Resident Fellow Thomas P. Miller
Last May, the Joint Economic Committee organized a hearing on many elements of this issue, and you can find that on the JEC Web site. Walton Francis was one of the folks testifying. You can also check out Michael Porter and Elizabeth Teisberg's "Redefining Health Care" or my shorter and more mixed review in Health Affairs in the September/October issue.
The problem with the current system is that it's a high-cost system. We get a lot of good things--but they are not matched by better value. This causes problems in terms of affordability for health care, access to it, and higher opportunity costs because the additional dollars we spend on health care might be going for something that could also be valuable as opposed to everything being spent within the health care sector.
We also have inconsistency in quality. We have peaks of excellence in our health care system but inconsistencies in where and when you get it. With respect to recommended care, the statistics are that a little more than half the time you might get the right stuff. Medical errors can be a problem. They can be a little overstated, but there is a real impact in terms of wrong delivery of care at the wrong time. It's a very complex system to navigate around; even very informed people find it hard to make heads or tails of what really matters and what's going on. It's less than an easy system to engage compared to other consumer goods.
There are poor incentives. We pay for stuff that is done; we don't necessarily focus on whether it's worth what is being paid and what it produces in terms of outcomes rather than inputs. And we have inadequate information. We just don't know enough about what works and who performs better, if not best. We lack sufficient data, effective measures, and standards. Even when they exist, they are not widely available or usable at the consumer level.
Why would adding improved consumer information to the health care toolkit matter? Third parties could certainly use some first-party help. In the post–managed care world, we found that various third parties--whether it's an insurer, an employer/payer sponsoring a plan, government program administrator, or politicians micromanaging on occasion--have given us the system we have. So let's at least think about giving a chance to consumers to perhaps make some of the same mistakes, but also different ones, and perhaps find out different results by having their hands engaged in the process more than they have in the past.
Consumers might begin to internalize and accept the limits on less than "everything, right now, at no cost" if they could understand better what the actual cost and quality trade-offs are and then could find more effective ways to optimize their choices within whatever constrained set of resources they may have. Consumer-driven health care, a relatively recent entry into the field, begins to encourage more cost- and value-sensitive decision-making in decentralized, individual-level decisions, but we're not there yet. We talk about empowerment, but information is all about putting some reality into the rhetoric of empowerment. Just giving folks more cost-sharing with some money funded into an account doesn't tell them how they're going to make better choices and end up better for it as a result of being more engaged in their health care. That's the information component.
There's also an issue of trust and legitimacy. Not only do we have folks not necessarily trusting insurers to tell them what exactly is right, or their employers, or a government program, but if we are going to have new information intermediaries or other incarnations of these past players providing this information, it's got to be done in a more transparent, verifiable, and trustworthy manner so that folks who want to engage that information think it's actually worth their while.
There are spillover effects in terms of the kind of competition that could ensue from greater information. More comparative information, as we know in other sectors of the economy, fuels competition, improves the overall level of services, but there's also a kind of dynamic effect on the provider's side, the supply side. It's somewhat "performing to the test," but there's also a positive effect within the physician and hospital communities if they know on a relative yardstick they're not doing as well as someone else: Even if it's not always being pressed to the consumer level in terms of taking their business elsewhere, it's likely to have some positive effects there. Also, as health care is increasingly complex and needs to be customized, the more you can drill that information down to a form in which people can actually use it would be very valuable.
We have some degree of transparency in health care, primarily at the hospital level, but it's more of a rear view than a leading indicator of what's ahead. So we need to go beyond that, which is kind of an open door policy.
The Bush Administration's Transparency Initiative is an executive order on the heels of some earlier provision of information on paying for procedures, what the reimbursements were in Medicare and the Federal Employees Health Benefits Program. It was issued last August, focusing on quality transparency, mostly through trying to measure the quality of services provided to beneficiaries in federal health programs, but hopefully enough claims and data aggregation so that it means something on a larger scale.
Pricing transparency was the greater focus: the prices paid for procedures by providers to beneficiaries and enrollees in the plan. It's unsure as to whether we're going to step up to the really important level, which is the overall cost per episode. It's one thing to know the individualized list price for a procedure; it's another thing to know what it means over the entire continuum of care. In the same way, there are hopes of eventually getting to overall cost of treatment for chronic conditions. These are possibilities, though, but they're not exactly on the board as yet, and I don't know that they’re going to be promoted as aggressively as the price side.
There was another component of this--to promote care, quality, and efficiency--but if you really look between the lines, it basically says that as long as they provide you with a consumer-driven health care plan product, that's good enough for us. So we probably need to go a little bit further in that regard, but that's the scope of the executive order--and, of course, the phase-in of interoperable health information technology, which has been an initiative of the Administration for some time, and trying to leverage that through other people participating with the federal programs.
This Transparency Initiative complements other earlier initiatives: consumer-driven health care, Medicare, health IT. It reinforces the emphasis on trying to switch to more of a market-driven approach to health care and health services and leveraging the federal role as the first big mover in the field. You really can't transform the health care system with half of it doing something else.
On the other hand, the federal role moves slowly. Everybody has to be on board; you have to get everybody around the table. It doesn't have the nimbleness and the flexibility and the experimentation you might have in the private sector. So it's trying to be a collaborative with other stakeholders. There was actually a line in the executive order saying this will all be done without incurring additional federal cost. I find that more of a hope than a dream, but we'll see if that can actually be done. At some point, you do have to pay for things.
Early assessment: The federal program prices are a nice opening measure, but they are just a crude start. These are not market-based prices in almost every case. There's a little bit more in the FEHBP. They're negotiated rates; they're telling you an initial setting of what might be a price for something, but far from what you'd like to know if you're out in the wider marketplace. Measuring the all-in market-based cost per episode or conditions? Much further away. How much transparency will really be provided to the public after this is aggregated at the federal level? That remains to be seen. How much data-sharing will actually occur with private payers? Private information may come in, but the question is what will come back out that private payers can use differently than federal programs do?
But you work with what you have. You attract some attention to the issue, and it's a start. On the IT side, my skepticism is just that building better, newer, and faster information technology pipes is terrific; it's important; but it doesn't ensure that we will use effectively what's in those pipes. So we've got to think about the incentive structure around that information, whether we're getting information flowing through the high-speed highway in the way we want: more relevant measures.
What do we need to know about what consumers really need to know and how can they learn it? The unit and the level of measurement really matter. There are real distinctions between health plan choice information, which we've had for some time--it could be better--and what we fundamentally need, which is better point-of-service decision information. It matters where you go to get your health care, what you do when you're told "follow this treatment," how it's followed up; that's what is going to turn around the delivery side more than even the aggregation of choices within an insurance bundle. That deals with the front-end discretionary care, the things that consumers can engage and use.
There are other important costs and issues, which are pointed out by critics of transparency who say, "What about those high-cost cases, emergency situations? You don’t have time to put thought into it." Then you need a bundled choice, and that's best done through insurance. But it means that we need better differentiated insurance options prospectively, more transparency about that, that not all insurance is the same; they do things in different ways. And if we can make those choices more transparent, that’s more than just knowing what your premium is and what your cost-sharing is and what the covered benefits are. It's how are you going to deal with that $50,000, $100,000 episode of care, and why are you better than another insurer in doing that? How do you go about it? What incentives do you provide?
We need physician-identifiable information. We have decent information on hospital-level performance on the quality side--not on the price side, which is much more of a murky area--but there is variation within hospitals among the physicians. So if you don’t know who is really making the decisions--what's that unit of analysis of the decision-maker, what their patterns are--even though you might, in the aggregate, think it's a better quality hospital or a more affordable hospital, it may not matter in terms of what's facing you in a particular set of decisions, and you want to make the relevant decision-maker, through this information transparency, more accountable. It also means the consumer is more accountable for making the decisions because you live with the consequences of what you decide when you have options.
An early assessment is that price alone draws attention, but it's not enough. We need the all-in cost by diagnosis or episode of treatment. We need to know a lot more about effectiveness. That doesn't mean that you've checked off all the boxes in terms of the latest recommended guidelines about what processes you go through, what we can measure in terms of the tests that we're provided. It really comes down to a better measure of the outcomes. What matters to people is whether they end being healthier or not as ill or somewhat better off, not that they did different things to you that a committee happened to recommend.
Efficiency is the same way in that regard. We want to know comparatively the resource use for one provider as opposed to another, not just the initial list price. Where you start isn’t where you finish. Consumer satisfaction is often neglected by the esteemed bodies that think about what the best quality care is, but in fact there's a conflict sometimes between what a consumer cares about and what experts think they need to know. Now it's a balance between those two, but we should not neglect the fact that if you're not happy as a consumer, that matters too, and the dimensions of that are on a subjective level.
We need better information but not perfect information that would stop us in our tracks. We need to drop the pretense that we're going to have predictive certainty and all this information is going to be precise, that there is this perfect quality threshold you have to meet or you don't meet. There is a continuum. In addition, it's the probability of this information. Nobody hits 100 percent every time out, but there are tendencies and trends, and that's what you're trying to give people: a better gauge as to, relatively speaking, where they are going to end up on these different types of measures across a range.
The continuum of cost and quality means that it's not necessarily only high-quality care that you pay for; you pay for the best value care that you can get. If you have fewer resources, you may not be able to shoot at the top, but you want to optimize that mix of cost and quality, and different people will end up in different places.
Data aggregation is very important. You want to collect it once, use it often. The problem here is that the private sector could do this more nimbly and effectively, but they don’t have enough denominator size and critical mass in terms of data. That means you have to get it out of the federal government, and electronic health records are a tool in this, but not an end in itself. We need to expand access to the Centers for Medicare and Medicaid Services physician-identifiable data.
Senator Judd Gregg introduced legislation last month; I recommend that you look at it carefully. It would provide a way to open up the vault for the CMS physician-identifiable claims data with appropriate safeguards through intermediary organizations that could provide some analysis of cost and quality and then, what's even better, make that more broadly transparent to the public as well as the requester.
Common data, common measures, don't require comprehensive consensus care standards. Different folks may end up in different places, and we’d like to have some competition in that regard. We want some pluralism and competition, and should not oversell sameness.
We want to link this to provider incentives because if the payment isn't there, people won't pay as much attention to it. We want to emphasize the trust factor. The challenges are that this data information is power; some folks are not eager to pool and share it.
I think also that you want to watch out for mission creep in this regard. Most of our health policy disagreements are being reimported back into what is said to be the information debate. People who have fought about other things are fighting about it in the context of information or trying to achieve their goals by saying it's all part of information, and that goes everywhere from trying to redesign the entire health care system to fighting over consumer-driven care.
Thomas P. Miller is a resident fellow at AEI.