both need vision care.
William Bennett, 47-years old, takes care of his disabled son, works as mortician for as many hours as work is available, and has a monthly income that averages 50 percent of the federal poverty level.
Medicaid is essential for these and the other plaintiffs, whose lives are ones of great hardship. Rather than promoting their ability to at least secure necessary medical care, HEALTH will undermine it.
Final Thoughts
For most of its life, social welfare experimentation has been part of the Medicaid policy landscape. Experimentation has led to some of Medicaid’s most important gains, including expanded eligibility for low-income adults, the managed care transformation, and a progressive approach to long-term services and supports. It is certainly the case that well-designed experiments can fulfill broader policy aims, and it is also the case that experiments—even those that promote access to social welfare benefits—typically carry risks as well as benefits.
Despite its historic role in shaping Medicaid policy, however, 1115 is a statute that grants limited powers. At its heart, 1115 is designed to fulfill a specific purpose: to enable Medicaid to better achieve its objectives. What makes the January 11 Medicaid work solicitation and the Kentucky approval such outliers is the singular absence of an experiment that promotes Medicaid’s core objectives. The relationship between work and health is axiomatic but this axiom alone cannot serve as the basis for threatening insurance coverage for people who cannot fulfill work requirements.
Indeed, so far afield from Medicaid’s objectives are the Administration’s work experiments that, in its solicitation, the Administration makes clear that it will permit no federal funds to be claimed for costs associated with work supports, nor will it permit states to keep savings attributable to the loss of eligibility for reinvestment in other Medicaid activities. These two fundamental caveats drive home the degree to which the Trump Administration has essentially admitted that, rather than advancing policy goals involving health care for the poor, Medicaid work experiments undermine them.
How can such practices impact your health? How? Why?
Share the wealth of health with your family and friends by sharing this article with 3 people today.
If this article was helpful to you, donate to the Shidonna Raven Garden and Cook E-Magazine Today. Thank you in advance.