• Health care reform after King vs. Burwell

    The Supreme Court ruled against the plaintiffs in King vs. Burwell. In the aftermath, Michael Cannon, a principal architect of the plaintiffs’ suit, wrote that

    the battle to set in place a health care system that works for all Americans is far from over.

    Cannon is absolutely right.

    King was a victory because it prevented millions of people from losing insurance coverage. But it did not advance the cause of health care reform a centimeter with respect to the status quo ante King.

    In The New Republic, I argue that the principal goals of health care reform remain to be fulfilled. Getting them fulfilled will require us to win new political fights to extend universal health insurance in every state. We need to keep working on innovative health care delivery models that control the growth in health care expenditures while improving the quality of care.

    Above all, we need to get more empirical:

    Health care reform has to be driven by results, not political beliefs. Programs should be selected based on evidence, such as the results of randomized clinical trials. Every new reform should collect rigorous data to determine whether it works.


    Comments closed
  • The Supreme Court just denied an ACA appeal. That’s a relief.

    The Supreme Court today declined to hear Maine’s appeal in a case involving a constitutional challenge to an ACA provision relating to Medicaid. That’s good news for supporters of the law, who had no desire to relive the fight over constitutional coercion in yet another case about the ACA.

    What’s the case about? The ACA says that, as of the date of its enactment, the states can’t restrict the category of kids and young adults who are eligible to receive Medicaid. Known as a “maintenance of effort” provision, the ACA freezes Maine’s eligibility rules for that population.

    Before the ACA, Maine chose to cover 18- to 20-year-olds in its Medicaid program. After the ACA, the state wanted to restrict its eligibility rules—but the maintenance-of-effort provision said it couldn’t. So Maine sued, arguing that the ACA placed an unconstitutionally coercive condition on the receipt of its Medicaid dollars.

    The First Circuit—quite properly in my view—rejected the argument. Under the Chief Justice’s controlling opinion in NFIB v. Sebelius, a federal spending condition is unconstitutional where failure to adhere to the condition not only threatens the loss of federal funds tied to a given program, but also the loss of federal funds tied to a substantial and independent program.

    In NFIB, the Chief concluded that the Medicaid expansion was a new program—call it Obamacaid. Because a state had to adhere to the rules of Obamacaid or lose all of its existing Medicaid money, the Chief reasoned, the expansion was unconstitutionally coercive.

    Here, however, the maintenance-of-effort provision applied only to Maine’s traditional Medicaid program. By the Chief’s lights, adjusting the rules for an existing program isn’t unconstitutionally coercive since it doesn’t threaten the loss of independent federal funding. If Maine didn’t wish to comply, it would lose all of its Medicaid dollars—but it wouldn’t also lose, say, its education funding.

    The Supreme Court’s rejection of Maine’s petition suggests that the Chief is not inclined to revisit his coercion decision in NFIB. If he was convinced that the First Circuit botched it, he probably could have persuaded his conservative colleagues, who have a more expansive view of what counts as unconstitutional coercion, to hear the case.

    More subtly, the Court’s refusal to hear the case suggests that the Court won’t use King v. Burwell to shift the law on what counts as unconstitutional coercion. Were such an opinion in the works, the Court would have held onto Maine’s petition and, after issuing King, given the First Circuit a chance to reconsider its decision.

    Does that portend anything about the outcome in King? I don’t think so. The justices who think that adopting the plaintiffs’ interpretation raises coercion concerns—Justice Kennedy, perhaps, or even the Chief Justice— could rule for the plaintiffs on statutory grounds, while flagging the constitutional concerns for future litigation. Alternatively, they could rule for the government in an effort to avoid potential constitutional concerns.

    Either way, the Court’s decision wouldn’t call into question the outcome of the First Circuit case. Beyond signaling that the Court won’t use King to rewrite the coercion doctrine—an unlikely outcome in any event—the denial of Maine’s petition doesn’t tell us much about King. We’re still in a waiting game.


    Comments closed
  • Some things we know about Medicaid managed care

    As I try to get my ahead around some aspects of Medicaid managed care, here are some things we know from various sources.

    Via Jay Hancock:

    • “Avalere Health projects that 76 percent of beneficiaries  in Medicaid and the related Children’s Health Insurance Program (CHIP) will be covered by private managed care by 2016.”
    • “Private insurers booked $115 billion in Medicaid revenue last year, according to data compiled from regulatory filings by Mark Farrah Associates and analyzed by Kaiser Health News.”
    • “Operating profit on those premiums came to $2.4 billion. Net profit, after accounting for taxes, depreciation and other expenses not directly connected to health coverage, would have been less.”
    • Among the new, proposed regulations of Medicaid managed care plans, “HHS now wants states to certify at least annually, perhaps based on direct queries to doctors, that enough caregivers are in the managed-care network and close enough to plan members to serve them.”

    Via Sarah Ferris:

    • “Nearly 30 million people are now enrolled through managed care, which amounts to about two-thirds of people with Medicaid. That’s up from just 2.7 million people with managed care in 1991.”
    • Both Ferris and Hancock discuss the new medical loss ratio rule. You can click through for that. It’s not my main interest at the moment.

    Via Kyle Caswell and Sharon Long, examining Medical Expenditure Panel Survey data on non-elderly adults from 2006-2009:

    • “[F]or nondisabled adults [] increased [Medicaid managed care or MMC] penetration is associated with increased probability of an emergency department visit, difficulty seeing a specialist, and unmet need for prescription drugs.” Some other work they cite, but not all of it, is consistent with these findings.
    • Consistent with prior work they cite (here’s some, the working paper version of which I mentioned here), Medicaid managed care penetration “is not associated with reduced expenditures. We find no association between penetration and health care outcomes for disabled adults.”
    • “[T]he primary gains from MMC may be administrative simplicity and budget predictability for states rather than reduced expenditures or improved access for individuals.”
    • Their analysis excluded Medicaid-Medicare dual enrollees and those with less than full year coverage. Separate analyses of the SSI population were generally not statistically significant, but the sample was smaller.

    Via a KFF brief by Julia Paradise and MaryBeth Musumeci, anticipating issues to watch for in the new regulations (written before they were out):

    • “[M]ore than half of all Medicaid beneficiaries are enrolled in risk-based managed care organizations (MCOs) through which they receive all or most of their care.”
    • “Not all state Medicaid programs contract with MCOs, but a large and growing number are doing so, and some states mandate that beneficiaries enroll in MCOs to receive Medicaid benefits.”
    • “In FY 2013, capitation payments to comprehensive MCOs accounted for about 28% of Medicaid spending nationally.”
    • “The federal regulations, last updated in 2002, set forth state responsibilities and requirements in areas including enrollee rights and protections, quality assessment and performance improvement (including provider access standards), external quality review, grievances and appeals, program integrity, and sanctions.”
    • Because I am hoping to see a new issue brief that summarizes what’s actually in the proposed regs, I’m not going go through this brief in detail. Suffice it to say, as things to look for in the new regs, the brief covers availability and accessibility of plan information, enrollee appeal rights, provider network adequacy, quality of care, long-term care services and supports, actuarial soundness of capitation rates, medical loss ratio, encounter data, and program integrity (basically, auditing contractors).

    KFF’s Medicaid Managed Care Market Tracker has some useful data. A report, based on it, provides some charts and key insights. Among them:

    • “Many state Medicaid programs are expanding their reliance on MCOs. In a recent 50-state survey of Medicaid directors conducted by the Kaiser Commission on Medicaid and the Uninsured, half the states reported taking action in 2014 to enroll additional Medicaid eligibility groups in MCOs. These states include California, New York, Texas, Florida, and Illinois – the five states with the largest Medicaid populations. A smaller number of states expanded their managed care programs geographically and/or shifted from voluntary to mandatory MCO enrollment. Nearly half the states plan to expand their risk-based managed care programs in 2015 as well.”
    • “A number of large health insurance companies have a significant stake in the Medicaid managed care market. Currently, 16 firms own Medicaid MCOs in two or more states, including five firms – UnitedHealth Group, WellPoint, Centene, Aetna, and Molina – that have Medicaid MCOs in 10 or more states. Eleven of the 16 multi-state parent firms are publicly traded; eight of these 11, including the five just mentioned, are ranked in the Fortune 500. The other five multi-state parent firms are nonprofit companies.”
    • Also, from the tracker directly, of the 38 states (plus DC) with MCO contracts, 25 have five or fewer and 14 have three or fewer. Suffice it to say, competition isn’t robust in many states.

    Via a Dan Diamond tweet, year-over-year stock prices of Medicaid managed care-offering organizations are way up:


    (May 2015-May2014)


    Comments closed
  • Adding a work requirement to Medicaid is a bad idea

    Some red states are considering whether they should accept the ACA’s Medicaid expansion, but only if they can get a waiver from the federal government allowing them to attach a requirement that recipients have or seek paid employment. In The New Republic, I offer five reasons why this is a terrible idea.


    Comments closed
  • Sidelining the courts in Medicaid enforcement

    The legal dispute at the core of last week’s Supreme Court decision in Armstrong v. Exceptional Child Center, Inc. is arcane, but the real-world consequences are not. After Armstrong, it’ll be easier for states to slash their Medicaid payment rates, even where the cuts make it difficult or impossible for Medicaid beneficiaries to find a doctor.

    What’s Armstrong all about? The states have loads of discretion in how they run their Medicaid programs and, specifically, in how much to pay for medical services. But that discretion is cabined by the terms of the federal Medicaid statute, which says, in section 30(A), that states can’t drop their payment rates so low that Medicaid beneficiaries find it harder than the rest of us to get needed care. (I first wrote about Armstrong-type cases a couple of years ago.)

    The question in Armstrong is whether the courts can enforce section 30(A) at the behest of private parties—typically providers who are upset at payment cuts. Before Armstrong, the answer was yes. The circuit courts were generally willing to invoke their equitable powers to enjoin state Medicaid officials from implementing spending cuts that conflicted with 30(A). The idea, conventionally traced back to the 1908 case of Ex parte Young, is that the federal courts won’t stand for it when state officials flout federal law.

    When it comes to Medicaid, however, neither the states nor the federal government were happy with this arrangement. In their view, the federal government—specifically, the officials within CMS who review and approve state Medicaid plans—has sole responsibility for holding states’ feet to the fire. If CMS says that state payment rates comply with 30(A), courts can’t second-guess that determination in a lawsuit against state officials.

    That argument makes a lot of sense to me. It also made a lot of sense to a five-justice majority on the Supreme Court. How are judges supposed to know if payment cuts will drive too many providers out of the Medicaid market? To decide that question, as Justice Breyer explained in a concurring opinion, you’d have to know something about “the actual cost of providing quality services, including personnel and total operating expenses; changes in public expectations with respect to delivery of services; inflation; a comparison of rates paid in neighboring States for comparable services; and a comparison of any rates paid for comparable services in other public or private capacities.”

    Courts aren’t well-equipped to undertake this kind of inquiry. Better to leave the question to an agency with expertise in payment rates. The decision in Armstrong elevates that idea into a holding: given the open-endedness of 30(A), Congress must have “implicitly preclude[d]” private enforcement.

    That’s where I start to have qualms about Armstrong—qualms that the four dissenting justices emphatically share. When Congress enacted 30(A) in 1989, it did so knowing that Ex parte Young would allow courts to enforce the provision directly against state officials. Congress could have—but did not—close the courthouse doors to such lawsuits. Nor has it done so in the twenty-six years since.

    A compelling amicus brief from former HHS officials explains that, as a result, “[e]very aspect of [CMS’s] administration of the Medicaid program—from its regulations to its annual budget—is premised on the understanding that private parties will shoulder much of the enforcement burden.” With private enforcement as a backstop, Congress hasn’t funded CMS at anywhere near the levels necessary to enforce 30(A).

    In other words, it’s fine and good to say that CMS should enforce the Medicaid statute. But what if it doesn’t? What if it can’t? And what if Congress never expected it to?

    That’s the quandary of Armstrong. For what it’s worth, I think it’s a genuinely hard case. I’m sympathetic with the view that the courts shouldn’t be responsible for enforcing 30(A) against the states. The question of how much to spend on Medicaid beneficiaries ought to be hashed out in negotiations between the states and the federal government, not in a courtroom.

    But I fear that the Supreme Court may have substituted its own views about sound enforcement strategy for the strategy that Congress actually adopted. In so doing, the Court has created an enforcement vacuum—one that cash-strapped states could exploit to make life even harder for Medicaid beneficiaries.


    Comments closed
  • AcademyHealth: How do changes in Medicaid affect access for kids?

    A lot of time, and a lot of ink, has been spent talking about access and Medicaid. Many who oppose the expansion of the program will point to the fact that sometimes evidence shows that fewer doctors accept Medicaid insurance than other types of coverage. There’s some truth there. Medicaid does often reimburse at a lower rate than other insurance coverage, and sometimes doctors don’t want to accept those lower rates. But there’s more to the story.

    That’s me in my latest post over at the AcademyHealth blog. Go read the rest!


    Comments closed
  • Healthcare reform is a marathon, not a sprint

    King v. Burwell will be argued before the Supreme Court this week. It won’t be decided, just argued. Then, we’ll parse how well the arguments went, and whether we can predict what will happen from that show. Last time, by the way, we couldn’t. Then, in June, we’ll get a decision.

    I know lots of people (including possibly everyone else at this blog) think the plaintiffs will win. Putting my cards on the table – I’m not one of them. I think the Justices will rule for the government, but I admit that’s a gut feeling. I may be wrong.

    But the world won’t end overnight, and neither will the ACA. It will still function in a subset of states. I think even more will take quick action to fix their exchanges. I have a hard time believing that any state that accepted the Medicaid expansion won’t find a way to accept subsidies for their citizens who are working. So we’ll be left with a minority of states that don’t have subsidies and have a broken individual insurance market.

    That’s where history comes in. I’ve said it before, but it bears repeating: All of this has happened before and will happen again. Medicare was, at one time, the death of freedom. Go read this piece on what people thought when “traditional” Medicaid was first passed in 1965. Then remember that Arizona, the last state to accept Medicaid, finally did so in 1982.

    Between 1965 and 1982, we had a country where a program existed to cover all poor children, all poor pregnant women, the poor elderly, and many poor parents – but only in some states and not in others. The Earth continued to spin on its axis. The country survived. It was ridiculously unfair for some people who lived in states that refused to accept Medicaid, but eventually, all the states did.

    The same can be said of the US if the Supreme Court finds for the plaintiffs. But, as with Medicaid, I think it’s likely the ACA will survive. I also believe, someday, that someday the Supreme Court will view the removal of the ACA as “coercive” as it viewed the removal of traditional Medicaid just a few years ago.


    Comments closed
  • Medicaid fees and access to care

    A new paper in NEJM by Daniel Polsky and colleagues sheds light on the impact of an increase in Medicaid payment rates to selected providers in 2013 and 2014. The increase of fees, which bumped up Medicaid payments to Medicare levels, was part of the ACA and designed to increase access to primary care for Medicaid enrollees. The fee increase expired on January 1, 2015.

    The fee bump was substantial. On average it raised Medicaid reimbursement for primary care by 50%.

    The investigators used the secret shopper method, calling providers to assess the availability and wait times for appointments for either new Medicaid or new privately insured patients in ten states. Their data spanned November 2012 through July 2014. They found that

    [t]he availability of primary care appointments in the Medicaid group increased by 7.7 percentage points, from 58.7% to 66.4%, between the two time periods. The states with the largest increases in availability tended to be those with the largest increases in reimbursements, with an estimated increase of 1.25 percentage points in availability per 10% increase in Medicaid reimbursements (P = 0.03). No such association was observed in the private-insurance group. During the same periods, waiting times to a scheduled new-patient appointment remained stable over time in the two study groups.

    Some comments:

    • One of the chief, longtime complaints about Medicaid, heard across the political spectrum, is that it pays too little, harming access and care. There’s even a big lawsuit about it.
    • This study is strong evidence that the bump down in payment that occurred at the start of this year will adversely affect access. “Currently, only 16 states plan to continue the reimbursement increases,” the authors wrote.  So, a problem has been identified (Medicaid pays too little), a solution tested (pay more), with clear outcomes (better access, by one measure). The policy implications are fairly clear.
    • There are a number of limitations discussed in the paper. I want to add or emphasize two:
      • I believe the observed effect is causal, at least in direction. It’s completely reasonable that higher payment begets greater access, as it was measured. It’s possible, though, that other factors also affected Medicaid access, even relative to privately insured patients. Perhaps greater education and outreach as part of coverage expansion played a role, differentially affecting providers’ willingness to offer care to Medicaid vs. privately insured patients. We can’t be sure. Still, a stronger study design for the policy as implemented is hard to develop. We didn’t randomize states to fee increases, for instance.
      • There are other, important ways to measure access, as I have discussed. It is possible that the degree of access as it was measured in this study (via secret shoppers) differs from that as measured in other ways (like surveying patients, both with new or continuing coverage). It’s important to examine both aspects of access, as well as consider access to specialists. This study only considered one type of access to primary care.

    For all that, it’s a very good study. To be sure, we should continue to collect data and study access. Still, about the Medicaid fee bump, I’m ready to say to policymakers, it’s your move.

    (See also Harold Pollack’s post, which makes some of the same points, only with a lot more style.)


    Comments closed
  • AcademyHealth: The return on investment for Medicaid

    My latest at AcademyHealth:

    There are many arguments as to why Medicaid is a good thing for children. Many studies have been done comparing outcomes for children with and without Medicaid. Many more have looked at how access to the health care system is different for kids with Medicaid.

    But concerns about Medicaid, and arguments about whether to expand it, inevitably devolve to the cost. Implicit in that concern is whether it’s “worth it” to have children on Medicaid. Are the benefits worth the cost? Might they be achieved by more efficient means? Perhaps money put into Medicaid could be used for other things.

    Many of these discussions, however, ignore some of the potential long-term return on investment of the program for children. In a recent NBER paper, David Brown, Amanda Kowalski, and Ithai Lurie attempted to get at that question. “Medicaid as an Investment in Children: What is the Long-Term Impact on Tax Receipts?”

    I discuss this paper, and what the results mean. Go read!


    Comments closed
  • Medicaid’s return on investment

    It’s time to face the fact that I’m not going to be able read, let alone write a full post on every interesting looking paper in my pile. But you’re hungry for content and want to know what’s potentially worth reading, right? So, expect a few posts that are little more than abstracts of papers I’ve at most skimmed. Maybe they contain a fatal flaw. Maybe they’re awesome. I just don’t have time to find out. Take with a grain of salt.

    Medicaid as an Investment in Children: What is the Long-Term Impact on Tax Receipts?” by David Brown, Amanda Kowalski, and Ithai Lurie:

    We examine the long-term impact of expansions to Medicaid and the State Children’s Health Insurance Program that occurred in the 1980’s and 1990’s. With administrative data from the IRS, we calculate longitudinal health insurance eligibility from birth to age 18 for children in cohorts affected by these expansions, and we observe their longitudinal outcomes as adults. Using a simulated instrument that relies on variation in eligibility by cohort and state, we find that children whose eligibility increased paid more in cumulative taxes by age 28. These children collected less in EITC payments, and the women had higher cumulative wages by age 28. Incorporating additional data from the Medicaid Statistical Information System (MSIS), we find that the government spent $872 in 2011 dollars for each additional year of Medicaid eligibility induced by the expansions. Putting this together with the estimated increase in tax payments discounted at a 3% rate, assuming that tax impacts are persistent in percentage terms, the government will recoup 56 cents of each dollar spent on childhood Medicaid by the time these children reach age 60. This return on investment does not take into account other benefits that accrue directly to the children, including estimated decreases in mortality and increases in college attendance. Moreover, using the MSIS data, we find that each additional year of Medicaid eligibility from birth to age 18 results in approximately 0.58 additional years of Medicaid receipt. Therefore, if we scale our results by the ratio of beneficiaries to eligibles, then all of our results are twice as large.

    So maybe, all in, Medicaid could be break even. Maybe. (Yes, this is the pure economist point of view, as if lives and quality thereof can be easily, unambiguously converted to dollars. And, yes, the framing here suggests — though does not demand — that things are worth doing only if they pay for themselves, which is ridiculous. Good things are worth spending money on.)


    Comments closed