One common argument from opponents of the Affordable Care Act is that an expansion of government into health care will result in “crowd out”, where private influences and coverage go down as public ones go up.
We spend a lot of time discussing health care spending for adults, especially the elderly. Some argue that’s because we spend the majority of money on those populations. But as a pediatrician, parent, and health services researcher, I sometimes get annoyed at our lack of attention to children. Luckily, I subscribe to some journals that focus on children. From Pediatrics, “Comparison of Health Care Spending and Utilization Among Children With Medicaid Insurance“:
Some conservative policy advocates claim either that people are worse off being on Medicaid than being uninsured, or that they would be substantially better off if the government fundamentally restructured Medicaid or replaced it with an entirely different program. The implausible (if not preposterous) argument that people are worse off with Medicaid than with nothing at all rests on a handful of studies reporting that Medicaid patients do worse in some particular medical settings than do uninsured patients. Highly regarded health economist Austin Frakt has thoroughly and convincingly debunked these studies, and this entire line of argument, as follows.
Frakt explains that observational studies that compare uninsured people with those covered by Medicaid are completely inadequate for drawing conclusions about whether Medicaid coverage causes worse health. This is because people do not sort themselves randomly between insurance conditions. All else being equal, sick people are more likely to seek out insurance, including Medicaid. As a result, uninsured people are, in general, substantially healthier than people with Medicaid. Therefore, it is almost certainly spurious to conclude that Medicaid is the cause of the worse health observed in those whom it covers.
I’m flattered. Nick has high regard for Mark, as do I, writing me that he thinks he’s “among the best health law scholars in the country, if not the very best.”
The work Mark cites by me is coauthored by Aaron, Harold Pollack, and Uwe Reinhardt, all highly regarded as well. Click through for the details. It appears to have been published in 2014, but I don’t recall having seen it.
Uninsured people who get Medicaid only gained from 20 to 40 cents in value from each dollar spent by the government.
A principal reason why the benefit of getting insured was so small is that when uninsured people received care, they typically paid only 20 cents on the dollar for those services. Safety-net providers, state or local government, friends, relatives, or someone else absorbed the remaining costs.
Because a large fraction of Medicaid expenditures financed care that recipients would have received anyway (for example, by leaving bad debt at hospitals), it is unclear whether recipients themselves would have been willing to pay the full costs of Medicaid.
Finkelstein and her colleagues were careful not to draw normative conclusions from these findings. But some Medicaid critics have argued, first, that Medicaid is an inefficient way to benefit the poor. If a Medicaid dollar results in only 20 cents in benefit to a previously uninsured person, wouldn’t it be more efficient to simply give that person a dollar? And, second, Medicaid is actually a subsidy for people other than those it ostensibly helps.
We see matters differently. One important reason why the value of Medicaid appears to be so low is that Finkelstein put a much lower value on the health of Medicaid recipients than is typically used in valuations of the health of other Americans. We also argue that in many cases, we should not be troubled that Medicaid payments are going to third parties who are, after all, providing care to Medicaid recipients.
After failing to persuade his Legislature to expand Medicaid, Gov. Bill Walker of Alaska said Thursday that he planned to unilaterally accept the federal funds available to cover more low-income residents under the program.
Mr. Walker, an independent who took office in December, said in a news conference in Anchorage that he could not wait any longer to offer health coverage to the roughly 42,000 people his administration projects will be eligible under the expansion. Expanding Medicaid — an option for every state under President Obama’s Affordable Care Act — was a campaign priority for Mr. Walker, who couched it as a “common-sense decision” for the state’s economy and for the health of its people.
Evidently Governor Walker tried to get the legislature to support the expansion, but when those options failed, he used his power of office to do it himself. From the tone of the articles I’ve read, he appears to be acting legally. At least, I’m not seeing any legislators calling him a tyrant and threatening lawsuits.
That makes Alaska the 30th state to accept the Medicaid expansion. Many of my colleagues think that is a sign of how weakly the ACA has been implemented. Me? I think the opposite. Healthcare reform is a marathon, not a sprint. It’s important to remember that the Medicaid expansion only went into effect less than two years ago. It’s really young. Traditional Medicaid was passed in 1965, and the last state to accept it (Arizona) did so in 1982. If that same gap held true for the ACA, then we should expect the last state to accept the expansion in 2031.
Even if you accept Arizona as an outlier, all states but that one didn’t start participating in Medicaid until 1972, seven years after it began. That would be 2021 for the Medicaid expansion.
And for anyone who thinks comparing something to Medicaid is a bad example, because Medicaid isn’t as “beloved” as Medicare, remember that just a few years ago the Supreme Court ruled that traditional Medicaid was so American-as-apple-pie and essential-to-freedom that threatening to take it away from a state was unconstitutional coercion.
Healthcare reform is a long-haul process. I know many people wish that every state would just accept the Medicaid expansion right now, but that’s unlikely to happen. The good news for them is that the fact that all the states haven’t come on board yet doesn’t mean they won’t in the near future.
I have obviously touched a nerve in many TIE readers. So let’s take a step back and revisit what I wrote and what is going on here.
There were somestudies that made predictions about the health and composition of the newly eligible for Medicaid. Those studies could be interpreted to mean the Medicaid expansion would cost less than predicted, because people entering would be healthier than those already enrolled. I WAS ONE OF THE PEOPLE WHO THOUGHT THAT INTERPRETATION MIGHT BE POSSIBLE. A recent report from CMS suggests that this lower-than-expected expense didn’t happen. I wrote that we need to figure out why and keep watching.
None of this was to suggest that the prior research was incorrect or poorly performed. It appears, upon further looking at the report, that there are some reasons that the differences are higher than estimated:
There are several explanations for the difference between the estimates in this year’s report and those in previous reports. First, most of the States that implemented the eligibility expansion are covering newly eligible adults in Medicaid managed care programs, and on average the capitation rates for the newly eligible adult enrollees were significantly greater than the projected average costs previously calculated.
Austin flagged this, by the way. In fact, some of the capitation rates may have been raised because states predicted pent-up demand because they assumed people entering might be sicker (in spite of the research I talked about).
Data for newly eligible adults are still limited. While CMS has reported some enrollment and expenditure data for this group, data on claims and managed care encounters, along with data on the health status and demographics of these enrollees, are not yet available. Thus, there is still considerable uncertainty about the health care costs of newly eligible adults in 2014, as well as for future years.
Medicaid still expects newly eligible adults to cost less in the future (Figure 6). It’s entirely possible they are correct.
But here’s the thing. It may be that this research turns out to be incorrect. I doubt it, in that I think the work is correct in that the newly eligible appear to be less sick than those already covered. It may be that this research is correct, and that still spending goes up for other reasons. Maybe people just spend more. Maybe docs find new ways to bill more. Maybe policy makers set capitation rates too high.* But if I report on research, and make some guesses as to the future, and they turn out to be incorrect, I will write that my predictions were wrong, that we need to figure out why, and that we should keep looking into it.
That’s what I did. That’s what I will continue to do.
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.
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.
“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.”
“[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.
“[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).
“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.
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.
Austin and Aaron are participants in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com.