U

P‑R‑O‑C‑E‑E‑D‑I‑N‑G‑S

8:33 a.m.

CHAIR JOHNSON: Good morning, everybody. Welcome back. Before we get into our hearing discussion with the first panel, Megan has some comments on some transportation logistics. We might hear from her at this time.

MS. GRIGGS: If anybody is going to National there is a free shuttle from the hotel. If you are flying out of BWI or Dulles, come see us at the break and we'll make sure that you get a taxi at the end of the day.

CHAIR JOHNSON: Thank you. Well, as we enter our now third day of briefings on the healthcare initiatives that we have and the needs of the Citizens' Healthcare Working Group, we are delighted to have with us Jennifer Jenson from the Congressional Research Service, Rick Foster, and Steve Heffler from the Centers for Medicare and Medicaid.

Just a brief background. First, we have your bios in our books so we won't read through those but your credentials are impressive and we've heard about you and that's why we've ask you to share your expertise with us. It's all been good, by the way. Let me just start there.

Second comment, as all of you are aware, in the legislation calling for the Citizens' Healthcare Working Group, there are a series of required subject matters for the hearings and for the report. As we looked at those questions and as we collaborated with Senator Wyden and Senator Hatch we realized that both CMS and CRS would be the best potential resources for questions that Senator Wyden and Senator Hatch have asked that we respond to.

Stephanie Kennan and Jamie Shimek of Senator Wyden's office are here this morning. They are kind of going to watch over us, I think, to make sure we get all that information or we answer the questions. Actually, that's not really intentional to suggest that but they are here and they are working with us to make sure we do have information. We appreciate your willingness to assist us in that respect.

As we did yesterday, we are going to ask you to share a presentation, each of you, about 20 minutes or so as we understand it. We have a tendency of wanting to break into the discussion and I'm going to ask our working group if you would be able to hold your questions until Rick and Steve and Jennifer are complete with their presentations because there is so much material that we need to hear from them. Then we'll have plenty of time to respond to questions that we have raised.

If we can proceed that way and, Jennifer, you're first on our agenda. Have you changed the order? You would prefer that Rick ‑‑

MS. JENSON: That would make more sense.

CHAIR JOHNSON: Okay. We're open to that. We want to be flexible.

MR. FOSTER: So we're all set then?

CHAIR JOHNSON: We're all ready.

MR. FOSTER: Good morning everybody. Happy Friday the 13th. For those of you who think about such things, you'll be pleased to know this is the only Friday the 13th all year long.

I'm Rick Foster. I'm the Chief Actuary at CMS. This is Steve Heffler who is the Director of our National Health Statistics Group. We are really pleased to have the opportunity to be here and meet with you and summarize the past trends in national health expenditures and also talk a little bit about the future outlook.

An awful lot of people are surprised when they find out that there is no single source of data collected on health expenditures in the U.S. There's no agency that does that directly. Instead, our little office with a dozen people or so collects data and information from every source we can find and tries to fit it together in a sensible way to put together our National Health Expenditure Accounts.

In the process we sometimes have to resolve discrepancies between conflicting sources of data as best we can. We are going to talk today primarily about the core Health Accounts.

We have some auxiliary studies that we do branching away from the Health Accounts that may also be of potential interest and use for your working group so we'll talk a little bit about those as well.

The National Health Accounts themselves are a two‑ dimensional matrix of expenditures. We first of all slice this data by type of service or type of expenditure. Here, for example, we show overall total national health expenditures that would include such things as research and construction of buildings and other investments. Also included are administrative costs. Then you can boil it down to personal healthcare, which are the things we normally think of like hospital services, physician care, prescription drugs, etc. That's one slice of this matrix.

The other slice is by source of funding. We show in the slide private versus public but within private, of course, there’s private health insurance, and there's individuals’ out‑of‑pocket payments that they make themselves. With public you have Medicare and Medicaid, of course, as well as some other programs.

For each year we create such a matrix and try to make sure that everything makes sense as it fits in there. It's not a straightforward process. In the end we get a pretty decent set of expenditure data by type of service, and by who is paying for it. Then we track that through time.

This chart shows total national health expenditures as a percentage of the GDP or economic output going back to 1965. The first thing you notice from this is that the curve is going up so that--with rare exceptions--the increase in healthcare costs outpaces the increase in economic output or GDP.

That's not a big surprise when you think about it. It's certainly not a surprise to this group. If you think about GDP, it increases with the number of workers, general inflation, and productivity pretty much. Whereas healthcare costs increase by the number of people in the population, also general inflation, also what we refer to as excess medical inflation.

That's medical price growth above and beyond general inflation. But then health cost increases also reflect growth in the utilization of services and growth in the “intensity” or the average complexity of services. Typically those factors outnumber the GDP factors and you get a curve like this that goes up.

You can notice for brief periods there are times in the curve where it levels off for two or three years. Most notably it leveled off pretty much from about 1993 through 1999. That reflected both slower growth in healthcare largely due to the introduction and expansion of managed care during that time. But also because the economy was performing pretty strongly during that period, so it kept up with the slower rate of growth of healthcare cost.

In 2003, which is the last year that we have historical data for in the Health Accounts, health expenditures were 15.3 percent of GDP. You can see from the chart that the public share of that was about 45 percent, roughly.

This next chart looks at the percentage rate of growth since 1980 in national health expenditures. You can see that the rate of growth slowed significantly in 2003 compared to the prior year. It had been 9.3 percent in 2002 and it went down to 7.7 percent. That deceleration was primarily due to things that happened in the public sector, which we'll talk more about in a bit.

Now, of course, we had been accelerating from a low point of about 5 percent in 1996.

It's useful to take these growth rates and decompose them into key factors. Here we have taken out population growth already so these are per capita average growth rates over three different periods. We split it up into three components: the top bar is economy‑wide price growth (or general inflation).

Then the middle bar is the excess medical price growth that I talked about, above and beyond regular inflation. Then the bottom bar is growth in utilization and intensity of health care services and essentially everything else. This is a residual category. It also includes any errors that we make in the process of measuring prices and so forth. Notice that the recent faster rate of growth over the last several years, the 6.9 percent total, relative to the 10 years prior--most of that difference, in fact, is due to more growth in utilization and intensity of services. The price factors were not very much different between the two periods.

In contrast, if you go back to the period 1980 to 1990, there we had significantly higher general price inflation and also higher excess medical price inflation.

In this next chart we are comparing the various sources of funding for healthcare in the U.S. and also comparing 1980 versus 2003 to see how things have changed.

One of the more significant changes is in out‑of‑pocket expenditures. This category includes direct payment for health services out of pocket either as a cost sharing requirement or because it's not covered by any insurance that an individual has. However, an individual’s health insurance premium payments are not included. If you look at the out‑of‑pocket share of total health spending, it's gone down from 24 percent in 1980 to about 14 in 2003.

Now, on the other side of the coin, private health insurance ‑‑ and these things are related as many of you know ‑‑ private health insurance has grown from 28 percent of total spending to 36 percent. We also had a number of expansions in Medicaid eligibility during this period, so you see growth in Medicaid’s share of the total. Also some (more moderate) growth in the share paid by Medicare, in part because the Part B deductible was not indexed and not increased very often.

Currently, private health insurance represents more than a third of the total spending, and Medicare plus Medicaid combined (including the State Children's Health Insurance Program) is just about one‑third of the total as well.

This next chart has a comparison of Medicare versus Medicaid growth rates over time. It's interesting to note the significant volatility in the growth rates of each of these programs. If you look at Medicaid, for example, you all remember what that peak in the early 1990s comes from, I trust. There were these so‑called “tax and donation” financing schemes that the states got clever on and that rather drove up the federal cost of Medicaid to the gain of the state cost.

What happened was the states took a lot of this extra federal financing and used it to expand Medicaid services, and to expand coverage, which is in part why we get that spike in total Medicaid expenditures. At the same time, of course, we had a recession in the early 1990s. That led to more people who were eligible for Medicaid, so there were reasons like that as well.

Now, notice the dip in Medicare outlays in 1998, the big deceleration where, in fact, the cost actually went down briefly. You remember what happened then, too. Everybody remembers the Balanced Budget Act. That was the biggest factor but that wasn't the only thing happening. We also had low general inflation rates at the time. Also we were doing a lot back then about certain areas of fraud and abuse that contributed to reduced expenditures, so it wasn't strictly the BBA.

Both Medicare and Medicaid slowed or decelerated in 2003. Medicaid in particular went from what had been over a 12 percent growth rate down to 7.1 percent. That reflected quite a number of states tightening their eligibility requirements and limiting their payment updates for providers.

It also reflects a change in the “upper payment limit” rules, again designed to help address creative state financing techniques to beat the Federal Government out of some more money.

In the case of Medicare, the growth deceleration was not as pronounced. It went from about 7.6 percent down to about 5.7. That was principally the expiration of some legislation, the so‑called “give‑back” legislation for the Balanced Budget Act, so that those provisions were no longer adding to Medicare costs.

Turning to private health insurance, this chart compares growth in private health insurance premiums versus benefits. As you would expect, the two track each other fairly closely, but there are differences--and the differences are often used in describing the so‑called underwriting cycle, which I think most of you are familiar with. At the start of the underwriting cycle, plans get a little carried away in expanding coverage, setting low, competitive pricing, and so forth, being aggressive, trying to capture more market share, and then they live to regret it. They say, "Well, we can't go on losing money and making it up in volume. We now have to actually make a little bit of money." That leads to the second phase of the underwriting cycle, where plans have to achieve a more conservative, sustainable financial position.

For the last five years we've had growth in private health insurance premiums that exceeded the growth in benefits, and that period followed four earlier years of the opposite. It wasn't so long ago, Mike [O’Grady], right?, that people said the underwriting cycle was dead, it was gone.

MR. O'GRADY: Right.

MR. FOSTER: And they were wrong. (I hope nobody here wrote those articles.) Growth in both the private health premiums and benefits slowed in 2003. This change is primarily reflecting the slower growth in underlying costs, principally prescription drug costs as we'll see in a bit. It's also worth noting that private health insurance enrollment has decreased to about half a percent a year, roughly, for each of 2001, 2002, and 2003.

In this next chart, we again compare 1980 and 2003. In this case we're showing where did the money go, how were the healthcare dollars spent, focusing on the major types of service in each case. What's interesting is that currently over half of all the spending shows up for hospital plus physician services. Those remain the two big categories in many respects.

The hospital share, however, has been steadily declining over time and by itself is down to a little less than one‑third of the total share now. On the other hand, prescription drugs have gone from about 5 percent of the total up to about 11, so drugs have been making up a lot of the difference. The physician's share has grown slightly.

The lower curve in this chart shows the rate of growth in total hospital spending in the U.S., including both inpatient and outpatient. You can see that there are several recent years of accelerating hospital spending growth, but then it decelerated in 2003 down to 6.5 percent.

That deceleration was driven largely by what happened in the public sector by Medicare and especially Medicaid. In 2003 a lot of state governments, in their effort to get Medicaid costs better under control, actually did things like freezing the update for hospital payments.

In other words, there was no increase from the prior year. Such changes contributed to this significant decline or deceleration in the growth rate. Because hospital spending is such a big portion of total national health spending, that helps slow down the overall spending growth rate that we saw earlier, just from what happened with Medicaid.

Notice on the prescription drug growth rates the sizable reduction from 2002 to 2003. It went from 14.9 percent growth rate down to 10.7. You remember what we've been seeing in recent years, and in 2003 that's when Claritin went over the counter. We also had increased use of generic drugs in part because of the increased use of tiered co‑payment systems that encouraged people to get generics or to use less expensive drugs where possible.

Now, on the other hand, look back at the peak in 1999 with a 19.7 percent growth rate for prescription drugs. That was the year we had a number of new drug introductions including Vioxx and Celebrex. Generally during that run‑up was when the FDA was accelerating their approval process. Fortunately that trend is now going in the other direction but where it goes in the next few years remains to be seen.

At this point I'll turn things over to Steve.

MR. HEFFLER: Great. Thank you. I'm going to touch on some of the ancillary products that we produced as part of our National Health Expenditure Accounts work that Rick had mentioned earlier. What is interesting about these different cuts of the data is it allows us to look at different pieces of the healthcare sector and sort of hone in on different pieces.

All these products are controlled in aggregate to the total national health expenditures, just a different way to cut the data, so we can look at sort of who is sponsoring healthcare and that is going to be the business, household, and Government analysis we do. We can look at who is getting services.

We can look at spending by age to see whether it's been on children or the working age population or the elderly. We can see what states are spending relative to each other across the country which I'm sure is of great interest to this group.

We can also look as we project this data out for 10 years at where we think things are headed which is particularly interesting this year since this was the first effort where we had projected the impact on the overall healthcare sector of the new Medicare drug benefit.

I'll have some charts at the end that have some interesting movements in them and I'll point that out. First looking at the sponsors of healthcare, the previous box was business, households, and Government. We can look at who is actually sponsoring healthcare.

Rick presented some data where we looked at private health insurance trends. What we can do is sort of split that data in more detail and look at who is paying private health insurance spending. Is it employers? Is it employees? Is it Federal Government employees? Is it private employers? And split the data that way.

You can see on the right‑hand side of this chart once we've done that analysis how the data can be cut a different way. Government spending, and this is 2003 data only, is about almost 40 percent of overall healthcare spending followed by household spending which is 32 percent and businesses which is 26 percent. The next sort of series of slides tells you how we kind of walk from our national health account data to this data just to give you an indication of sort of the effort that goes into splitting it.

If we look at private health insurance, I mentioned earlier employees, households pay for private health insurance. Private businesses also pay premiums for their employees as private health insurance. Government is also an employer so Government pays for private health insurance as well so the data is split that way.

Out‑of‑pocket spending, as Rick mentioned, is either those without insurance or co‑pay and deductibles. That's coming out of the household. Other private spending which is a small category mostly goes to this sort of other private category. There's a little bit on the business side.

This one is interesting in the Medicare. You have all these sort of different ways that the Medicare spending is financed. It's financed by Government out of the general revenue. It's financed by businesses and employees, their households out of the taxes that they pay so that money is split amongst the different categories.

Medicaid is strictly Government spending in the way we do this analysis. And there's a couple of other sort of smaller residual Government type spending categories, other federal, state, and local that's Government. That is sort of how we arrived at these statistics and you can see the complexity of this in sort of working with the data and trying to split it and take different looks at it.

What we can do then with that data is look at the burden that each of these sponsors is under in paying for their healthcare cost. One of the ways we do this is we look at spending as a percentage of revenue or percentage of income for each of these groups. Here over selected years we have federal, state and local, and household spending as a percentage of revenue or income.

You can see on the federal side there's a lot of debate right now about Medicaid and Medicare and how much healthcare is eating up the federal budget and where that's going and what is going to happen when the Baby Boomers hit and so forth. You can see here from a burden standpoint why that is an issue. We have seen that share rise and we have seen in 2003, you know, it's nearly 30 percent of revenue in that year is going to what we call Federal Government health spending.

On the state and local side there are a lot of issues with paying for Medicaid, the difficult situations that states are in from their budget perspectives. You can see over time the burden for states as a percentage of their revenue is increasing.

The household side, which is much smaller, as a percentage of income here but it's a smaller proportion relative to the Government pieces and has been increasing after sort of plateauing in the mid '90s when you had people switching from fee for service to managed care type plans with much lower co‑ pays at the point of service has now been increasing the last two years and particularly in 2002 and 2003. From a burden standpoint you can see a lot of pressures in the system as healthcare continues to eat more of the budgets of these different groups.

We can also look at health spending by age. Here we are looking at personal healthcare so in one of the earlier slides that Rick had showed you there's a progression down in sort of the detail of the Health Accounts. In this case it's personal healthcare so hospital and physician and prescription drug spending.

You can see sort of disproportionately how the population compares in counts of people versus what they spend on healthcare. This isn't news to anyone that the elderly, which is a much smaller proportion of the population than the other two groups, spends much more relative to its proportion. Likewise, children which is almost 30 percent of the population accounts for just 12 percent of personal healthcare spending.

This is 1999 data so it's a little bit older. This data is much more difficult to get and process and prepare. We only do it periodically this analysis so the data is a little bit older but over time as you'll see these distributions ‑‑ this chart here, these distributions would tend not to change too much.

If we look at sort of the financing piece this is analogous to the pie charts we looked at in total health spending in 1980 and 2003. Here, again, it's just 1999 data split out by age group and sort of who finances. There are some really interesting things you see across these groups, private health insurance for the children and the working age.

Over 40 percent of all the financing, for children is Medicaid a much higher proportion and this all makes sense. And smaller for the working age population and Medicare just a small piece for both those groups. Now, you look at the elderly 65 and over and almost half of spending coming from Medicare. Private health insurance not really paying for that much. This is consistent with how our health system is structured.

One interesting note here, and Rick and I were talking a little bit about this, is the out‑of‑pocket share actually is pretty constant across all the groups. One thing to notice in this, if you look at the bottom, the per capita spending from each of these groups, roughly $1,600 for children, $3,300 for the working age, and then you go to elderly $11,000 per person.

This is a pie chart so it adds to 100 but what you are multiplying these percentages by is much, much larger on that last chart. Even though the proportion of out‑of‑pocket is the same, it's a lot more money in each of those groups.

We can also look at for each of these groups where the money was spent and for what services. There's not a lot of surprises here but maybe that's a good thing when you're putting these estimates together. Physician spending is higher for children and then sort of as a percentage and then the percentage falls as you go across the age groups and you see less for the elderly than for the other two groups.

Then you look at nursing home which is less than 1 percent for children. It makes sense. Three percent for the working age. It makes sense. Then almost 20 percent for the elderly. You can see where the spending is different amongst these age groups.

MR. O'GRADY: May I ask one quick clarifying question? How do you count in terms on the children versus the working age, how do you count when a baby is born? It just looked high for hospital spending for me for little kids but is that a delivery? Does the kid get counted as the hospital cost there or does the mom?

MR. HEFFLER: Um ‑‑

MR. O'GRADY: Sorry. You can get back to me later on that one.

MR. HEFFLER: I think we'll have to.

MR. FOSTER: We'll provide that answer for the record there, “Senator.”

MR. HEFFLER: I would speculate but I don't know that specific answer. I think it may be in the children estimate but don't hold me to that. We can get back and verify that.

We can also look at health spending by state. We are going to have sort of a map of the U.S. here and point out some things that we think are interesting in these estimates. We have not updated our estimates completely yet so this data is 1998 data. It's a little bit older. We are working on updating those estimates.

What we did was we took the U.S. and we split it in sort of three groups and called them high, medium, low per capita spending. The important point to take away from this chart is this is spending by state by the location of the provider. That is the easy part to get. We can get how much hospital spending for hospitals that are in the state of Idaho.

What is harder to get is how much spending on hospital services in the state of Idaho goes for Idaho residents which when you are comparing across data is actually more the relevant comparison that you would want to make so we do an adjustment. This is the location of provider and we do an adjustment where we can switch it to the location of residence. You can see some of the states sort of move groups there so we have highlighted a couple ‑‑

PARTICIPANTS: Your arm.

MR. HEFFLER: I'm sorry. Oh, the wheel will work. I see. There we go. As you can see ‑‑ I'm trying to hide it here. So we've highlighted a couple states and we are going to start with Minnesota. On a location of provider it looks like Minnesota spends a lot on healthcare. It's in that upper third.

What you have in Minnesota is you have Mayo Clinic and you have a lot of people crossing into Minnesota spending for services so when you adjust for that, Minnesota actually drops back down to medium category so the per capita goes from 4,200 to under 4,000.

North Dakota is an interesting case because it does the exact same thing. There is a location of provider. It's high and then it falls into medium. The interesting thing about North Dakota, at least this is what we think is going on, is you have a big city right on the border in Fargo.

You have a lot of people in rural Minnesota that cross that border into North Dakota to get treatment in Fargo. The migration across the states can both be when you have a big clinic like Mayo as well as when you have these border cities and people from rural areas move into those states. That's two examples of where the per capita spending actually isn't as high when you residence adjust.

Now, look at New Jersey which is very high, or in the high ‑‑ I'm sorry, in the medium ‑‑ what is this? This is provider sort of in the medium category when you look at location of provider and the reason there is as it switches is you have people that actually go to the neighboring states, New York and Pennsylvania and so forth, to get their care.

This is just a different way to cut the data and look at things.

When we look at rankings, and this was a big issue the last few years, about states and how they compare, we think this is the better way to look at it when you are ranking states and look at what is spent per state, not just on the location or provider basis. It can change the story a little bit when you do that.

Okay. So the last piece of this is projections and where we're headed. Here is the chart that Rick had talked about with the healthcare spending share of GDP at 15.3 percent in 2003. We made this set of projections released in February. It's a current law projection so it doesn't assume that like the drug benefit is going away or is going to get expanded or any of that is going to change. It doesn't assume anything about Medicaid and so forth, just current law projection.

You can see that we are expecting that health will continue to out pace GDP as it has done for almost all of the history and eventually get to 18.7 percent of overall economic resources. The public piece here is up to almost half of the share. Rick mentioned earlier that in 2003 it was about 45 percent so it's almost half. Something is going on in those 10 years to move it. You will see as we get to the last two slides what's going on there.

Here is the growth rate. We are actually projecting that growth will moderate in the next few years at about the range we were in in 2003, sort of the 7.5 percent range or so. The interesting thing about that is we've got a lot of questions. Why would it moderate? When we break it apart and we look at what is going on in the public and private pieces, you'll see it almost falls out because they are moving so differently from each other they offset and the overall growth stabilizes.

We'll look at the public and the private pieces here and I'll focus a little bit on the periods first and then get into that big spike there you see in 2000 and 2006. What we're seeing in 2004 and 2005 ‑‑ I think I might actually have the wrong ‑‑ well, maybe not ‑‑ in the wrong place here. The public sector is supposed to ‑‑ we're expecting it to grow a little bit faster in 2004 and 2005 than in 2003 so the growth rates sort of bottom out in '03 and then accelerate.

As Rick mentioned, most of the deceleration in 2003 was from the public sector. This happens as Medicare deals with some non‑Part D MMA provisions that add money to the system. It deals with Medicaid which a lot of states that tried to reign in their Medicaid costs are now in '04 and '05 appears to be struggling with that and struggling with keeping things under control. Those costs are expected to accelerate.

On the private side we are actually projecting the trends you saw with hospital and prescription drugs that continue a deceleration in the growth rate of those pieces as there continues to be this push for additional cost sharing on the drug side and what might happen there and the expectation that there aren't major blockbusters coming to the market in the next couple of years.

Rick talked on that chart about utilization versus price and really it's sort of that residual utilization residual piece that we are projecting to slow somewhat, not so much the price piece. So 2006, there you go. There's the Part D impact as the drug benefit comes under Medicare in 2006. What you have is a shift amongst payers in that year. This pie chart illustrates that between 2005 and 2006.

Private health insurance, out of pocket and Medicaid all drop as a share of spending in 2006 and Medicaid picks up the remaining piece ‑‑ Medicare. I'm sorry. So Medicare goes from 2 percent to 28 percent in 2006. That is a big change so when I was talking earlier about that public share going from 45 percent to almost half, this has something to do with it, the shifting of drug spending from out of pocket and private health insurance into Medicare.

The only other point I wanted to add on this was in aggregate we actually are not expecting that large of an impact in total spending from this prescription drug benefit. That is because of the expectations that the additional use of drugs for those that are eligible for this benefit that either had no coverage before or didn't have as good a coverage as the Part‑D benefit will be almost offset by the price discounts that they are anticipating to receive once they enroll in the program.

In aggregate we have only projected an additional one billion dollars in spending because of the drug benefit which, to put into context, is less than one half of one tenth of one percentage point of total health spending so you don't really see it anywhere in those sort of growth charts because in aggregate the impact is not expected to be large. It's really a payer shift that we anticipate to be the major impact from that legislation.

That is the formal part. I don't know if you had anything to add to that.

MR. FOSTER: Yes, show them our website address.

MR. HEFFLER: Okay. Yes.

MR. FOSTER: Also let me just add that we've got our hands pretty full right now with implementing the MMA and so forth but if there is anything we can do to help your working group, don't hesitate to ask and we'll try our best. We won't make too many promises but we would be glad to try and help. You guys have a lot of work cut out for you but we can help perhaps.

CHAIR JOHNSON: Is the best way to do that to come through one of our staff members to your office?

MR. FOSTER: That would be just fine.

CHAIR JOHNSON: Okay. Thanks. Okay. Thank you very much. Lots of data. Lots of information. I'm sure we've got lots of questions. In fact, I've got a few myself but we'll go to Jennifer next and then we'll open for questions when you're done.

MS. JENSON: Thanks for having me today. It's a pleasure to be here and it's a privilege to follow Rick and Steve as well.

I'm going to follow‑up on some of the comments they made. I'm going to reinforce some of the numbers that they presented and I'm going to offer a few additional numbers of my own for you to take a look at. Then I'm going to make a few comments basically designed to nudge you to think critically about what the numbers mean and about trying to respond to growing healthcare costs and spending through policy.

Okay. As Rick already pointed out, U.S. health spending is a large number. From their projections it's estimated to be $1.9 trillion in 2005. That is about $6,400 per person and represents about 15 percent of GDP. I believe it's 15.6 percent of GDP this year. It's a big share of our national income that we're spending on health.

We are also spending a big share of our federal budget. Frequently people talk about Medicare, Medicaid, and the State Children's Health Insurance Program. According to CBO spending on these programs is estimated to be about $520 billion in 2005. This amount is about 21 percent of estimated federal outlays. Federal outlays for 2005 are estimated to be about $2.4 trillion.

In addition to Medicare, Medicaid, and SCHIP there's a bunch of other federal spending on health. There's tax expenditures for health insurance. According to the Joint Committee on Taxation, tax expenditures are expected to be about $90 billion in 2005. By far the largest part of this - $79 billion - is accounted for by the exclusion for employer provided health insurance from income. People get tax savings because they get their health benefits before ‑‑ you have a funny look ‑‑

MR. FRANK: They are not, strictly speaking, expenditures. They are revenues foregone.

MS. JENSON: Right, foregone revenues. That's true.

MR. O'GRADY: That's the term we use. Tax expenditure is like tax credit.

MR. FRANK: We're trying to talk to human beings.

MR. O'GRADY: Just a little translation.

MS. JENSON: How about this? Most people, two‑thirds of people, get their health insurance from their employer. Because health insurance benefits generally are paid before income taxes, most people don't pay taxes on their health insurance benefits. The $79 billion would represent tax savings to individuals because they get their health insurance from their employer. How is that? Does that work for you? Okay. Terrific.

All right. The $79 billion is a big number. Most of this is for tax savings for employer‑provided health insurance. The second biggest item in the same category is $8 billion in tax savings under the deduction for unreimbursed expenses. People who have health spending that exceeds 7.5 percent of their adjusted gross income and are able to deduct part of their health expenses receive about $8 billion in tax benefits.

There are several other items that are quite a bit smaller. The JCT estimates $3 billion in tax savings under the deduction for self‑employed workers who purchase health insurance, and about $400 million in savings this year under Health Savings Accounts. In the last case we're changing to an “m” for millions. Everything else so far has been a “b” for billions.

So, that’s $400 million in tax savings attributable to health savings accounts, and about $100 million in tax savings attributed to the tax credit for displaced workers and a few other people that was enacted in the Trade Act. The Trade Act provides a small tax credit for a small number of people to help them buy health insurance.

MR. HANSEN: What was that last number, please?

MS. JENSON: $100 million. In addition to these tax savings, health benefits for the military, veterans, and federal employees cost a decent amount of money. About $89 billion is the expected amount in 2005, according to the Office of Management and Budget. This amount includes $32 billion in spending for defense health benefits, $27 billion in spending for veterans' medical care, and $31 billion in spending for our health benefits as federal employees.

The final category that I have included–although if you are looking hard, you can find health spending many other places in the budget–but the final category I've included is the Public Health Service, expected to be about $52 billion in spending in 2005. Of that amount $31 billion is ‑‑ excuse me, $29 billion is spending for the National Institutes of Health.

All right. Health constitutes a big share of spending from an economy‑wide perspective,a big share of federal spending, and a fair amount of spending for individuals and businesses. According to data from the Kaiser Family Foundation, premiums for employer-sponsored coverage averaged about $3,700 for individual coverage in 2004, and about $10,000 for family coverage. These amounts include both the employer and the employee contributions for health insurance.

On average the employers pay about 85 percent of the cost of individual coverage and about 75 percent of the cost of family coverage. You can see that is a decent amount of money.

You also can see (on the slide) how costs have been affecting individuals over the past few years. I want to highlight two things. First, the source of healthcare coverage is changing for people. Employer coverage is down a little bit, from 67.8 percent in 2003 to 63.9 percent in ‑‑ I'm sorry, 67.8 percent in 2000 to 63.9 percent in 2003. That is about a 4 percentage point decrease in coverage under employer‑sponsored plans. I think that is attributed both to the fact that employers can't afford coverage,and to the fact that individuals don't always take up coverage. So, it's expensive for both of them. Second, you can see that the number of uninsured has increased over the same time period–by about 1.5 percentage points.

Let's start to think about these numbers and what they mean. I'm not going to offer any answers today. You guys have a difficult charge.

Although, I don't have too many answers, I do want to encourage people to think critically about the different kinds of numbers that they see.

To do that, I'm going to discuss three things. I'm going to put some of the numbers in international perspective. I'm going to talk a little bit about valuing healthcare and valuing other goods because, ultimately, whether it's national spending or federal spending or individual spending or spending by businesses, we do make choices between spending our money on healthcare or spending our money on other things. Then I'm going to talk a little bit about what we can afford.

Regarding the international perspective, the United States spends more on healthcare than other developed countries. This will be no surprise to anyone. According to OECD estimates, in 2002 we spent about 14.6 percent of our gross domestic product on healthcare.

This number compares with an average of 8.5 percent for other OECD countries, so the U.S. share is quite a bit larger. Other countries that spend above the median include Switzerland and Germany which spent about 11 percent of their national income on healthcare in 2002. Then there's a collection of countries that spent in the 9 range, including: Iceland, France, Canada, Norway, Greece, Portugal, and Sweden. All of these countries spent more than the OECD average.

So, in addition to total spending being above average, our per capita spending is more than double the OECD median. For example, the OECD does calculations where they adjust the different countries’ currencies to make them look similar and report all of the amounts in U.S. “purchasing power parity” dollars. According to such a calculation, in 2002, U.S. spending per capita was about $5,267 compared with a median amount of $2,220 for other OECD countries. So, the U.S spends quite a bit more.

This sounds like pretty bad news, right? Well, I'm not really sure. Given our wealth, the high spending may or may not be a problem. It really depends on how much healthcare we want to buy. Some researchers who have looked at international data have shown that about 90 percent of the variation in spending across countries can be explained by differences in the countries’ wealth.

Wealthier countries spend a higher share of their national income on healthcare is basically what the finding is. If you look at how wealthy a country is, that will explain 90 percent of the difference in their healthcare spending.

It raises a question about whether, as a wealthy society, if we want to spend more on healthcare–maybe that's fine. Ultimately, we might want to spend the same amount on healthcare. We might even want to spend more. It really just depends on our preferences for healthcare, compared with other things.

One of Catherine's colleagues at the University of Michigan put together what I think is a terrific table to help you think about this because it sometimes seems counter intuitive to people. On the top line of the table you’ll see gross domestic product per capita for a 40‑year period. Per capita GDP was about $13,000 in 1960 and about $32,000 in 1999, so it increased by about 2.5 times.

The next line is growth in per capita spending on health. It also went up quite a bit, from $646 per capita to about $4,200. This increase shows we are spending about 6.5 times as much on health compared with 1960. All the numbers are adjusted and reported in 1996 dollars, so the currency is the same.

The third line shows spending on all goods besides health, so it's the difference-the top line minus the middle line. You can see that spending on all other goods besides health also increased over the 1960 to 1999 period, from about $12,000 to about $28,000. So, it increased by about 2.25 times.

The data shows that a growing economy can potentially support both more spending on healthcare and more spending on everything else. The issue, of course, is that the difference in growth in health and other spending has been pretty big and probably isn't sustainable forever. But if the growth and health spending slows a little bit and growth from the economy increases a little bit, it's conceivable that we can continue to have more of everything for a little while to come. I'm here to be optimistic. Everyone is always so pessimistic that I figure that I might as well present the other side of the story, not because I think that you shouldn’t try to solve this problem, but rather because…

DR. BAUMEISTER: (Off mic.)

MS. JENSON: That's right. Okay. So the big question is what can we afford, right? Even if we could afford more health spending and growing health spending as a share of our economy. And even if we decided that we wanted to spend more and more of our federal budget on healthcare, which we may or may not want to do, such conclusions don't necessarily imply that everybody gets all of the healthcare that they need. Distribution matters. Even if our economy can afford it in aggregate, individuals might not be able to afford the healthcare that they need.

I understand you talked about the uninsured already, so you already know that 45 million people were uninsured in 2003. Low income people tend to be ‑‑ are more likely to be uninsured. Among the under 65 population, about a third of people earning less than 150 percent of poverty are uninsured. Of people earning between 150 and 199 percent of poverty, about a quarter are uninsured. These numbers compare with only one in ten for people whose income is 200 percent of poverty or higher.

I think that there is a general sense that regardless of whatever amount we might want to spend on healthcare, we could get more for our money. U.S. public health statistics are about average. Our infant mortality rate of 6.8 per thousand live births is about the same as the OECD average. Our life expectancy at birth is actually a little bit below OECD averages even though we are spending more than twice as much on health care.

For example, in 2001, U.S. women's life expectancy at birth was 79.8 years, compared with the OECD average of 80.4 years. The numbers for men were a little closer. U.S. men were expected to live 74.4 years, compared to 74.5 years for men in the other OECD countries.

In addition, there's variation in spending. Steve talked about (geographic variation) a little bit already. Variation in spending doesn't seem to be related to differences in outcomes or differences in satisfaction with care. In addition to variation in aggregate spending, people have shown some interest in variation in Medicare spending per beneficiary. The topic is interesting because Medicare beneficiares all have the same health insurance package, essentially.

MedPAC, the Medicare Payment Advisory Commission, did some analysis using 2000 data and showed that per beneficiary spending in Santa Fe, New Mexico was about $3,500, compared with $9,200 in Miami. It's a pretty big difference.

It's difficult, however, to interpret these types of numbers because the differences in the health statistics are due to a lot of things besides differences in spending for healthcare services. They have to do with nutrition and sanitation and housing and the prevention and control of infectious disease.

Similarly, even the Medicare spending per beneficiary can be tough to interpret. Part of the difference across states in spending per beneficiary is related to differences in the price of healthcare in different markets. The cost of labor and the cost of doing business in New York is different than in Utah, for example. So, some of the difference is actually what you would expect and what you would want in order to ensure access to care for beneficiaries.

In addition, when you look at the two numbers, if it's $9,200 in Miami and $3,500 in Santa Fe, you don't really know which number is best. Some people get too much care certainly, but some people also get too little care. It's hard to know what to make of such numbers.

As I've said, I'm here to be optimistic and it seems like there is some potential for making improvements. Certainly science and genetics offer opportunities for having more efficient healthcare services. For example, drugs that are targeted to particular individuals, or other efforts to provide better quality care, I guess, conceivably could reduce healthcare cost.

Care management techniques, including case management and disease management, can help to provide more efficient care for people who have chronic illnesses and who tend to be expensive. Most healthcare spending is for people who are sick, which should not be a huge surprise.

Information technology also offers promise both for improving the efficiency of healthcare by helping providers get the right healthcare to the right patient at the right time. IT also may help improve administrative processes, such as handling claims and all of the business of doing healthcare.

Finally, it seems that there is a trend toward increased discussion about health, including diet and exercise and other things that might change our health profiles and potentially change our demand for healthcare and our healthcare spending.

Of course, I don’t want you to think I’m out of touch with reality regarding the potential for reducing health care costs. I recognize that it also helps to be realistic.

In particular, although technology is being touted right now as a solution to many things, is also a problem in the sense of increasing healthcare spending. Technology creates lots of terrific cool new stuff that we all want to buy. Essentially, the more cool stuff there is, the more we are going to spend. Thisdynamic really highlights the tension between goals regarding access and spending.

We are spending a lot of money because there are a lot of things we want to buy. I think demand for health care is going to be one of thefundamental issues that you have to deal with in considering the types of recommendations that you might want to make.

Ultimately, I always like to make the point that spending is the product of the price that we pay for stuff and the amount of stuff that we use. We can try to reduce how much we pay, change prices, or ideally change the cost of providing things so that a lower price is affordable. Or we could change how many services we use.

So what's next? First my disclaimer is that I have a federal perspective. That's what I do so that's what I know best. I want to recognize formally here that there are so many other actors in this market: individuals, employers, states, insurers, and they are all important. There are the taxpayers, who pay for the public benefits and subsidize private insurance.

Individuals and families receive benefits from public programs. They benefit from tax savings, and they might be uninsured because of high cost. Employers need to balance the goals of providing an attractive compensation to their employees and keeping down labor costs.

Insurers need to provide products that people are willing to buy. They need to offer these products at a price that people are willing to pay, while also protecting themselves from financial lost and maybe even make a profit while they're at it. Healthcare providers, of course, depend on a functioning healthcare market and that people have insurance because payments for health care provide their income.

The fact that there are so many people involved makes things difficult. I don't know if different parties always have competing interests, but they do have different priorities, which makes doing anything quite a challenge. Nonetheless, I think there are three things that policy can offer. I lump things into three broad categories to help me think about and kind of simplify things.

Policy can help change healthcare; that is, how we provide healthcare. It also might focus just on the federal spending and how you could change federal programs. Or it could focus more on the private market and improving access to private insurance.

So in changing healthcare the goal is to –and I've already talked about this a little bit– the goal is to improve quality and efficiency and hopefully, if you're lucky, also reduce some cost in the process.

We need to think about what we're buying. As you think about changing healthcare products, you could focus on changing the mix of things that we buy for people who are ill who are expensive. But we should also think about changing the demand for healthcare and possibly also making people more healthy.

We already talked about improving the delivery system, the potential for technology, including information technology and other technologies, to help. I'm happy that people are optimistic about this, as you might imagine. But I also think that there needs to be somewhat of a paradigm shift in providing healthcare. The technology itself is not enough.

People need to think differently about what they are providing. Maybe the physicians need to think about the role of the physician in the process.

The physician certainly controls a lot of spending by sending patients off to get tests, and sending patients off to the hospital. There might be different ways for teams of healthcare providers to think about care, or even to think more broadly about the relationship between the public health system and the medical care system. I think there is potential here certainly because you can see ‑‑ well, at least it appears -- that we are inefficient relative to other countries in providing care. Certainly we could do something better.

The best thing about this approach is that if you are able to provide healthcare differently or change the way you think about providing healthcare, any potential cost savings that you realize would affect both public and private spending. Just focus on the whole system.

There are lots of tools for changing spending in federal programs very, very broadly. You could set a budget and just decide what you are going to spend. You could change eligibility and benefits for different federal programs. You could also change other program features.

In the first category you can set budgets for entire programs or for types of services or for beneficiaries. An examples of a budget for an entire program is the appropriation for veterans' medical care. We decide how much we're going to spend on veterans' medical care and that's how much we spend.

Medicare also has set budgets, sort of, for physician spending by linking the payment update for physician services to total spending for physician services. It is a way of saying that we are going to spend a certain amount on physician services. This approach has proven to be ‑‑

MR. FRANK: (Off mic.)

MS. JENSON: Yes. Well, it's proven to be a little problematic. However, it's interesting. One point about setting budgets is you have to set the budget right. You can set budgets too low and it can cause access problems and it can cause frustration with providers, but you can also set budgets too high and spend more than you need to.

Certainly the physicians have been under pressure in the last few years with payment cuts and threats of more payment cuts, but there were also years when the economy was growing rapidly and that affected their payment update. You might argue that their payment updates were quite generous, much more than they maybe should have been. Physician payment updates exceeded growth in costs by quite a bit. So budgets are tricky -- They can go either way.

You also can change other program features. Medicare uses this a lot. They change their payment methods. They change their update amounts. They move people from traditional Medicare into managed care plans. Certainly in Medicaid that has been even bigger. Lots of people get healthcare differently in Medicaid now. You can tinker with how much you pay for providers and how people get care.

All efforts to do this highlight the tension between competing objectives. On the one hand, you want to ensure access to benefits. On the other hand, providers are concerned about adequate payments. Of course, things you do to improve access or increase payment for providers make it tough to control spending.

Finally, I just wanted to talk for a second more about subsidies for health insurance. I think I went through most of this. We subsidize private health insurance through the exclusion and the deduction and the tax credit. These are things I mentioned already. But we also subsidize ‑‑ the Federal Government also subsidizes health expenses. In addition to the deduction for expenses exceeding 7.5 percent of adjusted gross income, there are several tax‑favored accounts that people can use to reduce their spending on health expenses not covered by insurance. These tax‑favored accounts include health savings accounts, flexible spending accounts, health reimbursement accounts, and a few people, I think, have Archer Medical Savings Accounts.

All of these things have the common feature of reducing the price that people pay for health insurance and healthcare. Or at least they reduce the apparent price. But they probably also, I think, increase spending. Subsidies encourage people to buy more insurance than they would otherwise.

Having more insurance can encourage at least some people to use more services, and the higher demand can drive up prices. Similarly for the expenses, I know that at the end of the year if I have some money left in my flexible spending account, I buy terrific new eyeglasses like I think a lot of people do. That is something that I might not do otherwise were it not for the tax incentives that I'm responding to.

Finally, I want to say that you guys have a real tough job and I want you to keep three things in mind as you take it on. The first thing is that shifting cost is not reducing cost. Particularly in the area of ‑‑ well, in all areas really. If you want to provide tax subsidies to purchase health insurance, that will certainly reduce the price that an individual pays for health insurance or an employer pays for health insurance. But the taxpayer still is effectively paying the rest. You can move around the spending but that's not the same as reducing the spending.

Spending less is one goal, but we might have other priorities. Ultimately, spending less probably means using less, or at least less relative to what we otherwise would have used. It is really a value decision. You have to decide ‑‑ we have to decide as a society which things we want to buy and which things we want to publicly subsidize.

I do think Government has some important roles and the Federal Government in particular has some important roles. I think that it can provide information and help with coordination. Everyone can benefit from information and it doesn't cost much more to disseminate information broadly so that it can be used to help improve the quality of care and help improve the way consumers seek care.

Certainly Government can provide guidelines for information technology so that systems talk to one another. It can facilitate coordination and help the system ‑‑ potentially help the system to work better. The Government also, I think, has an important financing role. The question is really about what the Government shold help finance. Should the Government be paying for care for certain populations or all populations? Should it be paying for certain services and not other services? Which part is the important part given that there are not unlimited resources. It might be helpful just to think about which parts of the problem might be the best parts for the government to help with. That's it.

CHAIR JOHNSON: Well, thank you, Jennifer. It is good to have some optimism. It goes a long way. If I could just start the questions on the CMS presentation, page 9 at the bottom of the page. The amount of drugs for the elderly is relatively low. It is at least 8 percent. I'm wondering to what extent you think that the Medicare Modernization Act will bring about substantive increase in drugs?

Second, to what extent that might offset other costs such as inpatient cost and so forth?

MR. FOSTER: When we estimated the cost of the MMA and the Part D provisions for the drugs, one of the things we paid a lot of attention to was what actuaries call “induced utilization.” In particular, studies consistently show the same result.

The more that an insurance product covers the cost of healthcare, whether it's drugs or anything else, and the less that the individual has to pay in terms of deductibles, co‑insurance, etc., then the more that person tends to use of that service.

An important question is to what extent is this increased use really necessary medical care, or to what extent is it frivolous or unnecessary? That is a much harder question, but the reaction by individuals is very well known and understood. We built into our estimates that for Medicare beneficiaries who had no supplemental insurance and no drug coverage to begin with (and who now would have the Part D benefit), they would be very likely to increase the use of prescription drugs. Similarly, other beneficiaries who would obtain more comprehensive drug coverage under Part D would be expected to increase their utilization. That response adds to the cost, of course. In fact, it was a fairly sizable factor.

As Steve mentioned, the better prices that many beneficiaries can get as a result of the Part D private plans, having pharmacy benefit managers working on their behalf, etc., largely offset the estimated cost of that induced utilization.

Now, your question had a second part that I've forgotten already.

CHAIR JOHNSON: Is there an offset of other types of expenses because drugs help provide therapy that reduced the need for other types of therapies.

MR. FOSTER: In general there certainly is. There are a lot of studies out there that have shown for particular types of illnesses or disease categories that the availability of modern prescription drugs can affect in a favorable way, cost-wise, the treatment of that disease. There are drugs that can keep people out of the hospital.

Now, the key question is to what extent do we think that is going to happen as a result of the new Part D coverage. The short answer is “not enough to measure.” I'll explain why. First of all, roughly three‑fourths of Medicare beneficiaries had some level of drug coverage before Part D anyway.

Sometimes this coverage was pretty good and sometimes not so hot, but most beneficiaries had some level of drug coverage already. So the availability of Part D is typically saving them some money on buying such coverage, but for those beneficiaries it probably wouldn't make a difference in their other health treatments.

The second thing is if you as an individual know that you can stay out of the hospital, you can avoid an in‑patient admission, by getting this drug, even if you don't have very good insurance, or any drug coverage, you have a strong incentive to get that drug somehow, some way. Borrow from your mother‑in‑law or whatever. There is already a strong incentive.

Finally, some studies indicate a favorable impact of having drugs available, while others show an unfavorable impact because, for example, you can get increased drug‑ to‑drug adverse interactions. The studies are sufficiently diverse that it’s hard to reach a consensus on the overall impact.

Our most recent Medicare technical panel, which was an independent group of expert health actuaries and economists ‑‑ in fact, your colleague, Michael Chernew, was on that panel ‑‑ came to the same conclusion. They reviewed all the literature and decided there wasn't enough there to hang an estimate on, so we assumed no net impact on other health costs as a result of the Part D coverage.

CHAIR JOHNSON: Pat and then Grant and then Richard and Mike.

MS. MARYLAND: I just want to see if I understand some of the data. My question is directed to Rick or Stephen. You have a chart that shows there has been an increase. If you look at the time frame of the '80s to 2000 there's been an increase in utilization in terms of intensity of services, both from quality and mix of services from 2.3 percent to 3.2 percent.

This does correlate. This makes sense to me because it correlates with the growth in the aging population. Yet, at the same time, if you look at that same time frame you've got other data that shows that from 1980 to 2003 the proportion of dollars spent for hospitals has gone down by 10 percent, prescription drugs up by 6 percent, and nursing home care dollars remain pretty stable at 7 percent.

This would suggest to me, and maybe I'm wrong, that we pretty much have on the hospital side managed cost pretty reasonably given the fact that utilization has increased and we've got an aging population increase. Am I wrong in terms of that interpretation?

MR. FOSTER: I think I would tend to agree with you for a couple reasons. One is most experts consider that hospitals can actually achieve productivity gains over time. Some other types of service like nursing homes, for example, or home health are so labor-intensive it's not that easy to get a productivity gain, but hospitals seem to have accomplished it. As a result, there has been some improvement in terms of the transaction cost or charges that have to be made reflecting the productivity gains.

Now, Congress introduced in 1983 for Medicare the in‑patient prospective payment system which moved from a cost‑based reimbursement approach to a prospective, bundled payment per admission. Almost every year since then, Congress has reduced the hospital payment update below the normal market basket increase that Steve and his staff calculate. For example, the update might equal the market basket increase minus 1 percent or minus 2 percent in a given year.

That has been a strong incentive for hospitals to improve their productivity. But it has also helped keep down Medicare hospital costs compared to what would have happened otherwise.

In addition, if you think about the role of managed care starting in the early 1990s, a high priority was to keep people out of hospitals however they could. This effort was pretty successful, especially in the “left half” of the country.

As a result of managed care, and partly just because of technology, there has been a shift away from in‑patient, expensive services and to outpatient services or even doctors' offices. So for all these reasons, hospitals, I think you're probably right, are not the big villain here if you want to put it that way.

MS. MARYLAND: I just wanted you to state that for the record.

MR. FOSTER: On the other hand, they are the major source of dollars still. Hospital costs don't always grow as low as we might like.

VICE CHAIR McLAUGHLIN: I just have one clarification question. On that chart, in fact, I was going to ask it and then I'm just going to tag on for Pat, where is OPD? Is this only hospital in‑patient or does it also include hospital outpatient?

MR. FOSTER: It's both. We also include in there the hospital‑based skilled nursing and hospital‑based home health.

CHAIR JOHNSON: Brent, we have a follow‑up.

MS. MARYLAND: My follow‑up question is to Jennifer. It's regarding the international perspective in terms of how many dollars are spent as a proportion of the GDP, has there been any effort to correlate outcomes for other countries? For example, we know that Sweden and Switzerland have some of the best international outcomes in relationship to the dollars spent. Is there any data out there that we can look at?

MS. JENSON: Well, Richard is nodding like he might have more information than I do. I don't have extensive information about that.

MR. FRANK: There was the special issues of health affairs that actually had a series of articles that took quality indicators and spending and sort of put them together for the OECD countries. I think it was in the last six months or so.

MR. O'GRADY: Part of my portfolio is also doing the OECD accounts. OECD now has done comparable spending between different countries pretty much funded by the United States and Japan over the last 10 years or so and now they are moving into trying to do this kind of quality measures across countries.

The one thing that I say, though, that we politely don't push to be put in the OECD reports, there are things like very long waiting times and other things that go on in these countries. When we sit down with my counterparts at the other OECD countries, nobody is getting too cocky about healthcare costs or anything like that including the French and the Germans and what not.

We are trying to share a fair amount of data on what works and what doesn't, disease management, different things like that. It's very much when you sort of sit with those other countries everybody feels they are pretty much in the same boat.

DR. SHIRLEY: When we compare our country with Sweden, if you make adjustments for the demographics and if you take all of the blacks out of our equation, I'm wondering what the differences would look like.

MR. O'GRADY: Yes. And I would say that in terms of Europeans, if you look at their immigration patterns over the last 10 years or so, all of a sudden a fairly large influx of Turkish populations, Moroccan, things like that.

All of a sudden what had worked kind of comfortably for them in the past, they are making a fairly serious transition to an immigrant society which we have had for 100 years or so at least. They are certainly going through their own changes with just the kind of dynamics you're talking about.

VICE CHAIR McLAUGHLIN: Aaron, one more thing, though, too, is that there are a lot of people who look at, as Jen said, the level of wealth and how that can explain variation and medical care expenditures as well as outcomes. But there's a whole school of people who also look at the distribution of wealth.

It's not just looking at racial and ethnic and immigrant differences. It's looking at the distribution of wealth. What they have found consistently is that the more unequal the distribution of wealth, the poorer the health outcomes so it's both of those things going on.

You can't really disentangle one from the other in a really rigorous way but I think conceptually when we do international comparisons and when we now do regional comparisons within this country, we have to keep that in the back of our mind that even looking at the different states in the United States, the distribution of wealth within states is not going to be the same.

DR. JAMES: I think my first was the same thing, just a little bit different. It was just the rate of increase in those other countries because having worked extensively they seem to be facing the same rate of increase that we are so that would be the first point.

Is anybody measuring waste in the healthcare system accurately in this country and those countries? That's another thing that really pops up is when you start to estimate the amounts of waste it is surprisingly high and fairly uniform in those other countries. Is anybody estimating that?

MR. O'GRADY: No. And one man's waste is another man's proper utilization. The one comparison that is made on the international is that what we have here is the skyscraper where the lights are left on all night long. We don't have waiting lists, we don't have this sort of time lag. You have chest pains and you don't have to wait three years for your bypass. At the same time that is a lot of resources being put online all the time to make sure you have that level of care.

DR. JAMES: I think when I say waste I mean a different thing, Mike. That would be part of it but a relatively small part of it. I think it may be much more extensive than just that. A better way of saying it is that in some cases one man's waste is another man's income perhaps.

DR. BAUMEISTER: I do a lot of negative endoscopies.

DR. JAMES: That's not what I mean either.

DR. BAUMEISTER: Think about that.

DR. JAMES: I understand.

DR. BAUMEISTER: I do a lot of screening colonoscopies.

VICE CHAIR McLAUGHLIN: Frank, microphone.

DR. BAUMEISTER: Maybe I don't want anybody to hear this. I mean, that's an issue, you know, that everybody faces.

DR. JAMES: It's a classification problem.

DR. BAUMEISTER: At our last meeting we had at the AHRQ headquarters, next door to our meeting was a meeting touting colonoscopy screening for every person that had a colon. It's a real dilemma for physicians.

DR. JAMES: Again, I think that's not what I'm trying to classify as waste, though. Something that has a clear indication has clear values.

DR. BAUMEISTER: Thank you. "Waste."

DR. JAMES: I realize it's part of the "waste" system. This is one I think we'll cover a little bit in Salt Lake City.

MR. FRANK: Well, I'm going to take this from the high‑minded to the nerdy. I have a question and then sort of a request for advice because we have this report we have to put together to inform the American people so I wanted to get some advice about certain types of data. Let me ask my question first.

There is a lot of mythology about fraud and abuse in the United States. Some people say it's 30 percent of everything we spend, etc. As I understand it, the National Health Accounts data because of the way they ‑‑

CHAIR JOHNSON: Richard, can I ask you, are you saying fraud and abuse or is that in addition to the waste that Brent was talking about?

MR. FRANK: No, I'm talking about fraud and abuse.

CHAIR JOHNSON: Thank you.

MR. FRANK: There are people who have made wild claims about it. As I understand it the National Health Accounts data don't include that because they are based on provider surveys. Is that right?

MR. HEFFLER: Yes. The spending data is from provider surveys. I'm not sort of linking “why.”

MR. FRANK: So, for example, I put in a claim for something that never happened, right? So that's fraud.

MR. HEFFLER: But if you get paid for that.

MR. FRANK: But I wouldn't report that.

MR. HEFFLER: But there's an amount of money that is paid by a private insurer or Medicare, Medicaid. A lot of the data that we use comes from the Commerce Department, the Bureau of Census, so when they go out and do their surveys, they ask, "How much revenue did you collect at the hospital? How much revenue did you collect this year?" They in that form would list all revenue whether it was for efficient care or other things.

MR. FRANK: Okay. So that's my question. I thought they went to the Commerce Department and you built the numbers off of the expenditure responses from the hospital which would not include that, right? I guess the distinction I'm getting to is that if there is sort of a revenue for something that isn't real, it would show up perhaps in the revenue data but not in the expenditure data.

MR. HEFFLER: Right, but we're picking up revenue. That's with the health account. We call it the National Health Expenditures but it's revenue data.

MR. FRANK: Okay. Thanks.

MR. FOSTER: And the other thing on that is certainly for Medicare and Medicaid which feed right into the accounts, those expenditures do include whatever is inappropriately spent because of fraud and abuse.

MR. FRANK: All right. Thanks. The second thing is advice. We need to present the variety of data on spending and coverage, etc. One of the things that I guess I've noticed that I think is well known is that if you sort of build up a picture of the way we spend money, say, from the MEPS you get quite a different answer than if you did it from the Health Accounts and so there is a big difference between the survey data and the health account data.

I think that apparently happens throughout national income and product account systems. The survey data always gets you a different answer. I'm just trying to figure out how would you sort of go about (a) reconciling and getting it so that if we have the Health Accounts data in our report and we have survey data, that somehow people might be able to crosswalk.

MR. FOSTER: When did we do our last reconciliation?

MR. HEFFLER: Can I refer this question to Mike? Actually, we were in a meeting not too long ago and Mike asked a similar question about the MEPS versus the NHEA data. We actually just had a conference about three weeks ago on the National Health Expenditure Accounts looking at sort of the future direction.

One of the projects in that was updating the reconciliation between those two surveys or different kinds of surveys. One is, like you said, a provider survey, establishment survey. We are getting sort of the macro. The other one is more of an individual sort of micro‑level survey.

It was a really interesting dialogue in that conference because I think people generally were recognizing that there should be differences because there are different types of surveys but there was some uncomfortableness with where some of the differences were. I would say both AHRQ and CMS shared some uncomfortableness with that.

That was the preliminary effort and what we are going to be doing over the next six months is finalizing that work. We are hoping toward the end of this calendar year to actually have an update of 2002 data and reconciliation of the exact reasons that those surveys are different. I think if you're interested, I think we can share the paper that was developed, the preliminary paper for that conference with the group.

MR. FRANK: That was going to be my question.

MR. HEFFLER: The estimates are going to change because the numbers are preliminary but the methodology and all that is ‑‑

MR. FRANK: At least so we can document that there are these known differences and things so that people can see that.

MR. HEFFLER: Right. And the folks at AHRQ actually took more of the lead on that effort so I'm sort of speaking for them and saying that we can share that. We put the conference on but they did the paper for us. I don't think there would be a problem with sharing that information.

DR. BAUMEISTER: Richard, what are the surveys you're talking about?

MR. FRANK: One of the ‑‑ you've been seeing data for the last two days on that survey that people have been called Medical Expenditure Panel Survey. It's a civilian population survey where they go into a lot of detail about healthcare use and spending, etc. That allows you ‑‑ much of the data we heard yesterday were profiles based on that.

We just heard a different set of profiles of the U.S. health system. It turns out when you put them together they don't quite look alike. In fact, the survey responses are quite a bit lower, as I understand it, than the Health Accounts data. The idea really is how are we going to ‑‑ we have to worry about how to present all that.

CHAIR JOHNSON: Okay. Mike.

MR. O'GRADY: Thank you, Randy. Couple different things. Rick, one of the things that you sort of picked up that you did in your presentation and also has been picked up as a theme here has to do with utilization. As some of us know, and some don't, Medicare the way they pay is they control the price, they don't control the utilization, and the spending is price times utilization.

Can you talk a little bit because of some of Frank's points and what not where you have seen over time? If my take‑away is wrong on your data, certainly straighten me out right away.

We see things that change over time not tremendously volatility, but then all of a sudden we'll see things like 2002 and whatnot where all of a sudden we'll see utilization seems much more volatile than some of the other aspects of spending. Can you just talk a little bit kind of how we think of utilization and how it fits in terms of the overall growth rate?

MR. FOSTER: Yes. It's sort of a general question and I can give sort of a general answer, I guess. You might, in fact, like to add to it, or Steve or Jennifer.

I'm almost tempted to say there is the good, the bad, and the ugly for a question like that. Let's talk about technology for a minute. Medical technology is a wonderful thing. It provides new techniques and treatments. The medical community can now do heart bypass operations or hip replacements for people in their 90s.

They couldn't do that even 20 years ago without killing the poor folks, so they wouldn't even try it. Then technology comes along and gives us much better healthcare, gives us better lives. It's a wonderful thing--and it's an expensive thing. New medical technology increases costs far more often than it decreases, in part because many of the new techniques are very expensive. Technology generally increases utilization, either because we can now do the same service for people in worse physical condition, or we can do brand new services that we couldn't do before.

In some cases, especially with prescription drugs, an existing drug can now be used to treat something else. What is the classic example? Epoetin, I believe, was used in dialysis for many years and somebody discovered it helps with the red and white blood cell counts for people going through chemotherapy. If you give them enough epoetin, you can give them even more of the chemotherapy drugs without killing them and have a better chance of curing the cancer.

So all that increases utilization. I would put that under the good, I think. With the bad, well, let me pose a question for you, something to think about. I don't know if this is really the bad or not but see what you think.

If you look at growth in Medicare cost per person over time, over about the last 10 years, a surprisingly large chunk of the growth is attributable to spending on people who die in that year.

If you take all the Medicare beneficiaries, the survivors and the decedents in a given year, and look at the rate of increase in Medicare spending for each category, the survivor spending is going up at a relatively moderate rate, I would have to say.

The decedent spending is going up two to three times faster. Why would that be happening? Is that good medical spending? Is that spending that comes under the category of wasted spending in the end? These are useful questions.

The ugly. I'll throw this in. This is not meant to offend anybody in the room but we've seen and we've measured on a statistically valid basis that if Medicare payments to physicians are reduced, utilization tends to go up. Now, maybe that's coincidence in some cases because it happened to be a bad flu season or there has been some other significant increase in health problems. On the other hand, it's not hard to find anecdotal examples where, in the year that the payment rates went down, suddenly a physician practice that used to have a thousand Level 1 visits a year next year has zero and they now have a thousand Level 5 visits or Level 4 visits.

Some utilization growth can result not really from medical need, I think, but rather, say, doctors and other providers not wanting a reduction in their incomes.

I've probably said enough for the moment. If I haven't irritated everybody yet, I could probably finish the job. I don't know if that really addressed your question or not, Mike.

MR. O'GRADY: Yes. Is it all right if I ask another one? I think it causes us all a fair amount of concern when you see the price either level or, in some cases, go down, but then we see the other part of the equation go up and the desired cost savings. I can't think of a time that they actually appeared at least the way we had hoped they would.

MR. FOSTER: We routinely expect an offset in our estimates, typically about 30 percent.

MR. O'GRADY: The other thing just to talk to Steve for a second about some of the things that were going on there. It was a very good presentation. Jen brought up the idea of these other sorts of tax expenditures. The notion that when a worker has his health insurance subsidized, that doesn't appear as income and, therefore, he doesn’t have to pay income tax on that amount of the employer's contribution.

The last figures I saw could be as much as a couple of trillion dollars over the next 10 years that doesn't appear as income. Although I think most economists would say that what we heard there was that this is, in fact, a non-cash form of payment that is going to a worker.

The other part being that the employer gets to deduct that as a business expense. The figures I saw, again, that's maybe not $2 trillion but that's maybe $1 trillion over the next 10 years or so. I just wondered in terms of the ‑‑ you know, Jen kind of nudged me on that one in terms of bringing that up. Really that's a lot of money on the table. When you're sort of doing your Government side versus private versus household, can you track that?

MR. HEFFLER: Much like the numbers that Jennifer presented, we can present these numbers. Whether we're tracking or putting any emphasis on it, the answer there is no, we don't spend much time looking at that side of it when we're doing our spending estimates.

The first piece of what I talked about was the business, households, Government analysis. That is where we've done our work on that one. When we do that analysis, which is not an annual set of numbers like the National Health Expenditure Accounts, the data that Rick presented, but when we do that analysis, we usually have a piece of that that talks about tax expenditures and discusses it in the context of how does the levels of that compare to other pieces of spending and where that is coming from, the $90 billion, I think you had, the $79, so we present the data that way.

That is how we have handled it up to this point. I will honestly say that there are some people that would love for us to build that into our National Health Expenditure Accounts because they want to see that as spending. The analogy we used is that the Bureau of Economic Analysis and the National Income and Product Accounts don't show tax expenditures as part of our gross domestic product.

To be consistent with that in that comparison, we also do not ‑‑ it doesn't mean that it's not there. I think they did do some separate studies and there are other groups that do present that data. That's why it has not been sort of officially incorporated in the NHEA estimates.

MR. O'GRADY: Okay. That makes very good sense. I think just for the members of the group who don't normally think in terms of tax expenditures and what not, it is important as you think about kind of overall design of the system, I think, anyway, to know that we have employee‑based health insurance, which many other countries don't, and it's a vestige of wage and price controls coming out of World War II.

It was the idea that you were looking for workers and an employer couldn't offer more money because the wages that they could offer were linked so they looked for extra fringe benefits to offer. That is when we first see hospital insurance and other forms of insurance.

So there is already this, I would call, fairly, at least, significant subsidization of what is going on for workers and employers in terms of through the tax code. In terms of any time when you do like the international comparisons, no, they don't do it. It's as if everybody had Medicare. They run it a different way.

As the guys pointed out here, we are coming up on kind of a 50/50 split between Government and private sector. At the same time there's an awful lot of this sort of additional kind of credit and deduction going on on the private side. That is an awful lot of money in the system right now.

CHAIR JOHNSON: Okay. Montye and then Therese and then Joe. We are at our end time so we will ask that we consolidate questions but let's get every question and maybe consolidate your answers as much as we can but get them complete.

MS. CONLAN: Okay. This is to Jennifer. First of all, I wanted to say that I'm grateful as a kindred spirit to hear the voice of optimism. But also I think you present to me a refreshing voice of youth and ‑‑

MS. JENSON: I like you.

MS. CONLAN: To me it calls to mind that if we are charged with thinking outside of the box, we need to bring a young, optimistic voice to the process, or hopefully many, to help us with this. Those that are not as heavily invested in the current system maybe can think outside of the box a little more.

One thing that you mentioned that is another thing that I am particularly interested in. It's the first time I heard mention of funds for medical research. You mentioned $29 billion being devoted to NIH. I guess if Pat can ask a loaded question, maybe I can, too.

I'm interested in the prospect, speculation what would happen if we just put a whole lot more money into medical research because personally I know a whole lot of people that are with chronic incurable disease or spinal cord injuries that would love to be off the dole if only there was a cure and that cure would only come through medical research.

MS. JENSON: Well, you might start liking me a little bit less because I think that medical research is a double‑edged sword. On the one hand, we can discover new treatments and we can use what we are learning in the genome project to provide more efficient care and better care. I'm anticipating better care for people with chronic problems. That's potentially really good news.

But there's a couple of things that Rick had mentioned in his good, bad, and ugly presentation, and that is when we create new things, we'll be able to do more. Doing more over time has costs ‑‑ I mean, the trend is pretty clear. All of these new things that we create that make lives better, they tend to cost money.

For individual cases, by providing more efficient care I think you can potentially save costs. But in aggregate I think more stuff is more stuff and that will lead to more spending. It might very well be worthwhile.

That's the whole reason why I include the front part of my talk. Health care might be exactly what we want to spend our national, federal, and personal resources on. It may be worthwhile for us to spend that money but I think it cost more in total.

MS. CONLAN: And I just had one more question.

MS. WRIGHT: I have a follow‑up on that, though, Jennifer. I think I would like to add just to Montye there is strict regulation with research also so that it's not willy nilly and everybody just doing any Frankenstein stuff.

MS. JENSON: Well, I'm talking about ‑‑

MS. WRIGHT: So, I mean, there are costs there, too, and it is strictly regulated so there is cost passed on to the research and development itself through your IRBs and the companies that release it whatever phrase trial we're going into.

MS. JENSON: Are you implying that the spending on research is targeted and that, therefore, it's more likely to ‑‑

MS. WRIGHT: I'm just saying there's a cost there also because it is regulated. We don't have ‑‑ she's asking why don't we spend more for research or have more research out there or more things coming out. It's because of the regulations of what we do with the consents for the patients to notify them of the clinical trials.

MS. CONLAN: Well, I guess I was talking about basic medical research like at NIH in terms of let's just say, surprise, surprise, neurological disease. From my understanding there is relatively little money being expended for neurological disease research.

Yet, you know, we have veterans who are being injured with spinal cord injuries or war wounds who could be repaired theoretically and people with chronic neurological disease who could be cured. Of course, then there is always the issue that some of these diseases may be predominately gender issues.

Are we putting appropriate amount of money towards research for diseases that women encounter and that kind of thing. That is what I was talking about is just at a lower different level of basic research before we get to the clinical trials.

MS. JENSON: I have one comment. I’ve had this thought in the back of my mind, and I don't really know what to make of it, but … As I've watched the NIH spending grow, I have considered going back and taking a look at growth in NIH spending to see if it predicts growth in healthcare spending; that is, to see if there is any relationship. Although I'm excited about all the things we're going to discover, who knows what's going to happen next. It's just a question that is in the back of my mind.

MR. O'GRADY: Real quickly, Randy, we have a natural test for that. The Europeans have no NIH and we saw how low their healthcare spending is. I think we liked the results that we get out of NIH.

You are absolutely right. If you see the doubling of the NIH budget over the last five years, so they have done what you said, and now all of a sudden we're seeing the pressure on the FDA because if you put that much money into investment, you are going to see that ripple effect down the road.

CHAIR JOHNSON: Therese.

MS. HUGHES: A couple of things. I think that I am here today as a result of investment in medical technology and for those of you that are at the speakers, I'm an end‑stage renal disease patient now transplanted. Forty years ago I'd be dead which is significant in several ways. Yes, dialysis is very expensive.

Transplantation initially is expensive and the meds are costly but I am productive. I'm in the work force. I'm off the dole. I'm no longer on Medicare so I think there is that balance that I would support on Montye's comments because certainly there is a personal side to it that numbers and statistics don't bring to the table. Having said that, I also wanted to talk about regulation.

Recently dialysis units have been required to have ‑ ‑ I'm drawing a blank. Frank, help me ‑‑ the heart machines, the defibrillators in every dialysis center. The reason that the defibrillators were put in there is because there have been patients who have been on dialysis obviously and who have had heart failure in the dialysis unit.

Now, the cost of putting that in the dialysis unit is, in my opinion, a shifting of cost but maybe I don't understand entirely the shift of cost. What has happened is that studies are being done and the first rough outcome shows that the patients once you go into heart failure on dialysis, you have approximately 40 days to live at best if you infuse the patient with all of this medical care.

The number of patients since this regulation has been implemented has shown that at best ‑‑ well, it's actually 38 days but I thought 40 sounded ‑‑ rounding it up. So here are costs into the system that clearly is an additional cost that is a regulation that has come down and is working at an end where, with all due respect, 38 days in the condition after you've needed a defibrillator on dialysis and having been there I can say this, the quality of life is not good.

I just wanted to say that some of this regulation that comes down that requires across the country to the thousands of dialysis centers to get these things in there. I think it's honestly a waste of money and I speak from experience. Not from the defibrillation part but from the dialysis part.

DR. BAUMEISTER: I think that this just brings up the complexities of these issues because there are people who would like defibrillators on every airplane and every taxi cab. There are defibrillators now that tell you exactly what to do that speak to you and can be operated by a six‑year‑old. They have been shown to be very effective. In the dialysis center you have people who have electrolyte disturbances that are more prone to arrhythmias. It's just a very complicated thing. I don't have an answer.

MS. WRIGHT: You know, the argument has always been, too, and, Frank, I don't know if you agree or not, but the definition of quality of life is individual.

MS. HUGHES: And I understand that.

DR. JAMES: You know, I have to say I spent most of my career in cancer, surgical oncology, and the idea that you could predict what you'll want when you are looking death in the eye just ain't true. What you discover is people have one very different view from being relatively healthy. You have people in that last end phase that come up with a very different approach. Quality of life is an important concept that we use fairly well but it's a very, very difficult concept to understand fully.

MS. BAZOS: I just want to add my 2 cents to this discussion in response to your comment, Brent. I think a missing piece in the conversation we talked about physicians and how when reimbursement for Medicare goes down utilization might go up. But what we didn't talk about was the consumer and I'm just wondering how we think the consumer if he plays a role at all in driving these costs because, actually, I'm pretty healthy.

I don't shop for healthcare but when I go to the physician. I'm assuming he's going to provide something that I need, not just something that I want. I do think at some level consumers still believe the healthcare system is giving them what they need.

When we go to our community meetings and we present to folks who are using the medical system to stay healthy, I'm wondering who really is driving these costs. Is it the consumer? Is it the system and how do we talk to patients about these issues that are pretty much invisible to them at some level?

DR. JAMES: You know, it's clearly both. For example, a lot of Jack Wenberg's recent work demonstrates this one thing called supplier‑induced demand. It turns out that specialists ‑‑ he was explicitly looking at internal medicine specialists ‑‑ seemed to have a particular income expectation.

He is fairly convinced and he has shown they adjust their level of practice to maintain that level of income for the same communities. One of the best predictors ‑‑ he has identified a set of conditions he calls supplier ‑‑ they are affected by supplier‑induced demand and he has shown the main predictor of the cost for those conditions in the community is number of specialists.

Elliott Fisher follows up on that and demonstrates that in a highly concentrated specialist community, Florida, for example, that you can account for the differences in Medicare health expenditures primarily by the number of specialists and that an increasing number of specialists means worse health outcomes for the population.

It seems to fragment the care is what the current belief is so you get uncoordinated care and you actually get worse health outcomes in Florida for two and a half times more outlays, you see. These ideas of supplier‑induced demand, preference‑sensitive demand, I think, are really important ideas.

CHAIR JOHNSON: And that's a subject that we will be talking about in some depth later on in some of our other hearings. Last question to Joe if we could.

MR. HANSEN: I kind of feel like I'm overloaded here with all the data that you presented, and presented very well. Just a couple of technical questions. On slide No. 6, I believe ‑‑ maybe that wasn't the right one. Anyway, it showed a reduction in out‑of‑pocket expenses from 1980 to 2003.

I guess my problem with that is I got the feeling, and maybe it's the size of the circle, the total expenditures are considerably larger. The people I see that's one of the major problems is that those numbers are increasing.

More importantly, I guess, to me is on slide 14 where you talk about health spending as a percent of revenue of the income, did you ‑‑ I know there are numbers behind all these numbers ‑‑ did you do any slicing of that by incomes, by people $30,000 and below or anything like that?

MR. HEFFLER: Not as part of this analysis. We didn't do any distributional type work. But I think in other projects we worked on we've done that and at the lower incomes a higher proportion of the income is spent on health than at higher incomes.

MR. O'GRADY: Once you hit Medicaid, then you see ‑‑ you know, it's not a high percentage so it tends to be ‑‑ you know, as you get up to that, that level, it gets worse and worse.

MR. HEFFLER: Okay.

MS. JENSON: If you give me a second, I actually have some consumer expenditure data for a different project that I happen to be working on. All right. This data shows that consumer expenditures on healthcare are about 6 percent of consumer spending. That's consistent with the number you (Mr. Foster) had, about 6 percent of income.

I have the spending data broken down; about half of the 6 percent it is for health insurance and half of it is for out‑of‑pocket expenses. For some other numbers that are comparable: consumer spending on housing is about 33 percent, spending on transportation is 19 percent, and consumer spending on food is 13 percent.

I found all these comparisons interesting. In addition, spending on apparel and services is about 4 percent of consumer spending; spending on entertainment is about 5 percent. When you look at the numbers, they raise the question of whether we are spending too much or too little as consumers. I was also curious about the breakdown by income so I looked at consumer spending by income quintiles. How about spending on health insurance first?

So, spending on health insurance … remember the share of spending was 2.9 percent for the overall population. In the lowest income quintile, the share of spending was 3.7 percent, so a fair bit higher than 2.9 percent. In the second income quintile it was higher still, 4.1 percent of consumer spending. The lower spending in the first income quintile (compared with the second) is explained by the fact that many of the people in the lowest quintile are on Medicaid. In the highest income quintile, about 2 percent of consumer spending goes for health insurance. In the highest quintile people spend about 2 percent of their income on health insurance.

For the category that's other healthcare, which I'm assuming is mostly out of pocket, people in the lowest income quintile devote about 3.7 percent of their spending. Those in the second quintile spend 3.9 percent, so a little bit more out-of-pocket on non‑health insurance health spending. People in the third quintile spend 3.5 percent, and people in the fourth quintile spend 2.6 percent. For people in he highest quintile, it appears that I have a typo, so I don't know the fraction, but I'm sure it's less.

CHAIR JOHNSON: Jennifer, is that in this report?

MS. JENSON: No, it's actually another report I'm working on that's coming out next week. The report is on private health insurance and how it affects consumers, employers, and other people.

CHAIR JOHNSON: Would you be able to provide that to us?

MS. JENSON: Yes, sure.

CHAIR JOHNSON: Thank you.

MR. HANSEN: I just had maybe ‑‑ I was going to ask a question about whether the NIH should be part of the healthcare cost but there's been discussion about that and I guess it should. I'm done with that. My last comment goes to what you said, Rick, about the utilization with physicians. I know when dealing with numbers that's the way it looks but I deal with enough physicians and sometimes you try to get organized around that.

I think there's a lot more that goes into that. When you look at physicians it doesn't go to their salary. Their expenses are really tremendous and I think we need to hear from physicians somewhere along the line. If they are making some of that up through utilization, I can't quarrel with that but the way it was just left on the table I think it left the wrong impression there.

CHAIR JOHNSON: Building on Joe's comment, the amount of administrative cost that you indicated that is part of the cost scenario was relatively small, less than 10 percent. I'm wondering if you have any way of getting at what is the cost of a hospital's administration to answer telephone calls and pay claims? And the same thing in the doctor's office. Do we have any estimates of that type of administration as well?

MR. FOSTER: Not within the health accounts. The question comes up from time to time. There have been special studies out there and the answer is “big”. In our case we are looking at the payments which include implicitly that cost of administration.

CHAIR JOHNSON: Is there anybody that you would know who would be able to help provide that information to us?

MR. FOSTER: Steve or Mike, can you think of any of these? We can dig around a little bit and see what there is.

MR. HEFFLER: I don't know that we have specific spending estimates like actual dollars. Rick mentioned that we do ‑‑ well, we develop in our office what are called market baskets but they are essentially ‑‑ Mike mentioned how in the Medicare system we sort of regulate the price and we let the utilization flow.

The price piece is the market basket piece and part of that formula and updating from year to year. When we prepare those indexes what we do is we look at the underlying expenses for each provider type so hospital, physician, home health, skilled nursing facility. We look at the distribution of their costs which is sort of what you're getting at there, sort of underlying to the physician how much of their cost is going to sort of patient care versus overhead cost.

We probably could do some cuts of the data and look at it where we could see some of these administrative costs. On the physician side it's about half or a little more than half that goes directly to what we call physician's income so their benefits weigh in. The other half goes for practice expenses.

Of that half about a quarter of that half is going to support staff and so forth. The rest is then split between things like pharmaceuticals, rents, other types of cost. That gives you kind of a rough indication. On the hospital side you look at the employees of the hospital and you look at where they are before you allocate overhead.

After you allocate the overhead salaries and sort of lump all salaries together, you end up with a little under half of all hospital costs going to salaries, so a somewhat smaller proportion of that is for direct patient care. The rest of it is things like food and pharmaceuticals and capital expenses, utilities and so forth. It depends on what you define as an administrative cost but that will give you sort of a rough idea of magnitude underlying the cost at least for hospitals and physicians.

CHAIR JOHNSON: Thank you. And maybe our executive director can work together with you to identify that a little bit more closely.

MR. O'GRADY: Randy, one quick comment on that. You see a wide variation as well and I don't know how comfortable most of ‑‑ about the rigor of that data but it is, I think, the point you brought up yesterday about, in effect, a return on investment because it is administrative cost that is really helping you better coordinate care, do better quality, and there is other stuff that is just not adding too much.

Certainly we have seen some insurers certainly who wouldn't want to go into hospitals or doc's offices where they have very low administrative cost and that is because they are not doing very much other than ‑‑ I mean, you know, a bill paying machine doesn't cost you a lot if you are just paying without thinking about it and what you're thinking about and managing, at least in the positive sense.

CHAIR JOHNSON: Agreed. Thank you. Well, we would like to have about two more hours with you but your schedule doesn't allow that, I suspect, and we need to move on. We really thank each of you for your time this morning. We thank Stephanie and Jamie and Senator Wyden for asking the questions that we need to hear from you so we'll look forward to maybe some further dialogue with you in writing or whatever in the future. Thank you very much.

MS. MARYLAND: My question would be is it possible for this group to help us put together that matrix of looking at the cost, the breakdown of ‑‑ you know, you've got a lot of that data in parts and pieces to put together that grid that we've talked about at the beginning of the last time we met. Would it be possible for CRS to provide that support to us?

CHAIR JOHNSON: There are some rules and regulations regarding CRS but if Senator Wyden might call on you for some help, that would be helpful to us maybe.

We'll take 15 minutes and then reconvene.

(Whereupon, at 10:38 a.m. off the record until 11:00 a.m.)

CHAIR JOHNSON: Well, good morning, Jim, Jack, and Jenny. We're glad to have you with us this morning. We've given you an opportunity and we're assuming that you know a little bit about the Citizen's Healthcare Working Group. We have here credentials or summary of some of your experience, at least. We've had an opportunity to look at that.

The topics that you're going to lead us through are very, very important to the discussion of the working group. What we would like to suggest is that each of you take maybe 15 minutes or so to talk about your subject. We are going to try to hold our questions until all of you are done but sometimes we get antsy about that and we may interrupt you for a clarifying question or something like that.

We'll have lots of questions on these very important subjects so we want to give you the full amount of time that was allotted to this section. We'll plan to go for an hour and a half if that is okay with you. Again, thank you very much and welcome. We'll turn our session over to you.

MR. HOADLEY: Thank you. I'm going to talk to you a little bit today about some of the ways that the Medicare program tries to control cost. As you can see, some of the things that Medicare does are fairly unique to the Medicare environment.

I know you saw a lot of numbers and graphs this morning. I'm going to give you only one slide with a graph on it and everything else is going to be words. I thought I would just put a little bit of context around how has Medicare spending growth per enrollee sort of compares with the private health insurance.

The answer is that it has been similar over time. Sometimes Medicare does a better job and sometimes the private health system does a better job. If you look at this graph, the green bars are Medicare's annual rate of growth from year to year per enrollee. We're not being affected by the number of enrollees.

A couple of things that are interesting in the '93 to '97 period Medicare spending was growing a lot faster than private spending and that was one of the reasons why there was a pretty big piece of legislation in 1997 called the Balanced Budget Act that tried to attack a whole bunch of these things. You see the results in the '97 to '99 period.

Medicare spending almost ground to a halt in terms of spending growth. Private sector growth was a little bit higher than it had been in the previous period but at this point Medicare was doing better. Then Medicare spending speeds up again from '99 to '03 as the Balanced Budget Act changes wore off and there were some what they call give‑backs to increase or give back some of the cuts that had been initially made.

Of course, general health spending was also going up faster. The point here is simply that Medicare and private both go up but they are not always rising in the same years, not always in the same patterns. One can do fairly extended analyses to try to say whether overall does Medicare do a better job than the private. I'm not here to try to make that point in one direction or the other.

I also thought it was important to say a few things in the beginning about what makes Medicare different than private spending in terms of some of the forces that affect Medicare's decisions. Part of it is the legal context. What is the circumstance that creates Medicare?

Actually, the very first sentence of the Medicare title of the Social Security Act provides a prohibition on Medicare practicing medicine. It says, "Nothing in this title shall be construed to authorize any federal officer or employee to exercise any supervision or control over the practice of medicine or the manner in which medical services are provided."

Now, when you hear that sentence it's hard to believe that Medicare does some of the things that it does but it is a constraint. It does limit the ability of Medicare to do some of the things that one might otherwise choose to do to try to influence the way Medicare controls its cost or tries to control its cost.

Also, in general, there are requirements for an open public process. There are certain things when you want to do a competitive bidding process that are going to look very different if you have to do it with certain public rules about how services are acquired than the way a private sector organization might be able to do it.

The administrative resources available. One of the interesting things in Medicare is that the administrative cost of doing something in Medicare comes out of the appropriated side of the budget, whereas the actual expenditures are on the entitlement of the trust fund side of the budget. Sometimes it's hard in budget terms to trade off those costs.

If you have to spend some money administratively in order to save money and services, that can be difficult to do because you've got to draw on an administrative budget that is limited through the appropriations process.

The size of the program. Medicare is huge and so for Medicare to do things, it leaves big footprints.

The potential, as I say in the last bullet, for market‑wide effective policy changes. If Medicare makes a change, it echoes throughout the healthcare system. The Medicare program has to think about what it does. If it changes prices for services, that is going to affect the incomes of providers. If it changes the way a particular kind of service is delivered, that's probably going to change how that service is delivered in the entire health system. Whereas a single employer can make adjustments, it may have only a small effect.

Of course, if you are General Motors in Detroit you may have a very large effect. In most cases the individual employer doesn't have those kinds of effects and Medicare really does.

Plus, of course, it's a political environment. All these decisions are political so if you make something that somebody doesn't like, they've got the political process to work back through.

You have the "not in my back yard" phenomenon. We saw that a couple of years ago when Medicare tried to do a competitive bidding system for managed care plans. Every community it was taken to basically said, "No, not here. Do it somewhere else." And it never happened.

Beneficiary freedom of choice. Beneficiaries are allowed in Medicare in general to choose their providers and that affects how you do things. It's a unique patient population. I think that is quite clear. The Medicare beneficiaries are just different and they have different expectations in the health system.

And the benefit structure is different. This whole division of Medicare into Parts A, B, and now C and D means that for some of the things you may do on the physician's side, you don't necessarily think about how they affect the hospital side because they come out of different pots of money. While in the end you can think about those things, it complicates the way we do it.

So what are some of the ways that Medicare tries to control costs? It's actually hard to divide these into boxes because most of the things you do end up overlapping in all these categories. One of the main things Medicare does is to constrain payment rates. That has been I would say over time probably the most important lever. I'll talk about each of these in more detail.

Medicare can also use beneficiary cost sharing and increase the rates that beneficiaries pay as a share of the cost. You can try to take steps to modify the volume of services either through financial incentives or information based incentives. You can try to influence the bundles of services, the way the cost of the services are structured, or limit the providers that participate.

In terms of looking at payment, Medicare payment in general operates on a number of principles. You are trying to make sure that the payment system is structured in a way that ensures beneficiary access to high quality care to give providers an incentive to supply their care efficiently.

You prefer to pay similarly for services regardless of the setting, although Medicare hasn't always done very well on that. Then doing all that in a way that tried to manage the spending of the program.

So how does Medicare's payment system work since this is one of the main things we're using? For the first 20 years of the program it was mostly a cost‑based reimbursement. Whatever the provider said the service cost, that's what Medicare paid. And there were some adjustments from that but that was the foundation of the system.

We've phased that system out now in most sectors. It still exists in a few corners of Medicare but quantitatively it's not very important.

What we have started going to in the early '80s was prospective payment. In 1983 with the hospital DRG system was when that really started. The principle was you are going to pay for a bundle of services. In the hospital you are going to pay for all the services related to the diagnosis with which the patient came into the hospital.

And you are going to try to give the hospital, the provider, and we now do this in skilled nursing facilities and home health agencies and various other kinds of providers, you are trying to give the provider organization the incentive to contain its cost within a bundled payment. Again, this has become a major way Medicare tried to look at cost containment in these institutional provider settings.

We now have prospective payment systems in most of the institutional provider settings. Some of them are still new and some of the wrinkles are being ironed out but they are in there. For physicians, it's a little bit different because in 1989 legislation was passed to institute its own version of prospective payment. It was a Medicare fee schedule, known as the Resource‑Based Relative Value Scale.

Again, instead of just paying the physician for their normal charge perhaps adjusted a bit through some “usual and customary” provisions, we set up a fee schedule to try to figure out what is the resource base that goes into a particular service and try to establish a set of fees that is, again, set in a way that will not let the cost go in an unlimited way.

How do you constrain those payment rates over time? In the institutional setting, all of the payment rates, all those prospective payment rates, are updated automatically through a market basket system. So if costs are determined to go up by a certain amount, you automatically get an increase to reflect that inflationary direction that the costs are going. That's locked in the statute.

When Congress wants to go in and say, "Spending is rising too quickly in the hospital sector," they actually have to go in and legislate an adjustment so you hear about market basket ‑1 percent or market basket ‑.5 percent. That becomes a very crude political process because it gets completely wrapped up in all the budget and congressional politics. But that becomes the way you bring spending down from that market basket increase.

It turns out that many of those market basket formulas would appear to be relatively generous. If you are a hospital provider or home health provider, you may not think that. But Congress has certainly made that judgment over the years so in many years we bring those increases down below the market basket either because they are simply deemed to be too generous, or because the Federal Government is trying to figure out how to save money and, well, you've got to go where the money is.

On the physician's side we had a different method. This was something originally called the Volume Performance Standard and then it morphed into the sustainable growth rate or the SGR system. It's a complex system as all of these are and I could spend the whole day if you really wanted all the details going through it, and I'm sure you don't.

But the notion of the system there was it was self‑restraining, self‑adjusting. If volume went up, the price per service went down so you tried to keep the system in check that way. It worked for a while and people thought the SGR system did a pretty decent job of keeping physician cost under control.

Right now it appears it's doing too good a job. It's actually leading to a situation where, if the formula is allowed to go into effect, actual cuts occur -- 4 percent, 4.5 percent cuts in physician payments. Again, Congress has to jump in because that seems to be too much. It seems to not be appropriate. Congress has to go in and actually raise the cost in order to get the prices to a better level.

I won't say a lot about the managed care side but, again, we have a formula in law to pay for the managed care plans under Medicare which was originally designed to pay at 95 percent of the average fee for service cost so built in a savings. The idea was you moved to managed care and you built in a savings.

Well, those formulas have gotten very complicated. Congress tried to do a lot of things with those formulas, partly to try to get plans to participate on the assumption over time they will achieve the desired goal of bringing cost down. Right now we're in a system that, according to MedPAC, pays plans at about 107 percent of fee for service. Again, there's a lot of complexity under that and I won't try to go into all that.

But we also have in the Medicare Modernization Act a provision that would move Medicare Advantage, the Medicare managed care system, to a bidding system in the next couple years so things will change there. It will remain to be seen how that will affect costs, but the idea of that proposal is to do something more to constrain prices.

On Part D the new prescription drug benefit, payment is done through a bidding system. The Federal Government has a fixed share of the total cost. Here we are doing something a little bit different. We are really relying on market mechanisms and hoping that the plans will be efficient in how they deliver services and people picking among competing plans will tend to hold those costs down but there's not a Government formula that brings those costs down.

The cost sharing side. Beneficiaries do pay a share of the health costs. You see some of the numbers here. We do this for a couple of reasons. It tends to reduce the Government cost. It also is designed to encourage cost conscious purchasing and reduce the use of discretionary services.

The existence of supplemental coverage through Medigap is viewed by many as diluting those incentives. This means that most of that cost sharing is covered by their supplemental so these incentives are not as visible to the average consumer.

But there is also the concern that cost sharing can impede the use of appropriate service. If you have to pay too much, sometimes you will give up not a discretionary service that you don't really need but maybe one that you really ought to have.

What are some of the things that Medicare has done or might think about doing to modify the volume of services? Well, we've got a bunch of steps that you can do to try to identify efficient providers and promote efficient care patterns. You can try to use provider profiling.

Look at the data that says which providers are being more efficient in how they provide care. What do you do? Well, partly you hope that if providers know they are less efficient they will realize that is something they should try to improve.

We have seen a lot of steps in the last two years to put these kinds of measures up on the web, publicize them in the newspaper, and tell people which are the more efficient hospitals, which are the more efficient nursing homes, and let the consumer go out and make some judgments and try to pick the more efficient institutions.

We haven't really done this yet but there has been a lot of talk and a lot of interest in creating payment incentives under a pay for performance approach. That, again, is something that is seen as providing some ways to create incentives to get people to behave and practice medicine more efficiently.

And we also have things like disease management and care coordination that are aimed at trying to get the care right, get the volume of care appropriately delivered. Now, the evidence from these various demonstrations, and we've done a lot of them, is that they work pretty well.

They improve the quality of services but it's not clear that they save money so it's harder to say this has been effective as cost management although we're, I think, pretty happy that they do a pretty good job of making care better for individuals.

We can also try to do certain things to make sure we pay only for appropriate care. These are some of the more technical and more direct things like prior authorization for certain services that we think might be overused or coding edits to try to look for things that shouldn't be done and try to eliminate some of those payments.

More things that I label under this other category (but it's really hard to sort of say what box they fit in) include trying to bundle payments in different ways. Put things together like we did on the DRG system to try to figure out more ways to pay for the bundle and let the providers try to find the efficient way to provide the services.

Looking at how we pay based on site. This has been an area that causes problems. If we do pay differently based on site, that sometimes means people go to the place that pays the most or the best or the providers recommend doing services in the setting that provides the best reimbursement. But you can also try to use that as an incentive to get people to the most appropriate site.

Use of hospitalists and intensivists, physicians that specialize in hospital care or the intensive care unit to try to change the way we define a service to try to cause it to be more efficient. These are more in the realm of ideas than things that have been extensively tried in Medicare.

Medicare has experimented a little bit with competitive bidding to establish price. Or, as you'll see on the next slide, to select providers among a larger set of providers. The best example of this is durable medical equipment where a demonstration was initiated in 1999 in Florida and again in 2001 in Texas and now is being rolled out under the MMA the main way we are going to pay for durable medical equipment. But the notion was that savings were achieved in the range of 20 percent.

Fees came down for the services. We did a better job at purchasing. The evidence from what's been studied suggest that there may be some adverse impact on quality or access but maybe not too much. Maybe within the realms of what is acceptable, although that is obviously something different people can look at.

Again, part of competitive bidding is to select providers to try to find the ones who are willing to provide the services a little bit less expensively. I throw this into this box as well. Medicare can also consider looking at the kinds of selective contracting, although this is difficult going back to those constraints.

When you try to do selective contracting, the one who doesn't get picked always can go to Congress and this makes it very hard. There are a lot of people who say selective contracting is not something Medicare is ever going to be able to do successfully as a public program with political constraints on it.

The Centers of Excellence. This is a similar area. This really wasn't done as much to limit providers but to target providers. We had a demonstration back in the early '90s on coronary artery bypass grafts that was viewed as being reasonably successful.

It created savings perhaps in the range of 10 percent but there were some issues that arose in terms of both administrative issues and practical issues. At the same time, there was a sense that there was no difference in outcomes between these settings and others. Yet, this demonstration ended and while there have been a number of attempts to try to either resurrect it or move it into law to do something on a more regular basis, those have not happened.

Then, finally, what are some of the things that might be in Medicare's future in the area of cost containment? Well, one thing you can be sure of is continued adjustment of the payment systems. The payment really has been the work horse of cost containment for Medicare. So we keep tinkering with the prospective payment systems.

We adjust those payment updates on the physician side. One of these days we'll probably replace the sustainable growth rate with some other systems because that one seems not to be working but it's hard to do because it costs money and gets caught up in the budget politics.

We are also continuing in Medicare to look for innovative approaches to purchasing services in the fee-for-service environment. Some more experiments along the line of the competitive bidding for DME or perhaps selective contracting or Centers of Excellence and other kinds of things, more experimentation with care management and disease management.

Other things that people are interested in and in some cases have specific legislative mandates include increased enrollment in managed care. We don't have anything that mandates increased enrollment. We have payment incentives to try new rules to try to encourage more managed care plans to come in that hopefully will be attractive to beneficiaries.

Starting in 2010 the Medicare Modernization Act includes what is called the comparative cost adjustment program which is really the premium support system that had a lot of talk during the debate over it leading up to the passage of the MMA.

It's a system that will be designed to try to come up with a different kind of system to put managed care plans more directly in competition with fee for service in certain markets. It is believed by its backers to be a system that will ultimately bring overall costs down. It isn't starting yet so we don't know how that will play out.

There is also a provision in the MMA that requires if Medicare spending exceeds 45 percent of general revenue, that there will be a review triggered. If we get to those higher levels of spending, it is supposed to force the Congress and the President to actually look at the agenda and figure out other ways to make adjustments. It doesn't actually force action. It just forces the discussion of alternatives. But, again, it's something that is out there as one of the ideas. With that I'll stop.

CHAIR JOHNSON: Thank you.

MR. HOADLEY: I'll switch over here.

MR. VERDIER: I think I'll try and put mine up on the projector as well as giving you the opportunity of looking at them. Does anybody know how to find mine? Oh, okay. You can either look at it this way or in the materials in your handouts.

Just to put this into the context of what Jack was just saying, one important ‑‑ one of the most important things to keep in mind about Medicaid is that it is a state‑based program. States pay on average about 43 percent of the cost. Wealthy ones pay 50 percent of the cost. Less wealthy 22 or 23 percent.

It's governed by a lot of federal rules and regulations but the tools that the Federal Government has to control costs in Medicaid are relatively limited. States themselves have plenty of incentives to control cost in Medicaid because unlike the Federal Government they have to balance their budgets every year. Most of them use a lot of the tools that Jack was going through that Medicare uses. A lot of the hospital and physician reimbursement systems in Medicaid, for example, are copied from Medicare.

So with that as context, what I'm going to be talking about are four things. First, national Medicaid spending trends over the last few years and then projected forward; the distribution of Medicaid spending by enrollment group, children, disabled, elderly, etc; then I'm going to go through very quickly the standard menu of options for cost containment in Medicaid.

A lot of this can be pretty arcane so I'm not going to get into the arcana. I'm going to give you kind of a high‑level overview and then to the extent we want to go into particular ones, we can certainly do that in Q&A. Then just say a word at the end about the potential to control cost in Medicaid by improving quality of care.

The national Medicaid spending trends in the last couple years have actually dropped a bit in 2003 and 2004 to around 7 or 8 percent and that follows two years of ten or 12 percent growth. Those 10 or 12 percent growth years came at the same time that state revenues were actually dropping year over year.

During that period of time in 2003 and 2004 there are very, very aggressive cost containment activities in virtually every state in Medicaid and that's why you see those declines in those years. But if you look forward over the next decade or so, you are looking at average annual growth in Medicaid of about 8.5 percent a year. Both CMS and CBO put it at right about that.

If you look at it historically and if you look at the structure of the state tax systems, you are not likely to see annual revenue growth of much more than half of that in states. There's a structural deficit built into state Medicaid programs going forward if these projections are anywhere near correct.

CHAIR JOHNSON: A clarifying question.

MR. VERDIER: Yes.

CHAIR JOHNSON: Are those aggregate cost or per‑person cost?

MR. VERDIER: Aggregate cost. I'll get to the per‑person cost in a second. This is probably the single most important graph to keep in mind about Medicaid programs. The largest number of enrollees, about three‑quarters, are nondisabled adults and children.

It's mostly mothers and their children. They don't cost very much so they don't account for a large share of the cost as you can see there. The most costly beneficiaries are the elderly and the under 65 blind and disabled and chronically ill population. Most of the cost for the elderly in Medicaid, probably around two‑thirds, are for nursing home care.

About 10 percent is for prescription drugs and that is all moving to Medicare starting in 2006. For the blind and disabled the big areas of service utilization there are hospitals, again prescription drugs, again moving to Medicare for about half of the disabled on Medicaid in 2006. Other big categories are residential services for people with mental retardation and developmental disabilities.

Then there are smaller groups of services after that for the blind and disabled category.

Looking at the cost containment options, provider reimbursement is always on the list and most states have done things in this area over the last couple of years. Nursing facilities are the biggest source of expenditures and cost in Medicaid. Managed care organizations are next. Then hospitals and then drugs and then all others are about a third. There's a nice little pie chart on the Medicaid at a Glance spreadsheet on the second page there where you can see all the other kinds of services and their contribution to Medicaid costs.

One of the important things you learn very quickly as a Medicaid director, which I was in Indiana in the mid‑1990s, is that the share of their reimbursement that provider groups get from Medicaid is a very, very good predictor of how well organized they are going to be to protect their reimbursement and their share of Medicaid expenditures from attempts to scale it back. Some hospitals are very, very heavily dependent on Medicaid. Children's hospitals in particular, specialty children's hospitals will get well over half of their revenues from Medicaid.

A lot of inner city hospitals will was well. Intermediate care facilities for the mentally retarded get almost all their revenues from Medicaid so they are very, very sensitive about efforts to change the reimbursement or the services that they get.

So that is one set of options. The other set of options are cutting eligibility or limiting eligibility in some way. States have not done a lot of this over the last couple years. It's kind of one of the last things you would like to do but some states have had to do it.

But what you get into is what I mentioned a couple of slides ago. Most of the costs are for the elderly and disabled and those are people by and large who have very, very heavy care needs and it's very difficult both politically and if you are trying to run a decent caring kind of Medicaid program, it's just very, very hard to go there for major eligibility cost. These are folks who basically have no alternative. They can't get private insurance. Medicaid is the high‑ risk pool for these folks so they don't have a lot of options.

Looking at benefits, scaling back benefits is another area that states have looked at. Again, most of the costly benefits are concentrated on the most needy beneficiaries and they are very well defended by well organized advocacy and provider groups so it's not an easy thing to do for a host of reasons if you are trying to control costs in Medicaid.

Co‑payments and other forms of beneficiary cost sharing are pretty limited under current regulations. The maximum co‑payment for a service with a few small exceptions for emergency rooms is $3 or 5 percent of the cost of services which for some prescription drugs is significantly more than $3 and for emergency rooms even more.

These have not been changed since 1982 and there is consideration now in Congress and by the National Governors Association of changing those. They could be changed without a statutory change. The statute talks in vague terms about nominal co‑payments and what was nominal in '82 is different now from what it was then.

Most Medicaid directors would agree that higher co‑payments are appropriate for some services. There's really not a lot of opportunity to change behavior and utilization of services with co‑payments and co‑insurance except for prescription drugs and emergency room use and your savings are not from the co‑payments you collect. It's from the change in behavior that the co‑payments lead either beneficiaries or physicians or both to do.

Some of the other cost containment areas, prescription drugs, there's a whole array of things that you can do with co‑payments, changes to pharmacy reimbursement. Some of those being considered at the national level now are so‑called preferred drug lists or formularies that give people incentives to use less costly drugs. Higher rebates from manufacturers are another possibility.

Manufacturers are required to give Medicaid rebates now of about 20 percent is what it actually works out to. Those could be higher and some states have negotiated for higher rebates. That is something that is also being considered nationally now.

Disease management which Jack mentioned, either stand‑alone or, in my view, more promising in a managed care context where it's not just a particular disease that is being focused on but it's the person with the array of diseases and co‑morbidities and the array of services that they use and can be better coordinated and managed in those kinds of contexts.

Managed care generally, lots of states are now looking at expanding managed care to the disabled and the chronically ill population. Most of the moms and kids in most states are covered by managed care. Unlike the private sector I think most Medicaid directors would argue, and I think most people would now agree, that managed care is actually better than fee for service for most people in the Medicaid system for reasons that I can go into if you would like.

There is a new opportunity in the Medicare bill for so‑called special needs plans which can specialize in serving people who are dually eligible for both Medicare and Medicaid. They can specialize in serving people in nursing facilities. They can specialize in serving other people with chronic illnesses and disabilities in Medicare. I think that is a very promising area that, again, I can talk about at greater length if you would like.

In the area of long‑term care reform, Medicaid is a very ‑‑ has historically had a very strong emphasis on providing long‑term care in nursing facilities. I can go into the historic reasons for that but people really do prefer if they can do it to live in their own homes or in a community‑based setting and there is a lot of interest in expanding the array of services that are available in a home and community‑based setting.

It's better care but I would argue that it's not a major opportunity to save money because even though the cost per person is less in a home and community‑based services setting, you'll find many, many more people wanting to use those services than wanting to go into nursing homes. You get smaller per‑person cost but more people.

Couple other options. I blandly labeled this creative financing and there have been other more provocative terms that have been applied to that. The various vehicles are disproportionate share financing for hospitals, intergovernmental transfers, provider taxes, and other more arcane forms of Medicaid maximization.

CMS has been cracking down on these over the last few years both by regulation and by statute and they have a group of auditors ‑‑ again, I'm choosing neutral terms ‑‑ that are out visiting all of the states trying to identify instances in which states may have been a little too creative.

It's made states pretty gun shy on a lot of these things. There’s not a lot of sign that there's going to be great potential there for savings from the state point of view, which would turn out to be higher cost from the federal point of view.

Fraud and abuse, everybody's holy grail. The crackdowns on fraud and abuse are pretty resource intensive because the people who are perpetrating the fraud are trying to hide it and so trying to find out what they're doing is not that easy but major areas of potential are in pharmacy.

So‑called Medicaid estate planning in which people will with the help of lawyers put a portion of their assets in forms that Medicaid cannot get at them for purposes of making them repay nursing home costs, for example, after death.

Billing for services not provided is another common area of fraud and abuse in Medicaid. Again, hard to define because Medicaid beneficiaries don't get a little explanation of benefits saying, "Here are the benefits you have received from Medicaid. If you didn't receive these benefits, please let us know." That just doesn't go on in Medicaid so detecting some of that is not as easy as one would like.

Just to summarize, the cost pressures in Medicaid are likely to continue for the foreseeable future and a large part of those costs really reflect underlying healthcare costs. Medicaid is not immune to the advances in technology and other kinds of things that are driving healthcare costs.

They are still going to be paying for a large share of prescription drugs. Not as large as before, although the clawback will have some effect on that. Anybody that wants to know about the clawback I can rant and rave about that for awhile.

But, as I indicated before, Medicaid does function as the nation's high‑risk pool and there are in terms of demographic trends the baby boom is going to result in larger utilization of long‑term care and Medicaid pays for about 60 percent of long‑term care and that shows no signs of going away.

As I indicated earlier, there are significant opportunities for improved care, especially through various forms of managed care and disease management and care coordination because it's a population that really doesn't have the resources to coordinate their own care in the way that other people may be able to do.

And there are opportunities for improved care through, again, I would argue, better managed care, better disease management, care coordination kinds of programs, but these don't save much money in the short‑term because, first of all, you've got to make investments up front to achieve savings down the road in hospital utilization and prescription drug utilization.

Also a lot of the people that are coming into these programs are people who have accumulated health care needs that have been unattended to for a long period of time. When they first come in just dealing with all of those accumulated needs can be fairly costly as well.

There are opportunities there for improved care but not necessarily major savings in overall Medicaid expenditures. Thank you.

CHAIR JOHNSON: Thank you very much. Excellent presentation. Jenny.

MS. KENNEY: Well, turning now to the State Children's Health Insurance Program which you'll see is much younger and smaller compared to Medicare and Medicaid, and it has a very different history with respect to cost containment.

As Bill Scanlon described for you the other day, SCHIP, which is how the State Children's Health Insurance Program is commonly known, is a new public health insurance program that has a far narrower purpose than either Medicare or Medicaid.

It was designed to address coverage gaps for low income children whose family incomes were too high to qualify for the Medicaid program, which Jim has just described, but too low to afford or have access to privately provided employer sponsored coverage.

It was enacted in 1997 with strong bipartisan support. It was funded as a ten‑year block grant which differentiates it from both Medicare and Medicaid. It's not an entitlement. Like Medicaid it also has a federal matching structure so there are both state and federal dollars that support SCHIP.

Because there was a strong interest in getting states to step up to the plate and expand coverage to low income children, the matching structure is more favorable to states than under Medicaid as states have federal matching rates that are higher in SCHIP than Medicaid. A hallmark of SCHIP was the latitude that states had over their program design in terms of the thresholds they chose and the type of programs that they supported.

Despite the fact that coverage is optional under SCHIP, all states actually have expanded coverage under SCHIP. In fact, at this point in time 39 states have eligibility thresholds for public coverage that are at 200 percent of the federal poverty level or above, which is almost double what it was before SCHIP was enacted.

Program structure under SCHIP varies across states. This flexibility was something that was exploited across the country as states chose state programs that were tailored to their individual markets and political realities.

In fact, over two-thirds of the states chose a non‑Medicaid program as part or all of their expansion and coverage under SCHIP.

And while SCHIP, with its focus on providing insurance coverage to low-income children, has a much narrower purpose than Medicare or Medicaid,--nine states are actually using SCHIP funds to cover adults. This has been somewhat controversial but, through waivers, states have been covering parents, pregnant women, and other adults. The most recent figures that have been published indicate that as many as 250,000 adults are being covered with SCHIP funds.

As I indicated, states had latitude over their program designs. While there is some variability in the benefits packages that they have chosen and in the cost‑sharing requirements, overall the benefit packages for children are fairly broad more closely resembling what you see in Medicaid than what you see in employer‑ sponsored coverage. And the out‑of‑pocket cost‑sharing requirements are relatively low for most groups covered under SCHIP.

Another differentiating feature of the program was that when it was started it was met with enthusiasm at both the federal and state levels. As I said, all states embraced the program and implemented expansions.

But along with the coverage expansions, came unprecedented levels of outreach to try to publicize the existence of public programs for children and enrollment simplification efforts aimed at making the processes easier for families to navigate There is evidence that the eligibility expansions, combined with the outreach and enrollment simplification efforts, reduced uninsured rates for children and by increasing coverage through both Medicaid and SCHIP. However, some of the cost pressures that Medicaid programs have been experiencing over the last several years derive from the fact that more children are enrolled in Medicaid as a consequence of these efforts.

Over time, SCHIP has become an important piece of our coverage patchwork for children, especially for low-income children.

At some point during the year in 2004, over 6 million children were enrolled in SCHIP. While SCHIP covers many fewer children than covered under the Medicaid program, SCHIP now provides coverage to a large share of children in its target group, many of whom do not have access to affordable employer‑sponsored coverage.

A key dimension of SCHIP is that it's layered on top of Medicaid coverage for children. Moreover, states that wanted to expand coverage under SCHIP, were prohibited from altering their Medicaid coverage for children. States had to build SCHIP on top of Medicaid.

Since SCHIP is not an entitlement, states have tremendous ability to affect the scope and scale of their SCHIP programs. Despite the optional nature of SCHIP coverage, we haven't seen tremendous cutbacks in SCHIP despite the fact that states don't have to go through a lot of hoops to cut their programs in most instances.

A major factor driving the lack of wholesale cutbacks is that states pay only between 15 and 35 cents on each dollar that are expended under SCHIP. For example, the very poorest states like Mississippi have to put very little of their own state dollars into the equation and if they cut their program, it's not saving the state that much.

Another factor is that the capped block grant structure, as with any block grant when there is a fixed amount of money on the table from the Federal Government, there's always a concern that those federal dollars won't be enough to go around. But to date, that has not been a concern, because SCHIP was funded in such a way that, up to this point in time, states have had adequate federal resources to cover their program.

This figure shows the federal funding structure and spending historically under SCHIP. The red bars show spending and you can see a progression over the years, with very low spending in fiscal '98, the initial year, and much higher spending in recent years, with FY 2004 SCHIP federal spending reaching $4.6 billion. The large federal allotments in the funding structure for the early years of this program represented full funding right out of the blocks, and a strong legislative commitment to show the importance of this program.

They put $4.2 billion on the table right in the first year even though everyone knew it was going to take a while for states to design their programs, to get approval from their state legislative bodies, and to negotiate with what was then HCFA and now is CMS. States needed time to get SCHIP plans accepted, let alone to implement their plans, especially given that so many states chose separate programs and had to build in many instances new administrative structures to support those programs.

As a consequence, in the early years of the program, federal funding was plentiful. It was on the table. In fact, many states couldn't even use all the federal dollars that were available to them. But you can see in the last couple years that spending is starting to outstrip the federal allotment.

The federal allotment in 2005 will be $4.2 billion, higher than the FY 2004 $3.2 billion allotment. But it looks like SCHIP is now at a point where on a yearly basis states are spending more than the Federal Government is providing. The fact that many states spent so little of their federal funds in the early years means that they actually have reserves on which they are drawing now, but that's not going to be the case in a couple years time.

For these reasons, cost containment pressures in SCHIP were rare in the early years of the program. SCHIP was launched at a time of unprecedented economic expansion. State budget surpluses abounded across the country and there were large federal dollar allotments available to states.

2002 really marked a turning point, with economic downturn and the budget deficits that many states were experiencing. That is when you saw states beginning to consider SCHIP cutbacks even with the high federal matching rates and the federal dollars on the table.

The first thing that happened was that states pulled back in outreach, not only because the outreach efforts cost money in and of themselves, but also, I believe, because they were effective at bringing in kids. They wanted a way to reduce enrollment and generate greater savings, to turn the spigot off a little bit.

Since that time cost containment pressures have increased and states have looked to other areas to reduce outlays. Because of the nature of the program, states aren't focused as much on reducing cost per enrollee which sets SCHIP apart from Medicare.

Although there have been some efforts to limit cost per enrollee, which I'll describe in a minute, when states have looked to save money, the real focus has been on limiting enrollment. Some states with separate programs have capped program enrollment, limited new enrollment, or reduced eligibility thresholds, but that has been rare. A larger number of states have increased premiums that families pay to enroll in a program. In addition, some programs have instituted procedural barriers such as waiting periods before you can enroll in coverage, especially if the child has had employer‑sponsored coverage, so they have increased the length of that time from three to six months or longer. These procedural barriers are a really unsexy area, but one that actually makes a difference in Medicaid and SCHIP.

In the last 10 years, we have learned that there are things that states can do that really lower the time costs and hassle associated with applying for public coverage. These procedures - the administrative systems that support enrollment and re‑enrollment processes - really seem to make a difference in terms of promoting enrollment and, on the back end, keeping kids in coverage. In response to budget issues, some states rolled back some of the simplifications that had been introduced to the enrollment and reenrollment processes during the last few years.

As I said, there has been less focus on limiting the cost per enrollee, but a couple of states have cut SCHIP benefits, increased cost sharing on services in terms of co‑payments, or reduced provider payments. Since SCHIP isn't a dominant market force in any state, reducing provider reimbursement levels hasn't been a tremendous area of activity.

As I said, cutting back on outreach has been the most commonly used tool for restraining SCHIP costs and there are very few states that are still actively engaged in outreach at this point in time. Premium increases were the second most commonly used strategy. A number of states have imposed enrollment caps to hold down spending. It is a really direct and blunt instrument but capping enrollment assures savings.

There are interesting questions whether limits imposed on programs affect the mix of children enrolled. The kids who stay on the program are likely to be higher cost per child, so there could be adverse retention. The fact is that it could raise the cost per child going forward if the more costly kids stay in the program.

The net effect of outreach cuts and related strategies may have slowed SCHIP enrollment growth. While over 6 million children were enrolled at some point during 2004, we actually saw a decline in enrollment in late 2003 and early 2004. This was driven by just a handful of states, especially Texas, which has one of the largest programs in the country. Texas undertook a number of changes to their administrative processes that seem to have dramatically reduced enrollment.

Let me just describe what they did. They lowered their eligibility thresholds by counting income in a different way. They also decreased the period of continuous eligibility. In many programs when a child enters SCHIP they have coverage for a ful year and they don't need to have contact with the state over that 12‑month period to retain coverage. If the state reduces that coverage period to six months, at six months there is a reassessment of whether the child is still eligible, which may require more frequent action on the part of the parent to keep the child enrolled, a disconnect which leads to a greater risk of disenrollment.

Finally, Texas imposed a 90‑day waiting period so each child applying for coverage would have to wait 90 days before enrolling. In the nine months following this set of policy changes, enrollment in the Texas SCHIP program dropped by about 150,000 children, which is about a 30 percent drop.

That is not definitive – one can’t conclude that the policy changes were responsible for the enrollment drop, as we don't have a counter factual for what would have happened in that time period without the changes, but it does suggest that these kind of administrative changes can really affect enrollment.

As I said, premium increases were a very common strategy among the states that were trying to hold down spending. For example, Wisconsin went to the highest cost‑sharing amount permissible under the SCHIP statute -- they raised premiums from 3 percent of family income to 5 percent of family income for children whose family incomes are between 150 percent and 200 percent of the federal poverty level.

Following that change, enrollment dropped by about 2,500 kids in the couple of months following the premium hike, which represented about 13 percent of the enrolled kids in that category. Again, that's not definitive, but it is suggestive of the kind of cost savings that states could experience when they make these types of changes.

I would like to close by just sharing a couple of thoughts with you about the future. The first question, I think, is why haven't we seen deeper cuts in SCHIP at this point? You saw SCHIP spending continue to increase during the recessionary period that we have experienced the last couple of years.

In fact, while many states made changes in response to budget pressure, SCHIP programs have largely survived this period more or less intact. The first factor that may have protected SCHIP from deeper cuts is that SCHIP programs enjoy popular support both at the federal level and at the state level. When you hold your hearings around the country, I think you are going to observe the ownership that is felt at the state level over this program. I think that is one reason that SCHIP programs have been spared.

Also there is widespread support among the general public large for covering children, so it's hard to cut back on coverage for them.

The second factor, though, is that federal spending is capped, so there haven't been pressures to reduce SCHIP spending at the federal level. Since federal SCHIP spending is capped, it's a known budget item. In fact, in the past year, unspent federal funds were actually returned to the Treasury. That didn't sit well with some states and there are concerns that there may not be adequate federal funding in a couple of years. But that is suggestive of the fact that the way SCHIP was funded in the early years of the program led the federal funding piece not to be a limiting factor up to this point.

But the other thing to really keep in mind under the current system is that reductions in SCHIP enrollment don't generate large scale savings to the state. Since the states are the prime decision makers with respect to their SCHIP programs, that's what you have to think about in terms of cost pressures and incentives. In particular, the large federal match is a big factor.

Some of these kids, if it weren't for the SCHIP program, would actually become eligible for Medicaid through the medically needy spend‑down provisions. These kids are among the highest cost cases in SCHIP. Some economists at AHRQ have estimated that there is a real offset both in state and federal dollars from cutting back on SCHIP coverage. If you cut SCHIP, state dollars are still going to cover many of the costs of treating these very high‑cost children.

So I think these factors explain why we haven't seen a more serious effort to hold down costs in SCHIP programs. But as I indicated before, the federal funding and spending picture is changing. A number of states are projected to face federal funding shortfalls in the coming years and are going to be dependent on other states not spending all their money and transferring those unspent funds to them.

However, many states that hadn't been able to spend their SCHIP money have figured out ways to do it either by covering parents or by expanding coverage to children. Therefore, the pool of unspent resources is declining.

Finally, there is the whole issue of what tools the states have at their disposal to cut costs. They have these different tools, but they really don't have good information on the cost and benefits of alternative cost containment measures. We just don't have the evidence base for states to understand the merits of different approaches. Going forward, I think that's a big frustration for states facing tough choices.

In terms of what the future holds, how Medicaid is changed, to the extent that it is changed, would have profound implications on SCHIP because the two programs are so intertwined. And SCHIP was authorized as a 10‑year program. It is coming up for reauthorization in two years. There has been some talk of reauthorizing SCHIP at an earlier point, even this year, and in the context of Medicaid reform. All those things, I think, would make the historic experience of SCHIP programs not necessarily indicative of what is to come. Thank you.

CHAIR JOHNSON: Thank you very much, Jenny.

We'll open our time now for questions. Jim, if I might ask a question to you. You talked about some ‑‑ I have a whole bunch of questions actually but just one at this time.

You talked about a cost management program. I'm not sure what your words were but I'll just call it the give‑down programs. In other words, if my mother has some assets and she needs some medical care or let's say assisted living care, we would spend‑down her assets and find a way to give them away. What is the potential savings from that and what would be the approach that some have contemplated?

MR. VERDIER: It's very, very hard to get any kind of decent data on the extent of the so‑called Medicaid estate planning that is now going on. There are certainly an awful lot of lawyers who seem to be making a good living doing this. They have conferences at quite cushy places but that's about the only real evidence that I've seen and there have been attempts to research that.

How you would scale it back and what kinds of savings you would get from doing that, I haven't seen any good data. The standard ways of scaling it back are simply tightening the current limitations which basically say that you have to dispose of your assets within ‑‑ if you dispose of your assets within three years or five years from the time you go into a nursing home, then some or all of those assets are subject to recovery after the person dies.

If you dispose of them before that, then Medicaid doesn't look at them so that is one way of doing Medicaid estate planning. Another is taking advantage of the exemptions that are in there in the statute for particular types of assets like homes, for example, are exempt. Cars are exempt so you can have lots of very expensive homes and very expensive cars and they are all exempt.

Those are the kind of standard plain vanilla ways of doing it. There are lots of more complicated ways of doing it. Because they are complicated, because they are being done by fairly sophisticated people, it's pretty hard to get a handle on the extent of it that is actually going on.

A lot of it may be happening for people who are just never going to get on to Medicaid but they are just doing this out of an excess of caution and maybe their children who might get their assets are urging them to do it but they are not actually going to really get on Medicaid.

If I had to say whether there is a significant amount of savings likely to be obtainable in that area, I would say probably no but it's worth tightening up anyway because it gives a bad impression of the Medicaid program. It is one of those things that is just hard to defend in a public forum in some context if you've got some fairly egregious examples.

CHAIR JOHNSON: Based on your follow‑up question, based on your experience, to what extent is a political will to do something like that an issue?

MR. VERDIER: It's a major issue. Whenever you propose to tighten up on those kinds of areas, just as there are examples of egregious abuses, there are examples of people who would be in very dire straights if some of these rules were tightened up.

It's the examples of people who would be very, very severely disadvantaged by these kinds of moves that you hear most about in these kinds of debates. It's very, very hard for legislators or governors or Medicaid directors to respond to some of these kinds of very compelling stories.

CHAIR JOHNSON: But we have other folks who will have questions but it just seems to me as a common observer that is a potential wealth of opportunity if we had the political will.

Were you going to comment on that, Mike?

MR. O'GRADY: It's just because my office has done a number of studies in this area. There are a number of things to keep in mind. One, we see this vary a lot from state to state. New York is always sort of the poster child of where this is viewed as being a big problem. We're not totally sure what that is.

It may have to do with a fair number of seniors downstate New York who do not own, they rent, and, therefore, the notion of moving assets is a little bit more flexible. But there is another theory out there, hypothesis that says New York state also has fairly heavy taxation that is going on as well as what the feds do and, therefore, again, you are in a position where you are giving people strong motivation to take some of these steps.

It is quite true that as states have seen some of this squeeze we are hearing more and more about ‑‑ there are some states who are not as aggressive as the law allowed in terms of going back and actually when the children are selling this home after mom has died in the nursing home going back and saying, "No, no, no. Excuse me." The people in the state of Virginia take $80,000 and then you guys can split the rest.

That's a nicety they can't afford anymore so they are really getting more aggressive in terms of they are going to do the estate recovery. They are going to go in and make sure that the taxpayers in the state of (fill in the blank) are repaid before the children get anything.

MS. STEHR: My comment is for Jim. I have a 22‑year‑old son who is a Medicaid recipient. He has cerebral palsy. He's in a wheelchair. He's one of the really big expensive high‑need people. I have an appreciation and a very deep appreciation for the move for home and community‑based services.

John has been on a Medicaid waiver since 1993 or 1994. We just used minimum services when he was young and I've seen how helpful that has been to families and to keep our families intact so I have a deep appreciation of that. As he's gotten older we could place him in an ICFMR or a group home but we're not going to do it. It's not what he wants.

I see like the cost savings. I really see the cost savings, home and community‑based versus institutional. I also see just how by ‑‑ we now have kind of had a move toward contract providers where family members are now allowed to be hired to be their care giver.

I'm seeing where just as I'm his provider, and it saves $10,000 a year as opposed to using an agency, even though the agency would be taking the money and paying a direct care worker probably less than what I'm making an hour. I really appreciate the flexibility within the Medicaid program and I wanted to comment on that.

Then also on fraud and abuse. It would be very helpful as consumers to know exactly what we are being billed for. It would be so helpful because I'm real cautious about that, particularly like on prescription drugs. If it's advertised on TV, don't put my kid on it. Don't put me on it and don't put my kid on it.

I have argued with doctors, you know. They have written out a prescription for a drug and I'm going, "No. This other drug is cheaper. We would prefer to have this one." I think really educate the consumers on the actual cost. Even on durable medical equipment.

John has a power wheelchair but he also uses a manual chair as a backup chair. His manual chair is almost 10 years old. It's a piece of junk. It's falling apart. It needs to be replaced. It needs to be grown out and we can't get a second chair under Medicaid so we are looking to buy a cheaper chair just slightly bigger so the seating system from the power chair will fit into the manual chair.

I got on E‑bay because I knew Medicaid wasn't going to pay for one and I couldn't believe how you could buy wheelchairs, and fairly nice wheelchairs, new ones, for $200 on E‑bay so there is like this tremendous markup. I think we need to be aware of the huge differences between what is being marketed and how you can obtain it cheaper, too.

There would have to be that flexibility that if we do find it cheaper. Or even on lifts where you could go buy a used lift on E‑bay. I hate to say it but I actually found one for 5 bucks but I couldn't go get it. It was the really expensive lift. Couldn't go get it. It was in Kansas City and I would have had to go to Kansas City to get it. There is just a tremendous markup and I think we definitely need to address flexibility in everything.

MR. VERDIER: There are some initiatives in that area that I can probably say something about as well. One is called Cash and Counseling. Independence Plus, I think, is what the administration calls their program.

The theory behind these is that you provide people like yourself or your child with a fixed amount of money to provide the array of services that they need to continue living in the community and give them also the resources in terms of financial and other counseling to the extent they need it, helping with paying for care givers, helping with finding better deals on wheelchairs.

There is this flexibility but it's flexibility within a relatively fixed dollar amount per person per year. The early indications of the results of that, and some of my colleagues at Mathematica are doing evaluations of these programs, are pretty positive.

Again, they don't save huge amounts of money. They do save some money, especially in the second, third years of the program but the beneficiary satisfaction is just overwhelming like 95 percent approval of these kinds of things. They certainly are promising when the people who are on the waiver have the kind of capability that you obviously had to do the management and the coordination of those kinds of services.

MS. STEHR: And I think that's great but it's like an optional service, too, because not everybody is going to fit into that. We need a wide range of services but I am aware of that.

MR. O'GRADY: My office does the Cash and Counseling stuff and a lot of the evaluation of it. Again, it falls into that sort of category that Rick Foster was talking about, things that may cost you more money but are worth it. Your return on investment is good.

One of the things that is a tough nut here is we see an awful lot of people in exactly that kind of circumstances who on paper have wonderful benefits. The Medicaid program will cover the services but the services aren't available. They are way out in the country. On paper the entitlement looks tremendous.

If you go to the flexibility in your first ‑‑ if your first line of thinking is why would we spend ‑‑ you know, they are not going to like it and why would we spend $60,000 a year on a nursing home when this person can stay in their home and have that kind of services. We can do it cheaper, better care, they're happier, we're happier and everybody wins.

The reality is that it does if you had somebody who was going to go into a nursing home you are going to save money but there's an awful lot of people that, as I say, the benefits are there but there's not a nursing home bed for those people in some states. There's not the room for them.

There's not the direct care workers so it does end up probably netting out totally. It may end up costing you more money if you’re a state and the feds. But you are also talking about you had whatever, a certain number of people who were eligible for your care and you never were able to serve them.

If this allows you the flexibility to actually serve people in need who are eligible in your state, that's the kind of extra spending I can certainly live with but we can see the pressure that both the states and the feds are under for overall cost. It is a good return on investment even if it does in some cases cost more where we weren't providing the benefits at all.

DR. SHIRLEY: Question about the distribution of the Medicaid dollars. The FQHCs are frequently referenced as major safety net providers. When we look at the charts of expenditures it would be interesting to know where or how those expenditures are posted. They are not fee for service and they are not physician ‑‑ which category are they in? If they are grouped with some others, is it possible to sort it out?

MR. VERDIER: Yes. If you look at that graph on the Medicaid at a Glance thing, the pie chart, they would be classified under outpatient/clinic so that lumps them in with a whole lot of other clinic‑like entities but they are considered to be in the fee‑for‑service program as opposed to managed care.

It is possible, I think, to find out how much Medicaid spends on FQHCs certainly nationally and probably even state by state. I'm not exactly sure where the data are but I wouldn't think it would be that hard to get.

DR. SHIRLEY: I know it will be coming up as we move around the country and it would be very helpful if we had a fix on that.

MR. VERDIER: It's certainly less than ‑‑ it's probably 2 percent or something like that at most, I would guess.

DR. SHIRLEY: It's very small.

MR. HOADLEY: It's certainly true that the National Association of Community Health Centers tries to put together a lot of data along those lines. You also can start to get into the issues of the federally qualified health centers and then a lot of the other kinds of clinics that do similar things but for one reason or another are not in the classification of FQHCs so those are interesting things.

Of course, they have other sources of revenue separate from Medicaid that help them treat some of the patients who are not Medicaid eligible but I think there are data available on most of those breakdowns.

DR. SHIRLEY: Could we put a request in?

MS. CONLAN: I have one question for Jack and one for Jim. I don't know which to go first. Jim. Well, maybe both people can answer this one. Is there a way ‑‑ I live in a county where there are a lot of senior citizens and a lot of providers and practitioners that accept Medicare, but there are very few who accept Medicaid so that presents a problem for the dual beneficiaries as well as those who are just Medicaid eligible.

Is there any kind of way to say if a provider accepts public funding for Medicare that they need to accept Medicaid funding at a certain level? Maybe not the same level but a percentage.

MR. VERDIER: You could pass a statute at either the federal or the state level to say that if you could get political support for it but I doubt very much that you could. If Medicaid paid for physician services at the same level that Medicare did, then obviously some of that disparity in access between Medicare and Medicaid would be diminished. Most states don't do that. They pay some percentage of what Medicare pays, 80, 90 some percent.

It varies enormously from state to state because that is one of the many areas in which states have considerable flexibility so there may be some states where the Medicaid payment is very close to the Medicare payment or even close to what private insurance would pay but other states where there is a very great disparity.

MS. CONLAN: Well, that triggers another question. A physician once told me that the trouble in my county is not so much the compensation but a liability issue. He said that particularly specialists don't want to sign a form saying that they assume all liability for that patient. Is that a factor or is that true?

MR. VERDIER: That is somewhat of a new one on me. I've heard variants of it that physicians think that Medicaid beneficiaries are more likely to file malpractice suits. Most of the studies that have been done show that is not the case but it hasn't dispelled that belief.

There are situations in which some Medicaid pregnancies and births are extremely costly because the mother may not have had access to appropriate prenatal care. She may have been abusing drugs or smoking or other kinds of things that would have increased the risks of a problem pregnancy.

It is probably the case that very, very expensive problem pregnancies and high‑cost births are more likely on Medicaid and so a physician that specialized in that area may be reluctant to participate in Medicaid for that reason.

MS. CONLAN: And then my question to Jack. You mentioned hospitalists. How does that make the system more efficient? Just educate me because I don't know.

MR. HOADLEY: It's probably one of these areas where it may not be clear that the direct additional use of the hospitalists necessarily leads to saving money. It's one of these things that people study and try to figure out. The concept is if there's a physician that is paying closer attention to the patient's needs while they are in the hospital that it may make it easier to get that person out of the hospital in a shorter time.

It may make it possible for the hospital stay to be handled more efficiently. Of course, with the prospective payment system that doesn't always translate directly into savings for the program because we bundle the fixed payment around that hospital stay.

It's probably like the situation with disease management and the care coordination, that specialist can pay more attention to the patient and perhaps that patient will have a better experience in the hospital that should in the long range not only benefit quality of care and quality of life, but should also reduce some of the possibilities of complications and rehospitalizations. Typically when these things are studied we don't see much of the dollar savings evidence on the other side of that.

MS. CONLAN: Well, would it improve the quality of care for the patient if that doctor was not as familiar with the case? Say it's a complicated case for a senior citizen.

MR. HOADLEY: It can be a tradeoff. But the concept is typically when you are going into a hospital for care you may not have a doctor familiar with your case unless you are going in for repeated treatment of some long‑term existing ailment where you have a specialist. If a person shows up with a heart attack, they may not have a cardiologist so their primary care physician will stop by and check and be involved but the primary care physician may find it difficult to be at the hospital on a regular continual basis.

A person who is based there at the hospital can monitor the case more consistently. I haven't studied this myself, however, so I don't know a lot of the instances and how this is used.

If you've got somebody who is making another in a series of visits for some kind of serious chronic condition that keeps causing hospitalization needs, then I suspect it is more often the case that the specialist who has been following that patient's care is going to continue to be the lead practitioner throughout that hospital stay as well.

Perhaps the hospitalist will check in and help coordinate between visits from that specialist. It's going to depend a lot on the individual circumstances clearly. If you've got situations where the specialists are in the building right across the street, this is probably less of an issue than if it's a situation where the primary care doctor is 20 miles away.

You have a lot of different circumstances but it's one of these things that people have been trying and thinking of that can, at the very least, improve quality, they hope, and potentially have some spill‑over into savings and efficiently as well.

CHAIR JOHNSON: Thank you.

VICE CHAIR McLAUGHLIN: I'm going to break with the trend and I have a question for Jenny. You gave the figure of 6.1 million kids participating at some point in time and I assume at some point in time is in part because kids entering in over the year.

But I wanted to know about exiting. Do you have some idea of sort of average length of stay? Do people exit? Do kids exit off of SCHIP? Is there some average length of stay? When they exit, are they going into Medicaid or are they going into ESI?

MS. KENNEY: There is certainly a lot of evidence that the 6.1 million overstates the number of kids at any one particular day. That is probably closer to 4 million. Many kids are entering the program and staying for short periods of time and leaving over the course of the year.

In terms of average length of stay on the program, I have seen estimates of 9 months to well over a year. It seems to depend on the state. In terms of the evidence on where kids go when they disenroll, it looks like they are going to all sort of three different types of coverage possibilities.

A small chunk, maybe 15 percent, are going to employer‑sponsored coverage. What's interesting about that is that appears to be about as many who came from who came from employer‑sponsored coverage on the front end, or 20 percent. A much bigger group is going to Medicaid. But a nontrivial group appears to be uninsured.

Some of that is because the kids are aging off the program. They are 19 and they are no longer eligible for public coverage. Some of it is that their family circumstances have changed and put them outside of the income eligibility for coverage. But there is also concern and evidence that the parents don't realize that the kids are no longer enrolled - they think the children are still enrolled.

It appears some of these kids cycle back so they have a gap and then they re-enroll. We’re not sure why - whether it's premium nonpayment or just a period when the child was healthy and they weren't using services. We are not really sure what is going on there. Clearly, SCHIP is filling in gaps between Medicaid and employer sponsored coverage.

There's just no question that this is happening on both the front and the back end. I think the work that is being done with longitudinal databases with the SIPP over a four‑year time period will give much more information than we have right now to tell us what those coverage profiles look like and how they are changing over time.

VICE CHAIR McLAUGHLIN: The follow‑up was do we know how the length of stay and the entrance and exit differs from Medicaid for kids?

MS. KENNEY: That is knowable and I think there's work that's been done on two different evaluations that might speak to that. I haven't seen a side‑by‑side in the same states but I could definitely get back to you. I know there is some research going on in four states that looks at Medicaid and SCHIP together and I believe a CMS evaluation is also looking at that. I just don't think they have come out with definitive findings on that yet. I think it really depends on the group of children in Medicaid that you're looking at. For example, the SSI group is completely different from the poverty related group.

VICE CHAIR McLAUGHLIN: No, I absolutely agree. Then the last piece of this is the sort of the average cost. We didn't get average cost per kid from Medicaid or from SCHIP presentations today. With the Medicaid it's really tough because you give us the percentages of the aggregate cost going to the blind and disabled and the elderly and the kids.

But it's really hard for me. Being the numbers junkie that I am I'm looking at the 52 million and looking at the $252 billion and then looking at those same kind of figures for Jenny. I'm trying to figure out are they comparable in terms of how much is spent by states on average ‑‑ there's going to be variance across states ‑‑ per kid or is one program more expensive than the other?

MS. KENNEY: Again, I think we could put some of the pieces together on this issue, but it hasn’t been done to date. It's a difficult issue because the case mix is so different in Medicaid and SCHIP. We don't have data systems that really allow us to answer this question cleanly. We have spending data and we have enrollment data but we don't really have the two connected. Some researchers have used the MEPS to try to estimate cost, but we know that the MEPS understates cost levels compared to the administrative data. Maybe Mike has some insights on that.

MR. O'GRADY: No, not particularly. I just think you're just between a rock and a hard place on the data without getting to something. Certainly the actuaries will always give you an estimate and they can try and control for as much as they can but we should probably have asked Rick Foster that when he was here.

They are spending a lot more time. He's hired more actuaries to work on Medicaid and SCHIP than he ever has had in the past. One question. Just a real quick follow‑up. When you brought up on your slide about Wisconsin, one of the things I wondered when Wisconsin was going to 5 percent for a premium on the highest, I guess if I was sitting in Madison trying to decide how to control this, I would be a little tempted once we're up to 200 percent of poverty so we are talking maybe 35 grand a year or so if I'm not trying to gently nudge people back towards employer.

I mean, we heard other things when we were hearing about the employer, the premium increases and what not. If you have a family who is sitting there who has some options.

Therefore, in all fairness, I was wondering about employers in what has been described as a heavy union city like Milwaukee, thinking you know, we're trying to nudge people onto the state roles. I think I would be a little tempted to push back a little bit and see if those folks wouldn't take up what was going on with their employer. I was just wondering if you had any thoughts on that.

MS. KENNEY: I think we need to have a better understanding of how premiums affect people at different income levels. It's just amazing what we don't know about how families of different income levels respond to different premiums and to public premiums in particular.

But the great benefit of the last several years has been that states have changed their premium schedules at different income levels so there is a lot more variation in premiums and it has changed a lot over time. I know there is research going on right now that is really going to speak to that, assessing the impacts of these premium changes.

I think the point that you're making is that increasing premiums for kids between 150 and 200 percent of the fpl, as was done in Wisconsin, could have a very different impact than increasing premiums for kids who are at 100 percent of poverty. Their access to employer‑sponsored coverage is much lower so the implications are quite different.

MR. O'GRADY: But, in that case, they didn't do the 5 percent for the poorer families?

MS. KENNEY: No, they did not.

MR. O'GRADY: They did it ‑‑

MS. KENNEY: Nine states have increased premiums for children between 100 to 150 percent of the FPL, but the impacts on coverage are not known. There may be a tradeoff there. You could perhaps increase premiums more for the higher income kids, but we don't really know what the market will bear there. And we don't know what happened to the kids that lose public coverage. We don't know whether they gained private coverage when they left. We don't know what their service use experiences were. There is so much that we don't know that is relevant.

Let me just say one more thing, Catherine, on the issue of who is coming onto SCHIP and where they are going when they leave it. The evaluation we're doing for ASPE has some very useful information on this topic, which should be available in the not‑too‑distant future.

MR. VERDIER: Yes, just on the average cost per year for nondisabled Medicaid kids. In 2005 it is $1,800. You've got in that Medicaid Program at a Glance thing 2003 data that shows it at $1,700. Most of the actuaries I've talked to who have looked at both Medicaid and SCHIP for purposes of managed care rate setting would say that the SCHIP kids are a little bit less expensive but not huge amounts. Maybe a couple hundred dollars or something like that.

VICE CHAIR McLAUGHLIN: Well, I mean, just from my looking at this, I wasn't sure all the disabled kids were out of that chart.

MR. VERDIER: They are not all out of it. Most of them are.

VICE CHAIR McLAUGHLIN: So it may ‑‑ we don't know how long that tail is and how much it pushes up the average then. If it's not really a long tail and the average is close to $1,800 and nine months is the average length of stay, it does look like the SCHIP kids cost less. That is what's of interest to me is how much of that is case mix and how much is programmatic?

What can we learn from the different programs because one of the things we are supposed to be doing is saying what can we learn from the existing programs that are out there to yesterday expand coverage and today to contain cost. It would be interesting to know if there's something that's working better out there.

MS. MARYLAND: I have just a few questions to Jim first about disproportionate share dollars and other creative ways of states bringing in additional dollars. I saw that 5 percent of the total Medicaid dollars that have been paid out goes towards DSH. What about provider taxes? We don't have a lot of information on that. Can we get information in terms of a distribution by state? Is that possible to have that type of data available?

MR. VERDIER: Unfortunately, I don't think so. I was at a presentation in which the Government Accountability Office was presenting on a study they had done of this issue and they had done case studies of six to 10 states or something like that. They had reasonably good data that they were comfortable with for those states but they really didn't ‑‑ they weren't confident that they could extrapolate to other states and come up with good state‑by‑state estimates. I do recall that the Congress has asked HHS for those numbers and the secretary ‑‑ no, they didn't. I don't think there are state‑by‑state numbers that are real reliable.

MS. MARYLAND: Okay. That was the first question. The next question is to Jack regarding the Medicare issues, the Part D prescription piece. I noted that you said you are going through a bidding process. Are they looking at creating a formulary that will be required to be used in generics?

MR. HOADLEY: No. The plans are free to establish their own formularies in the new Part D benefit. There are a variety of rules that protect access to drugs so each formulary has to go through a review by CMS and has to meet certain requirements in the statute and additional requirements in the regulations.

Part of the design of this benefit is that each of the private plans that are offering the drug benefit can come up with a different formulary, can have stricter or looser rules, it can use generic substitution, it can use prior authorization or not, it can use step therapy or not, and so forth.

Part of the assumption is that this will lead them to having higher or lower premiums, and consumers can make the tradeoff. If they want to get a cheaper premium but put up with more limitations, narrower formulary, they can choose that, or if they want to go for a broader formulary and potentially pay a higher premium.

We don't even know at this point whether even that relationship will end up being true. Especially in the first year we have a lot of unknowns and plans are having to do some guesswork as to what is the right premium to charge.

MS. MARYLAND: Well, the only reason I even bring this up is our earlier presentation was about the increase and just the cost for certain components of the overall health expenditures, pharmaceuticals 6 percent increase. So to me this would naturally be an area where I would think there would be some scrutiny and perhaps looking at how can you reduce the cost of pharmaceuticals. Standardizing formulary seems to be something that would be reasonable.

MR. HOADLEY: It's definitely an area of importance because, as you say, the spending trend has been high. There are questions of what amount of it is price and what amount is utilization. There certainly is a lot of both going on. But the Congress did make a choice when they passed the MMA to not go with designs that would have more regulation in the formulary or more of a nationally structured benefit to go with this privately provided benefit and let the market place try to address it. Of course, it's too early to say. We don't even know what the plans are going to look like. We do know there will be a lot of plans offered.

We also don't know enough about the effectiveness of formulary restrictions and prior authorization and all those things. I finished a literature review about a year ago trying to see whether there is literature to support the potential for cost savings with various other kinds of methods. There just aren't a lot of studies, although there are a lot of anecdotes.

There are a lot of reports that say, "Well, we instituted this system and our cost went down,” but it wasn't really a study that said what else was happening and what other changes they made at the same time. There is certainly some sense that tighter formularies have the potential to save, but we also have other studies that suggest that tighter formularies lead some people to skip doses of drugs or not fill prescriptions.

But certainly plans are making decisions along these lines. If you are putting up a choice of several drugs on the formulary for ulcers and the clinical evidence suggest that all these drugs are pretty equivalent and for most patients it's not going to matter which drugs they use, then you get leverage by picking one product and trying to get a better negotiated price with the manufacturer around that product. That brings the premium cost down and that brings the cost of those individual prescriptions down.

We have a sense of the market dynamics that supports that. We don't really have the kind of overall scientific evidence that says this kind of formulary design leads to this kind of savings in a consistent way. But, again, the Medicare benefit is designed around the ability of each of those plans to pick products.

We see the same thing in Medicaid, a move towards use of preferred drug lists. They can't be formularies as such but try to go with certain preferred products and to try to build a lot of this on the evidence base about which drugs are more effective.

MS. MARYLAND: And then my final question ‑‑ I know you will be happy to hear that ‑‑ is directed to Jenny. You indicate nine states were using their SCHIP program to include adults. What kind of experiences have they had in terms of that and cost to the SCHIP program?

MS. KENNEY: As far as I know, costs are not available for the adult population but you can get enrollment data. You can get total expenditures at the state level but I haven't seen any reports that actually take the adult costs out. However, other evidence would suggest that the cost per adult would be higher than the cost per child.

But I do want to make the point that the statute did anticipate coverage of parents to an extent. There is evidence that covering parents through public programs actually stimulates greater enrollment among the children and may actually improve their access to services so there is an issue there.

In terms of the states that are covering parents and other adults, we know what enrollment levels have been over time but we really don't know much about the care that these adults and their children are getting. It's not a population that has been studied very carefully.

CHAIR JOHNSON: Thank you. And our last question to Richard.

MR. FRANK: I'll pass.

CHAIR JOHNSON: You'll pass? Okay. Did you have one more? Okay.

MS. CONLAN: This is to Jack. There's been a recent change in the Medicare appeal process and part of it, I think, might be a cost‑saving measure in terms of limiting the live, I don't know, centers to come for personal appeal and then some video conferencing in its place.

I can understand that in terms of saving cost. The part I'm concerned about in terms of ‑‑ maybe this is the wrong perception and correct me if I'm wrong ‑‑ representation. I'm wondering if that's going to affect access and be detrimental, particularly to seniors.

MR. HOADLEY: I have not directly studied some of these changes in the system and can't really speak to a lot of the detail. I've only read the newspaper articles basically that talk about some of the changes.

I do know that there were a broader set of changes out of which these little pieces that you're talking about come. There was wide agreement that the structure of the appeal system needed some fixing, some of the complexities of using Administrative Law Judges and the Social Security Administration to do Medicare cases who didn't necessarily have all the right training.

These are changes that over a number of years had broad bipartisan agreement in Congress and both the previous administration and the current administration pushed those kinds of measures. Now, the details of how the changes are being implemented is still a work in progress. I think is probably the fairest thing to say.

I don't want to try to comment on whether those particular instances are things that are necessarily locked into stone. They are decisions the Administration made at the moment perhaps for reasons of administrative cost and other choices being made but are things that are certainly worth watching. But I can't speak to the exact issues there.

MR. O'GRADY: I can if you want. I was involved in that decision as it went to the Secretary. A couple of things to keep in mind: 90 percent of the appeals are from providers, not from seniors. What we are talking about is tremendous lags in terms of people. If you are working off you had to stand in front of an Administrative Law Judge to have your case here.

We are talking a tremendous lag, a couple of years before you could be heard. Especially once you started getting out into places like rural Montana. God help the Alaskans in terms of this sort of a situation. What you've got now is you've got a more flexible approach. Will people still be able to see an Administrative Law Judge in person if they want to? Yes. Would I say seniors in particular would probably get a little nudge up that line? I would think yes as well.

That is not an attempt to deteriorate in any sense or just to save bucks on the deal. It's an attempt to make sure that it is the old justice delayed is justice denied notion. And, as Jack said, now you are moving it out of Social Security. You are moving it into the Medicare program where people are a little more facile in terms of some of these issues and you are trying to address it, as long as Congress made the change, it was an attempt to bring it up to latest science, industry standards, make it work in ways that would do.

If you are talking about somebody who is willing to go before a video conference sort of a thing and they can have their hearing taken in two months rather than two years, it is considered a good deal. I think anybody who still wants to see an administrative law judge I have not heard of any plans that they would not be allowed to if they wanted to.

CHAIR JOHNSON: Well, thank you very much, Jim and Jack and Jenny, for your informative presentations and response to our questions. We appreciate it a lot.

I would just like to turn to the working group at this point. We are 15 minutes before the start of our next session ‑‑ the scheduled start of our next session. Is that a better way to put it? Let me check on departure times for you as a working group. Is there anybody who has to leave before 3:00? What time are you thinking here? 2:30? Somebody over here?

So what are your suggestions in terms of process between now and the end of the day? How would you like to proceed in giving everybody affected a chance to hear the next panel and still take care of our biological requirements, including food?

VICE CHAIR McLAUGHLIN: I was just going to say yesterday we went ‑‑ a whole bunch of us went next door and it's a lot faster. The food is not great but it's edible so I guess I think in the interest of time we try to do that today and not take an hour lunch and have a shorter lunch.

MR. FRANK: I would suggest that we be ‑‑ we've been very tolerant about sort of the length of presentations and that perhaps we be more disciplined in keeping people at 15 minutes or 12 minutes or whatever.

CHAIR JOHNSON: Okay.

VICE CHAIR McLAUGHLIN: Well, the good news is there are only two.

MR. FRANK: I know. That's what I'm saying. We only have effectively an hour if we eat lunch.

CHAIR JOHNSON: Okay. What time would you like to reconvene?

MR. O'GRADY: Fifteen minutes.

VICE CHAIR McLAUGHLIN: Thirty minutes.

MR. O'GRADY: We can bring our lunch back and eat here.

MR. O'GRADY: There's no place close to get a takeout. That's the problem.

CHAIR JOHNSON: Well, building on Mike's comment, is there an ability to order something from next door and bring it back on a takeout basis?

VICE CHAIR McLAUGHLIN: I don't know.

(Whereupon, at 12:47 p.m. the above-entitled matter went off the record for lunch to reconvene at 1:12 p.m.)

A‑F‑T‑E‑R‑N‑O‑O‑N S‑E‑S‑S‑I‑O‑N

1:12 p.m.

CHAIR JOHNSON: Well, good afternoon and welcome back. Thank you for taking an abbreviated lunch time. Alice and Helen, we would like to welcome you. As at least Alice is aware, we just took a break from our prior panel a few minutes ago and we ran over to Subway to grab something to eat. If it's okay with you to help you with your schedules and help us with ours, we'll eat while you're talking. We really appreciate you coming.

We all have had a chance to look at your bios. I'll just say in general that both of you have come to us with high recommendations and we are just delighted that you are able to be with us. It's really an important topic to talk about. Just before you came, of course, we were hearing about public sector initiatives to help manage our cost and now we have a chance to hear about private sector initiatives to handle our cost.

We would like to target for adjourning shortly ‑‑ very close to 2:15 if we could. If we can ask you to keep your presentations initially to some place between 12 and 15 minutes and then we'll ask questions of you. We'll try to hold our questions until you are done with your presentations.

Alice, it looks like your material is on the screen so why don't we start with you and, again, welcome.

MS. ROSENBLATT: Thank you and thank you for having me here. I'm looking forward to some really good questions and I'm going to breeze through the presentation. Please eat and I'm used to having working meetings where everybody eats lunch so it doesn't bother me at all.

You all know I work for WellPoint. I'll talk a little bit about that. If you haven't read my biography, just one more word of warning, I am an actuary. I am a fellow of the Society of Actuaries so keep that in mind.

I'm going to talk about a couple of different ways that companies like WellPoint as well as most other managed care companies control cost and it's through product innovation, network design, healthcare management, pharmacy management, and leveraging technology.

With that, just a couple of words about WellPoint. We are the leading health benefits company in the nation by virtue of our membership. We have about 28 million medical members. We have Blue plans in 13 states. Where we don't have Blue plans we write health insurance business through UniCare.

We also have HealthLink which is a company that does network administration. We also have some specialty businesses including pharmacy, life, disability, and others. We are the second largest Medicare contractor.

As part of our 28 million medical members we have about 2 million that we call state sponsored members. I heard you talking about Medicaid and SCHIP when I walked in so we are a big player in that. We have about 38,000 associates.

I'm going to talk a little bit about product innovation starting with the current buzzword, consumer‑driven healthcare, consumer‑directed health plans. You will see it abbreviated as CDHP. There's been a lot in the newspaper recently. I know Helen is probably going to touch on this.

The idea here is to put information in the consumer's hands so that the individual consumer can make wise decisions about both cost and quality. Part of the incentive for making those wise decisions is for the consumer to have skin in the game, to have a cost impact for those decisions.

I think there has been a lot of talk about this. I think it's relatively a new thing. I think it's an unproven thing. Early results would say, yes, there is an impact but I think we are all wondering right now is if we're seeing an impact because the healthiest people are selecting this option, something that actuaries would call positive selection.

I don't think we really know all the answers yet but it certainly seems to be a step in the right direction towards getting consumers to think about cost and quality when they make their healthcare decisions.

One of the things WellPoint does to assist members in making those decisions is we have teamed up with a company called Subimo. What you are seeing on the screen is an actual look at something that would be on a website if you were one of our members and it would give information on various hospitals, physicians, etc.

This is another look at what some of that information would look like so that if you were thinking about having a particular procedure at a given hospital, you could do some research with this tool and see how many times that particular procedure is done in that institution and the kind of outcomes.

One of the other product innovations that WellPoint has done is a new product called Tonik that actually received some press. It's aimed at what we call the "young invincibles." Those are people 19 to 29 years old who think, "Nothing is ever going to happen to me. I don't need insurance." What we have found is about 12 million of the 41 million uninsured are in that age group.

Young adults 21 to 24 have the highest uninsured rate of any age group. Most of them are healthy but they think insurance costs too much, that it's complicated, and they procrastinate. What we have attempted to do is make this product appealing to that particular age group getting them to enter the insurance market. This would bring the cost down for everyone because we would be bringing in young healthy people, plus avoiding the uncompensated care issue. If those people do end up in the hospital, we all end up paying for that. There are multiple reasons for getting that group insured.

We are doing the same thing with small businesses. Many of the uninsured also work for small businesses where the employers do not provide healthcare. I think Helen is probably going to focus on the large employer side of things. We do have a lot of people in the U.S. working for small businesses. The rate of their employees being insured is much lower than the rate of employees who work for larger employers being insured.

We are trying to make it easier for small businesses to offer coverage through one of our products in California called BeneFits. We have reduced the participation requirements which would be a normal underwriting requirement of a health insurance company. We have reduced the employer contribution requirement.

We are providing coverage for part‑time employees. We have just started offering this pretty recently so we don't have a big block of business there but we have found that 84 percent of the groups buying this were previously uninsured.

This is a diagram of the next thing I'm going to talk about, network design. Managed care companies are all about creating networks. This is an example of the numbers and types of providers ‑‑ I know it's a little bit hard to see ‑‑ in our Blue Cross of California network.

Each one of these little circles is a different type of provider like laboratory networks, physicians, imaging centers, etc., and the slide shows how many are in the network. Part of what the managed care company does is to select the providers that go into the network and then determine the reimbursement for those providers. What we are actually going to pay those providers is usually determined by contractual payment. Both the mechanism and the rate are important to determine what the cost will be.

Just talking a little bit more about the reimbursement side of network design, one of the things that WellPoint is doing is creating Centers of Excellence in a couple of areas where we are looking for particular hospitals, for example, that have experienced more of a particular kind of procedure like CABG.

In California, for example, we have identified 70 hospitals that have had the best surgical and patient recovery results. For transplants, we have also provided our members with recommended transplant Centers of Excellence. Right now we have an application pending before the Department of Managed Healthcare in California that will create a network of approved Centers of Excellence for bariatric surgery.

Our member co‑pays and benefits will be tied to the use of these facilities. Our benefit design will incent our members to use those particular facilities.

In general our HMO and PPO networks do not cover all hospitals, all specialists, etc.

We are also getting involved through our reimbursement mechanisms in focusing more on paying for performance, P4P. In general, we select clinical measures to do that. We are using measures that have been endorsed by professional societies like the American College of Cardiology, the American College of Radiology, and the National Quality Forum. The programs also measure patient satisfaction. We have evidence‑based indicators ranging from preventive screenings to treating chronic conditions such as asthma and hypertension. Our goal is that as the programs mature, we'll shift from process oriented measures like how many mammograms have been done to outcome measurements.

Just to give you some idea of the numbers, in California we paid $57 million through these programs to 134 provider groups. Our Medicaid program in California paid $7 million to 185 medical groups. In Virginia we paid $6 million to 16 hospitals.

We also do healthcare management. This focuses on what can we do to make sure that we are getting good quality and that we are using the resources in the healthcare system appropriately. What this shows is that we insure lots of different types of people. We insure people who are very healthy and they want to remain healthy all the way to people that are chronically ill. We want to address that full continuum of our members.

Here are some statistics for you. Five percent of our members generate 50 percent of our healthcare cost. One percent of our members generate 25 percent of our healthcare cost so you can see when you think through those numbers that it's very important that we try to use our resources appropriately for those high‑cost members.

Down at the bottom of the slide I'm not going to go through all this in the interest of time but each blue bar there is a different program that we provide, from providing information to members to having disease management programs. I just mentioned the pay for performance program, and there is a specialty pharmacy program that I'll talk a little bit about.

As an example of some of the programs that we have, we recently introduced a radiology management program. You can see on the graph the trend of usage of CT and MRI was on a slope going toward very high utilization. We put in some radiology management programs. You can see that since the program inception, that usage has leveled out. We are actually expecting it to now start dipping down a little bit.

It's not that we are denying care. It's that we are redirecting care to more appropriate procedures. In our northeast region, which is Connecticut, Maine, New Hampshire, and in Colorado where we have implemented this, we have seen a reduction of about 11 to 15 percent.

We also do a lot of disease management work. We've recently submitted some of the disease management results to a special program with Harvard, the Harvard BlueWorks program. I heard in your previous session a lot of healthcare research is very difficult when you try to measure the effect because you don't have a control group. Here is a group with the intervention we're talking about versus the control group.

Usually, you have the situation where “this is what happened” but you don't know if it's just due to some extraneous factor. These were studies that actually had control groups. In the first one called our Midwest Care Counselor with a control group we had savings of 14 percent or $3,500 per year for the members that were enrolled in the program. And with another disease management program called Health Management through our subsidiary Health Management Corporation, we focused on diabetes, asthma and coronary artery disease. We had 11 percent savings there when you compare it to the control group.

We also have some programs that deal with behavioral health. Basically, the types of programs we do here are designed to increase the number of members receiving outpatient care after they have been discharged from a psychiatric hospitalization to reduce readmissions. It might include telephone outreach, active discharge planning, or case management of high‑risk members. That is just another type of healthcare management.

We also do pharmacy management. You can see here in the slide in the small type all around the circle there are lots and lots of different ways to control pharmacy cost and to reduce the cost. Pharmacy trend has been one of the biggest contributors to the high rate of medical cost trend. Pharmacy just two years ago was running at double digit trend rates.

It's now down to about 9 percent for most healthcare companies. It's been done through a lot of programs, programs that are done by the PBM (Pharmacy Benefit Management) Company, and programs that incent the consumer to make decisions like to buy drugs on formularies or to buy generic drugs instead of brand drugs. That is usually done through benefit design where there is a lower co‑pay if you buy a generic drug -- less payment out of pocket if you use a generic drug versus a brand drug.

We also have formulary management. We try to move people to mail order. There is a savings on mail order. We have a P&T, Pharmacy and Technology Committee, to determine what new drugs are approved drugs and things like that. There are a lot of different ways that we are working on the pharmacy issue.

Lastly, leveraging technology. I used to be on the Medicare Payment Advisory Commission and last year's report from the Medicare Payment Advisory Commission had a very good chapter on the lack of use of technology, primarily by physicians. Hospitals are using it.

The individual physician is pretty much still a cottage industry and they are not using technology the way they should. At our insurance company we are trying to use technology to keep our administrative costs as low as possible.

We are also trying to get physicians to use technology more and in doing that to reduce their costs. We spent $42 million in 2004 on an initiative that supplied 20,000 physicians with either PCs or e‑prescribing hand‑held tools.

The idea there was to get more of the claim information to us electronically, untouched by human hands, which would have reduced everybody's cost and probably improved the quality of the information we were getting.

On the pharmacy side there's a big quality issue in that. In the last five to six months we've had 60,000 electronic scrips written from 400 to 500 physicians. This is an example of illegible handwriting on a prescription being replaced by legible, electronic submission. One of the quality aspects of an initiative like this is to avoid adverse drug reaction.

This type of mechanism would be the only place drugs from multiple providers who might be treating the same patient could be reviewed and could indicate: "You shouldn't be taking this drug with that drug." Very often, the patient is asked when you visit a doctor, "What drugs are you taking?" They don't know unless they bring all their little bottles with them.

Finally, WellPoint is participating in both national and regional efforts that are trying to improve the coordination of healthcare information. We believe we can improve quality and reduce cost through this data provision. We are actively participating, for example, in the California Regional Health Information Organization which has just recently been started to try to identify some of the ways that we can reduce cost and improve quality. They are sharing information.

Here is one of the examples that was recently discussed by the committee. I don't sit on that committee but I have received information from other people at WellPoint who do. The example is: can we supply some of our information to emergency rooms throughout the state of California so that when somebody is admitted, they have that information.

Here is an example one of our doctors presented to me recently which was very unfortunate. Somebody who didn't speak English actually died in the emergency room while they were waiting for the results of cardiac tests to get done, which take about five to six hours. If they had the information that this person had cardiac conditions in the past, they probably could have saved that person's life, so there really is a quality aspect.

With that, I don't have any summary or conclusion but the idea was to give you a taste of the various things that are being done by insurance companies.

CHAIR JOHNSON: We just want you to know that Dr. Aaron Shirley's handwriting would never look like that drug prescription up there.

Mr. O'GRADY: Now, Frank's might.

CHAIR JOHNSON: Alice, we feel like you've run the 400 meter dash and made it in record time. A lot of great stuff that you have presented so thank you very much. We will come back to you for questions in just a little bit.

MS. DARLING: Good afternoon and thank you for the opportunity to be here. By the way, Dr. Shirley and I were involved in a study many years ago for the Institute of Medicine on Health Services Integration. It is very nice to see you again.

Actually, what I have to say is I think somewhat different, although I'll be moving into some of what Alice talked about because I am from the National Business Group on Health which used to be called the Washington Business Group on Health. It is a 240 plus member organization of mainly large employers.

The messages that we are communicating these days have to do with what we see as an affordability crisis and a quality and patient safety crisis. Our members and, of course, many other very smart people in this country who spend a lot of time on these issues, feel that we don't really get our money's worth for what we're spending, that we could provide a lot more care to people who have either nothing or not nearly enough if we simply spent the amount of money we are already spending more wisely.

That includes having higher quality care and safer healthcare. Our perspective is from the employer perspective. Frankly, most of what we say if you believe the front page of the newspapers, almost every day these days apply to the Medicaid program, to state governments, to local governments, and to the Federal Government because the affordability crisis is by no means just in the private sector.

Employers provide about 60 percent of all the healthcare for people under 65 including children. That is actually a number that has gone down and one of the reasons it has gone down is because as healthcare costs have risen, more employers who are smaller employers or middle‑sized employers have not been able to offer coverage. They haven't felt they could afford it so they haven't done that.

The second reason is that as costs have soared and generally the employee pays about 20 percent of the cost and the employer pays 80 percent, that amount (the 20 percent) has, of course, has soared as well. Absolute incomes, wages, our standard of living have either stayed flat or gone down while healthcare costs have gone up.

This particular chart which many of you may have seen actually many times shows you can see on the bottom, the lowest one is worker's earnings. You have at least two economists, I know. Maybe more. I know two of you so I know you are bona fide economists.

VICE CHAIR McLAUGHLIN: We can't hide.

MS. DARLING: I'm sorry?

VICE CHAIR McLAUGHLIN: We can't hide.

MS. DARLING: You can't hide, no. But basically worker's earnings in this country have been essentially flat for a very long time, or have grown very slightly and at least have, in fact, at some point gone down. Over all inflation has been relatively low but healthcare inflation has not. The percentage of what the country spends of all goods and services on healthcare alone soared and almost nothing else has relatively speaking.

It's not a very comfortable picture and here is another way to look at it. This shows the share of our total goods and services, the gross domestic product in the country, and the share that healthcare is taking.

The country hasn't grown very much and at the rate we're going, the standard of living that our children and grandchildren will have will be considerably worse than we are enjoying and is certainly a lot, lot worse than what was available to many of us who came out of school in the '70s if you look at wages and what people get paid and what they can afford.

A lot of the stories right now that you hear about housing and cars and anything else you talk about what working families can afford would have been impossible if we didn't have two adults in most households. It used to be that we had one adult who made a very decent living and now it takes usually two adults to essentially equal what we used to have.

The importance of that in terms of healthcare is more and more of the money is going for healthcare which means all sorts of other things that people might want. Buy a car, pay rent, buy a home, make a down payment. Anything that they might want to do they can't do because they are spending more and more money on healthcare.

Employers are, as a consequence, also creating fewer and fewer jobs in this country. This next chart shows you total health spending per capita and this is adjusted for the different prices and value in different countries.

But you can see that the next most affluent country which has a very high standard of living, very high everything, which is Switzerland. The United States spends 47 percent more in what it spends on healthcare per person than even Switzerland. We're not comparing ourselves to poor countries or even average countries. We are just spending so much more.

It wouldn't matter if we didn't have companies that have to make goods that then they sell abroad, or they compete with companies in this country who can sell goods and services for less money because back in their countries they don't have the high healthcare costs.

The stories that you see mostly these days are about General Motors because of this, but it's almost any industry or anything where a product or service is being produced in this country. We have to become more productive in order to pay for just the healthcare cost that companies in this country have to pay for.

Now, if it were just about cost, I think we could all say, well, all right, so we'll just spend more on healthcare and we'll just have less of other things. Who wouldn't want health and who wouldn't want us to spend money on healthcare? We would all want to do that.

The problem is the healthcare system also has a quality and a safety crisis. I hesitate in the room with Brent James to say anything about either of those topics. There is probably no single person better known or more prominent in the field of health quality and safety than Brent James so I'm sure he can help the Committee to understand that much better than I can.

But this one chart shows you, and this is a recent study, relatively recent, about two and a half years ago, published in the New England Journal of Medicine which shows that even now after many years of trying to improve quality and safety in this country roughly 55 percent of the time in some pretty common problems for which we have had years and years of work trying to improve the quality, we still don't get best practice medical care throughout the country.

So it's not just the cost, it's how good the quality is and, therefore, what is the value that we are getting for this expense. Corporate America and, frankly, the public sector as well as taxpayers, cannot make enough or sell enough in order to sustain these kinds of cost. They just can't. It means that other things, education, the states, the counties, localities, Federal Government can't spend money on educating children.

We have terrible foster care probably in almost every place in the country. I mean, there are dozens of things that are not done for people that we would all want to have done because we are spending more and more money on healthcare and not all of it wisely.

We know that, for example, the last three or four years we have had the worst jobless recovery that even as the recession began to recede and we began to have the economy pick up just a little bit, employers were not creating jobs. They were holding back as much as they could to not fill positions.

They certainly weren't creating new positions but they weren't going back and filling ones that had been emptied either through retirement or they had to lay off people. They are not filling the positions because, quite simply, the minute they started filling the positions they had on average $6,000 to $8,000 per employee average healthcare cost.

If that individual chose family coverage, we estimate that family coverage this year in the United States, again on average across the nation, would be between $11,000 and $12,000. The average wage in this country is about $30,000 average wage per employee.

At the rate we're going, it won't take us long that if somebody walks through the door and you hire somebody who wants family coverage, and if our wages aren't growing any faster than they are growing, we will hand somebody more in the benefit package for healthcare coverage for family coverage than we will give them in cash wages.

Now, that's an unsustainable situation by any standard. While I think most people, in fact, are very eager to have health benefits, at some point if they have to pay 20 percent of that, one, they can't afford it and, two, they would say, "I'll take my chances." There's no surprise that we have a lot more uninsured today. And, of course, together we have to find new ways and new resolve to try to deal with this affordability crisis.

Now, one of the things that our employers figured out pretty quickly about three years ago when we began an initiative to try to figure out what all large employers could do to make a difference is they decided, and this is very analogous to what I know you all are trying to do, and that is that you have to do two things.

You have to do some very specific sort of detailed tactics when you change things. You have to decide what are you going to pay for. A lot of the things that Alice talked about, pharmacy benefit management. Are you going to have step therapy. Are you going to cover some things and under what circumstances.

Are you going to pay more if they go to Centers of Excellence or less, what we call plan design changes. We know that all of us whether it's the Federal Government, the state government, or the private sector have to make those decisions and have to make very wise ones if we are going to have solutions.

But at the same time, we can't do things like that without changing the delivery system we're buying into because the delivery system itself is so flawed and so out of control in terms of cost we just can't say we'll just keep paying in all these different ways because it's unaffordable. We cannot afford what we've got.

Again, it's the kind of things that Brent James and his colleagues in the quality and safety world are trying to do to make the structural and strategic changes that will make it possible for the tactical changes to be effective and not lose the patient or the consumer in a difficult situation.

We don't want the patient, the consumer, to get caught by all these things. We want to make certain that those of us who are not in the middle of it can help to change everything and make it better for them for those who are having to deal with things.

Now, cost sharing, which is one of the things that is used quite a lot in the private sector, and increasingly in the public sector, is really only part of what we recommend. We are really saying that we need incentives, and Alice talked about this, to set up systems to have contracted networks and encourage people to get excellent care, to go to places that provide evidence of doing everything the right way, the pay for performance programs being examples.

Then we also want both purchasers again, whether governmental or private sector, to reward those organizations, those hospitals, those surgery centers, the physician offices, the medical groups. If they are doing the right things in the right way, we should really be paying them more.

We should be giving them rewards whether it's to help them to have computerized systems that make it easier for them to live with quality, if it's to help to pay for personal coaches for people who really need a lot more help than, say, some other people.

But to make certain that we are rewarding the providers that are doing an outstanding job, not just treating everybody the same because there's enormous variability out there. Of course, the other thing we need is transparency.

We need all of this to be done in a way that people can trust and can see and can understand where the numbers come from, where the data come from, how they are collected, what they mean. Then finally we need these things to be done urgently. We are increasing at about 10 percent a year.

It's estimated that just two years ago, three years ago healthcare costs were in 2011 going to double so we are about three years into that doubling and we are right on target. We are going to double. So by 2014 the healthcare cost in this country will be about $3.6 trillion.

Just the portion of that number that the Federal Government says all the public sector will pay will be equal to what it was last year in total. Think about that and think about the taxes that will be raised on everybody to pay for that including payroll taxes unfortunately.

Now, you've heard from Alice some of the things they are doing which is the kinds of programs and services that employers are asking for including decision support tools, data, how to control your own health conditions where that's appropriate, help with how to treat asthma, how to have some self‑management, and even feel that you can do those things because that requires some training.

Most importantly, evidence‑based, benefit design, evidence‑based practices. What we are hoping to do is to make available, again, through a lot of the tools that Alice talked about, you can find out what is the best treatment for the problems that you're dealing with. There is a lot of that available now.

I'm just going to go quickly through these last points because I know we want to leave time for questions. We need good technology assessment. We need excellent chronic disease and complex case management tools. They can work. I've talked quite a bit about quality and patient safety. The cost trends have gone down slightly but that should give no one hope. It's down to only 10 percent a year. That is not a good sign.

We also know that the private sector, and I think the public sector is really beginning to do more of this, is defining what works, what the best employers, the best public sector examples, where things are working, and try to recommend those approaches and to help train people to share information and that kind of thing.

The best performers are doing better. We know also that the best performers are able to manage much more effectively than those who are not so we should really disseminate best practices. We also know that they are working on quality, cost management, and in particular getting information into employee's hands. They are also looking at hard return on investment calculations.

Increasingly the private sector will be saying we will pay for things that make a difference and we are going to stop paying for things that don't make a difference. If you want to spend some of your own money on things that don't make a difference, you're welcome to but we're not going to support that.

We also know that the best performers use quantitative analysis. There are these very large data warehouses now, programs. Again, I think Alice mentioned some of that. But they are available. Even the Federal Government is doing more of that, focusing on data to understand what makes the difference and what works and what doesn't.

We also know that speed is really important. The successful employers work really fast. They actually make decisions in weeks, not months, not years. They know they really don't have a choice because they're not going to survive. You will see more and more American legacy corporations going belly up because they cannot afford to compete in the global economy.

If you look at the airlines, and if you want some examples, just make a list of the companies that were Fortune 500 20 years ago and see how many of them are no longer around. Some of them got acquired so you don't see them but some of them went under.

We also know that best performers give their employees choices. Best performers rely on health management. Almost all the large companies now are developing comprehensive health improvement programs.

They are saying if we want to control cost, the only way we are going to do it is to help people be healthier, to make them healthier by giving them information, giving them subsidies to take health risk appraisals. There are a whole slew of things. All these things, again, have to be done and they are being done with a lot of energy and commitment.

Now, Alice mentioned also the high‑deductible plans. By the way, the new term is high‑deductible plans with or without health savings accounts, with or without health reimbursement accounts. This actually shows you what is being offered. It's about 18 percent offered in 2006 and 8 percent offered in 2005. You are talking about about 25 percent of employers are offering a health savings account as part of a high‑deductible plan.

Then I have some comments which I'll just leave with you on the positive and problematic aspects of HSAs. I would be happy to talk more about that. I have said most of these things but it's a nice checklist. If you want to think about how to do some of these things the way the best performers, the ones that have the best results, you've got two slides on a checklist. Together I would just sum up by saying we all have to work together.

The only way we are going to have a safe, efficient, high‑quality healthcare system that we can afford is if the public sector at all levels and the private sector work together and figure out how to make the changes we absolutely have to make. We don't really have a choice and we have to do it as quickly as possible. Thank you.

CHAIR JOHNSON: Well, that was the second leg of our 440 run at full speed. I'm going to yield to the rest of you on asking questions of Helen and Alice. Richard, you want to go first?

MR. FRANK: I have one question for both of you. In a sense you have told us, actually very nicely and very clearly about kind of the emerging principles of consumerism, transparency, and personal responsibility which are sort of underlying a lot of this.

Last night when I went to my room ‑‑ we heard some presentations on this topic ‑‑ and so I said, gee, Harvard is probably going to have a high‑deductible health plan next year. I started trying to write down how I would make the choice, okay? I have to worry about a medical savings account, which is part of all of it.

The fact that the prevention is going to have a particular co‑payment that is really different from everything else; that there is going to be a tiered formulary on the side, okay? There's going to be a tiered network in some of the choices because there are centers of excellence.

Tufts is trying to do that. Suddenly I realized that you have this incredibly nonlinear price schedule both for choosing the plans, but once you get in the plan for figuring out what you ought to do in terms of going to hunt down for care. I'm saying how am I going to make this choice?

It turns out that, at least, the choice of plan depends critically on my guess about what I'm going to spend next year. I have learned this year that I suck at that and I'm probably better than most of the other people I know.

MR. O'GRADY: A chance to edit the record.

MR. FRANK: That's a technical term in economics. And then there's balanced billing that is growing so, in fact, I'm not sure about what the base price is that I'm getting my co‑payment applied to, and I don't know what my doctor is going to balance bill me very often.

I'm trying to figure out as we explain all this to the American people in our first report, how are we going to say, "Well, we're on the road to transparency, personal responsibility, and clear choice." It's boggling. I understand how each piece is motivated by that. I'm struck by the fact that the hole is shockingly different from that goal.

CHAIR JOHNSON: Helen and Alice, I'm going to ask you to be as precise as we can in answering the questions. I'm going to ask our colleagues around the table to be quick in asking the questions as well so we can get as many in as possible. That's not a comment. It's just for Helen and Alice right now.

MS. DARLING: Good reminder for me.

MS. ROSENBLATT: Number one, I agree with a lot of what you said. I think some of these plans are extremely complex. As an actuary I would say it's good that you couldn't predict because the more people that can predict, the more we get adverse selection. And we do try to factor that into our calculations.

It is extremely complex. The balanced billing issue depends on the carrier. Generally with a Blue plan you don't have balanced billing issues so that's a “commercial plug” there. And I do think that some employers, and maybe Helen can speak to this, have actually introduced tools where employees can enter what they expect their expenses to be and the tool will direct them toward one of the plan options.

I think on the issue of the consumers making informed decisions we're not there yet. We're moving towards being there but I think it's a long way off. At WellPoint, we want the total cost to the consumer to be available on our website. We are absolutely not there yet but it's certainly a vision.

MS. DARLING: First, most large employers do have something that helps and plans help to figure that out. The reality is most employees actually don't even open their booklets each year. They don't. They mostly do not make changes. When they are forced to make changes, they don't necessarily even read them.

They sometimes learn it after the fact. I don't think things are going to change that much that directly. The one thing the plans have going for them is they often cost a lot less. We do see people choosing them because they don't want to pay the higher monthly premium that they would for the other plan.

On the other hand, we also know that most Americans over insure. I mean, they don't make good economic decisions for their own purposes. You have those two things working against each other. There's a lot we don't know.

CHAIR JOHNSON: Montye.

MS. CONLAN: Alice, I was interested in your slide on MRIs. I recently went to a patient program conducted by a radiologist and received an education. I didn't know there were such things as different strengths of the magnet measured in Tesla and how to pick the appropriate size.

But I recently went to a hospital that had been restored after the hurricanes and there was a women's health center and they were showing their new MRI. I felt like I was buying a car. They were talking about, "Look at this. You can sit there and have peace of mind because the technician is right there to hold your hand or speak to you."

I said, "What is the size of the magnet?" He said, ".3 Tesla." I being a good consumer could decide that wouldn't be appropriate for me. I would need 1.5 at least. I'm wondering because I know a lot of consumers are feeling that they are being denied. You are making decisions probably on those bases or something like that. Have you considered educating the consumer in that way on technology?

MS. ROSENBLATT: I think that's a real good point. I used to work with a consultant when I was in the consulting business. He used to tell stories about the amount of radiation ‑‑ not X-rays but whatever the radiation is that people were absorbing when they had too many MRIs or CT scans or whatever was done. People don’t think about the radiation impact of some of these tests that they are getting.

He also told stories about people using dated machines to do totally inappropriate things. There is a lot of inappropriate usage going on. I think we need to have the health insurance company doing the kind of programs that we are putting in place which is geared towards making sure that the right imaging is done at the right place by someone that knows what they're doing. We also need to get information into the hands of the consumer.

There is definite overuse of imaging. That is one of the other really high trend factors. I mentioned pharmacy as a heavily increasing component. This is another extremely high component ‑‑ you saw the graph. It's a tremendously increasing component.

CHAIR JOHNSON: We're going to have to go. If we have more time, we'll come back.

DR. SHIRLEY: In your pay for performance feature when you move into the outcome measures, let's say diabetes, asthma, congestive heart failure, what are some of the measure you're going to be looking at to make a determination, Alice?

MS. ROSENBLATT: What we are looking at right now are the basic measures like have we reduced the amount of emergency room treatment. Have we reduced inpatient admissions. Then what we are also measuring for diabetes, for example, is there an annual eye exam done. For some of the congestive heart failure programs, are the patients on beta blockers.

There is right now a combination of some clinical markers combined with some process measures. We would like to get to the point where we are actually looking at long‑term studies showing outcome results but that is another thing where I would say we are just not there yet.

MR. HANSEN: Two quick questions and maybe the first one could kind of be answered as a follow‑up to Richard's question. You talked about one of your programs, Alice, of educating individuals on choice and I thought you were headed toward quality there, not necessarily cost or expense. I wonder if you had any data to back that up?

The second question is you alluded to it and I think we are dancing around something here, your Centers of Excellence. I think, Helen, you might have talked about it in a different kind of way. Are you talking about a different distribution of equipment ‑‑ I'll use the word rationing because I don't know of any other word ‑‑ where you limit this high‑cost equipment to just a few places in a community?

MS. ROSENBLATT: I'll take the first part of that question. We are not getting involved in limiting anything within the community. What we would be trying to do through our Centers of Excellence would be to say, for example, with our bariatic surgery program, these are the centers that you should use if you are one of our members.

If you don't use one of these centers, you are going to end up paying more out of pocket. We are trying to put some incentives to have people go to the ones that we believe are going to have the best quality outcomes as well as the best cost.

In terms of educating the consumers on quality, I showed some examples where we partnered with a company called Subimo where you can look at hospitals and at least see the number of that type of procedure that is done. Again, it's kind of a raw quality measure but there is something out there that the consumer can look at to get a feel for quality.

CHAIR JOHNSON: Is the Subimo report more than Leapfrog? Can you build on what Subimo includes?

MS. ROSENBLATT: All I can tell you is what is on our website. I can't compare that to Leapfrog. I'm just not familiar enough with the two but maybe Helen can. One more point. We also have on our website, for example, for members in our HMOs in California, a comparison of certain criteria for each medical group. This is based on measures like what percent of the female members had mammograms. There are some quality indicators on the website on particular physician groups.

MS. DARLING: Well, the data on the Subimo site include measures like Leapfrog’s. All of the products use essentially the same set of measures and almost all of them ‑‑ a lot of them come out of CMS from the Medicare program and from what is known as QIOs, quality improvement organizations, which are the ones that do the quality measurement.

There's an amazing amount of overlap, HealthGrades, HealthShare Technologies. There's a whole bunch of them. They all use the same sets of data for the most part. They may array them or multiply them in slightly different ways and it's a growing database. It's something that is eventually going to have a tremendous impact if only because it makes people pay attention to quality and performance.

I would like, if I may, just to respond to the second part of his question. When we talk about Centers of Excellence, we do mean them just the way Alice talks about them as being specific to certain conditions or certain procedures. For example, a hospital may be a Center of Excellence for orthopedics and another for cardiovascular disease or cardiac disease or something like that.

But in all instances we want it to be based on quality and documented performance and then efficiency measures and not just a question of efficiency or cost alone. I think that is where we are headed as a country. It's not going to be easy to get there.

Certainly not everybody is going to be immediately satisfied with every detail. I think there is agreement that we have to move in the direction of having rewards for the places that do things better because we can't just pay ‑‑ we can't afford to pay everybody whatever they want which is the way we are doing it right now.

CHAIR JOHNSON: Okay. You talked about ‑‑ go ahead, Richard, and I'll follow you. Well, I was just going to follow‑up your most recent comment where you talked about reporting on efficiency. Today what we do is we contract ‑‑ and I suspect WellPoint still does but may be expanding ‑‑ we contract based on who is going to give us the good discount on a per‑procedure basis. That is what most of the health plans have been doing up until now. Talk to us a little bit about your focus on efficiency and what that means. Does that mean just plain cheap?

MS. DARLING: Well, we're not saying that so you're right. We are saying that all of the selection should be on the basis of a combination of quality and efficiency. But efficiency ideally measured would capture not just price by any means. It would capture the cost of, say, admission.

I'll give you one example. This is an analysis I did for a company down in Tennessee. This was based on Medicare data and it was case mix‑adjusted and severity‑adjusted admissions for specific procedures and things like psychiatric disorders. We could compare the Vanderbilt University Hospital with another hospital owned by a for‑profit company which shall remain unnamed.

The difference was dramatic. With all the adjustments the cost of the two hospitals were very, very different. The for‑profit hospital was about $1,000 more for the stay, one particular one I looked at. If anything, most people would have assumed that because Vanderbilt is a teaching hospital, that it would have been more expensive and maybe not particularly more efficient. They both may have been very costly and one was only slightly less costly.

At least in that instance we had reasonable measures that adjusted for different case mixes. There were people who believed that the Vanderbilt Hospital was the higher quality and there were some quality metrics that were used, imperfect to be sure. Things like that is what we are talking about for Centers of Excellence.

Also, bariatric surgery is another example. If an employer covers it, and fewer than half of the employers in our group cover bariatric surgery at all, but if they do cover it, or if they continue to cover it, we would recommend that they find centers who, among other things, are willing now to report outcomes data and willing to have their patients followed and have physicians work with a registry to report on readmissions and then longer‑term complications and consequences. There are some very significant differences by location for bariatric surgery, and by surgeon, too.

MR. FRANK: This is for Alice. First of all, I thought your comment on adverse selection was just right. On the other hand, it's sort of ironic that what we have to do is count on people being confused. We fix adverse selection by making sure that they almost randomly select plans and half of them get the wrong plan so that hardly makes my heart go pitter‑pat.

But I want to talk about what you learned about pay for performance. My understanding, at least from looking at the literature and seeing the way Pacific Care and other California people have implemented this, I see very clearly how you can get higher quality because the measures tend to focus on under‑treatment so are you screening?

Are you getting a beta blocker where you wouldn't have it? Are you getting follow‑up? Those kinds of things. The connection to cost containment is less clear to me. I was just wondering how you saw it (a), and (b), what your experience has been.

MS. ROSENBLATT: First of all, I did say that comment in jest.

MR. FRANK: I know.

MS. ROSENBLATT: I just want to get the record straight there. We are trying to get to clarity and there are ways an actuary can adjust for the adverse selection but clarity is definitely the goal.

In terms of pay for performance, you're right. When we are doing pay for performance, we are doing screening for breast cancer, cervical cancer. We're doing a little bit of something more specific. We are able to measure the use of generic drugs as opposed to brand name drugs, for example. That is getting a little bit more specific. The beta blockers, the eye exams, it's all of that.

Even in the design of our networks, Randy, to one of your points about, are we picking the providers that just have the biggest discounts. First of all, you could have the biggest discount off a big starting point and that doesn't help you. Let's all keep that in mind. But we are trying to create smaller networks that are based on not only cost but HEDIS measures, examples of chronic case management like we talked about so we are trying to do all of that.

MR. FRANK: At least at this stage of development you don't expect to save a lot of money there except on the generics.

MS. ROSENBLATT: Actually, what we have found is if we compare the PMPM, per member per month cost that we are expecting in our smaller networks versus our larger networks, we do expect to save cost. We have the data and we can find two hospitals where one will charge $8,000 for a procedure and the other will charge $2,000. To us there's not a lot of difference.

DR. JAMES: Just a follow‑on comment, Alice. There are too many models that people are using for pay for performance. One that you measure some quality parameter and then you offer a premium. You know, if you hit this decile, we will give you 1 percent. If you hit this level, we'll give you 2 percent additional.

The other class of models that are coming out are called shared savings models which require much better measurement systems. It's based around the idea of waste elimination, quality waste elimination mostly, that you demonstrate better clinical outcomes and usually expect cost savings to fall out. Are you using any shared savings models or are you mostly using premium models in your pay for performance?

MS. ROSENBLATT: We are generally using like an add‑on to the fee in the future or something like that.

DR. JAMES: So a premium.

MS. ROSENBLATT: Yes for most of our pay for performance at this time. Although, the typical HMO model usually has a shared component for the capitation. There is often a capitation paid and a withhold and then there is a sharing of the withhold based on the results. That has been around a long time.

DR. JAMES: Capitation is a shared savings model inherently.

MS. ROSENBLATT: Yes.

DR. JAMES: The one big shared savings model is the new CMS demo project, the PGP project for outpatient settings. I think that is just getting started but I think everything else has been premium based.

MS. ROSENBLATT: I think it's harder to do all of that in a PPO type environment versus an HMO where there is capitation so that is part of what you're probably seeing.

DR. JAMES: I think you can do it but you have to have really good data systems.

MS. ROSENBLATT: Yes.

MS. DARLING: Let me just say my impression is this is the only model that we are going to see in the future. There's not a lot of tolerance among the payers right now for paying more money and not having some kind of take‑away. Since that is harder to pull off for obvious reasons, a combination of, say, rewarding those who are doing an excellent job in whatever the definition is with an update and not updating others even on the private sector side. We'll see a combination of that and then shared savings rather than what we're seeing right now.

CHAIR JOHNSON: We don't have a lot of time but yesterday we heard some comments on need for mental health care. I'm wondering if both of you could just share very briefly again what your organizations are doing with respect to mental healthcare and what your recommendations would be for our system in general with respect to that. Again, it's a big question. Just get us into it a little bit.

MS. ROSENBLATT: I actually meant to mention something when we talked about bariatric surgery and Helen commented that a lot of the employers are not paying for bariatric surgery. In some states bariatric surgery cannot be omitted from your benefit plan. I think with mental health you're touching on an issue of state laws and what we call mandated benefits and what should be covered and what shouldn't be covered.

I did have a slide on behavioral health so I think, to answer your question, I would just go to that behavioral health slide.

MS. DARLING: Most of our employers have ‑‑ I would say virtually all of them have very comprehensive benefits including mental health benefits. We don't have any ‑‑ we think that is a good thing. We don't see that changing. In fact, you could probably argue that mental health care, at least on the private side, has not only not grown, it has actually proportionately gone down compared to some of these other expenses. It's a little bit different world. Obviously much different than the public sector when you talk about mental health care.

The one area that we worked very hard to make sure that our friends on Capitol Hill and the Treasury Department understood it was important to protect from the high‑deductible plan model for the services of employee assistance programs.

As you know, because Motorola is a good example of a company that has an excellent employee assistance program, most large employers provide support services through EAPs that tend to ‑‑ they are not intended to be a substitute for mental health benefits at all.

They are intended really to be benefits that help all employees cope with the kinds of problems that may in fact lead to or exacerbate mental health problems. There are family issues, marital issues, things that would not normally be covered under mental health benefits, so‑called V‑codes for the technical people in the room.

What we did was we wanted to make certain, and I would urge you to do the same, that people understand that there can be a set of services that among other things help people, but not for psychiatric disorders. If you have a psychiatric disorder, you would be assessed and referred for mental health services.

All of those other services, especially in jobs, hotels, places like this, I mean, an awful lot of employees in some of these jobs have just horrendously complicated lives and have a lot of family problems. Some may or may not have mental disorders but they certainly have enough issues that would put most of us close to the edge under any circumstances. We see mental health services as a combination of those two sets of services. One set should be virtually unlimited.

It usually is. EAP benefits usually are somewhere between four and 10 visits per year per person per family member. It's a lot of services that are available. Right now they are allowed to be outside the high‑deductible plan because of work that we all did together.

CHAIR JOHNSON: We would like to thank you for your participation with us this afternoon. We have learned a lot and we appreciate you coming.

MS. DARLING: Thanks for the opportunity.

CHAIR JOHNSON: We'll just take a minute to close our meeting. Before we do, we have a couple who have to leave. How would you like to proceed with the rest of our discussion?

VICE CHAIR McLAUGHLIN: I had a couple of things. Larry, I don't know whether the list of things that several of us have come up with since yesterday is administrative stuff in an executive session versus an open. Let me throw out a couple of things.

Yesterday several people made the comment that they didn't feel as though we had enough time to talk about what we had heard. I know that I as a member of the Subcommittee on Hearings and Pat also, and I can't speak for the other three members, would like a little bit more feedback from the group of whether we stick with the five hearings or we rethink that so that we have more time for working groups. In other words, how can we struggle. That was one issue. A second is that ‑‑

MR. PATTON: Can we do them as you go? The first one seems to be a decision about process and that strikes me as administrative.

VICE CHAIR McLAUGHLIN: Okay.

MR. PATTON: If you were talking about what you heard the substance, then you should be on the public record.

VICE CHAIR McLAUGHLIN: Okay. Good. The second was we never had any time to talk about America Speaks. I know that the staff was very anxious and I am sure that whoever ends up being on the Community Meeting Subcommittee will be anxious to get some feedback about how the group felt about that.

MR. PATTON: Again, I think that is more of a process issue because it's a question of you thinking through just from this approach whether you want to hear from other approaches and all of that. Again, that is not decisional in this sense.

VICE CHAIR McLAUGHLIN: Well, in response to Randy how we would like to handle it, we were supposed to go to 3:00 and I know a couple of people had to leave early but I guess I would like, as soon as we finish whatever public stuff we need to do, to ask for, again, administrative issues. While we are all here given all the energy we put into talking about it, try to come to closure on a couple of things before we leave.

CHAIR JOHNSON: Could we go into ‑‑ is there anything else that we need to discuss on the public record? Okay. Why don't we just go into executive session then and we can discontinue the recording and so forth.

But before we do, just a word of thanks to you who have served as staff. Andy, thank you for your time and effort in collecting minutes and so forth, taking minutes and putting the report together. We know that will help us financially as opposed to outsourcing that, so thank you.

Caroline, thank you again for all your work in putting the hearings together.

VICE CHAIR McLAUGHLIN: I'd also just like to thank Andy for getting that website up so that we could post the hearings and stuff. I personally know he spent an enormous amount of energy going back and forth, back and forth, back and forth with the AHRQ staff correcting things that were wrong. Thank you, Andy.

(Whereupon, at 2:23 p.m. the meeting was adjourned.)

Submitted by Cygnus Corporation, Inc., for the Agency for Healthcare Research and Quality under Contract No. 290--1-0001.