Impact of the Evidence Shortfall on
Health Care Costs, Public Budgets, and the Economy

October 8, 2007

CBO Director Peter Orszag's Presentation at the Institute of Medicine's 2007 Annual Meeting

Thank you I find myself in the common quandary of the final speaker that everything has been said but not everyone has said it. I will try to calibrate my remarks a little bit to reflect that reality and give you a little bit more of policy perspective.

And that really has to do with many of things that we are discussing this morning are central to the nation's long term fiscal challenge and in fact that long term fiscal challenge has been largely misdiagnosed and misrepresented in popular descriptions. It is typically described as being driven mostly by the aging of the baby boomers, lower fertility rates, higher life expectancy, when in fact it is driven almost entirely or to a first approximation by excess health care cost growth. That is the rate at which health care cost grow compared to income per capita. You can see this differential arising even over the next decade, these are our projections for spending on social security, Medicare, Medicaid, through 2017. And you can see over the next decade social security rises by about .5% GDP from 4.2% to 4.8% GDP that is the dark blue bar, the light blue bars are Medicare and the federal share of Medicaid which rises from 4.6% to 5.9% in other words roughly twice as much as a share of GDP even over the next decade.

If you look at over longer periods of time this is accentuated. This chart shows you on the top line what will happen if healthcare costs continue to grow at the same rate over the next 4 decades as they did over the past 4 decades. Under that scenario Medicare and Medicaid would rise from 4.6% of the economy today to 20% of the economy by 2050. Which is the entire size of the federal government today.

I think the most interesting part of this graph is the bottom line. Which isolates the pure effect of demographics on those two programs. The only reason that that line is going up is that the population is getting older and they are more beneficiaries on the two public programs. I think what should be immediately obvious is that the increase between today and 2050, in that bottom dotted line, so aging does affect the federal government fiscal position. But that increase is much smaller than the difference in 2050 between the bottom line and the top line. In other words the rate at which health care costs grow whether they continue to grow at 2.5% point per year faster than income per capita, or 1% or 0.5% is to first approximation the central long time fiscal challenge facing the United States. This graph is why CVO is increasingly becoming the congressional health office out of necessity because our future will be dominated by the rate at which these costs grow.

It is common to say that the sooner we act the better off we are, just to calibrate that, if we slowed health care costs growth from 2.5% percentage points to 1% percentage point, starting in 2015 which is almost impossible to do but lets assume it for the moment, the result in 2050 would be a reduction of 10% GDP in Medicare, Medicaid expenditures for the federal government. That is half of what the federal government spends today. Obviously a very substantial fact.

This all sounds pretty challenging and depressing frankly. And it is further complicated by the fact that it is in plausible to me that we will slow Medicare and Medicaid growth unless overall health care spending also slows. Because if all you did was reduce say payment rates under Medicare and Medicaid and tried to perpetuate that out over time without a slowing of overall health care costs at growth you would create sufficient access problems that would be inconsistent with underlying promise of the programs.

Therefore one needs to think about changes to Medicare and Medicaid in terms of the impact that they can have on the overall health care system. And I am going to come back to that in terms of whether Medicare and Medicaid can lead in evidence based medicine and provide incentives for a lot of the things that some of the former panelists were talking about. Before I get to that I just want to highlight again a point that Elliot made, which is and that others made, which is that there seems to be very substantial opportunity imbedded in this long term fiscal challenge facing the United States. To take costs out of the system without harming health. Elliot put up, by the way I have an updated slide of Elliot's own work, so one piece of value added.

On sees significant variations across different parts of the United States that does not translate into differences in health quality or health outcomes. I recently formed a panel of health advisors on which several people in the audience and on the panel serve and ask them what share of health care costs could be eliminated without any adverse consequences for health outcomes, if we could abstracting from political economy constraints. Just in terms of what is the opportunity. And the answers ranged from about 5% to about 50% with most answers centered around 30% figure that Elliot had put up. 30% of health care spending in the United States today is 5% GDP that is a lot of money to be reduced without any adverse consequences for health outcomes.

The question then becomes why is this happening and I think that it is basically a combination of two things. One is the lack of information specifically about what works and what doesn't. And the second thing is a payment system that accommodates a low value health care delivery, especially in terms of practice norms that deliver unnecessary or very low value care. And that occurs on both the provider and the consumer side.

On the consumer side just to start with that for a moment because it is a little bit easier. Despite media portrayals to the contrary if you look over the past several decades the share of health care expenditures paid out of pocket, which is basically the relevant factor for evaluating the degree to which consumers are faced with costionary has plummeted, from about 33% in 1975 to 15% today. And all available evidence suggests that at least on the margin to the degree that shared declines health care spending overall is increased and then we collectively all pay the burden. So because we each individually don't face that much of a incentive to forgo low value or unnecessary care from a financial perspective health care spending overall is raised and we all then bear that burden.

Now that leads some people to believe that the way forward is more cost sharing and a health savings account type of approach, and that can indeed help to reduce cost. But there are two things that need to be taken into account or kept in mind in evaluating putting all of the oneness on the consumer. And that is first that there is a significant amount of cost sharing that is involved in existing plans and moving to universal health savings account type of approach would have some reduction would entail some reduction in overall spending but it is not as big as you might think in part because you are not moving from free plans you are moving from existing plans that have some cost sharing in them.

And secondly, there is an inherent limit to this because so much of health care costs are concentrated to the top 25% of Medicare beneficiaries account for 85% of cost, that basic healthcare fact that healthcare costs are very concentrated among a small share of the population is replicated in every different setting, Medicaid, the private health care system. And to the extent that we in the United States want to provide insurance and insurance is supposed to provide coverage against catastrophic costs, because those catastrophic costs are accounting for such a large share of overall costs there is an inherent limit to the traction that you get from increased consumer cost sharing. And you even see that embodied in the design of health savings accounts which do provide catastrophic insurance. So the traction that you get from increased cost sharing on the consumer side is there and can help but it does not seem to be sufficient to capture the full potential of say 30% cost reductions that many experts believe are possible.

That leads us to the provider side and there accommodation of additional information and changes in incentives presumably would be beneficial. There is sufficient interest as has already been noted in comparative effectiveness research that is looking specifically at what works and what doesn't, and the house version of the S Ship Legislation actually had some additional funding for an expanded comparative effectiveness effort I think there other sectors of interest in the Congress in expanding comparative effectiveness research. But we need to ask some hard questions about what we mean by comparative effectiveness research and how it would be implemented.

First is what kind of research is undertaken and what is the standard of evidence used. I want to really underscore something that Mark McClellan mentioned earlier on which is, the tension in using nonrandomized evidence which seems like it is going to be absolutely at the heart of an expanded evidenced based medicine effort. It seems implausible to me that we are going to build out the evidence base across a whole variety of different clinical interventions, and practice norms, and what have you, using randomized control trials, especially if we want to study subpopulations. But as an economists I am well aware of the limitations on panel data econometrics where you try to control for everything possible to control for and you still don't succeed and that tension between using statistical techniques on panel data sets of electronic health records which seems like the only cost effective and feasible mechanism for significantly expanding the evidence base. And the inherent limitation on being sure that you have got causation will be the central dilemma that this community and others involved in evidence base medicine will have to struggle with.

In terms of the budgetary effects from comparative effectiveness research a lot depends on both what is done and then how it is implemented. If all you do is a bunch of literature surveys and then release that to the world you would expect the effects to be relatively modest. If you are doing new research through registry's, through analysis of electronic health records and what have you, and just put that for consumption by the medical profession there may be somewhat larger effects. But the real traction will come I believe from building the results of that research into incentive payments for providers. That is moving from a fee for service system which is what we have today, to something like a fee for value system, where we are providing more financial incentives for higher value care, and weaker financial incentives or perhaps even penalties for low value or negative value health care delivery. The design of that is going to be very complicated and difficult to implement but that is where the long term budgetary savings potentially could come.

And it is plausible to me that the combination of some increase cost sharing on the consumer side and the a very substantially expanded set of comparative effectiveness effort combined with changes in the incentive system for providers is the best hope that we have to capturing that 30% opportunity. So it will be difficult to implement but struggling with it seems to me the best promise that we have to helping to reduce long term costs in the health system and again that is the central long term fiscal challenge facing the United States. So what you are struggling with this morning and throughout the panel maps over into the nation's long term fiscal challenge in a way that is much closer than most media accounts would suggest.

Thank you very much.