• What should Biden do about Medicaid work requirements?

    I’ve got a new article out at The Atlantic digging into that surprisingly thorny question. At stake here is whether the Biden team can move quickly enough to forestall the Supreme Court from deciding two pending cases involving work requirements in Arkansas and New Hampshire. It’ll be tricky, in part because of some last-minute shenanigans to protect work requirements from reversal:

    On January 4, … the Trump administration announced that it was changing the rules. In a seemingly innocuous letter to state Medicaid directors, the director of the Centers for Medicare and Medicaid Services, Seema Verma, offered “additional details of the process” for withdrawing waivers. One of those new details is that no withdrawal can take effect for at least nine months.

    The change is a brazen, cynical attempt to protect work requirements long enough for the Supreme Court to rule on them. And while it’s dastardly, it’s also clever. When the states agree to the terms of Verma’s letter—and Republican-controlled states certainly will, if they haven’t already—its terms arguably become enforceable as a kind of intergovernmental contract. I say “arguably” because the letter itself may be legally defective, as two Democratic congressional leaders have already argued in an angry missive to Verma. But the possibility that the courts might treat it as binding means that it’d be risky for the Biden team to withdraw the waivers before nine months are up.

    The Biden administration still has options, however—and here’s where the creative lawyering comes in.

    Read the rest here. I’ll add that it speaks well of the The Atlantic that they were willing to run a piece digging into the nitty gritty of administrative law. It’s important stuff, but complicated.


  • Fighting the Institutionalization of Racism in Medicaid

    Paul Shafer is an assistant professor of health law, policy, and management at the Boston University School of Public Health, follow him on Twitter @shaferpr. Nambi Ndugga is a policy analyst at the Boston University School of Public Health, follow her on Twitter @joerianatalie.

    In January 2020, the Centers for Medicare and Medicaid Services (CMS) issued a letter to state Medicaid directors that let them convert Medicaid programs from the current state-federal partnership model to a block grant model. The block grant model would introduce caps to Medicaid funding. These changes could have detrimental impacts on historically marginalized populations who have already been adversely affected by the pandemic.

    In an editorial in the American Journal of Public Health, we discuss how block grants and other recent pushes, like work requirements and elimination of retroactive eligibility, in Medicaid policy in combination with established structural discriminatory practices negatively affect Black Americans and other people of color.

    As we wrote,

    HAO [Healthy Adult Opportunities] could have devastating impacts on lower-income families who may be unable to pay ‘up to 5% of their household income’ to cover out-of-pocket costs or will be punished by suspension from Medicaid for failing to pay any required premiums. States will also be able to adjust benefits, cost-sharing, and other requirements to stay within budget without additional CMS approval. Due to the large and persistent structural inequities faced by Black Americans and other people of color daily, this could have a particularly devastating impact on their health.

    The persistent inequities experienced by marginalized populations and their ties to poor health outcomes has been extensively documented. As we navigate a pandemic and economic recession, now, more than ever is the time for public health professionals and policy-makers to call out these injustices.

    Read more here.

    Research for this piece was supported by the Laura and John Arnold Foundation.

  • While the US is reeling from COVID-19, the Trump administration is trying to take away health care coverage

    Paul Shafer is an assistant professor of health law, policy, and management at the Boston University School of Public Health. Nicole Huberfeld is a professor of law and health law, policy, and management at the Boston University Schools of Law and Public Health. Follow them on Twitter at @shaferpr and @nhuberfeld1.

    The Trump administration has had its hands full responding to the coronavirus pandemic, but that hasn’t stopped it from taking steps to reverse much of the gains in health insurance coverage since the Affordable Care Act was passed in 2010. In an article today on The Conversation, we discuss two major actions by the Trump administration that may would typically have made huge headlines but may have gotten lost in the COVID shuffle – 1) attempting to block grant Medicaid and 2) supporting a Supreme Court case that could take down the ACA.

    Despite shaky legal footing, the administration has moved ahead with its Healthy Adult Opportunity guidance, issued to state Medicaid directors in January, that allows for states to opt-in to a per capita cap or program-level block grant for Medicaid. Oklahoma was going to be the first to implement this until a ballot initiative to expand Medicaid passed in July.

    Block granting Medicaid has been a goal of Republicans for years, including during the ACA repeal efforts in Congress, but has never been able to be passed into law. This end run around federal law has been loudly challenged by legal scholars but is only one plank of the administration’s plans.

    Texas v. California will be heard in November, a case in which 17 Republican-led states are trying to take down the ACA through again dubious legal arguments about severability of the individual mandate from the rest of the law. The administration has abdicated its role to defend the law and is arguing in favor of striking it down, trying to accomplish through the courts what it has failed to do through Congress.

    As we wrote,

    If the ACA is struck down, that means the loss of coverage for preexisting conditions, the return of annual or lifetime caps, or policies being revoked for cancer patients. Those who don’t earn much money still deserve good health care. Nearly everyone will feel it if the Trump administration and Texas are successful, regardless of whether your health insurance is through your work, HealthCare.gov, Medicaid or Medicare.

    This fall, the Supreme Court and the voters will have a lot to say about how access and affordability of health insurance coverage look in 2021 and beyond.

    Read the whole piece here.

    Research for this piece was supported by the Laura and John Arnold Foundation.

  • In Pursuit of Value-Based Drug Prices in the US: New York and Beyond

    Despite overwhelming agreement among voters that the cost of prescription drugs is too high, Congress has yet to pass substantive legislation in recent years to tackle drug prices. In the absence of federal action, New York lawmakers stepped up to the task in 2017, establishing a drug spending cap in the state’s Medicaid program. Setting an annual limit on the state Medicaid program’s prescription drug spending, however, meant that the legislature also needed to give Medicaid officials a way to keep costs down. So how has New York gone about cutting excess drug spending?

    The answer is that the New York legislature gave officials the power to systematically limit the price of high-cost prescription drugs to their therapeutic value

    More specifically, the state authorized its Medicaid Drug Utilization Review Board to conduct reviews of the value of expensive drugs when officials project that the program’s annual drug spending cap will be exceeded. After the board determines a value-based price for a drug, the state then negotiates additional rebates with the drug’s manufacturer. 

    To compel manufacturers to come to the table and concede supplemental rebates, the New York Medicaid program may implement a number of strategies intended to lower manufacturers’ profits if they don’t comply. These bargaining chips include instituting prior authorization requirements, directing Medicaid managed care organizations to remove the drug from their formularies, and limiting reimbursement for provider-administered therapies that are billed under the medical benefit. Under the terms of the Medicaid Drug Rebate Program, however, New York may not remove a drug from its Medicaid formulary. This means that using a state-run approval process, beneficiaries can still receive Medicaid coverage for drugs that are not on their managed care plan’s formulary.

    Though the New York Medicaid program was the first public payer in the US to limit reimbursement for prescription drugs based on therapeutic value, this practice is common in many European countries. The New York Medicaid Drug Utilization Review Board utilizes many of the same tools used in health technology assessment in Europe, including quality-adjusted life-years (QALYs) and reference pricing, though in a more limited capacity. 

    To date, the New York Medicaid Drug Utilization Review Board has completed three pricing reviews. In these reviews, the board determined target supplemental rebate amounts for lumacaftor/ivacaftor (Orkambi), infliximab (Remicade), and nusinersen (Spinraza). Following the board’s reviews of lumacaftor/ivacaftor and infliximab, the New York Department of Health was able to reach confidential supplemental rebate agreements for both drugs with their respective manufacturers. 

    In the case of nusinersen, the board only recently announced the completion of their review at a public meeting on July 23, 2020. Though negotiations between New York officials and the drug’s manufacturer Biogen will be kept confidential, the state’s prospects for securing additional rebates seem promising in light of the drug’s billing under the medical benefit in Medicaid. This is the case because of the board’s ability to direct Medicaid managed care organizations to reduce reimbursement for provider-administered drugs billed under the medical benefit, which includes nusinersen, if manufacturers refuse to concede adequate supplemental rebates. It’s possible that this same leverage helped prompt Janssen Pharmaceuticals, the manufacturer of infliximab, to agree to pay additional rebates to New York after the board’s review of the drug last year. 

    Efforts to conduct assessments of the value of drugs and other medical goods and services in the public sector in the US have been met with fiery political opposition. After all, who could forget Sarah Palin’s infamous and wildly inaccurate claims about how the Affordable Care Act would create “death panels”

    More than a decade later, the consequences of rejecting cost-effectiveness research are starting to come into focus. Today, rising brand-name prescription drug prices (and increasingly-prevalent high-deductible health plans) pose a threat to Americans’ financial security and to the health of our communities. Political controversies notwithstanding, using cost-effectiveness research and analyses to pay for high-value prescription drugs can help US patients afford the drugs they need as prices continue to rise. While value is difficult to measure or even define in health care, working through those difficult questions is well worth the reward of maximizing patient health and lowering costs. Value-based pricing frameworks, such as those used by the Institute for Clinical and Economic Review, are valuable tools that can help policymakers rationalize the way we pay for pharmaceuticals. 

    Taxpayer dollars have bankrolled the pharmaceutical industry’s inflated bottom line for far too long. To protect Americans’ health and financial security, lawmakers across the country should follow the trailblazing example of the New York Medicaid Drug Utilization Review Board and join in the pursuit of systematically-achieved, value-based drug prices in the US. 


  • Express Lane Eligibility, so why haven’t states used it?

    Paul Shafer is an assistant professor of health law, policy, and management at the Boston University School of Public Health. He tweets at @shaferpr.

    A few weeks ago, I wrote about an often ignored pathway for streamlining enrollment into Medicaid and the Children’s Health Insurance Program (CHIP) called Express Lane Eligibility (ELE). ELE allows states to use information collected by other agencies to determine eligibility for Medicaid and CHIP. This follow-up post explores the why, why ELE is underutilized and what capabilities we are missing as a result that could be game-changing in the midst of this pandemic.

    States didn’t have the authority to do this sort of information sharing before, so granting it should lead to all sorts of excitement at the prospect of administrative efficiency and better reach at serving those targeted by these programs, right? Not exactly.

    Less than a third of states (14) are currently using it. Why?

    A 2016 report from the Department of Health and Human Services (HHS) highlighted several barriers to states implementing and continuing use of ELE. Eleven of 14 states surveyed noted difficulties related to collaboration with relevant agencies and information sharing. These were often overcome but the short-term nature of these partnership agreements led to several states discontinuing ELE after a few years.

    Very few states implementing ELE were able to partner with their state tax agency because information sharing would have required new legislation and other changes. Other states noted that ELE was a short-term plan to boost enrollment and/or upgrades to enrollment and eligibility determination systems as part of the Affordable Care Act coverage expansions had “reduced the need for ELE.”

    Another concern could be less than perfect information sharing resulting in eligibility errors. A pair of 2016 reports from HHS estimated that over 95% and 88% of eligibility determinations for CHIP and Medicaid through ELE were correct, respectively, based on very small audit samples. Even with little guidance, states were implementing ELE in a way that yielded generally accurate eligibility determinations.

    But perhaps the elephant in the room is simply the cost to states of having more people enrolled. A 2012 letter on ELE from the Government Accountability Office (GAO) to Senator Baucus, then chair of the Senate Finance Committee, noted

    significant budget pressures that states are facing, which might make them reluctant to undertake options that would increase Medicaid or CHIP enrollment and costs, including costs for the services additional enrollees would receive.

    Making government operate more efficiently could have a real cost in this case, fiscally and politically, if Medicaid and CHIP rolls grow.

    In the same letter, GAO added that allowing ELE for adults makes it a much more attractive option for states. As currently legislated, it can only be used for children. 1115 waiver authority can be used to extend ELE to adults, but has been rarely used despite promising evidence (e.g., Massachusetts). Still, numerous states have embraced ELE, and we have an idea of what makes it work well (e.g., automated processes, permanent information sharing authority).

    Even if a state could get past those barriers, the transient nature of ELE makes states reticent to invest the political capital and dollars to set it up. ELE has been historically tied to CHIP reauthorization, which was most recently extended through fiscal year 2027 in the Bipartisan Budget Act of 2018. The Medicaid and CHIP Payment Advisory Commission (MACPAC) has recommended “permanently extending the authority” for ELE but Congress has yet to do so.

    States are experimenting with other mechanisms for streamlining enrollment. The Maryland Easy Enrollment Health Insurance Program allows anyone to check on box on their state tax return to get more information from the state exchange and approval for a special enrollment period, which more than 18,000 Marylanders had taken advantage of as of early March. An opt-in system isn’t nearly as powerful as an opt-out, as demonstrated by the Louisiana experience discussed in my first post, but it is nonetheless a step.

    A crisis is not the time to overload social service agencies and the American people with paperwork and documentation requirements. We should consider robust ELE-like information sharing across state agencies as part of our disaster preparedness strategy, allowing us to pair short-term stimulus, like checks and expanded unemployment benefits, with the capability to efficiently get those affected covered by Medicaid when they qualify.

  • Express Lane Eligibility, a missed opportunity to patch holes in the safety net?

    Paul Shafer is an assistant professor of health law, policy, and management at the Boston University School of Public Health. He tweets at @shaferpr.

    We spend a lot of time debating who should be eligible for public assistance and how generous it should be. But we spend a lot less time on how we get those who are eligible into the programs they qualify for. You can be dropped from Medicaid for missing a single letter, an example of the precarious nature of our ‘safety net’.

    The introduction of the Marketplace gave states a reason to update their enrollment infrastructure, all states now accept online applications for Medicaid and nearly all able to determine eligibility in real-time (within 24 hours). However, that doesn’t solve the problem of getting people to apply or keeping them from dropping out when they are still eligible, which has historically been a big problem.

    An underused and poorly studied provision of the 2009 law reauthorizing the Children’s Health Insurance Program (CHIP) is called Express Lane Eligibility (ELE). It gives states the authority to use information from other public programs, like Supplemental Nutrition Assistance Program (SNAP), Temporary Assistance for Needy Families, and others, for determining eligibility for Medicaid and CHIP coverage for kids. The authority was extended through fiscal year 2027 in the Bipartisan Budget Act of 2018.

    This sounds like an obvious thing to do, but states previously did not have the authority to use data from other agencies to make enrollment determinations for CHIP and Medicaid. For example, if states could use income tax data, then there would be no need to prove income or lack thereof for eligibility purposes. This makes intuitive sense, a way to make it easier for applicants and more efficient for the state.

    California passed a version of ELE back in 2001, with local philanthropies funding a pilot to link the National School Lunch Program and Medi-Cal in seven school districts from 2003 to 2006. The state struggled with data matching and when combined with already high Medi-Cal enrollment rates, about half of ELE applications generated were for children already enrolled. There were also concerns that worries about immigration issues led to families not completing the simplified application. Though implementation was problematic, the authors of the evaluation concluded that a ‘”no wrong door” approach to health care’ still held promise.

    Congress gave states a lot of latitude in how to implement ELE with lots of policy options available to enroll the one in five children who were eligible for CHIP or Medicaid but unenrolled and to keep them enrolled. So what happened? Only 14 states are currently using it, and they are all doing so in different combinations of programs. Some for CHIP, some for Medicaid, only four for both and a lot of variation in which programs or agencies they ‘talk’ to for ELE.

    Unfortunately, we don’t know very much about the consequences of states’ ELE implementation. I only found only a handful of commissioned reports and four papers in peer-reviewed journals that rigorously evaluate ELE.

    A report commissioned by the Department of Health and Human Services (HHS) found that across the adopting states, ELE increased enrollment by about 6% and saved millions of dollars in administrative costs. By analyzing each state’s implementation as a separate case study, the report found that ELE increased retention of those eligible and reduced administrative burdens on state social service agencies.

    South Carolina had ranked 45th in health insurance coverage for low-income children before its ELE program was implemented in 2011. Afterwards, it helped enroll another 92,000 children and keep 276,000 on CHIP and Medicaid. The Louisiana experience shows why automation is so important. It was the first state to implement the automatic enrollment option for ELE, sending out more than 20,000 Medicaid cards in 2010 alone. Then, it switched to an ‘opt-in’ approach via a checkbox on the SNAP application, which resulted in a 62% drop in monthly Medicaid enrollment through ELE.

    A study published in 2014, whose authors were part of the HHS-funded evaluation team, echoed the enrollment findings of the commissioned reports. They found that ELE was associated with a 5.6% increase in Medicaid and a 4.2% increase in combined CHIP and Medicaid enrollment. Their results also suggest that this effect may get stronger in the long-run. Another study from 2014, from authors also part of the HHS evaluation team, noted children enrolled via ELE used care less and less intensively, and were therefore less expensive to cover.

    Massachusetts incorporated ELE for Medicaid renewals for parents and the expansion population, finding that “ELE participation was the strongest predictor of continuous coverage during the 90-day period following MassHealth annual review”. The rate of coverage loss within 90 days of review was more than ten times higher in the non-ELE household group (34%) than in the ELE group (2-4%). Though ELE was initially conceived as a way to reduce the number of uninsured children, using the same process and data infrastructure for the adult population not surprisingly also increases retention in Medicaid.

    Under the Trump administration, we have seen the number of uninsured children jump by 400,000 and proportion of adults who are uninsured or avoided health care due to cost inch upwards. SNAP is also increasingly hard to get, with hundreds of thousands of food insecure families kicked off the program. A nationwide embrace of ELE coupled with a streamlining of the eligibility criteria for certain programs could go a long way towards helping low income populations stay healthy, housed, and fed.

  • Enforcing Work Requirements in Michigan is a Colossal Waste

    Or so I argue today in an op-ed in the Detroit News:

    State officials in Michigan … are still subject to the Snyder-era law requiring them to roll out work requirements in January. So far, the state has spent $28 million implementing them and is poised to spend $40 million more in the coming year. This month, for example, the Whitmer administration will have to spend $1 million just to send detailed compliance information to about 200,000 Michiganians.

    Even if you believe in work requirements, it makes zero sense to spend millions on a program that the courts are going to halt anyway. That’s why a bunch of deep red states, including Arizona, Indiana, and Montana, have all put their work requirements on pause while the litigation works its way to the Supreme Court.

    Michigan should do the same. The Legislature and Whitmer may be at each other’s throats over roads and budgets, but this is not a partisan issue. It’s about fiscal prudence.

    The Legislature, however, appears content to let taxpayer money burn.


  • Tennessee wants to block grant Medicaid. Is that legal?

    Earlier today, Tennessee released a draft proposal to introduce block grants into its Medicaid program. Setting aside the dubious policy merits of block grants, however, I don’t think the proposal is legal. I don’t even think it’s close.

    Under section 1903 of the Medicaid statute, the federal government must pay a fixed “match rate” (known in the statutory lingo as “the Federal medical assistance percentage”) to every state that participates in Medicaid. In Tennessee, the match rate is 65.21%. That means that, for every $1 that Tennessee spends on its Medicaid program, the federal government kicks in about $2.

    Tennessee wants to change that financing structure. Instead of matching dollars, Tennessee would like an up-front, lump-sum payment—a “federal block grant”—that’s calculated to cover the anticipated expenses of its Medicaid population. Because Tennessee would have a fixed sum of federal dollars to spend, the state would have an incentive to economize on Medicaid. Any savings, Tennessee says, would be split 50-50 with the federal government.

    As Tennessee recognizes, it’ll need a waiver from HHS to make these changes. And section 1115 of the Medicaid statute does allow HHS to waive lots of the law’s restrictions in connection with experimental projects that are likely to assist in promoting Medicaid’s objectives.

    Now, I’ve written before that I’m not at all sure that block granting Medicaid counts as an experiment that serves Medicaid’s purposes. But there’s a more fundamental problem with Tennessee’s proposal. You can’t use section 1115 to waive section 1903. To the contrary, section 1903 is pointedly omitted from the list of statutory provisions that HHS is empowered to waive.

    So you can’t use Medicaid waivers to change Medicaid’s financing structure. And that’s exactly what Tennessee is proposing to do.

    Under the proposed waiver, the state would no longer get $2 for every $1 in Medicaid spending. It would instead get a lump sum every year that’s calculated to cover its anticipated Medicaid expenditures. If Tennessee spends more than it expects, the effective match rate would be lower than 65.21%. If it spends less, the effective match rate would be higher. Either way, the proposal would violate section 1903.

    I’d been thinking that Tennessee would find a way to conform its block grant proposal to section 1903. But it looks like it didn’t even try. Indeed, it’s telling that Tennessee doesn’t even mention section 1903 in its proposal—a glaring omission for a state that wants to make a fundamental change to how the federal government pays for Medicaid.

    Maybe Tennessee has an argument for why this proposal is consistent with the Medicaid statute. I’m all ears. For now, however, I don’t see how to square what Tennessee wants with the language of the law.


  • Healthcare Triage: The Implications of “Public Charge” on Immigrant Children

    The Trump administration has put forth a rule change for immigration saying that if an immigrant gets one of a number of benefits from the government, it could lead to their being denied legal permanent residency or entry to the US. Will such a rule lead to some immigrant parents disenrolling their families from safety-net programs? How will this affect children? We’ve got data.


  • Healthcare Triage News: Infants and the Medicaid Expansion

    What can we do about infant mortality and disparities in health care? This week we take a look at a recent study in JAMA that may have an answer.