Blood Safety Transcripts

DEPARTMENT OF HEALTH AND HUMAN SERVICES
ADVISORY COMMITTEE ON BLOOD SAFETY AND AVAILABILITY

Fifteenth Meeting

9:11 a.m.

Friday, August 24, 2001

Hyatt Regency Capitol Hill Hotel

400 New Jersey Avenue

Washington, D.C. 20001

P A R T I C I P A N T S

Arthur Caplan, Ph.D.
Larry Allen
Michael P. Busch, M.D., Ph.D.
Mary E. Chamberland, M.D.
Rajen K. Dalal
Richard J. Davey, M.D.
Jay Epstein, M.D.
Col. Michael G. Fitzpatrick
Ronald Gilcher, M.D.
Edward D. Gomperts, M.D.
Paul F. Haas, Ph.D.
W. Keith Hoots, M.D.
Harvey Klein, M.D.
Dana A. Kuhn, Ph.D.
Karen Shoos Lipton, J.D.
Lola Lopes, M.D.
John Penner, M.D.
Jane A. Piliavin, Ph.D.
Jerry A. Winkelstein, M.D.
John Walsh
Stephen D. Nightingale, M.D.
Capt. Lawrence McMurtry
Virginia Wannamaker

C O N T E N T S

AGENDA ITEM

Call to Order, Conflict of Interest

Current Status of Monitoring Blood Supply and
Demand - Stephen D. Nightingale, M.D., Office
of Public Health and Science

Summary of 8/23/01 Blood Demand Contractor's Conference, Darrell J. Triulzi, M.D.

Break

Public Comment

Committee Discussion and Recommendations

Lunch

Immune Deficiency Foundation Proposal to
Monitor Demand for Plasma Derivatives -
Jason Bablak, J.D. Immune Deficiency
Foundation

Plasma Protein Therapeutics Association
Comment - Christopher Healy, J.D., Plasma
Protein Therapeutics Association

Public Comment

Committee Discussion and Recommendations

Adjournment

P R O C E E D I N G S

CAPT. McMURTRY: Good morning, ladies and gentlemen. Regardless of what this says in front of me, my name is Lawrence McMurtry. I'm the Deputy Secretary--Deputy Executive Secretary of the Advisory Committee for Blood Safety and Availability. I just gave myself a nice promotion there. I wish I had gotten the pay for it.

DR. CAPLAN: When Steve is not here, he is in charge.

CAPT. McMURTRY: I'd like to begin by calling the roll, if I could. Mr. Allen?

MR. ALLEN: Here.

CAPT. McMURTRY: Dr. Busch?

DR. BUSCH: Here.

CAPT. McMURTRY: Dr. Chamberland?

DR. CHAMBERLAND: Here.

CAPT. McMURTRY: Mr. Dalal?

MR. DALAL: Here.

CAPT. McMURTRY: Colonel Fitzpatrick?

COL. FITZPATRICK: Here.

CAPT. McMURTRY: Dr. Gilcher?

DR. GILCHER: Here.

CAPT. McMURTRY: Dr. Gomperts?

DR. GOMPERTS: Here.

CAPT. McMURTRY: Dr. Guerra is in San Antonio today. He had--I think it was grant reviews today.

Dr. Hoots?

DR. HOOTS: Here.

CAPT. McMURTRY: Dr. Klein?

DR. KLEIN: Here.

CAPT. McMURTRY: Dr. Caplan?

DR. CAPLAN: Here.

CAPT. McMURTRY: Ms. Lipton?

MS. LIPTON: Here.

CAPT. McMURTRY: Dr. Lopes?

DR. LOPES: Here.

CAPT. McMURTRY: Dr. Penner?

DR. PENNER: Here.

CAPT. McMURTRY: Dr. Piliavin?

DR. PILIAVIN: Here.

CAPT. McMURTRY: Captain Snyder, unaccounted for this morning.

Dr. Winkelstein?

DR. WINKELSTEIN: Here.

CAPT. McMURTRY: Dr. Davey?

DR. DAVEY: Here.

CAPT. McMURTRY: And Mr. Walsh?

MR. WALSH: Here.

CAPT. McMURTRY: Ms. Pehuja is also not with us this morning. She started law school this week. So she is going to remain active with the Kulis (ph) Anemia Foundation and will be at subsequent meetings but is not here today. And, in fact, I'd like to ask for somebody to propose a resolution commending her on being accepted to law school, if I may.

DR. PENNER: So moved.

CAPT. McMURTRY: Second?

DR. GOMPERTS: Second.

CAPT. McMURTRY: Nobody opposed to that, I assume. Thank you. I'll have that in the notes.

The next thing, this is the part that I have been looking forward to since Steve told me I was going to open the meeting, and that is, getting to read his conflict of interest statement. I'm telling you, this is exciting for me.

The following announcement is made part of the public record to preclude even an appearance of a conflict of interest at this meeting. General applicability has been approved for all committee members. This means that unless a particular matter is brought before this committee that deals with a specific product or firm, it has been determined that all interests reported by the Advisory Committee members present no potential conflict of interest when evaluated against the agenda.

In particular, as specified in Title 18 of the United States Code 208, subsection (b)(2), a special government employee, which all committee members are, may participate in a matter of general applicability, for example, advising the government about its policies related to hepatitis C epidemic, even if they are presently employed or have the prospect of being employed by an entity, including themselves, if they are self-employed, that might be affected by the decision of the committee, provided--and this is the key point--that the matter will not have a special or distinct effect on the employee or the employer other than as a member of a class.

The example given at 5 CFR 2640.203, which implements the U.S. Code, is as follows: A chemist employed by a major pharmaceutical as been appointed to serve on an advisory committee established to develop recommendations for new standards for AIDS vaccine trials involving human subjects. Even though the chemist employer is in the process of developing an experimental AIDS vaccine and, therefore, will be affected by the new standards, the chemist may participate in formulating the Advisory Committee's recommendations. The chemist employer will be affected by the new standards, but only as a part of a class of all pharmaceutical companies and other research entities that are attempting to develop an AIDS vaccine.

In the event the discussions involve a specific product or a specific firm for which a member has a financial interest, that member would exclude him- or herself from the committee discussion, and that exclusion will be noted in the public record.

With respect to other meeting participants, we ask in the interest of fairness that they disclose any current or previous financial arrangements with any specific product or specific firm on which they plan to comment. I will note also for the record that each voting member of the Advisory Committee has in a packet before them or had in a previous meeting in a packet before them a specific waiver for the purpose of participating in the meeting under the terms that I have just described.

When the committee was established, it had been considered that the appointment by the Secretary himself constituted that waiver. Where changes of administration rules undergo review, as do formal policies and action, and the waiver is an additional measure of protection for the committee, it confers no new rights. It does not require your signature. We perhaps have completed the piece of paperwork which we'll be glad to discuss with any member of the committee if they wish to do so, but the bottom line is your rights, your responsibilities, and your protection, which is really the bottom line here, are unchanged by that.

And I think that we have gone through as much of that as we need. So, with that--

DR. KLEIN: Where's our chemist?

CAPT. McMURTRY: Actually, he's out in the hall because there were no chairs.

DR. KLEIN: I see.

CAPT. McMURTRY: So, with that, the agenda, which none of you have--but I do--calls for a welcome by Dr. Arthur Lawrence, the Acting Principal Deputy Assistant Secretary for Health, but he's not here either. So we'll move along.

I'll let you take over.

DR. CAPLAN: Do you want to do the summary?

CAPT. McMURTRY: Yes, let's do this.

x DR. CAPLAN: There was a meeting held yesterday on the contract to try and develop a plan to monitor the blood supply. As you know, with all the discussions of deferrals, with all the discussions of what can be done to ensure safe and adequate supply of blood, it's been difficult to do that without numbers and without having an independent assessment of what the situation is with respect to inventory of blood.

There are lots of ways to compile statistics. I'm always despondent when I find out that I thought we'd get an answer by looking to numbers, and then it turns out that numbers are flexible. So there are a lot of ways to count, and there was a discussion yesterday. Who's going to do the contractor--who's going to recap? That's what I'm not sure, who's doing this meeting.

CAPT. McMURTRY: Dr. Triulzi.

DR. CAPLAN: Okay. And you're going to hear me today say make sure to hit your mikes, the button, when you're talking later. Also, I'm going to ask you to say your names today because we have a new transcriber who's desperately trying to keep tabs on who's who.

CAPT. McMURTRY: Dr. Triulzi, could I get you to come up to the front? We have the press that is wanting to put you all over the television and newspapers.

DR. TRIULZI: Thank you very much. My name is Dr. Darrell Triulzi, and I'm the medical director for the Institute for Transfusion Medicine in Pittsburgh and one of the 29 participating sites in this effort to collect data on blood demand.

I think it would be appropriate to first thank Captain McMurtry and Steve Nightingale, and especially Virginia Wannamaker, in their tremendous successful effort in being able to pull together this group so quickly and so successfully.

Yesterday's meeting was critical because this was really the first opportunity the participants had to get together and really define the parameters for this multi-center study, which is essentially what this is. And it's important to remember that we were really defining how to do this study, that we're not nearly at the point of collecting data in any amount sufficient to provide meaningful information to this committee. We don't anticipate that that data will be available for many months.

We're really at the stages of defining how to report data, which data to report, how it would be analyzed, and that was really the function of yesterday's meeting, a critical part of the meeting. So to that end, I will tell you some of the things we were able to accomplish.

First, we came to agreement on a form for Web-based reporting to make it easy for the 29 centers to report the data being requested. The second, and very important, was that we needed to be sure that we were reporting apples and apples. So we had to come up with a common definition of what is total inventory, what is exported or shipped, should we be including autologous and directed, how do we count neonatal units. So we discussed that and came to agreement on what standard definitions would be so that the data that we are getting from each of the centers would be standardized.

Secondly, there was a recognition that assessing only the inventory in the hospital was probably not the complete answer of what we were looking for in trying to assess the impact of blood supply in the hospital level. So we realized we needed some information that more directly monitored impact on patient care. And we actually came up with quite a long list of potential candidates, including how often a transfusion might be delayed or cancelled, how often surgery might be delayed or cancelled, how often Rh-positive red cells would be used in an Rh-negative recipient, how often physician consultations were needed for shortages. So we came up with a list of parameters or endpoints that we would report that really assess what are we having to do with the transfusion service level to effect an impact of blood shortages.

We also recognized there was a need for a standard operating procedure for this ongoing study, which tells you how early we are in the process, that we really don't have yet a standard operating procedure in which we can codify all the definitions and reporting requirements that we discussed yesterday.

On a very positive note, many centers, I think 25 of 29, have already begun reporting data. The definitions that we came up with and the data reported so far will be useful and still we'll be able to put those in the standardized format. But we don't anticipate having a meaningful amount of data--I want to emphasize that--for many months.

The office of the Department of Health and Human Services has been extremely helpful in beginning the coordination of this, and we in the hospitals are extremely thankful to have the opportunity to participate in what I think is a critical part of assessing the impact of any decisions that might impact the adequacy of the blood supply.

I'd be happy to take questions if there are any questions.

DR. CAPLAN: Questions? Karen?

MS. LIPTON: Can you let us know or do you have an approximate idea of what proportion of blood being transfused is represented by those 29 hospitals?

DR. TRIULZI: Okay. I did a quick calculation of this because this came up yesterday, and the Department of Health and Human Services office, in their wisdom, included the three centralized transfusion services that are in the country--Pittsburgh, Seattle, and Tampa. Just between those sites alone is in excess of a quarter of a million units. Then there's the other 25 hospitals participating, and one of the tools that we realized we needed is for each site to submit to the office a profile of the size of those sites. But using a rough estimate, I would say it's probably going to be about a million units, or about 10 percent of the blood supply, in that range.

DR. CAPLAN: Rick?

DR. DAVEY: Darrell, just following up on that point, the 25 hospitals, do they represent a range of types of facilities--pediatric facilities, small hospitals, academic centers, et cetera?

DR. TRIULZI: They appear to be all represented, although we don't have the specific data on each of those, which is why we want to have a profile of each of the hospitals participating. But we have cities from as small as Bismarck to New York City represented.

There's also geographic representation from the four regions intentionally, so there's Northeast, South, Midwest, and West Coast. And there was an attempt to include both university and private hospitals.

DR. CAPLAN: Colonel Fitzpatrick?

COL. FITZPATRICK: Mike Fitzpatrick. What's the mix of suppliers? About what percentage of blood is going to come from Red Cross and what from non-Red Cross?

DR. TRIULZI: I don't think we know that yet. We'll have to get that from the profile information.

DR. CAPLAN: Harvey?

DR. KLEIN: Yes, Harvey Klein. I have two questions for you, Darrell. One, I'm sure that the committee, in following up on Rick's question, would like to have some kind of analysis as to whether the sample of hospitals selected in any way represents the country? And perhaps by the next meeting, people with survey expertise could tell us whether this is a representative sample for a sentinel-type study.

And the second question is whether or not the suppliers--we didn't hear any supply data, but hospital data, except for the three centralized transfusion services--whether the suppliers will actually be in the same areas as the sentinel hospitals so we can kind of match collection and distribution usage.

DR. TRIULZI: I'll address both points. The first one, regarding the representativeness of these hospitals, I think is why we need to collect the profile data, and I'm sure the office would be happy to present that to this committee, and the committee could give its view on whether that's representative or not.

Regarding the blood center issues, we currently are not structured or don't have a mechanism to collect the data on inventory levels at the blood center at the same time that we're collecting inventory levels at the hospital, although I have talked with my colleagues at Puget Sound and at the institute in Pittsburgh where, because we're centralized, it's easy access to that data. So that we will collect the data at our own two sites, which does represent about 30 percent of the transfusions for the entire group so that there will be some effort to collect that data, but currently not a plan to collect it on a systematic basis.

DR. CAPLAN: Mike?

DR. BUSCH: Darrell, obviously this is a very rapid attempt to bring this program online, I think in great part because the hope was that it could be in place before the expanded CJD deferrals kicked in so you could actually attempt to measure the impact.

You mentioned that there were sites that were accruing some data. Do you think there will be a capacity to look at the impact over the next few months of the expanded deferral?

DR. TRIULZI: If the deferral were to go in in mid-September, we probably would not have more than four to six weeks of baseline data. I can't answer whether that would be a statistically valid sample. My tendency would be to say no. So that is unfortunate. I think that still the office did a tremendous job just to get people to start reporting, even by August 1st. But I think we probably will not even know the first six-week data in a meaningful manner until long after the CJD deferrals change.

DR. CAPLAN: Ed?

DR. GOMPERTS: This is Ed Gomperts. Could you give a brief overview as to how the data is actually going to be managed?

DR. TRIULZI: There's a Web-based reporting form, and the data will be collected by the Department of Health and Human Services office. They will be creating an aggregated spread sheet for the data. We did talk about how to express the data, and we at least came to the following decisions: one, that it needed to be expressed as aggregate data; that participants were comfortable with expressing it as regional data, breaking it down by Northeast, South, Midwest, West; and there was also a consensus that it would be helpful to present the data as the centralized transfusion community's as one data set versus the single hospitals as another data set.

That is as far as we've gotten at this point, and I would say that we really don't have consensus on whether the endpoint data that we want to discuss, for instance, daily inventory over the amount of units transfused, so we need more discussion on how to analyze the data.

DR. CAPLAN: I have two questions. One, how will the data be available to those who want to see it? In other words, who can access it? And, two, has there been a discussion yet of kind of quality control audit to make sure that the reporting stays on target from the different places?

DR. TRIULZI: We didn't discuss the latter point. The first point on how it would be expressed, Dr. Nightingale did say that this would be available on the Web site for public view. Again, the data is not at a stage at which it's ready to be--any of it is ready to be put on a Web site.

DR. CAPLAN: By the way, the Chair is going to take a few questions out of the audience, if there are any. So if you do want to get in after the committee has asked some questions, that will be an opportunity for you.

Ron?

DR. GILCHER: Darrell, perhaps I missed this, but how are the inventory levels established for the hospitals? As an example, in our region this is a joint effort between the blood center and the hospital, which includes medical oversight, to establish what the inventory level for the hospital should be. And so I'm interested in whether there really is medical input into the establishment of the actual inventory levels.

DR. TRIULZI: The study does not intend to define what the inventory level should be. The institutions are defining their own inventory levels, and we're taking basically a daily picture, a daily snapshot of what the inventory levels in the hospital are on each consecutive day. So that it's really at the discretion of that institution to vary their inventory levels as they see fit.

So we're really not trying to affect the endpoint. We're just taking a picture of it.

DR. CAPLAN: Jane?

DR. PILIAVIN: Jane Piliavin. I'm just wondering, in terms of the sample of hospitals, why an actual representative sample wasn't attempted. I mean, now you're saying after the fact that you're going to go and see whether the group of hospitals you've got appears to be representative. I mean, there are in place some rather sophisticated sampling techniques that could have been employed in order to make sure of that in the first place. And I was wondering why this was not done.

DR. TRIULZI: There's a couple factors that Steve Nightingale brought up that go into that. One was the speed at which we were or he was requested to bring this project together.

The second was the financial limitations of the number of sites that could participate with the amount of money available. The sites participating are receiving less than $5,000 per site to participate in this study. So very small sum of funds for each of these sites.

The third is that I think there was an effort to obtain as representative a sample as could be done on first blush. And it may be representative. We don't know yet until we get the profile data back from each of the sites.

And I think Steve also made it clear that if there was a need to change or alter or add sites, that they would be amenable to that possibility.

CAPT. McMURTRY: I'd like to comment on that for just a minute, if I may, also, at the risk of wading in over my head here. We were not really looking for a representative sample of hospitals. We were looking for sentinel sites, and there's a difference. We were unclear and still are unclear what would constitute a representative sample, if you're talking about a scientifically representative sample, of hospitals. That's not something we were looking for.

The information that we're trying to get from our sites now are more descriptive terms, once again, not in an effort to be able to say, well, we represent hospitals, here's what they are, but just so that we can describe who our hospitals--what our hospitals are. That's been part of the problem that we have seen, is that we haven't had any transparency. We don't know where we're getting our data. And so the request to find out who these sites are is just an attempt to be a little more transparent than we've been.

DR. CAPLAN: Karen?

MS. LIPTON: Karen Lipton. I still am concerned, though, even in using the word "sentinel," that I don't think that we have an appreciation that these are truly sentinel sites. And I guess as a committee member I understand there are financial limitations, but I would hate to see us set policy on bad data. And I think that we should make some effort to ensure that we do move towards having either a representative sample or understand what we mean by sentinel, if it's valid, because, I mean, I do feel as if the study is kind of going ahead, and I think it's good to get all this information. I just don't know if we're going to know what we're looking at when we finally get the data.

DR. CAPLAN: Ron?

DR. GILCHER: Ron Gilcher. As a follow-up to the comment that I made earlier, one of my concerns is that if the inventory level that's established for a particular hospital is invalid--and what I mean by that--and let me give you as an example. In terms of shortages or chronic shortages, what hospitals tend to do--you know this and I know this as well--they will increase what they call their inventory level because they begin to hoard. They're worried that the central supplier doesn't have enough reserve.

Now, I've taken a long time in my own system to try to weed that out, and so what we can have is really an invalid inventory system in the hospital which would make it appear that their needs are not being met, when, in fact, they are. So we have to look very carefully at what they're transfusing as well.

Can you comment on that?

DR. TRIULZI: I think that goes to the point that Dr. Klein made, which is to look at the transfusion services in a vacuum may not give you the whole story.

Now, I think there are a lot of communities that, for instance, in centralized one where we would be able to know whether that's happening. But that would be an argument to get the data on the blood center side to assess whether that was happening.

DR. CAPLAN: Could you tell us a little bit about the supervisory structure for the monitoring? In other words, when we're offering some ideas about making sure that this is going to work as a sentinel and exactly how to keep tabs on the overall supply side, who is it that is in charge, or what's the structure of the contracting--of the blood demand contractors? Who's riding herd on this?

DR. TRIULZI: I think that's a Captain McMurtry question.

DR. CAPLAN: What I mean is when you were there yesterday, to put it into slightly more English, did you come up with a steering committee?

DR. TRIULZI: No.

DR. CAPLAN: No. Okay. That would probably be good.

Mike?

DR. BUSCH: Yes, to me, more important than the representativeness, et cetera, I mean, you've got 10 percent of the recipient pool you think representing 30-some hospitals. It's a pretty good sample. To me, much more important than sort of what it is is the continuity and the trend over time piece of this. And in that context, I'm wondering what's the duration of these contracts? What's the anticipated, you know, long-term stability of the study?

And then this whole issue of the interface between the blood centers and the hospitals, one concern--I hate to raise it, but if blood centers know that a hospital in their region is one of these sites, what's to prevent them from oversupplying that blood center? And how can one keep, you know, stable processes in terms of inventory management both at the transfusion service and the hospitals in place or understand any changes so that trends over time can be understood?

DR. TRIULZI: Those are good points, Mike. First, the contracts allow for currently data collection through December. We did discuss yesterday already submitting budgets for continuing to collect data through the fall of 2002, so we'd have at least a full year of data. So it's anticipated that we will be collecting data through October 2002.

Your question about bias is one that we did discuss, that blood centers may treat a hospital differently knowing that they're in the spotlight, so to speak. And we didn't discuss how to either assess it or prevent it, other than to recognize that that may be a potential limitation in interpreting the data.

COL. FITZPATRICK: Mike Fitzpatrick. So as we--if we concur that these are sentinel areas and that you then discover a sentinel event, was there discussion by the group as to what intervention can occur when the data indicates a sentinel event has occurred and there is a critical supply shortage in a region or somewhere else? Or is that just left up to industry to handle?

DR. TRIULZI: I would say that's the next level of sophistication. We're right now just trying to define the parameters for data collection. How to use the data to impact decisionmaking either at the public health level or even at the regional level is something that we have not addressed yet.

DR. CAPLAN: Make sure we identify ourselves.

DR. SANDLER: My name is Jerry Sandler. I'm the Director of Transfusion Service at Georgetown University Hospital, and I'd just like to offer perhaps a slightly different perspective from the hospitals that are participating.

We are very concerned that certain initiatives may impact on the shortage of blood at the hospital level over the next several months. While I think we all participating in this recognize that the quality of data would be very much improved if we could get the community blood centers to report their data, we see this as a hotline to the Office of the Secretary of Health and Human Services at a critical time. If we run short of blood in October, November, and December, this is a way that we can get to someone and say this is a serious problem. The quality of data may not be as terrific, but the ability to communicate that message is very important to us.

Thank you.

DR. CAPLAN: Keith?

DR. HOOTS: Keith Hoots. Your comment raises a question that I already kind of tried to formulate a little bit in terms of the response to the data. One of the things I guess that we've been concerned about as a committee over time is how to, obviously, work nationally to increase donations. And this could be either a very positive impact if it's used very carefully, but it could also be used as a crisis intervention repetitively and actually work to the detriment of the overall continual donation supply for the country.

I think one of the things we probably ought to make sure that, as we monitor not only the data, but the way that response to the data is--how it impacts overall strategies must be part of this schema. Otherwise, I'm a little concerned that in times of shortage, if we whip the horse repeatedly and then it gets to be habitual, sort of up and down in terms of crisis management instead of a more continuous developmental strategy for enhanced donations, that we may actually--it may actually work to everyone's long-term detriment. So I think we need to keep that in mind as well, particularly in light of the issues that were raised about how representative, particularly in the short term--maybe over time we'll be able to refine this sentinel event for the entire organization of data collecting centers and say, well, yes, it is representative so, therefore, we really can respond rapidly to it.

On the other hand, in the short term, we don't know that, so if we use it--I think we all want more people to donate, but we've got to be very careful how we do it.

DR. CAPLAN: One comment I would like to ask about, too, is it seems to me, in listening over the years to blood banks, they know a lot about their inventory and what they've got and what's around. Part of the way to power up this sentinel reporting, then, is not so much to make them do something they don't already do. It's to standardize the reporting.

So it seems to me that if we could get agreement from these 29 centers and then perhaps make plans to disseminate this style of reporting standardization and so forth, it wouldn't be too hard to recruit lots of other places very fast to come on line and report in the same way. It's not that no one knows what's going on out there. They all know individually. So this seems to me to be a standardization issue, and I'm sure it hasn't come up yet, given the speed with which we're trying to pull this together. But I would urge that the group that is not running this thing yet, the non-existent steering committee, but soon will deliberate about what they can do to encourage standardized reporting on this.

I understand that each hospital has its own needs and habits and customs, but if this just became the state of the art, I suspect going up from 10 percent to a bigger number would be pretty easy.

Jane?

DR. PILIAVIN: I was quite concerned to hear that this is only solid through December. You don't have to be a blood banker or a hospital administrator to know that there are seasonal variations in blood need--I mean in blood usage and in blood availability, particularly the summer problem, the Christmas problem, and so on.

Again, getting back to sampling issues, which as a social scientist is about the only thing I have to contribute here, sampling isn't just in terms--I mean, you have to think about sampling not just in terms of the reporting units, but also in terms of time periods. And unless you have a good sample of both, you don't know what you've got.

CAPT. McMURTRY: I'd like to talk for a minute about the duration of this. I had a conversation with Dr. Chamberland yesterday about how long we were going to be able to monitor to blood supply, and to say that this is firm only through December is true because of budgetary limitations you can only count the eggs that are actually in your basket right now.

As a young man in Texas messing with the Texas Legislature, I learned that there's nothing more permanent than a temporary tax.

I think that this is going to be similar to that. I would hate to be the bureaucrat that says, no, we're not going to give you money to monitor the nation's blood supply any longer.

So, yes, we are only firm through December, but I'm not uncomfortable with that.

DR. McCARTHY: Dr. McCarthy from Indiana. I agree with most everything that's been said, but I just want to say from a large transfusion service in the Midwest, having been in this now for 30 years, we've all seen the biannual shortages. And I think what perplexes many of the recipients is that why is there a shortage in one area and not another. So I think this is the first step in gathering facts. And the facts may not be enough, but I think the facts should be friendly, should be regarded as friendly data, and not be threatening to some people, and that we may get more facts. But I think it will be good to know if there's a shortage in one area and another, and what can be done with the facts is a different echelon of decisionmaking.

But I think this is a good start, and it's been long needed. This thing really started with the American Blood Commission back in the early '70s when the disparate problems about paid donors and labeling and so forth started out. And I think this is just a last aftermath, if you would, of it.

DR. CAPLAN: I hope that that's partly wrong, that it didn't take 25 years to get the thing rolling. But I suspect you may be right.

The other question I have--and I didn't hear much discussion of this when we were sort of getting ready to launch this effort--that the Chair made clear is I'm very happy that we're going to have this. I think we've got to make sure that it's done right and well, but moving on this at a time when there's a lot of discussion about adequacy of the blood supply, what the impact is--not only of CJD but, you know, we've been in this group to listen to leuko-depletion, NAT testing, and many, many other safety measures. So we're always fishing to try and find out what the impact on supply--it would be wonderful to have--it will be wonderful to have some standardized information on that.

That said, did it come up as to what other countries were doing in terms of monitoring their blood supply? I don't know if Mary's encountered that either, Canada or U.K. Is there any other--do you know, Mike? I'm just curious what the--has anybody standardized out to keep an eye on overall--I mean, in a sentinel-type fashion?

DR. TRIULZI: We didn't discuss it, and I'm not aware of it being done.

DR. CAPLAN: Do you know, Jerry?

DR. SANDLER: I directed a blood bank in the State of Israel for six years. Every morning we made an inventory report, and we sent it to the central one blood collector in the country, and the concept was that the blood was given by the people of the country. It was on consignment to the hospital. And for reasons that we never knew, blood would be moved in and out of our hospital inventory which we had established to other locations. And, obviously, that's a goal for your committee to achieve that in the United States.

DR. BUSCH: There's a number of countries that have introduced hemo-vigilant systems, but to my knowledge, they're mostly focused on risk assessment. But just in the last day, I saw--I think it was distributed by the Web, but Ireland is in the process of reversing their British deferral policy because of unavailable blood supplies for patients right now.

COL. FITZPATRICK: We deal with most of the NATO member countries in Europe, and because most other countries have a national blood system as opposed to a civil blood system, they have some ability to assess their inventory availability depending on primarily the size of the country. Germany is regional, so you had to go to each region to find out an inventory. Belgium is national, and you could go to one point to find out an inventory. France was regional. So it varied from country to country.

But because of the nationalization of the system, there were single points of contact or small numbers of points of contacts we could go to within those countries to determine if they have inventory that could assist us during a war.

DR. CAPLAN: So the steps here may be--I don't know to what extent DOD or Armed Services is free to share its own inventory templates. But either there or in other situations like Ireland or Israel, it would be useful, again, if we're talking standardization and keeping an eye on things, to at least draw upon other experiences and maybe push a little bit toward some international standardization as well.

John?

DR. PENNER: I'd like to just return very quickly to what Mike Busch and Ron Gilcher brought up and, that is, the act of examining the data can change the data. It's the Schrediger-Katt (ph) association.

Is there any attempt to do spot checks with, say, other centers that are collecting blood to assure that the data you're getting is not being manipulated by asking for the data?

DR. NIGHTINGALE: I think it might be easier for me to give a fuller answer after I give my delayed presentation. But in the--immediately, it is clear that sentinels are just that. Sentinels, when they're identified--if you have an outpost somewhere that you're guarding a frontier, you anticipate that the attacking army is not going to come straight at the sentinel.

We've tried to correct for that in two ways. I think the first way is by geographic distribution, and the second way is by making this an evolving process. You've got to start somewhere. We're starting with a six-month contract where we value--we're re-evaluating it right now as we're up and running. But I think that what we're--the basic question we ask is: Is something going wrong in our sentinel sites? Yes or no. Do we have reports that something is going wrong outside of our sentinel sites? Yes and no. Depending on the answer to those two questions, we go in one of four different ways.

If nobody is having problems, we go home. If somebody outside the sentinel sites are having problems, we ask why outside and not inside. There may be a simple answer. There may not be. If both are having--if it's the reverse, well, we've got an early-warning system. And if both are having problems, we ask ourselves why didn't we pick this up earlier?

DR. PENNER: I'm just reflecting on some studies we did on platelet utilization, and our control group, it seemed, once they realized they were a control group, suddenly started improving on their use of platelet in the hospitals. So that sort of game playing is obviously part of what we do. So sometimes what we were able to do is just spot check a few other institutions and find maybe the data we were getting was not really true.

DR. NIGHTINGALE: Well, I believe that--well, not believe. I recall Dr. Linden saying yesterday that New York State planned some activities of its own. That would be an obvious check against us. I haven't talked in any detail with Dr. Linden, but that statement was made yesterday, and we've had conversations with New York State about intensifying surveillance in that state at this time.

DR. CAPLAN: All right. Let me thank you for that presentation.

Do you need a break in order to get to your talk, or are you set up to go?

DR. NIGHTINGALE: Not at all.

DR. CAPLAN: All right. Then we'll heard from--

DR. NIGHTINGALE: I guess the question is: Does anybody else need a break?

DR. CAPLAN: Not yet. I know.

[Laughter.]

DR. CAPLAN: I can tell.

x The next speaker is Stephen Nightingale, who is going to talk to us about the current state of monitoring and perhaps fill in a few of the details from yesterday's meeting. You know, I was up on the Web today, and I saw this article in the Baltimore Sun. And some of you may have seen some other press coverage on this move toward monitoring the blood supply. I just want to, again, make clear we're going to probably spend some time haggling and niggling a little bit about how to do this.

The general push to do it is something that I think the committee is absolutely supportive of and wants to see all organizations cooperate with. It's just vital if we're going to be in situations where we might have steps taken that impact the blood supply, that we have some ability to know what's going on. So at some point, it may be wise for us to simply pass a resolution or motion to that effect. I'm just suggesting that, but that may be something you want to say this morning.

DR. NIGHTINGALE: Thank you very much, and I apologize for the lateness of my arrival.

Dr. Penner?

DR. PENNER: Is that on?

DR. NIGHTINGALE: Is the microphone on? Can everyone hear me in this small, and not uncrowded, room? Yes, I think so.

Where I would like to start is with a brief review of how we got to where we are today. And how we did so really started out in April of 1998 and, before that, in December of 1997. In December of 1997, we became aware, a couple of months after the fact, I might add, that there was a very acute shortage of plasma derivatives in the country.

The Advisory Committee held its third meeting in April of 1998 on this topic and on what the government might do to help alleviate it and, as importantly, to prevent its recurrence in the future. One of the recommendations that came out of that meeting, and one of the responses that came from it as well, was a monitoring of the supply of plasma derivatives in the United States. It was instituted by the industry group that was then known as the IPPIA and now known as the Plasma Proteins Therapeutic Association. I hope that I got that correct.

The monitoring of the supply began in October of that year and was really fully functional in January of 1999, and it had several benefits, but I think as great a benefit as any was the degree of confidence that it provided to the user community. There are many representatives in the community, and they could acknowledge or question that, but I think most would acknowledge that.

It also provided us with some useful information. What we got from that, and continue to get on a monthly, and in fact now every fortnightly basis, is both a measure of the inventory that is on the manufacturers' shelves at a specified time and the distribution of the product over that month. So we get a ratio of inventory on top to distribution on the bottom. The higher the ratio, particularly if it's greater than one, the better, at least subjectively, we feel things are because that means you have, at least, a 1-month supply of product. For products in short supply, that ratio has been well below one, particularly for recombinant factor 8 in recent months.

The second event that got us to where we are today concerns monitoring of the blood supply itself, as opposed to the plasma derivatives, and that came to our attention in February of 1999, when the National Data Blood Resource Center published a summary of its biannual survey that included a projection the supply would fall below demand for blood by the year 2000. We held a meeting on that, had made a variety of recommendations.

And after that meeting, the surgeon general, who is then also the assistant secretary of Health, Dr. Satcher, convened an intergovernment panel that Dr. Epstein led to see how we could go beyond the recommendation of the Advisory Committee. One of the recommendations that came out of that was the monitoring of the blood supply. And, in fact, in October of that year, a contract was let to the National Blood Data Resource Center to monitor the supply from a representative sample of blood establishments throughout the United States, and that contract, in fact, continues to this day as well. We get supply, we get inventory and we get distribution from that.

The review of the monitoring efforts had picked up steam really in January of 2001. Responsibility for that activity was transferred to the Office of Public Health and Science, which is the Office of the Assistant Secretary of Health, for whom I work. And in April, we had a full review of that, on April 20th, here at the Advisory Committee.

The basic outcome of the review was that we found both provided very useful information, but there are issues for both about the timeliness of the report. We get the information somewhere between 4 to 8 weeks after the events that are being counted occurs. And there is a question, if you get monthly or even bimonthly data, what constitutes a trend. So timeliness was an issue.

I think the second, and perhaps most important issue, was that both for the monitoring of the blood and plasma supplies, the question has repeatedly arisen of the relationship between supply of blood or plasma products and demand for those products.

And, finally, there is also an issue that has been raised about the transparency of the review. To the extent that these are designed to provide information to the public, the more transparent, the more visible the process, the more rapid the distribution of the results, the more effective they will serve that purpose. And also for the people who are using the information, the more easily they can verify the sources on which the information is based, the more confidence the government, for example, or any other interested party will have in using the data

A reason why we started with supply, rather than demand, was very simple. Supply is easier to measure than demand. We've had many conferences about that, both public and private. I might add that Dr. Haas, a member of our committee, has been extremely helpful, certainly to me and I think to many others, because of his expertise in this area. But the original way that we measured demand was the ratio of inventory to release at the producer level. And for the reasons that I just mentioned, we felt that that was suboptimal. We didn't necessarily have a better idea.

The second way of measuring demand is the classical economics way of measuring the price of the product. For a variety of reasons that Dr. Haas reviewed at the April Advisory Committee meeting, price is not the best way to do that. You can have a product that is in demand, but still be cheap, depending on market factors, and you can have the converse as well.

The second is inventory--the inventory release at the distributor level. This is not a simple market. The distributors play an important role, and one of the pivotal thoughts that we've had is that the transfusion services in the large hospitals, in the sentinels, if you will, might, in fact, be functioning as distributors on behalf of individual patients, and that's pretty much where we are.

The inventory release at the end-user level, of course, is the optimal one, but there are a lot of problems with measuring that, not the least of which is the privacy of the individual. Others I have a question here. We, by no means, exhausted the topic, but at the moment this is where we are intellectually at the distributor level.

And how should inventory be measured is another question that is very much on the table right now. In the very preliminary data that I'm going to show you, we're just measuring inventory over daily use. For example, if one blood centers that supplies, for example, all of greater Seattle has--let's give a number--500 units, and one hospital supplies, say, half of Bismarck, North Dakota, has 50 units, you've got to factor the difference in the size of the institutions.

There is, for those of you who follow or perhaps lead the business section of the newspapers, you'd be familiar with trailing averages. This is a way of smoothing out daily fluctuations, but there's also a way of filtering out information that you may not want filtered out. So we are still in the process of trying to figure out, as we go along, whether or not it's appropriate to use trailing averages.

Very briefly, what it would appear use of a trailing average does in the data that we've collected so far, is to push the curve a couple of days or 3 days to the right, depending on how long the trail is. If there's a blip, the trailing average filters the blip out, so it takes a couple of days before the curve shifts. So, to some extent, the use of a trailing average negates or at least counteracts the daily collection of data. This is an ongoing topic that we'll be working on.

The third is a definition of a trend. In this business, we're also going to have to learn this as we go on. Looking at data over a year, it's clear, for example, from the NBDRC data, that in the past couple of years, both transfusions and blood collected have gone up. There are trends, if the curve is significant, if the confidence interval at Point A and Point B do not overlap you have a trend.

Of course, when you're looking at data every day, you're sampling an infinite number of times, and conventional statistics are not particularly helpful. Bayesian statistics are what you make of them. So we will need a definition of a trend, which we don't currently have. Certainly, there are weekly, as well as seasonal cycles, in the data that we need to factor up, but since we haven't looked at the data yet, we're going to have to find that out as we go along.

What happens when you have a transient effect, for example, in the plasma business, when a manufacturer shuts a line down for preventive maintenance, and what happens when you have a nonlinear event, for example, when there's a change in the number of manufacturers or a major change in policy, are things that we also need to learn as we refine our usage--plus, an adequate inventory.

At the moment, then, we're looking at a variety of ways of measuring the inventory at the distributor level. And for our benchmarks, initially, we are asking the question, what is the inventory at a time when there's consensus that there is an adequate supply at least nationally. Much of what went on yesterday, and, gosh, I hope I'm not going to repeat more than a couple of words Dr. Triulzi gave to you, is we got a number of very, very valuable suggestions for how to define an event that would make us think that the inventory was inadequate to meet demand, and we're going to pursue that very aggressively over the next month.

There are questions, of course, whether or not an event is a random or a sentinel event. If we had guidelines, well, hey, the stock would be shorter. What's an adequate sample? This was something that was raised several times by several speakers yesterday. When you get started with a sentinel system, I believe Dr. Chamberland said you, when you start to use the example of monitoring the AIDS epidemic in the early days in New York City, I think there's two things that you do is, one, you start out with the best information available, and number two, with experience and with as much of the scientific method as you possibly can, you try to make it as much better as you can, as quickly as you can. I think our current intent is to bootstrap, rather than to restart the calendar every 6 months. I don't think we really have time to do that.

To the extent that a sample is representative of an entire population, well, that's very difficult, unless you know exactly what the population that you want to sample is, and this is the long answer to Dr. Penner's question. I mean, clearly, what we want to do is to put our sensors at the most sensitive parts of the system. By putting the sensor there, at least in the part of the government that I work for, you attract attention to the sensor immediately. In other parts of the government, apparently you attract attention to it sooner or later.

This is a problem that we're going to have to work with. I think that the way to deal with it is to maximize public information rather than try to put it under a blanket and hope nobody finds it for 90 days. I do not think that's the right way to deal with a sentinel system.

There being 5,000 acute hospitals in the United States, can an N of 30, which is our budgetary limit right now, truly be reflective of the population? And the answer is probably not. The next question is, if it isn't perfect, should you do it anyway? And the answer is, yes, at least that's our answer right now.

I mentioned about bias from disclosure of participation. We're going to learn some things from this, at least if we conduct the study correctly.

John, would you like to make any further comments?

DR. PENNER: I think you have to start someplace.

DR. NIGHTINGALE: I'll tell you where we're going to start.

DR. PENNER: And then you can modify it.

DR. NIGHTINGALE: That segues me into someplace. Where did we start? These are the contracts--let me say that while I'm not sure that all of the contracts have been signed, we certainly have handshakes, and we have data coming in from 25 of the 29 sites so far.

In the Northeast, we have two hospitals in Boston, St. Elizabeth's and Brigham and Women's--hey, we're grateful for everybody--the Institute of Transfusion Medicine in Pittsburgh, in--oh, my God, I did these slides this morning, and there's only two of the four here, and I apologize--Mt. Sinai and Columbian Presbyterian are in Manhattan. There is more to New York than Manhattan, and Maimonides Hospital in Brooklyn and Jamaica Hospital are also participants, and my apologies to them. I knew there was something wrong with this slide. In Washington, there are two, the Georgetown University Hospital and the Washington Hospital Center.

One of the initial thoughts, as we started this process, was to have two hospitals in a particular geographic region, and it was not particularly important to us which they were, but we didn't want carbon copies of each other. If we didn't want sort of the University of North Manhattan and the University of Midtown Manhattan were not what we were looking for, and we do have Sinai and Columbian Presbyterian there, but I don't think they would fit that description too precisely.

In the South, we have Emory and Grady. I think Emory and Grady pretty much mirror what the idea was initially. And as you can see as the distribution changes, you'll see how our distribution matured, as we started making phone calls and actually listening to the people on the other end of the phone.

We have Mt. Sinai Hospital in Miami, and we have the Florida Blood Services from Tampa-St. Pete. You can't sample every small town, and Mobile is, by no means, a small town, but one of the concerns was are you just going for the big cities, and the answer is, yes, predominantly, because that, to paraphrase Willie Sutton, is where the money is, but we tried not to go just there. The Ochsner Clinic, Baylor, and Parkland in Dallas are the two other contractors that we have in the South.

In the Midwest, Indiana. Leo, is that the correct hospital?

DR. McCARTHY: It is.

DR. NIGHTINGALE: Thank you.

Northwestern, University of Illinois, Chicago Circle, the University of Iowa, the University Medical Center in Minneapolis, and St. Alexias Hospital in Bismarck. We will make allowances for the weather in North Dakota in the winter when we analyze the data, but it is not without interest, by the way. There was a reason for Lackington, North Dakota, to at least identify, albeit as a sentinel site, what happens in the part of the country when it gets rough in the winter.

And in the West, we have the Puget Blood Center in Seattle. Puget, Pittsburgh, and Tampa-St. Pete, of course, are three communitywide networks for blood banks. And Denver General is now Denver Medical Center and the University of Colorado. We have Cedar-Sinai, and Harbor, and University of Arizona, Tucson. Those complete the list of our 29 contractors.

Preliminary data. Presented this yesterday in what really is a public session. There are caveats here. Number one, we only have, so far, in the data bank, 25 of the 29 sites. This is not final data. We only have 537 total days of observation. Yes, that's a lot. This is a big system. The database has been unscrubbed. A technical term is have we gone back and verified every entry for accuracy? Have we done our best? Yes. Have we scanned for obvious errors? Yes. Would this be ready to submit to FDA for a new drug application? No.

And for those reasons, the only slant that I have on these slides, if you look at the placements of the Bermuda Triangle, and they say, "Not for use in navigation," this is not for use in navigation, but it is an indication of where we are. And I am also bringing this out so that we can get feedback from you. If you don't know where we are right now, you can't help us get better, and that's the reason for going public at this point.

This is total inventory of daily distribution of red cells. What this is is the average inventory over the amount that went out, that was either transfused or it was exported to another site. For example, if you're in Bismarck, and you have blood, and Fargo doesn't because it snowed in Fargo, and you send some blood to Fargo, that's blood that went out the door. And it also includes any blood that might have outdated, although there's not a lot of blood being outdated these days, and I think that that is an observation that we need to follow up on.

If we're flush with blood, you'd expect a percentage of it to be outdated. We really have to think about what that percentage might be, but right now there is not a lot of blood that is being outdated. And there's a modest amount, it would appear, that the exporting is in line with traditional numbers, maybe about 1 out of 7, 1 out of 6 is being exported, for the median. This is not an average, this is a median. Half are below and half are above. I'm giving this number because this data is unscrubbed.

Yes, Jane?

DR. PILIAVIN: What are the units here? I mean, 7.4 what?

DR. NIGHTINGALE: Actually, these are pure numbers, but what it is is this is the number of bags on the shelf divided by the number of bags that were released by the system in 24 hours. So what you are looking at could be construed as day's worth of inventory. Just like, if you are selling cars, and you have 900 cars on the lot, and you sell 100 cars a day, you have 9 days' worth of inventory. That is a way of looking at these numbers, and I think--

DR. PILIAVIN: So it's basically days.

DR. NIGHTINGALE: Yes.

DR. PILIAVIN: Thanks.

DR. NIGHTINGALE: That's right, it's basically days.

I'm not going to show you a complete analysis because I don't have comfort in the data sufficient for a complete analysis. I'm showing you where we are right now. If you look at two of the segments, the Northeast is a little bit above the median, the South is a little bit below the median, but we're tracking in the same area. And if you just look at the data that was collected for really the first week where we have 25 out of 29 reports in the computer right now, the median is very close to the overall median for the data. So I think we're up and running.

This is a number that if you just look at the total inventory over the amount transfused, you'll see you actually have a 9-day inventory because one of your questions is don't let any blood outdate, and the second is don't export blood. This will give you a measure of the slack in the system, but it remains for further observation to see how good this measure of slack will be. But, certainly, if anybody has any comments about this measure, I would be more than happy to receive them.

DR. DAVEY: Steve?

DR. NIGHTINGALE: Rick?

DR. DAVEY: One thing, obviously, I think you are aware of that the inventories for different blood groups will vary considerably.

DR. NIGHTINGALE: Dr. Davey has anticipated my next slide. We had dinner together, and I guess I discussed that because here we have for A-positive--

Don't worry. I'm not going to go over all of them, I'm going to go over most of them, and I'm sure the blood bankers were interested in this.

Again, while these probably correlate with expectations, this is still preliminary data. For all observations, as opposed to 7.4, there's a little bit more A-positive in the system than there is for O-positive. Is that a meaningful difference? I do not believe so on the basis of the data that we have right now. But, again, this is for discussion and for input from the public about how we should anticipate it.

And, again, I've compared the other areas that we're tracking pretty much the same. The median for the past week is a little bit higher. Is this statistically significant? My expectation is no, but my warning is that this is not data on which you want to put statistical test. This is an exploratory data analysis that I'm presenting to you.

So a little bit over for A-positive. B-positive, there's a lot of B-positive in the system. Is it statistically significant? I don't think so, but it tracks about as well. AB-positive is the next one. There's not a lot of AB-positive in the system relative to the others, but what is "not a lot of"? There's reason: AB-positive is blood that outdates. And overall you can see for the last weeks it pretty much tracks this.

The two items that you want to look at--and, again, I'm cautioning you not to look at it too deeply--are O-positive, where the amount of O-pos in the system would appear to be right about the same level as the mean, and there's not a whole lot of variation in there. And of such interest at the moment is O-negative inventory, which is actually a little bit high.

Now, one more time, this is a snapshot of preliminary data. This is being presented so that we can gain input from you, the committee, from you, the public, and from those who will have access to this presentation later. We want suggestions for how we can do this better. We do not feel that we have a template which we can simply plug in and monitor the blood supply in a manner that would suit everybody's purposes.

The platelets are a separate issue. Platelets, for those of us who are not blood bankers by training, which includes myself, outdate a whole lot faster than red cells, so there'd be a lot more turnover. For the random platelets for all else, it looks like we have about a 1.4-days' supply of platelets. Again, what that means depends on the experience of the user. The platelet market is much more active than the blood market. For the apheresis single-donor platelets, we have a number that is pretty much in the same line.

What we asked on our samples is please provide any comments on any actions that you took in response to finding that supply was inadequate to meet demand. Most of the comments so far about platelets, and I think the preliminary analysis would appear that this would be perhaps more closer-to-normal business practice than it is for blood because of the short half-life of the platelet.

I'm up to the last slide now, and this is the summary of the preliminary data that we have. Where we are with the monitoring system is we have a single-point estimate. We are just about to get daily averages. We will consult with our consultants, we will consult with the committee about how to analyze and to improve the collection of this data. We have a second round of contracts for the data that will be coming up for the first year and that I anticipate, after the second round, after we found out how much we've learned from a 1-year analysis of the data, we will be prepared to put out a formal contract pending approval of the Secretary, but I think that the support for continuing this process is pretty strong right now, and I don't expect that to change.

I am done. Ms. Lipton?

MS. LIPTON: Just a quick question. When you're doing inventory, and I think you mentioned this, but I couldn't quite recall, you are including directed autologous, and then what about special inventories, you know, for a red-cell antigen specific. Does this include everything? Are you tracking those separately?

DR. NIGHTINGALE: We had an extensive discussion of that yesterday afternoon. Unfortunately, I had to be out of the room when that discussion took place.

Ms. Wannamaker? Ginny? Ginny, would you like to comment on that question?

MS. WANNAMAKER: These were all issues that we knew that we had to discuss, and we wanted to discuss them together, and we did that yesterday. This preliminary data that Steve has has--it's not distinct as to whether or not there is--there is autologous and there is directed in these numbers. We did discuss yesterday how we're going to handle that so that our numbers may not reflect those later on. But this particular data has everything that people were collecting.

We had initially asked people to include those in. Some people, I believe, had done it. I'm not sure that all were. But we have had discussions about that, and we will be looking at this a little bit differently. But for those numbers, yes.

DR. CAPLAN: Rick?

DR. DAVEY: Steve, just to revisit a topic that the committee has already discussed, to some extent, this was the first time I think we had a listing of the hospitals that are participating, and I understand the necessity to have big hospitals. As you say, that's where the money is, and they need to be very clearly represented. But I was struck that there were no, for instance, military hospitals. There weren't very many, maybe except one, kind of community hospitals, like in our area, Sibley or Suburban, included. I know it's going to be tweaked, and you're thinking about it, but could you comment on that a little bit?

DR. NIGHTINGALE: Yes. I think I'd reiterate the comment that I made before I got up to the podium, is that it is not practical for us to put a monitoring in each of the 5,000 hospitals. I think that we started with a plan that was really based on geographic distribution, duplication within a metropolitan area, and a budget that was $113,000 for sites, and we busted that. We went up to--I had to stop when I hit $155,000. That was how it was taken.

So the plan is to see how this group performs. I think that the best way we'll see how the group performs is we will either receive or not receive reports, comments or, indeed, criticisms, just the point of having this talk, that you're not shining your light where the problem is. If we find that our group of sentinel hospitals is flush, say, over Thanksgiving, over Christmas, during a traditionally low period, and we're receiving reports from other institutions that, say, we're not flush, then we will revise our plans accordingly.

Short answer to your question is we started where we thought the money was most likely to be, and like any good investigator of money, we are prepared to react and respond to events as they reach our attention.

DR. PENNER: Steve, I wonder, as this data emerges and becomes public, if some hospitals may not use it as a standard, looking at the data and saying, "Well, we only have 5 days' worth. We better go out 7 days' worth. And those hospitals that have 10 or 12 might even decline in their amounts. But looking at this kind of sentinel group, they think maybe they should be in line with it. And I'm wondering whether that impact will really be good or bad.

DR. NIGHTINGALE: Once we have the data scrubbed with an N of 29, I think that we can give confidence intervals. And the question is I don't know whether the distributions will be normally or plus on--or uniformly distributed across an area, and I don't know what the variance will be.

What I do know it will have is a measure of what inventories were at a time when shortage events were reported. Remember, a shortage event at the moment is any action taken in response to discovery that supply was inadequate to meet demand.

We talked yesterday about making a typology, about classifying those different responses, and several of the members presented specific events that should be used, delaying elective surgery, for example. We will have a great deal of data, I think, within 90 days, I would hope, by which we could improve the reporting system. Honestly, two ways of doing this are to go into a dark room and figure it all out by yourselves or come in the bright lights and request comments like that, and we have chosen Plan B.

DR. CAPLAN: I've got Karen, Harvey, Jay.

MS. LIPTON: Steve, we've talked a number of times about some of the hospitals that aren't in the reporting and using them to see if there are some other problems. How are you going to get that set up. Will that be through the website, encouraging people to, if you're not within this range, and you know, if it's showing that these hospitals aren't having a problem, but you are, how are you going to--how is that going to work?

DR. NIGHTINGALE: Yet another hanging curve. I am, using forums such as this to publicize the existence of this system and to encourage individuals who wish to report, to do so in whatever format they like. For the record, I don't have it up here, but I could type it in, my address is 200 Independence Avenue, Southwest, Room 736, Washington, D.C. 20201. My e-mail is posted on our website, which is www.dhhs.gov/Blood Safety. And the direct line to my office is 202-690-5558. Thank you.

And I mean that very seriously. I think one of the ways we're going to find out about the shortage is simply going to be the number of events that are reported. If nobody is calling up, we're going to make the assumption that the problem is less than it would be if we were getting a lot of telephone calls.

One other thing that I have not mentioned here, of course, is what happens if you find that things are tight in North Bismarck, but they are not tight, say, in South Bismarck. We have had discussions with America's blood centers. And I think Ms. Fredrick and I raised this topic the last time we met, is that there is data on supply that the blood establishments do collect. And if we found an imbalance, we would certainly want to go to the suppliers to see if there was a technical fix. We anticipate a great amount of cooperation from those, as the need for it arises. So we haven't given up on measuring supply. What we're trying to do is complement the measure of supply by a measure of demand.

DR. KLEIN: Yes, Steve, I'd like to make a comment. I noticed a number of my colleagues writing down numbers. These really are numbers. These are not data. There was no common definition of even what an inventory is, and I know that some inventories, that people call inventories, weren't included in the so-called inventory collections. Now that's going to be straightened out. It's going to be tweaked, but I think it's important not to try to think that you know something about the O-negative supply in the Northeast.

The second point I think is related to that, and that is that let's take the Northeast, where one of the sentinel hospitals is Brigham and Women's. It's one of the few hospitals, one of the 8 percent or so, that collects its own blood. And knowing something about their supply is important, but really tells you nothing about the regional Northeast and the New England Regional Red Cross Center, which supplies virtually all of the rest of the blood north of New York City.

So I think we can be misled by these numbers now. I'm sure they'll get a lot better. But if you've written them down, you may not go out of this room knowing any more about the status of the blood supply than you did when you came in.

DR. NIGHTINGALE: Yes, and let me make one additional comment. The existence of a contract between the government and an individual is a matter of public record or at least FOIA-able, perhaps not all, but certainly all Health and Human Services' contracts are. But we do not plan to report data on an individual center. We never did. There's a variety of descriptions for it, one of them is trade secret, the other is plain old-fashioned privacy, and the other is that's not what we're interested in. The government does not perceive it to be the business of the government to know, on a daily basis, what B and W's blood inventory is.

What we have done is chosen 10 hospitals, and we've oversampled New York City, to get a first-pass estimate of what's going on in the Northeast. We have alternate sources for what's going on in the Northeast from the blood establishments, the producers of blood. And we will have a third source from our contacts, formal and informal, the telephone, about what's going on in the parts that we haven't sampled.

As I said earlier, if the State of New York does its own sampling, we will have a fourth. I wouldn't be surprised if other regions might do something of the same. Again, Harvey's point bears reemphasis. These are preliminary data. I think that my point also bears reemphasis, and that point is that there are two ways to get off the ground, and one is to do it by yourself, and the other is to go public and ask for help, and we have chosen B.

DR. CAPLAN: Jay, and then we'll go to Jane.

DR. EPSTEIN: Steve, I apologize I came a little late.

Have you attempted also to capture percent of inventory outdating on each day? I understand that outdating is a phenomenon that's multiply determined, but on the other hand, in the end, it might prove to be a useful indicator. If having a larger daily inventory turns out to correlate with a larger outdate, it would be one of way surmising you're collecting too much, for example.

DR. NIGHTINGALE: That information is being collected, I think. Until I have a scrubbed set, I'm not prepared to do any analysis of that as an indicator, but I am very hopeful that that will prove to be a useful one.

DR. CAPLAN: Jane?

DR. PILIAVIN: I want to second and third the comment about this not being data that people should take away and quote as gospel. There is a saying among those of us who work in the data business is that the plural of anecdote is not data. But what I originally wanted to ask was whether you have, indeed, observed in this set of anecdotes any of these shortage events?

DR. NIGHTINGALE: Yes, there are comments. I am not in a position to go further than that. But I think the truth is this is not anecdote, this is not data, this is a progress report. This is a seminar writ large. For people who were starting a new investigation, the ones who survive usually go not only to the people right around the corner and ask for advice, but the people who might have some information that they would not otherwise have access to unless they went public with their concerns.

I think one of the things that we have observed that I think is a preliminary observation, and I think it's an observation rather than an anecdote, is that the platelet business is different from the blood business. What we may end up observing is that the AB-positive business is different from the A business. And if anyone has comments about that, particularly if anybody has suggestions for an exploratory data analysis, we would be most grateful for them.

DR. KUHN: Dana Kuhn. Dr. Nightingale, prior to your presentation, this committee had some concerns, and I just wanted to follow up on some of those concerns and maybe you can address them. They were concerned that these 29 sentinel entities, there was a need for a standardized reporting, and also there was a need for a sampling throughout the seasonal time periods. Also, there was a concern of conducting spot checks to avoid manipulation of data or to avoid, as one person put it, hording or probably speculative obtaining of product.

Will this entity be addressing this or has it already addressed this?

DR. NIGHTINGALE: Yes. I think we've addressed it. You usually do questions in reverse order and hope that you remember number one. I do remember number three, which is the validity of the data.

We have asked each of our contractors to prepare a standard operating procedure for this, and include it in their standard operating procedures. And I think that the blood banks, as just about everybody in the room, I believe, will agree, is a very heavily-regulated business, and you regulate adherence to standard operating procedures pretty carefully. So I believe that that was the most efficient way to meet our fiduciary obligation to ascertain the accuracy of the data. I don't know of a better way, honestly.

Now, having done number three, I've got to ask you what one and two were again.

DR. KUHN: The need for standardized reporting and then the sampling--

DR. NIGHTINGALE: Okay. Stop there. It's been a wild week, and it's almost over.

The need for standardized--the reason we had a contractor's conference yesterday, so we could get everybody together, after they'd done it for a while, and said, "Okay. How are we going to standardize this?" If you have one of these conferences before you get the thing going, you're likely to need another one, and we scheduled a teleconference for September 25th, and I anticipate we'll be doing them on a regular basis.

That was number two, and I'm sorry.

DR. KUHN: The sampling throughout the seasonal time periods.

DR. NIGHTINGALE: I think, originally, we cut contracts for 6 months so we could get off the ground, but not be constrained if we found that we were going the wrong way or, for that matter, if a lot of other people found we were going the wrong way. The expectation right now is that we will continue pretty much the way we're going, depending on budget, honestly, among other things, for a 12-month period.

The thing I'm not certain of yet is whether or not we'll have full confidence that we have a full data set from October 1, '01, through September 30, '02. I would like to have one full month of lead-in before I started really collecting data for the record. That's pretty much been the experience with the plasma and with the blood supplies. They started in October, and looking back on them a year later, they were really fine by January of the following year, but a 90-day roll-in is what I need.

COL. FITZPATRICK: Steve, Mike Fitzpatrick. Just to kind of respond to Dr. Davey's, part of his question, I think we discussed on either teleconferences or previously at this meeting that since DOD collects as much as it transfuses, that we wouldn't be representative of what's happening nationally, and so we weren't included in the model.

Sort of going along with what Karen asked, though, if the committee felt it desirable for us to mirror and report similar data, as you are reporting, I have the means to do that, but it would help if I were asked to.

[Laughter.]

DR. NIGHTINGALE: I understand the comment. I think I had similar reticence. I will just say that the conversations I've had with the other suppliers, and with Red Cross and Celso was particularly helpful yesterday at the contractors' conference. I hope to develop those.

I don't want to scare anybody off.

COL. FITZPATRICK: Well, I wasn't implying that we felt left out. What I was meaning was that in order to obtain the resources to mirror what you are doing and provide the committee that data, if the committee asked, I could probably get the resources to do it.

DR. NIGHTINGALE: Understood. Of all of the branches of government, yours is certainly the one most likely to get the resources it needs.

DR. McCARTHY: Leo McCarthy. Just a couple comments. Again, to reinforce what Dr. Klein said. I think the contractors, many of us, we hadn't seen this data either and we haven't manipulated it, so this was new to us. This data was handled by the office, Steve's office.

One of the things that I think one needs to be very cautious when we leave this room and return to our areas, is not to think that there's a week's supply of blood on the shelves, because, as we might find out--and I'm not trying to be Cassandra-like--that may not be proven to be true over this next group of holidays.

Lastly, is just a question, which I guess we should have dealt with yesterday, that just follows up on Karen's comments and I believe the Colonel's. Steve, how would one respond or your office respond to a hospital who feels left out, who wants to participate n a program, provide the data on an every-day basis, and would not expect to be paid for that? I don't know that we discussed that contingency, but I can imagine, as we just heard, some of the people, even outside the government, may want to do that in some part of fly-over America, in Omaha or something like that. I don't know that we discussed that.

DR. NIGHTINGALE: Actually, your next-door, St. Vincent's, called up a couple of days ago and asked that question. And I believe that here is--I'm going to throw the question back out. One of the reasons why we looked at Indiana University as well, it's a large bank. It also is a major transplant center, and one of the inclusion issues is--that we would like to look at--and we'd go back to Dr. McCarthy to look at--is what's inventory right in the middle of transplant season, which is usually Christmas, which is usually when things go tight.

We really don't know the answer. We have a couple to that. I think that St. Vincent's--and correct me if I'm wrong, Leo--does a lot of heart transplants?

DR. McCARTHY: Open-heart surgery.

DR. NIGHTINGALE: They're a heart center. And this is not to say that that's all one hospital does. This is information that I would like to obtain. I think it's something that we could--we should all think about--is: data that comes voluntarily, the same as data that comes on contract? I haven't got there yet. I think perhaps by September 25th we will get there. Where I have gotten is that there's a finite amount of money in the pot, and some of that money we do anticipate, at least keeping it available for monitoring the plasma supply, which is a very, very different business, the demand for plasma.

Dr. Chamberland.

DR. CHAMBERLAND: While I understand people's concerns about trying to get additional data, more sites enrolled, trying to address this elusive goal of representativeness, I think having participated in yesterday's meeting, we have a big job ahead of us just to put in place a standard operating procedure, protocol, case definitions, criteria, data collection form, have that disseminated, and as uniformly as possible be followed by the 29 participating sites, and people shouldn't be discouraged by that or think that that's unusual because when you get a multi-center study up and running off the ground, it takes a lot, a lot of hard work and time to do that in a careful way so that you have confidence in the data that you are collecting. And so as Steve and others have emphasized, and Dr. Truelzi, as a participant, this is at a very preliminary stage, and if data collection were to be expanded to other sites, particularly, as was brought up by Dr. McCarthy's sites that might have an interest in volunteering their data, I think that would raise potential concerns that unless other sites were willing to follow a uniform protocol using the same definitions, et cetera, it really would make it very difficult to aggregate these data and analyze them in a way that you have confidence.

So I think there's a lot to do just to get the 29 sites up and running, view it as a pilot that's something that we need to, at the end of the 6 months, take a step back and look at, see what worked, what didn't work, how can we improve, do we need to change the participation by individual sites, you know, add, subtract, get more, et cetera. But I think, you know, caution is the word here. This is very much needed and very just laudatory that Steve and his office--as someone who works in government, I know how difficult it is to get contracts let, and they did this in an extremely short period of time.

DR. NIGHTINGALE: Quick and dirty?

[Laughter.]

DR. CHAMBERLAND: But I think we have to take a step back and sort of do some of the things that were outlined in yesterday's meeting, that participants very much wanted to do.

The other thing that was of interest to me is that many--the hospital transfusion services all have--operate using computerized data collection systems. There apparently must be a finite number of these software packages that they operate on, and some are more user friendly than others, that allow you to manipulate the data and to extract the data to be reported to the system. So again that's another caution for hospitals outside the 29 that might be interested. There was a lot that I certainly didn't have an appreciation for going into this meeting.

DR. NIGHTINGALE: If I could just thank Dr. Chamberland for expressing what I had hoped to express myself. Dr. Chamberland's words are the guidelines under which we are proceeding, and they are the guidelines under which I hope you have received the information that I've presented.

DR. WINKELSTEIN: Jerry Winkelstein. Steve, I understand, or I hope I understand that the reason you've made the presentation is to get feedback.

DR. NIGHTINGALE: Absolutely.

DR. WINKELSTEIN: That it's a work in progress. I'd like to suggest that committee members spend some time thinking about this today and actually produce some written suggestions for you as a housekeeping suggestion.

DR. NIGHTINGALE: Thank you.

DR. WINKELSTEIN: Because I think that would be useful. I certainly have a few, and I'll just, rather than publicly give you the advice, I think probably most of us should spend some time writing out a few suggestions. If this is truly a pilot program we should assume that it will be changed in January and our role is to make that a better program.

DR. NIGHTINGALE: That is the reason for having this meeting today.

DR. CAPLAN: Mike?

DR. BUSCH: Steve, I commend you definitely for bringing this on board, and I think it's an excellent pilot activity, but I think a lot of the comments and criticisms are that the transition to it, a sustainable program, is not clear. The commitment, and, you know, having worked in a number of multi-center studies, these are typically 5-year programs with the well-funded coordinating center that manages a variety of these issues and has policies about publications, forms a steering committee that really oversees the program. In most studies, presentation of data like this in such a preliminary context simply wouldn't be allowed. It's so preliminary. Not that it's--I think it would be useful now in the context of a very pilot-oriented program.

But the question is: do you intend to continue to manage this program through your office on an annual allocation program, or is there some mechanism to set up a real sustainable study group?

DR. NIGHTINGALE: I think the first answer is, obviously, given the position that I am in, if this did not reflect a commitment of my superiors, including the Secretary, to continue this program for as long as it was beneficial to the public health, I wouldn't be up here giving it. I think the details of where this program will--before that, this is also a component of the Department's Blood Action Plan that was signed in on November 23rd of 1999, and my presence at the podium reflects a commitment of the new administration to maintain that policy.

Where we go with this in the long run is something that we have discussed extensively within the Public Health Service and among the Advisory Committee members. It is not clear that this has a convenient home in a particular spot with the Public Health Service. It's come to the Office of the Assistant Secretary for Health because since October of 1995 the Assistant Secretary for Health has served as the Blood Safety Director for the Department, and that continues in this administration. So we clearly plan to keep this going as long as it is useful. That is my understanding from my superiors, and I have no expectation that that would change.

Where do we do it? Do we keep it in this office? I hope not. This has been a long summer. There are several other things that I have to do at the same time. There's three of us and we're hiring a secretary. It's tough. I hope that the computer will make it a whole lot easier.

Where does it site? Does it site within the Public Health Service? Does it site within the private sector? That's a very complex question. But let me add the one reason that is implicit in what I'm saying, but explicit. Is when an entity, whether the government develops a program, whether AABB, ABC, Red Cross, Immune Deficiency Foundation, develops a program, that entity will face a very legitimate challenge, and that challenge is: did you develop this system for the public good or did you develop this system to spin your own version of the public good? And that above all is the reason why this is being presented to the public for this time, for their input, with the tape running.

The final version of this program, when it comes out--and the January version may be it, that's what I'm shooting for, and that's why I'm holding this program right now--should at least address the concerns that are raised at this meeting, raised at the meeting yesterday, and that I'll hear about later. If we had a lot of time to put this in progress, we might not be pushing it so fast, and that's another reason for bringing it to the public early rather than late.

It is my own view that we don't have a couple years to fool around with this program. We need this information right now. There is major policy change in donor deferrals that's going to be coming fairly soon. We need to know the impact of that policy change on the capacity of the blood supply to react. And when we put the UK donor deferral in, we missed an opportunity to get data, and I don't want to miss that opportunity this time.

DR. WINKELSTEIN: So I have another housekeeping question. Forgive me. Have our previous resolutions in any way addressed this problem? Do you need something more specific? Is that an agenda item for today? Should we be thinking in those terms?

DR. NIGHTINGALE: I think we got a pretty strong boost in April. This is a different meeting from the one in April. This is much more a nuts and bolts meeting. It's nuts and bolts this morning. It's nuts and bolts this afternoon. There may be a big picture into it, but I think we've dealt with the big picture before, we'll deal with the big picture again. This is--if you have a thought, a constructive, or just a plain old-fashioned suggestion or criticism of this, this is a real good time to give it because this is a time when those criticisms are going to be addressed.

DR. CAPLAN: One issue that I think has come up time and again is, as we sort of build this sentinel system, want to make sure that there is an adequate steering committee, oversight and accountability for the system. So one issue is where is it house? Another is sort of who's steering the ship and ultimately making decisions?

So what thought's been given, Steve, to the sort of structure, executive structure on the sentinel system?

DR. NIGHTINGALE: I think the toughest critic is the public, and that's the critic we're eliciting at this time. We are using the Advisory Committee, once again, as the surrogate for the public, but we're doing that in a public rather than a private session.

At the time same time, I should say that there have been extensive private discussions with people in the community, both those who are participating in the study, and those who are not. I would mention in particular, Dr. Brendan Moore, of the Mayo Clinic, who's been here before, declined to participate in the program. He felt his center was not representative, but he sent us a 5-page very detailed critique which was extremely helpful.

At the moment, supervision comes from my superior, who's the Assistant Secretary for Health. The immediate supervision are my colleagues in the Public Health Service who are on the Advisory Committee as well, and many members of Dr. Epstein's office. There is a working group on blood safety where I think that the--the supervision of this with Dr. Williams--there's a light in my eyes and I can't see Alan if he's in the room, okay, I see his hand--Dr. Williams, who has a very extensive, very valuable background in this area--

DR. CAPLAN: Here's my point: I think if we're going to be in a situation where there are major concerns about inadequate blood supply in the face of the pursuit of increased safety and how that debate is going to sort out, it's going to be very clear--it has to be made very clear who is going to be at the helm of the sentinel system. So I didn't really mean to put you on the spot--

DR. NIGHTINGALE: No, no, no.

DR. CAPLAN: --for running the names out, but I think that should be there. We can sort of suggest, perhaps, that as you--part of the standardization and reports and so forth, should include some attention to what's the best way to make sure that the public, which donates the blood, understands who's keeping an eye on the supply of it in terms of really administering the program, making the decisions. That's, to me, taking a public perspective, I want to know names and zero numbers to speak.

DR. NIGHTINGALE: That's easy.

DR. CAPLAN: And they may all be within the Departments. That's okay. I just want to know where they are, so if Jerry's sending suggestions in, he knows sort of who's reading them at the other end, and then who's accountable.

DR. NIGHTINGALE: Okay. On January 8th of 2001, Dr. Satcher signed a directive that established the responsibility for monitoring the blood supply in his office at the time he was the assistant secretary for health, as well as the Surgeon General. So to where does the buck stop, that stops here. It stops in his office. And the person who is implementing it is myself.

DR. CAPLAN: All right. Well, if we've exhausted the subject here, maybe we can take our 15-minute break. Thanks, Steve, for the update on the details. We'll get back here in 15 minutes.

[Recess.]

DR. CAPLAN: We're going to go into the public testimony part of the meeting. Our first 5-minute set of comments, let's hope 5 minutes, Jackie Fredrick from the American Red Cross. So if it could get the room to quiet down so we could hear her.

Jackie.

MS. FREDRICK: I'm getting older. I need glasses.

Mr. Chairman and members of the Advisory Committee, thank you for allowing the Red Cross to make some public statements here.

As you know, the American Red Cross is committed to developing a stable and sustained blood supply to meet increasing patient needs and hospital demand for these life-saving products. We believe it is incumbent upon the blood banking and transfusion medicine community to commit to a new way of doing business by accurately forecasting the demand for blood and ensuring that blood collections are geared to meet specific patient needs. The Red Cross has instituted a series of new initiatives that are enhancing our ability to monitor the amount of blood collected, distributed, and in inventory at each blood center nationally and nationwide. With this information, along with the results of market research that enhances our understanding of how to effectively reach our generous blood donors, we are working to make chronic cyclical shortages a thing of the past.

We have been asked today to comment on HHS' plan to monitor the demand for blood products. We believe in order to ensure blood availability, it is critical to have an effective means for monitoring hospital and patient needs as well as available inventory. The recent announcement by HHS regarding a sentinel system with real time information from 29 hospitals on the supply and the demand for blood and blood products will be useful for all of us in the blood services enterprise.

At the April meeting of this Committee and at the June meeting of the TSEAC Committee, the Red Cross shared with the Department of Health and Human Services our short- and long-term plans to increase blood collections. Today we are pleased to share with the Committee and our blood services colleagues, the initial results of our efforts to monitor and increase the blood supply, as well as our comments on the present HHS data collection effort.

Before I outline our recent activities and results, we would like to update the Committee on research to determine the further impact of our expanded donor deferral criteria related to variant CJD. This information is directly relevant to the discussions today on blood availability.

The Red Cross commissioned Wirthlin Worldwide to perform a telephone survey of a nationally representative sample of Red Cross donors to determine the number of individuals that would be deferred because of our expanded variant CJD criteria. The findings of this survey indicate a total of 3 percent of our current Red Cross donors will no longer be eligible to donate under our expanded guidelines with a margin of error of 0.6 percent. In addition, approximately 1 percent of eligible donors will erroneously self-defer even though they are actually eligible to donate. Take together, the results of the survey indicate about a 4 percent loss of our donors in the American Red Cross, equivalent to approximately 235,000 units from the expanded deferral.

Based upon our collections experience this July and forecasted collections through mid September, the Red Cross believes that donations will cover this anticipated loss. Although the Wirthlin survey shows the Red Cross national system experiencing a loss of between 3 to 5 percent, our collection goals are based on prior modeling that indicated an 8 percent loss in donors. The Red Cross is working hard to ensure an adequate blood supply to our hospital customers. Our recent summer initiatives have already shown positive results towards increasing blood collections.

There are those who have speculated that donations are actually decreasing in this country. We have found exactly the opposite. In fact, presenting donors to the American Red Cross surged to 7-1/2 million in our last fiscal year or a 6 percent increase compared to the prior year. This corresponds to a potential 8-1/2 percent gross increase in productive units if one uses fiscal year 2000 deferral rates that did not include our loss of donors due to finger sampling. So an apples to apples comparison would have actually had us growing 8-1/2 percent in presenting donors. This would have corresponded to over a half a million units of blood collected.

The Red Cross has seen an increase in our blood collections for this summer compared to past. We embarked on a targeted advertising campaign, personal contact with our donors, and new programs to monitor and forecast blood collections and distributions. Our July 2001 collections of 551,949 reflect an almost 8 percent increase over July of last year. That's over 39,000 more units collected this July than the past July, or an equivalent increase in two days of transfusion need for our 3,000 hospitals. In addition, 31 of our 36 blood regions also increased their supply from July to July.

These increased collections have had a direct impact on our inventory. Our total red cell inventory is 33 percent higher this August than last August. Type O inventory has increased 83 percent over last August. This shows that our targeted efforts to contact Group O donors, over 2 million of them, paid off. Distributions of red cells are up 3-1/2 percent in July compared to last July, as well as increases in Type O distributions.

Our recent campaign highlights our ability to increase blood collections by using the right strategy and resources. Our goal now is to make this sustainable.

We are moving forward with long-term initiative to build upon our positive experiences this summer, the lessons learned and what we are hearing from our donors. We are increasing our collection staff, our collection sites, and our collection goals. We are developing specific appointment scheduling methods that will expedite the donor experience, and are planning to expand our telemarketing and call management systems as well as our blood donation record process. All of these initiatives are geared to making the donation experience quicker and more enjoyable for our volunteer donors.

The Red Cross is also working on its projection and demand models so we can forecast where blood is needed before there are shortages. We have been forecasting collections for over a year and we believe out models will continue to accurately predict what has been collected.

I heard the prior conversation about what happens if you find out there is blood needed in Bismarck and it's in Madison, Wisconsin. In fact, the Red Cross has been moving blood all summer because of our inventory model and our demand model. And that will allow, in the Red Cross, inventories to be balanced absolutely across the country.

All of us in blood services are challenged by meeting growing patient needs for life-saving blood. Chronic, cyclical blood shortages have long plagued this enterprise. HHS' announcement about the sentinel system for monitoring the blood supply is commendable. This is a good step towards better understanding issues of the blood supply and the need for blood in the 29 participating hospitals. This information will be useful to everyone in blood services and will provide a snapshot of certain variables that impact the blood supply.

The Red Cross would like to thank the Committee for the opportunity to provide our views on this important public health issue. The safety of the blood supply and its availability are top priorities of the American Red Cross. We believe it is a shared responsibility of blood collection organizations and hospitals to collect data on supply and demand, and therefore, institute programs to ensure blood availability.

I'd be happy to answer any questions.

DR. CAPLAN: A couple of questions we have time for. Karen?

MS. LIPTON: Karen Lipton. Jackie, we were talking earlier about the HHS monitoring effort, and people were talking about it would be important to have some data if we could from the blood suppliers. Have you been asked to participate in that?

MS. FREDRICK: No, we haven't been asked. I think Steve came about end of June and we discussed the agenda here, but as I said, I think it was at the April meeting when we talked about the same thing. We would be glad to provide data. We'd like to be asked to provide the data. And we'd like to participate in how that data's going to be used also.

DR. NIGHTINGALE: Jackie, you can anticipate all of the above. Thank you very much for the offer.

MS. FREDRICK: Thank you.

DR. HOOTS: Of your new donors, do you have a sense of how many are either first-time or nonchronic donors? Are you harvesting a new pool, or are you moving pools of donors around?

MS. FREDRICK: I can't tell where that question's coming from.

DR. HOOTS: Right here.

MS. FREDRICK: Oh, I'm sorry. I can't see the red light.

We haven't analyzed the July data yet to know what percent are new donors and whether we've increased the frequency. We're going to actually monitor donor behavior over time because one of the theories we all have is do you actually cause donors to donate more frequently or do you just move them more current in their donation process?

So right now I actually can't tell you whether the first-time donors or the frequency have changed.

DR. CAPLAN: Let's do two more. We'll do Ed, Dr. McCarthy, then we'll get done. Ed first.

DR. GOMPERTS: Mr. Fredrick, the estimate from the point of view of deferrals, the percentages and estimates that you've given, on what--was this an estimate or did you actually have some hard data? Did you do some surveys? How was that done?

MS. FREDRICK: I'm sorry. We did a representative survey of our blood donors who have donated in the past 12 months, and it was done to determine the national impact, though I do have geographic data, but it's large geographic data. It was done by telephone surveys, statistically significant sampling, detailed questionnaire. So, yes, we have very detailed data, and I would say that that has been shared in its detail with FDA, with Alan Williams and Jay Epstein.

DR. GOMPERTS: Thank you.

DR. McCARTHY: Leo McCarthy. Jackie, in your ramped up efforts to increase donations, could you comment on what singular efforts are being made to recruit the minorities, which have been under recruited for a long time. Is there a focus on minority recruitment?

MS. FREDRICK: Yes, there is a focus on minority recruitment, but it tends to be very focused locally, where I believe it should be. We can provide guidance from national. But, for instance, in Detroit, which I think two weeks ago came out as the country's largest African-American city, we have a three-year initiative there that will grow collections by 58,000 donations and 24,000 donors. We have a large hispanic initiative going on in the Los Angeles area.

But clearly, going forward, those have to be a focus for us to be successful, particularly in urban areas.

DR. CAPLAN: I've got Rick, then I've got Harvey, then I think we'll go.

DR. DAVEY: Rick Davey. Jackie, in your survey of the deferred donors, I wonder if you got information on the number of deferrals or the percentage of deferrals that may include repeat donors and platelet pheresis donors? There's some indication that the deferral criteria may more heavily impact those particular groups.

MS. FREDRICK: Yes. We did. We specifically asked a question about apheresis, and apheresis donors or platelet donors will be impacted by a higher rate, and I want to say I think it's about 4 or 5 percent of platelet donors. We also did look at first-time donors and regular donors and I believe first-time donors had a higher incidence of deferral than repeat donors, which would make sense, because we've kind of done the UK deferral already to those. But I have that detailed information. I can get you those numbers.

DR. KLEIN: Jackie, first, if no one else is going to say it, I'm going to say how encouraged I am that the data on collection looks so good. I think this is wonderful news, and the organization really needs to be commended.

Having said that, this is the other half of the data. I mean, this is the collection side. We heard a little bit about the hospital side, and I certainly would hope that at least some of the regional center, if not the whole system, would be able to supply that kind of information so we can put the two together. Do you have the kind of information that might look at need, such things as orders that were partially filled or orders that weren't filled. Even though the collections are going up, that may not be supplying the hospitals, and perhaps you have some information that might be of some help.

MS. FREDRICK: I do. And we are monitoring that. We call it fill rate. And it's not perfect, because just like the discussion about what's an order and what's filled and if people are over-ordering, but, yes, I have that by blood region and by blood type, I believe. I will tell you, I'm a little leery about the data, but what we have done is we've already set our goals for the next five years for blood collection, specifically in detail for the next three years, and we have taken the fill rate loss, that I call it, and we've actually built it into our collection number such that within 24 months, we would meet 100 percent of the orders placed today. Now, I don't personally believe that we are under transfusing the country by huge percentage points, but--so, yes, I have that information.

DR. NIGHTINGALE: This is Steve Nightingale. I would just like to state for the record that I fully concur w