Avastin, Rationing, and Death Panels

President Obama’s health reform legislation empowers a new Independent Medicare Advisory Board (IMAB) to curtail spending. Republicans fear that IMAB restrict access to costly technologies, going so far as to call IMAB a “death panel.” The appointment of Donald Berwick, an advocate of cost-effectiveness research, to run Medicare has fanned the fires of discontent. But most everyone thought that this would remain a theoretical issue for political talking heads, at least for a few more years. Until yesterday.

Yesterday, an FDA advisory panel recommended rescinding approval of Avastin for treating metastatic breast cancer. The recommendation followed new research showing that Avastin extended the time until cancer worsened by one month; previous research had showed benefits of five months on average. That is right, Avastin is efficacious, but the benefits are small. Citing the high monthly cost ($8000), the panel ruling was nearly unanimous.

If the FDA follows the panel recommendation, this would be the first time to my knowledge that cost effectiveness trumped efficacy and safety. The U.S. would join other nations in saying that there is a price that we are not willing to pay in order to improve health. Republican Louisiana Senator David Vitter wasted no time in saying that the FDA was rationing healthcare. “I shudder at the thought of a government panel assigning a value to a day of a person’s life,” he said.

Such hypocrisy. Republicans have wasted no time trying to trim Medicaid budgets, nutritional programs, housing subsidies, and other expenditures that improve the health and happiness of low income Americans. They do this on the grounds of affordability. If we can’t afford to keep low income Americans healthy, sheltered and well fed, how are we supposed to afford Avastin?

Despite what Mick Jagger says, we can’t always get what we need, let alone get what we want. We ration all the time, making implicit judgments about benefits and costs. The FDA advisory panel is making these judgments explicit. I haven’t seen all the data and I don’t know what benefit/cost threshold was used by the panel. And I wonder if the panel valued the hope that Avastin gives patients and loved ones. But these are accounting issues and should not deflect from the fact that resources are scarce and we are always rationing.

I would much prefer to see the market ration access to Avastin. Perhaps Senator Vitter does as well. (If he truly believes that all Americans should have unfettered access to all health care services, regardless of cost, then he must believe that the sky is the limit for health spending.) The FDA should stay the course with Avastin and let individuals decide whether they want to purchase it themselves (or sign up for a health plan with very generous drug benefits.) But those who can’t afford it, or who choose a health plan that does not cover it, will not be able to receive the drug. This is rationing, but it is through private choices rather than government coercion.

If we don’t privatize Medicare (and I am not holding my breath), then we will still have to face up to government rationing, because Medicare is publicly funded and will therefore require public decisions about how to spend scarce dollars. Why demonize those who understand the reality of rationing and want to openly debate how to do it?

Senator Vitter and all of your Republican colleagues, stop the demagoguery. You are big supporters of traditional Medicare, an indemnity insurance program in which tens of millions of Americans spend other people’s money with little direct financial accountability. So please answer this simple question: How do you propose to rein in Medicare spending? You oppose cutting payments to providers. Your only other choice is to cut the quantities of services they provide. Isn’t that rationing?

Donald Berwick and Rationing

Let me take time out from book writing to comment on the recent appointment of Dr. Donald Berwick to run Medicare. This was a recess appointment made by President Obama apparently to prevent Republicans in Congress from reviving the debate about health reform. Dr. Berwick seems vulnerable because he has frequently commented about rationing of medical services, even admiring programs for rationing that have been instituted in England and elsewhere. Dr. Berwick’s critics want to argue that rationing is un-American, even evil. But rationing simply means that individuals do not receives all the services they would like regardless of cost. All goods and services are rationed, even in market economies. Health care services are no exception. There is no way around it – America rations health care services.

Unfortunately, U.S. market-based rationing of health services is highly problematic, for reasons that I have described in previous blogs. One problem is moral hazard due to health insurance. The only way to fix this problem is to completely eliminate health insurance (high deductibles slightly mitigate the problem), but this exposes individuals to unwanted financial risk, a solution that may be worse than the cure. This is the Catch-22 of health care. The other problem is inadequate and asymmetric information about medical need and outcomes, which leads to demand inducement and practice variations. Health insurers could intervene to correct these problems but they might have incentive to encourage under treatment. In any event, last decade’s backlash against managed care has tied insurers’ hands. For these reasons, U.S. market-based rationing is woefully inefficient. Health services researchers are in near unanimous agreement on these points and Dr. Berwick stands on solid ground.

Dr. Berwick looks to other models for rationing. England’s National Institute for Clinical Excellence offers perhaps the most sophisticated model. NICE rations by relying on cost-effectiveness data and restricts access to medical treatments that cost “too much” – generally more than $60,000 per “quality adjusted life year”. Interestingly, if we had true market-based rationing without insurance, many of the technologies rationed by NICE would be unaffordable to the average American so there might not be much difference. English-style is not without its faults. It is a one size fits all solution to medical decision making. It does not adequately account for the learning curve associated with many new, costly technologies and does not account for innovation spillovers (when using a technology for one purpose sparks other innovative uses.) It also begs the question of how to choose the cutoff – is a year of life worth only $60,000? Who gets to decide?

We should all agree that market-based rationing with health insurance is problematic. Doing without health insurance may be even worse. We can also agree that English-style rationing has its own faults. Who is to say which form of health care rationing is the lesser evil? The Republicans think they know the answer without really understanding the question. The President wants to avoid the question altogether.

Isn’t that a debate worth having?

Health Information Technology — So Near and Yet So Far

In the current issue of Health Affairs, which is devoted to health information technology (HIT), one article describes key issues for successful adoption of HIT in medical homes. (Medical homes are integrated, team-based primary care practices.) The list of issues is daunting, including the need to integrate HIT across providers, the need for clinical decision support, and the need for outcomes measurement. As I read this and other articles in this issue I was struck by an overwhelming feeling of déjà vu. Ten years ago I described the key issues for HIT in my book, The Economic Evolution of American Healthcare. I had learned from hands on experiences with physician hospital organizations (PHOs) and integrated delivery systems (IDSs) that success depended on good management information systems, which in turn required HIT. I also learned that successful HIT would require integration across providers, decision support, and outcomes measurement. Deja vu indeed.

So what has been going on in the past decade? Seemingly everyone has become a believer in HIT. But no one talks about PHOs or IDSs very much. The medical community has embraced “new” organizational forms like medical homes and accountable care organizations, which aren’t all that new when you look at them closely. They are largely PHOs and IDSs in new clothing. But these new organizations are struggling and the reason is simple. It’s the HIT, stupid. It doesn’t matter how you dress up the health provider organization. If the organization cannot measure its performance, and if the market does not reward the organization for delivering value, then the organization will never fulfill its potential.

What makes the article on medical homes particularly interesting is its juxtaposition against an interview of David Blumenthal, current national coordinator for HIT (our nation’s “HIT czar”), by David Brailer, who held the same post under President Bush. The interview is remarkable in part because it is remarkably short on specifics, even for politically appointed policy makers. But the part that really caught my eye was Brailer’s effusive praise for Blumenthal’s creation and implementation of the “meaningful use” criterion, as in individual providers can get subsidies for adopting HIT if they can establish a meaningful use for it. Several other articles in the same issue described various creative ways of establishing meaningful use.

This is when I realized that the two Davids, and by extension the national leadership of the HIT movement, have entirely missed the point of HIT. HIT is not a tool for the individual provider. It is a tool for the market. In economics parlance, HIT displays a huge externality – when one provider adopts it, others benefit. More importantly, HIT increases the effectiveness of managers, payers, and patients, for example by facilitating quality evaluation. In a nutshell, it does not matter whether individual providers can find a meaningful use for HIT. The market will make meaningful use of HIT. I could not find any mention of this meaningful use in the interview or any of the other articles in Health Affairs.

And therein lays one of the biggest problems with our health care delivery system. It is run by doctors who believe that all problems have medical solutions. (Blumenthal and Brailer are both doctors.) And so they require that providers have their own private “meaningful use” justifications for HIT when the most important meaningful use of HIT is by the market. This requirement stymies HIT adoption and forces HIT to accommodate the needs of providers, rather than the needs of the market. Dr. Brailer should not have praised Dr. Blumenthal for creating a new bureaucratic tool (that is what it boils down to, as “meaningful use” is a tool for doling out federal subsidies.) He should have excoriated him for setting up such an unfortunate road block to successful market-based health care.

I admit that we are closer than we were a decade ago. At least we have an HIT czar. The next step is for the HIT czar to understand in the broadest sense why HIT is important. I fear that that will not happen until the HIT czar is a manager, not a doctor.

The Cambridge Cabal

It is hard to imagine that fifteen years ago the private sector seemed to have figured out how to contain health spending. One third of privately insured patients were in HMOs and nearly all the rest were in PPOs. While insurers were experimenting with tight networks and capitated payments, providers were responding by forming integrated delivery systems and offering capitated models of their own. Private health inflation was nearly flat, even as Medicare and Medicaid spending continued to grow by double digits annually.

Everyone should know what happened next. Americans rebelled against managed care (the Boston Globe quipped, “People hate its guts.”) HMOs fell by the wayside, PPOs expanded networks to include even the least efficient providers, and powerful hospital systems emerged to dominate markets from coast to coast. Providers struggled to implement integration and capitation due to a lack of performance data. Why make the effort when insurers had reinstated no-questions-asked fee-for-service payments? We now suffer through the purgatory that is managed care “lite.”

Today’s Wall Street Journal calls for the private sector to again take the lead on cost containment. I am afraid that ship may have sailed. Private insurers have been made out as the villains in this drama and are reluctant to do much more than implement pathetic pay-for-performance schemes while standing back and waiting for the government to act.

And act it will. Don Berwick, a Harvard professor who was just nominated to take the reins at the Center for Medicare and Medicaid Services, is a big admirer of the British National Institutes for Clinical Excellence. NICE is a rationing board run by academics who study cost effectiveness data before deciding what services the Brits can receive. David Cutler, the brilliant Harvard economist who is likely to head the Independent Medicare Advisory Review Board, is another scholar of cost-effectiveness methods and is philosophically opposed to price controls. So when it comes time for Medicare to trim spending, you can bet that they will ration a la NICE. This will provide cover for private insurers to do the same. (I predicted all of this in my 2003 book What’s Your Life Worth? which was read by about 25 people who were not family members. I also called for a public discussion of rationing methods; it appears that NICE-style rationing will instead be implemented by stealth.) Medicare is also experimenting with bundled payments and stronger quality incentives; expect the private sector to again follow suit.

The future of cost containment is no longer in the hands of the private sector. Instead, a few bright, well-intentioned academics will call the shots at Medicare, with private insurers playing Monkey See, Monkey Do. Only the academic community cannot agree on the best course of action. Many of my colleagues believe that the state of the art of cost-effectiveness analysis – what they do at NICE and are likely to do at IMAB – is rather pathetic. I tend to agree. Academics also disagree about whether and how to implement bundled payments.

So a few Harvard scholars will soon get to decide what is the best way to control health spending. Except they don’t know what is best, they can only guess. Whatever they do will be an experiment; the entire health system will be their petri dish.

In my world, private insurers and providers would conduct their own experiments. Some would succeed, others would, and resources would flow to those that worked in practice and not just those that made sense in the seminar room. But I afraid that world no longer exists. By demonizing private insurers, the Democrats have made sure that cost containment will be the sole purview of the federal government. I hope that Don Berwick, David Cutler and the rest of their Cambridge cabal get it right. Unfortunately, there will be no way from them or anyone else to know whether they did or not. We will soon find out where their ideas will take us. We can only imagine where else we might have gone.

They’ve Reformed Health Insurance. Now It Is Time to Reform Health Care

I’ve commented about the absence of any meaningful cost containment in the new health reform legislation. It’s not that I favor direct government command and control over spending. So what could the government do to promote efficiency? Very little, and yet quite a lot. After decades of talk and years of effort, it is time to give us integrated EHR. Not the type that the politicians and doctors are talking about, which is EHR for doctors only. We need EHR for managers. Think of EHR not merely as a medical decision making tool, but as a management decision making tool.

I cannot understate the importance of management-focused EHR. Without EHR, we cannot hope to measure either costs or quality. The lack of this basic information doomed physician hospital organizations and integrated delivery systems and is posing insurmountable obstacles to report cards, pay for performance, and accountable care organizations. The idea is so simple and yet no one wants to accept it: if you can’t measure something, you aren’t going to improve it.

The Obama administration is continuing the EHR initiative begun under Bush. But these EHRs are for doctors’ eyes only. And they will not contain the cost or outcome information required to make them effective as management tools. If we don’t fix this, we may as well give up on reforming healthcare delivery. We will have the same messed up fragmented system where no one is accountable to anyone for their costs or quality.

What will it take?

Integrated health records that link patient information across all providers. We are getting there, but not everyone is on board. EHR providers fear being driven out of business (this is a common worry when there is a standards battle.) More problematically, doctors don’t want the kind of transparency that EHR will bring.

Linking clinical and administrative data. We need risk adjusters to do proper quality measurement and this means making clinical data available to managers.

Outcomes data. The outcomes data currently available is shockingly limited. We can measure the quality of an automobile in a hundred different ways. Why can’t we get decent health outcomes data? It can be done, and there is some amazing ongoing research that points the way. Getting this information into EHRs should be job #1.

Won’t this all create concerns about privacy? I don’t think so. Employers and insurers can learn any evil tidbits they need to know from administrative claims data. There have been no privacy scandals thus far. Adding de-identified medical information to the mix wouldn’t threaten privacy, but how it would dramatically improve care delivery! I don’t see how we can reform healthcare delivery without it.

Bending the Curve Versus Shifting the Intercept

Time for another guest blog. This from my long time collaborator Will White at Cornell University.

Recently there has been a lot of discussion of “bending the curve” on health care costs. The basic notion is straight forward. Health care expenditures are currently growing much faster than GDP. Project current rates forward and it’s easy to conclude that it won’t be too long before health will gobble up the entire national product. (The CBO estimates that at existing rates, we will get there around 2082). But slow down the rate of growth to a sustainable level by permanently flattening the trajectory of the health cost curve and bingo, problem solved.

Bending the curve is certainly an appealing idea. But taking a look at some of the kinds of strategies actually being proposed, it seems “shifting the intercept” may be a more appropriate metaphor. Take the recent 2009 Brookings piece “Bending the Curve” by Joseph Antos and colleagues. They call for improved use of IT, better use of comparative effectiveness research and improvements in the health care workforce. All good ideas that could potentially substantially improve efficiency and slow the rate of cost growth. But permanently? Past history is replete with innovations like the Medicare Prospective Payment System whose implementation was accompanied by substantial savings, but where eventually the potential for gains was largely exhausted and cost growth returned to its old levels. The real elephant in the room seems technological change. The introduction of new technology could certainly be better rationalized; everyone has stories of expensive new tests and treatments yielding questionable benefits. However, raising the bar and requiring a higher cost benefit ratio to introduce new technology by no means guarantees there won’t still be substantial cost growth. This will depend on the pace of medical innovation and how it impacts on costs and that seems hard to forecast. If we really want to bend the curve, there is a simple solution, but I don’t think most people would like it: move to a budget driven system that caps growth at the level of growth in the GDP.

Make No Little Plans or, How Republicans Can Be Meaningful Players in the Health Reform Debate

In a virtual carbon copy of their previous Wall Street Journal op-ed pieces, John Kogan, Glenn Hubbard, and Dan Kessler recently (9/25/2009) offered the standard Republican critique of Democratic health reform proposals before laying out their own agenda.  They are long on anti-government ideology and often fall short on economic principles.  But mainly, their ideas lack the power to bring about real change.  Daniel Burnham once said, “Make no little plans. They have no magic to stir men’s blood.”  Republicans have become the anti-Burnhams.

They start by trashing mandates to purchase insurance.  I would agree with them if the uninsured, upon falling gravely ill, had the good graces to forego charity care and die cheaply.  Thankfully, that isn’t how the system works.  Many, perhaps most, providers are willing to treat the uninsured and worry about payment later.  The uninsured are free riders waiting to happen.  We all ought to pay our fair share without forcing providers to go to court to press for payment.

Kogan et al. also trash the idea of uniform coverage, instead arguing that everyone should be allowed to buy the coverage that is best for them.  This type of argument almost always makes sense.  I like the taste of Ghirardelli chocolate, and when I purchase their candy bars this creates value for me and Ghirardelli alike.  And I don’t have to buy vanilla.  But if I feel prone to heart disease and load up on cardiovascular insurance, I am merely engineering a transfer of wealth from my insurer to me.  There is no value creation to speak of.  And what if I skimp on cancer coverage and my prostate goes bad?  Will the hospital turn me away?  Some benefit mandates are silly: acupuncture, podiatry, and so forth.  But these are a sideshow to the main event.

I agree with some of their concerns about the Democratic proposals.  It is foolish to predicate cost savings on future Medicare cuts.  And it is dangerous to predicate funding of insurance expansion on tax increases.  Add the Congressional Democratic tax proposals to proposed increases in state taxes and marginal tax rates will quickly exceed 50 percent.  Maybe this is a clever way to cut health care spending.  What surgeon would want to operate on a Medicare patient when fees are 25 percent lower and taxes eat up more than half of what is left?

All of this is window dressing to Kogan et al.’s own three-pronged proposal: allow insurers to compete across state lines, give a bigger push to high deductible health plans, and fix malpractice. None of these ideas are silly, but none will have much of an impact.  Most local healthcare markets already have 4-6 major carriers, including a Blue plan, United, Aetna, Cigna, and Humana, not to mention some local plans.  A few markets are down to 2-3 choices, but I don’t see how letting someone purchase insurance from Aetna in Massachusetts will improve competition for consumers in Connecticut, who already have a local Aetna option.   Besides, successful insurers must manage local provider networks.  Out-of-town plans won’t make a dent in most markets.

There is nothing wrong with high deductible health plans and the Democrats are committing economic malpractice by trying to limit out of pocket spending.  But unless we are willing to expose individuals to extremely high deductibles, these plans will not influence the marginal purchases of the most severely ill Americans.  And malpractice reform?  We already have it in many states and the best research suggests that if reform goes national, spending could drop by as much as 3 percent, giving us a four month respite from a decades-long cost spiral. This is hardly revolutionary.

It is time for Republicans to abandon their little plans.  Capping or eliminating tax deductibility of insurance is a good start.  But our best chance to make markets work is to reinvigorate managed care, which the Wall Street Journal and many Republicans once equated to socialized medicine.  Fueled by a revolution in health information technology, managed care in the 2010s would be a far cry from the much reviled HMOs of the past.  There have been many proposals to harness competition among managed care plans, from Alain Enthoven’s managed competition, to Stephen Shortell’s vision of competing integrated delivery systems.  These are big plans, worthy of those who truly believe in market-based healthcare.

Uniquely Inefficient

David Dranove:

Will,

The Eternal Problem

Have you read Alan Garber and Jon Skinner’s new NBER paper, “Is American Health Care Uniquely Inefficient?”  The title is provocative enough, but the answer – yes – and the research that they critique – mainly Cutler’s work on innovation – make for a fabulous read.