The Incidental Economist

I gave two talks at the end of last week. The first was on Accountable Care Organizations for a local chapter of the AAP. The second was on health care reform and the health care system in general. By the end of the second talk on Friday, I was pretty much feeling despondent.

As I reviewed ACOs on Thursday, I was talking about how many hoped they would bring about cost control at the physician level. But as I reviewed the progress of how the rules have been modified over time, I could feel cynicism creeping into my voice. There was the removal of sticks in October. The elimination of rules requiring EHRs. The reduction in quality oversight. The ability of patients to seek care outside the ACO. And let’s not forget the counterproductive side effect of provider concentration that the formation of ACOs encourages (Austin helpfully posted a FAQ on this balance of power earlier today).

So I’m skeptical that ACOs will be what gets us out of the ditch in the future. Not that I see many other helpful ideas from others. Tort reform won’t do what people say it will. Medicaid block grants are just passing the buck. Medicare “sorta” premium support is just cost-shifting. None of these things will get us there.

So what will, I was asked? I wish I had a magic wand to just fix things. But, more and more, I’m becoming convinced that we need to identify areas where we can save money at the local level. Wennberg’s work shows us that there are huge variations in care that don’t seem linked to quality. That stuff has to go. Berwick and Hackbarth went further and identified how much “waste” there is in the US health care system right now:

I’m drawn to the third line alone. We’re spending somewhere betweeen $156 billion and $226 billion just on overtreatment. If we could just figure out a way to stop doing that, we’d be well on our way to bending the curve. Not that we can’t fix the other things as well. But stopping overtreatment seems like it would be the easiest to do, and it would have the added benefit of not requiring that much new added infrastructure.

We may need to focus on where these things are occurring locally, though, instead of thinking there’s some large lever to pull at a national level. We also need to identify ways to change physician behavior, which isn’t easy. I’ve been talking much more about that in my own corner of the world. I hope others are, too.

@aaronecarroll

One of the perennial complaints about Medicaid is that many doctors won’t accept it as insurance. That’s primarily because it under-reimburses. Some have used physician avoidance of Medicaid as a reason to argue that we should scrap the program. I’ve argued that physician avoidance is a reason to increase reimbursement. That’s about to happen:

Primary care doctors could get a pay raise next year for treating Medicaid patients, under a rule announced by the Obama administration Wednesday.

The proposed regulation implements a two-year pay increase included in the 2010 health-care law. The increase, effective in 2013 and 2014, brings primary care fees for Medicaid, which covers indigent patients, in line with those for Medicare, which insures the elderly and some disabled patients.

Although Medicaid is jointly funded by states and the federal government, the pay boost would be covered entirely with federal dollars totaling more than $11 billion over the two years it would be in effect.

Congress automatically appropriated those funds when it adopted the health-care law, so it will not need to act now.

Granted, I’m a pediatrician, so feel free to accuse me of bias here, but this is fantastic news for those of us who see a lot of patients on Medicaid. It also has a secondary effect of increasing payments for a lot of primary care providers, also a problem in the US. There’s not much to hate here, other than the cost.

But let’s be real here. We’re talking about $5.5 billion a year. Yes, that’s a lot of money, but compared to overall health care spending?

That’s almost infinitesimal. But what about if we only compare it to Medicaid spending?

It’s still not that much. Moreover, it’s entirely covered by the federal government for two years, so in the short term, states have little to complain about.

Now I’m not suggesting there aren’t problems here. This is all part and parcel of the ACA. If that gets overturned by the Supreme Court in June, all of this goes away. The same holds true if the whole law were repealed next year. And, in two years, it’s going to be a struggle to find the additional money to keep this going, especially if some of it falls to the states.

But let’s face some facts. Medicaid is still a pretty lean program, and costs less than putting people on private insurance in the exchanges. We face a primary care shortage, in large part because we don’t reimburse those physicians as well as we do specialists. And many docs do avoid Medicaid because the program pays so much less than Medicare and private insurance. This small change impacts all of those problems at the same time.

Sometimes good things cost money. Yeah, there’s no money now, but wait until they threaten to cut Medicare reimbursements in a couple months during the next “doc fix” crisis, and you’ll see how quickly people’s tunes change. Shoring up Medicaid in the eyes of providers while simultaneously making it more enticing to be a primary care doctor? This here is a bargain.

@aaronecarroll

From Health Affairs, “State Spending On Dual Eligibles Under Age 65 Shows Variations, Evidence Of Cost Shifting From Medicaid To Medicare“:

Roughly half of Medicare beneficiaries under age sixty-five are also eligible for Medicaid. These “dual eligibles” have been the subject of much research because of their low income and poor health status. Previous studies suggest that some states seek to shift costly health care services for this group out of state-run Medicaid programs and into the federally funded Medicare program—for example, replacing nursing home care with hospital care. Using state-level data on dual eligibles under age sixty-five, we found support for this hypothesis. In states with below-average per capita Medicaid spending, corresponding Medicare spending was above average. These state-level estimates also revealed a nearly threefold difference in total—Medicare plus Medicaid—price-adjusted spending per person, ranging from $16,309 in Georgia to $43,587 in New York. Such large variations among people with serious diseases suggest inefficiency. Some states may be spending too little for Medicaid, meaning that some patients’ needs are not being met, or some states may be spending too much, meaning that more services are being provided than needed. Such inefficiency exposes patients to unnecessary risk, drives costs up unnecessarily, and highlights the large potential gains arising from improved care coordination for dual eligibles.

Can we start with the fact that half of people over 65 can qualify for Medicaid? That speaks pretty poorly about the financial stability of the elderly population. (ed – Sorry – I’m tired and I misread. This is besides the point anyway so ignore – Aaron)

Once you get past that, though, this manuscript speaks pretty well to the potential pitfalls of keeping Medicare and Medicaid financed at different levels of government. Medicare, of course, is a federally run single-payer system for everyone over 65. Medicaid, on the other hand, is a state run system where the federal government matches some percentage of payments so that states aren’t totally responsible for financing. The trick, though, lies in the dual eligibles, or people who are on both Medicare and Medicaid.

States, of course, would like the federal government to pay for more so that they can pay for less. So whatever they can get shifted to Medicare for dual-eligibles, the better. Medicaid, for instance, pays for nursing home care (which Medicare does not), so if they can get a patient admitted to the hospital instead, then they can get Medicare to pay. But this probably isn’t good for patients. If that was occurring differently in different states, you’d expect great variation in the amounts states spend on Medicaid versus Medicare:

If people are unhealthy, you’d expect more health care spending period. But as you can see, the proportion spent on Medicare versus Medicaid is all over the map. This means that states are likely finding ways to get Medicare to pay for a larger percentage of care in some states. Here’s a better way of looking at that:

There’s really no positive relationship between Medicare spending per person (x-axis) and Medicaid spending per person (y-axis). You’d expect that if increased spending was just a measurement of “illness” in the population. There is, however, a negative relationship:

We also found evidence of a negative association between Medicaid and Medicare spending for dual eligibles under sixty-five (Exhibit 3), with a weighted correlation coefficient between Medicare and Medicaid spending for this group of −0.40 (). That is, for these beneficiaries, Medicare expenditures appeared to substitute for Medicaid expenditures. States with lower rates of Medicaid spending experienced higher rates of Medicare expenditures, and vice versa.

A short while ago, I argued that it would be very difficult to cut Medicaid without significantly attacking benefits, beneficiaries, or physician reimbursements:

Make no mistake about it, under the block grant plan in the budget proposal approved by the House, states must innovate. Even if the ACA’s Medicaid expansion were eliminated, the House budget would reduce federal Medicaid spending by $163 billion in 2022. That’s a 34% reduction 10 years from now.

How can states possibly account for that difference? Where’s the magic in innovation? If states refuse to cut benefits and spend the same per enrollee, then even if the Medicaid expansion of the ACA never takes place, an additional 19 million people need to be dropped from the 2021 Medicaid rolls to meet budget cuts. That’s about one-third of all people on Medicaid. If states cut benefits or somehow slow spending to that of GDP growth, they still need to remove 13.8 million people from Medicaid in 2021, in addition to forgetting the ACA Medicaid expansion. If states act to protect the elderly and blind or disabled persons by holding their spending/benefit reduction to 10% (which is still a large cut), then 27 million people, most of them children and pregnant women, need to be dropped from Medicaid in 2021 even if ACA’s Medicaid expansion never occurs.

Yesterday at Politico, Matt Dobias gave us a glimpse of how “innovation” might occur:

House Republicans want to stop rewarding states for finding and enrolling low-income children in Medicaid and the Children’s Health Insurance Program, and public health advocates are livid.

The Republicans say it’s a smart fiscal move that will better protect the program against fraud; their critics say it’s undermining years of progress states have made in identifying and enrolling a hard to serve population…

Energy and Commerce Committee Republicans voted last week to strip about $400 million earmarked for a bonus program created by the 2009 law to extend the Children’s Health Insurance Program. The bonus payments run through 2013.

The provision, sponsored by Rep. Joe Barton (R-Texas) and passed without Democratic support, is part of a $115 billion package of health care savings included in the House’s reconciliation process. The House Budget Committee plans to mark up the reconciliation package on Monday. And although the package won’t make it through the Senate, some of these health cut proposals could resurface this year.

Some of the supporters of this measure have argued that provisions like this bonus program lead to “fraud” in Medicaid. But none of those reports of fraud have linked the potential waste to this program, or to children’s health care. Others have argued that while the federal government is subsidizing the increased rolls now, when that support ends, the states will now have to pay for the new children enrolled in Medicaid.

Ironically, that’s exactly the problem with shifting Medicaid entirely to the states for “innovation” and “block grants”. States will be left on the hook, unable to run short term deficits when they need to. They’ll be responsible for all coverage, as the amount given to them shrinks and shrinks.

I’ve long argued that there are no easy answers here. We’ve seen an early example as to how Medicaid might be cut – by making it harder for kids to become beneficiaries. And there are still many more cuts to come.

@aaronecarroll

This post is coauthored by Austin Frakt and Aaron Carroll.

Christopher Flavelle has put together a fascinating Bloomberg Government Study on the allure and growth of Medicaid managed care and the recent trend in Medicaid spending by states. It’s the first of three pieces in this area and, unfortunately, is behind a paywall. If you can get your hands on it, it’s worth a full read. If you can’t, here are a few details we thought worth highlighting.

Flavelle writes that advocates of turning Medicaid into a block grant program often claim it would “increase spending predictability.” Given this and other rhetoric from states about their “out of control” Medicaid growth, you’d think that spending has been growing exceptionally rapidly recently.

According to analysis by Flavelle, that’s not the case.

Inflation adjusted Medicaid spending per capita by state general funds increased just 3.8% between 2002 and 2011. This is illustrated by the dotted line in the chart below. Per capita Medicaid spending by each of the five states with the largest Medicaid programs is also shown. Though they gyrate up and down, they all end up at the end up in 2011 close to or even below where they started in 2002.

Some might object to dividing spending by the state population (per capita), because the population grows over time.* Of course, the state’s population reflects its potential tax base too, so it is fair to divide by it by that standard. A State with a growing population should to be able to afford commensurate growth in its Medicaid spending, though that can depend on how different sectors of the population grow relative to each other (more wealthy or more poor people).

In any case, total (not per capita) real Medicaid growth was just 12% from 2002 to 2011. That’s not as high as we expected, though there is variation by state. While spending in Illinois actually decreased by 1% over this period, spending in Texas went up 58%. Moreover, this is growth above inflation, so there is room for improvement. Still, Medicaid has held spending growth below that of other payers. Flavelle quotes Vernon Smith, former Medicaid director for Michigan:

“When you look at the rate of growth for all the major payers — Medicaid, Medicare, employer-sponsored insurance, National Health Expenditures — what you see is that no other payer has constrained the rate of growth in spending as well as Medicaid has. [] The reason is that no payer has been as motivated to undertake cost containment as state governments.”

In total, it looks like states have done a pretty good job. Based on spending trends alone, it isn’t clear why officials in some states think the program needs major restructuring. That’s not to say it is perfect or that there aren’t other justifications for reform. Of course there are. In fact, given its low reimbursement rates, some might reasonably argue we should spend more on Medicaid, not less. At any rate, the data don’t support claims that state spending on Medicaid has been growing in an especially concerning way, particularly relative to other payers.

* Another potential objection is that Medicaid spending by state general funds does not count “other funds and revenue sources used as Medicaid match, such as local funds and provider taxes, fees, donations, assessments.” True! However, Flavelle justifies a focus on general fund spending by quoting Alan Simpson who characterized that type of spending as a ”tax gimmick” and that states just use “that additional ‘spending’ to increase their federal match.” If you collect a dollar and then give it back to those you collected it from in order to get another dollar (or more) from the federal government (which, of course, you also spend on the same providers you taxed), have you really spent your own funds?

The House budget reduces Medicaid funding and proposes to deliver it to states in block grants in an effort to get them to “innovate”. But what does that really mean?

I have a post-up at the JAMA Forum on this that I hope you’ll find worth your time. Go read!

@aaronecarroll

I’m a day late to the game on this. Jared Bernstein, Dean Baker, and Paul Krugman have all already sounded the alarm as to Robert Samuelson’s piece declaring social security unrecognizable to FDR and a significant part of what’s going to bankrupt us. I can’t add anything to the excellent responses already mentioned in terms of facts.

I can make charts, though. So  hopped on over to the CBO for their data on the long term budget outlook. Here’s how things look in the Alternative Fiscal Scenario, when Congress and the President do everything in their power to break the country:

So what do we have here? Well, interest on the debt is horrific, because we extend tax cuts and refuse to do anything about health care spending, leading to increasing deficits. But I’d like you to notice two things. The first is that for all the complaints about discretionary spending (green), it’s pretty stable over the next 75 years. The second thing is social security. Does it look like it will be the thing that drives us off the cliff?

No.

But let’s say we wise up. We increase revenue, we stop with the doc fix nonsense, etc. Then we have the Extended Baseline. Please note that this involves no social security reforms at all:

Interest is much better, because revenue is up and overall spending goes down. Still – discretionary spending stabilizes. And social security isn’t nearly the budget buster some would have us believe.

We can stand on the edge of the cliff, look over the edge, and worry that social security is going to doom us. But that’s silly. It’s health care. That’s what’s gonna kill us.

*Bonus points if you got the West Wing reference

UPDATE: for the purists who demand that every slide have the same axes regardless of the message or content (yes, I’m having a frustrating day), here’s the first chart with the same y-axis:

@aaronecarroll

I have a post up at the JAMA Forum:

So much attention has been paid to the individual mandate that relatively few have bothered to focus on the other questions that will be debated tomorrow in front of the Supreme Court. One involves the expansion of Medicaid, and it is absolutely worth some time.

I don’t want to minimize the significance of the individual mandate discussion today, but ironically, tomorrow’s Medicaid arguments have consequences that are much more far-reaching. We should pay attention.

Go read my post!

@aaronecarroll

Last summer, when I was guest blogging over at WaPo, I wrote about some developments in my home state of Indiana concerning abortion. Specifically, Indiana was trying to pass a law that would no longer allow a provider to receive funds for any services at all if they also provided abortion. Specifically, this was a bill to defund Planned Parenthood. I was concerned, because this could be a potential problem:

[A]ccording to the Indiana Legislative Services Agency (pdf):

The Family and Social Services Administration (FSSA) reports that federal law requires state Medicaid plans provide any eligible individual medical assistance and that they can obtain such assistance from any institution, agency, community pharmacy, or person, qualified to perform the service(s) required. This also includes an organization which provides such services, or arranges for their availability, on a prepayment basis. Federal law permits states to define a qualified provider, but requires that this definition is related to a provider’s ability to perform a service and not what services are provided.

It seems that federal law stipulates that state Medicaid programs have to allow recipients to obtain services from any qualified provider. States can define “qualified,” but only with respect to whether they can actually provide the service. In other words, it appears they can’t be ruled “not qualified” because they also provide abortions.

Medicaid regulations are pretty specific. They’re set up so states can’t cherry pick providers. I feared that if Indiana went through with this, they they could lose Medicaid funding.

I should have been watching Texas. They did it:

After Texas blocked abortion providers’ participation in its Medicaid Women’s Health Program, the White House officially notified the state Thursday afternoon that it will pull all funds from the program, which totalled about $39 million last year.

Twenty-nine states participate in the Medicaid’s Women Health Program, which extends Medicaid coverage for reproductive health services to lower-income women who do not qualify for the rest of the entitlement program’s benefits. In Texas, the program served about 130,000 women, with the federal government footing 90 percent of the bill. About half of the clinics participating in the program would have been disqualified by the new legislation.

Gov. Perry reports that he will come up with the money to cover the services provided to these women elsewhere. I have no idea where he will find it, as Texas is already running a $4 billion Medicad shortfall, but I wish him luck.

When the federal government is paying 90% of something, I’d take what they say seriously. I said they were playing hardball last June. They still seem to be.

@aaronecarroll

I asked Chapin White to comment on Avik Roy’s post about his work. In reply, Chapin sent me the following by email and authorized me to post it.

The author of the recent Atlantic blog post did not contact me while writing the piece, and I found that it missed several key points.

1. The Atlantic blog misrepresents the ACA in two key ways.

a. It represents low provider payment rates in Medicaid as a major problem with the program. It ignores the fact that the ACA begins to address this problem by increasing primary care physician fees in Medicaid beginning next year (sec. 1202). This feature of the ACA is pointed out in the third sentence of the HSR article.

b. The blog claims that the Medicaid expansion in the ACA will mainly impact people in the lower-middle income group (i.e. not the lowest). In fact, the Medicaid expansion in the ACA will occur more or less exclusively among childless adults with very low incomes (below 138 percent of the federal poverty level). The crowdout phenomenon, which drives much of the discussion in the Atlantic blog, is smallest when coverage is expanded to people in the lowest income groups.

2. The Atlantic blog also misrepresents the paper’s findings:

a. The paper finds that the effects of CHIP expansions on indicators of access are mixed. Non-cost related access problems (e.g. waiting for appointments) appear to go up in one of the four income groups, whereas the share of children having 1 or more ER visits in a year appears to go down for the lowest of the four income groups. This mixed finding does not support the blogger’s contention that “[the ACA’s] expansion of Medicaid coverage … may actually reduce those individuals’ access to health care.”

b. The evidence in the HSR paper suggests that CHIP did not change aggregate physician utilization one way or the other (i.e. the point estimate is near 0), but the HSR paper points out that “the results on doctor visits are not precisely estimated due to the variability in the underlying measure.” The Atlantic blog incorrectly characterizes the HSR paper as finding that “physician utilization was lower in the states with the largest CHIP expansions.”

Here are the last three paragraphs from the HSR piece—these properly summarize my take on the paper and its findings:

“In general, these findings argue strongly against the idea that the effect of expanding coverage on utilization can be deduced simply from the reduction in patient cost sharing. The nature of the coverage—for example, does the coverage consist of a tightly managed product? does the coverage pay providers generously?—appears to be critical.

“From a federal budgetary perspective, these results are good news— if we extrapolate from the results in this article, the expansions of public coverage called for in PPACA will not have any effect on aggregate utilization of physician services. From the enrollee’s perspective, the results are mixed—the benefits of expanded public coverage may lie primarily in improved financial protection, rather than a sheer increase in services received. These findings also support the idea that public health insurance plans can have spillover effects on children who do not themselves gain coverage, and that those spillover effects can either increase utilization (if the public plan’s reimbursement environment is made more generous) or reduce utilization (if coverage is expanded without making reimbursement more generous).

“As it is conventionally understood, our policy options are either to expand coverage and increase health spending or to leave coverage gaps and hold the line on spending. That dilemma is false. Coverage expansions by themselves do not necessarily spur increases or decreases in overall utilization—what does appear to matter is the nature of the coverage and the generosity of provider reimbursements in the public program. The policy questions that we should be focusing on are as follows: (1) the degree to which we want the rationing of medical services to occur based on out-of-pocket costs and the ability to pay versus nonprice factors such as queuing, and (2) the degree to which we want our financing of the health care system to be redistributive. Expanding public coverage clearly moves in the direction of redistributive financing. Depending on how we choose to set reimbursement levels in our public programs, expansion coverage may or may not move in the direction of increased utilization and increased system spending.”

An ungated pdf of Chapin’s paper is here.

@afrakt