Skip Navigation
  • Text Size: A A A
  • Print
  • Email
  • Facebook
  • Tweet
  • Share

Blood Safety Transcripts

TABLE OF CONTENTS

Risks of Transfusion, Dr. Busch, 1
Dealing With and Communicating Risk, Mr. Maibach, 32
Open Committee Discussion, Mr. Caplan, 53

P R O C E E D I N G S (8:00 a.m.)

MR. CAPLAN: -- first on the risks associated with transfusion by Dr. Michael Busch, second by Dr. Ed Maibach, dealing with the topic of how to communicate about risk. Then we are going to move to readdress the language on the recommendations that the committee wishes to offer in response to the questions it has been asked, and we will work on that for the rest of the morning, or until such time as we are satisfied that we have language that we are willing to pass back to the Public Health Service and the agencies that have asked us to comment on the issue of hepatitis-C. So, let me at this point turn the forum over to Mike.

Agenda Item: Risks of Transfusion

DR. BUSCH: Good morning. In a brief half an hour this morning, I am going to review current risks of transfusion and actually, this is an outline of the presentation. I am going to first give you two quick slides, just to give you an historical perspective about how much infectious risk the blood supply used to represent compared to the way we are, and the issues we will be dealing with this morning.

Then, review briefly the kinds of approaches that have historically been used to measure risk, and lead you to what is really, the current only option, which is model estimates of risk. I will give you data on the current estimates of risk for the four major viruses that we are concerned about, and then liberally borrow from the GAO Report, which is a document that I think each of the committee members received.

It is really an outstanding effort by these government individuals to compile estimates of risk, both for the infectious -- major infectious, and what we might call minor infectious agents, as well as the noninfectious risks, and then put all those together and estimate risk for the general public from transfusions, and then contrast that with both medical and nonmedical risks in different settings.

I think it is very useful to briefly present a few examples about how new initiatives to improve safety further can be examined and particularly, a couple of slides from Jim on cost-effectiveness of some of these new initiatives, and then briefly touch on approaches and viruses and agents of concern with respect to new and emerging agents.

We will go quickly. Again, all of this material, a lot of it, I can give you a paper -- in fact, I will probably give to Paul a recent review paper for distribution that summarizes all of these risk estimates from our work as well as point you back to that GAO Report, which is really outstanding.

Just, again, for retrospective considerations, the risk of blood with respect to hepatitis, specifically, hepatitis-C, was in excess of 20% per transfused recipient, back in the sixties. In the seventies, with introduction of surface antigen and the concurrent discontinuation of paid commercial donors, there was a dramatic, nearly two-plus-fold drop in risk, and then stable risk at around 10% of all patients getting HCV through the seventies and into the early eighties, when, serendipitously, the AIDS risk exclusions resulted in a modest drop.

The evidence came forward that this was a more significant post-transfusion phenomenon(?), and could be a more significant problem than previously thought, and the associations with ALT and anti-corp appreciated. The introduction of these surrogate tests, and then eventually the discovery of HCV and progressive improved assays. So, you see, we are now dealing with risks well below 1%, and this is what we will be struggling with to estimate today.

For HIV, it is a very different picture. It is not a longstanding endemic virus, but rather an epidemic explosion of occult infection in the absence of symptoms for about five to ten years. First cases of transfusion transmission documented back in the 1977, 1978 time period, and then really clearly, now well-documented exponential rise in risk in some areas, like San Francisco, exceeding 1% of all units infecting with HIV.

The peak actually occurred in 1983, in the high risk areas, and declined well before testing became available, again as a consequence of education efforts and exclusion efforts. So, this illustrates, as did the hepatitis-C experience, how measures other than screening -- education, safe donor selection -- can have a dramatic impact on the safety of the blood supply, well before a specific agent is appreciated.

In terms of how we can measure today, there are three sort of approaches, two of them, as I will show you, somewhat impractical because of the low risk, and yet still important. One is recipient follow-up, enrolling large cohorts of recipients prior to transfusion, following them prospectively for evidence of seroconversion or clinical disease.

A second is to actually focus on the donor products that are leaving the blood bank, have been screened as safe, and one can subject those samples to culture or PCR to estimate the residual rate of occult infection that is missed by the standard methodology.

Then third is the approaches using models that both focus on window period risk, as I will show you, which rely on understanding the rate of new infections, or incidence and the duration of the infectious window period, to estimate the risk of seroconverting donors, but also, more recently, as I will show, we have added in estimates of risk due to variant viruses, due to test failure, things of this nature, to compile an overall modelled estimate of residual risk.

To illustrate some of the classical approaches, probably, certainly, the most recent and large-scale study of recipient seroconversion was one funded by NHLBI and conducted in the immediate post-HIV period. Over a six-year period, almost 10,000 recipients and additional non-transfused patients were enrolled, who were undergoing cardiac surgery, in the Hopkins, Houston, Ken, Nelson and colleagues; 120,000 units were transfused into these recipients. They observed actually six of these patients -- again, they collected pre-transfusion samples, and then samples six months post-transfusion.

They started with the strategy of testing the six-month post-transfusion sample. They found eight infected patients six months post-transfusion, but when they went back to the pre-bleed, six of those eight were actually infected prior to the transfusion.

They had six prevalent infections, but they did have two documented transmissions, both of which were linked to donors who subsequently seroconverted. And both of these transmissions occurred in the fairly early phase of the study, in the late eighties, and before we moved into contemporary, more sensitive antibody tests, and certainly prior to antigen screening.

These estimates certainly are out of date, but the point estimate from this would be basically the two transmissions over the 120,000 units for one in 60,000, but an upper confidence bound of about one in 20,000. So, not terribly reassuring.

This same population of samples from that same cohort was subsequently tested for HCV and HBV. This slide summarizes the HCV risk and if you would just focus on this column, the per unit risk, you can see that, prior to surrogate tests, a half a percent of these units were causing recipient seroconversion.

After surrogate tests but before HCV was introduced, .36%. So, very much the predicted efficacy of surrogate tests, about a third reduction in risk per unit. And then, HCV first generation test kicked in, dropping the risk fairly dramatically to .06 or six per 10,000, which is a low risk -- one in 1500 -- but this has clearly been dramatically reduced further by the second generation assays, which actually detected in clinical trials, one in 1,000 units that were missed by the first gen tests were detected by the second gen tests, and confirmed by PCR.

This is funny, because actually, that would suggest that the second gen test picked up more infections than the first gen test was missing; one in 1,000 is a bigger number than one in 1500. So, we are dealing now with residual risks that are so low that studies are having difficulty matching themselves up.

This just shows HBV now. HBV transmission, same population of recipient samples. Prior to surrogate markers, about one in 2,000. Modest drop with anticorin(?) ALT, perhaps 10%. A more dramatic and somewhat surprising drop with introduction of HCV, almost a reverse surrogate effect, where HCV screening dramatically reduced the risk of HBV transmission, something that is still not explained. But, very low residual risks from the studies, and those of that recipient seroconversion study, I haven't got a firm number, but Paul could probably confirm this, but I would suspect it probably cost $20 million to conduct over the course of that five- or six-year period.

In parallel there was a large study in San Francisco where we actually performed PCR and cultures on 200,000 donations for HIV. Again, you know, probably a $20 million dollar study over the course of five years. Every donor had an extra tube collected. After the routine testing was done, the tubes that were screened as safe, corresponding to safe units, were subjected to pooling of the cells, eventuating in pools of 50 donor cells each.

Those pools were then tested in parallel by culture and PCR, and the results of that study, there was a two-phased component, but the bottom line over the entire study was, we identified one infected donor who had been missed by routine screening, over the course of the study, during the two phases, a total of 200,000 samples.

A confidence interval, an estimate of -- a point estimate, when you adjust for the pooling issues in the sensitivity of about one in 160,000. But again, an upper confidence bound of about one in 25,000, which is not reassuring to the general public.

Clearly, these approaches are very tedious, very expensive, for measuring risk, which has led to our estimating risk through modelled approaches, and as we realize that there are continued rare transmissions of blood infectious viruses, despite screening, we can step back and ask, why does that happen, and that can then let us focus on estimating the contributions of these various sources of risk, to the total picture.

In terms of viruses, there are four major potential reasons for failures. One is pre-seroconversion window period donations, and we will talk about that. A second is the concern about viral variants, type O HIV, HCV, rare variants, things of this nature, that tests are simply not adequately representative of viral diversity, and so they miss some rare infections.

A third consideration are what we term atypical or non-seroconverters, also called silent carriers. This would be not a problem with the virus being variant, but a person who is infected and has an unusual or incomplete immune response, so the virus is a standard strain, but the host does not form a typical immune response and is a long term immuno-silent carrier.

Then the fourth consideration is that we simply screw up, that the markers in the blood and the testing is performed inaccurately, and as we get into very low residual risks due to all these other things, you will begin to see that test error can begin to contribute a significant contribution to risk, especially with viruses that are very prevalent, so you have a fair frequency of positive donors in the tested population, a rare test error occurring on a highly prevalent virus can begin to contribute proportionately to some of these other sources of risk.

Window risk, though, has been the major focus of the last decade and I want to just give you a little bit of examples of how we approach estimating residual risk from window period. Basically, this slide summarizes the very simple relationship that we use to both estimate residual risk, and to project the value of a new test designed to reduce the risk.

The residual risk is simply estimated by multiplying the incidence rate, which is expressed as the rate of seroconversions over person time of observation in a population, and we can measure this in the donor pool -- times the infectious window period, or the time prior to standard seroconversion that the individual is infectious or viremic.

If we want to evaluate a new test, as we will do toward the end of the talk, we simply similarly multiply the incidence rate, the rate of new infections, times the reduction in the window period that the new test achieves. What fraction of a year does that new test cut off of that residual infectious window period?

For the various viruses, we can measure incidence -- this is data from the red study published in the New England Journal last year, that estimates the incidence rate for each of the major viruses that we currently screen for, and you can see, they range from a low of one in 100,000 person years for HGLV, to incidence rates of about four to five per 100,000 for HIV and HCV, and actually, not surprisingly, in the broad public health context, but somewhat surprising to blood bankers, is that HBV is actually the most common virus occurring in our donor population. It is by far the most frequent virus of these transmitted in the general population. So, we are running at about one per 10,000 incidence for HBV.

Toxins(?) themselves, but the window period summary slides here, HIV we believe that there is a phase prior to detectable RNA, a brief phase during which the virus is beginning to disseminate at very low levels, so we have an overall estimate of a 22-day infectious window period, with RNA becoming detectible and ramping up for about half of that phase.

Perhaps this theoretical RNA negative infectious phase is an area of consideration in current experimentation. Whether this exists or not, whether RNA screening could eliminate infections is an important question that is being studied. But RNA can be detected for usually about three or four days before P-24 antigen is detected, transiently, then antibody and improved antibody tests have successfully now detected IGM antibody, and we will talk a little bit about this when we talk about P-24 antigen screening, but in terms of residual risk, it is this ten or so -- 10 or 20 day pre-seroconversion window that is a concern.

For HCV, there is actually a much longer, both total window from exposure to seroconversion of about 80 days, of which nearly all of that is viremic with high levels of detectible HCV RNA. Hepatitis-B, we screen for surface antigen as well as currently anti-corp(?). This I am sure will be an area of focus for this committee, whether this test is still worth retaining.

You can detect HBV DNA for in the range of 10 to 25 days prior to surface antigen and seroconverters, and there is an overall nearly 60-day window between exposure and when you detect surface antigens. So these are the kind of window period information that we need to plug into formulas to estimate residual risk of seroconverters, and that is simply -- as we showed in that earlier slid -- achieved by multiplying the incidence rate, times the fraction of a year that the person is in this infectious window period of 22 days, and you get these estimates of two per million. So this is where these kind of numbers come from, through these calculations of this nature and are subject to confidence considerations, etcetera.

Now, aside from the window period risk, we have to consider those other three sources. Now, this is a slide that just illustrates the distribution of HIV subtypes around the world. As you all know, the U.S. was seeded and subsequently promulgated Type B infections, really, through the blood supply and other mechanisms, throughout much of the developed world. In other areas of the world, though, in Central Africa, and seeding out into various regions in Asia, other subtypes are by far predominant. In Central Africa, Type B is a very rare variant, and you have an incredible mixture, and really endemic within different countries, different subtypes.

The greatest concern with respect to subtypes in our discussion is the failure of some of the tests to detect some of the variant subtypes, particularly, there was a problem with HIV-2, which led to the introduction of HIV-1-2 screening tests. It is worth noting that in the last now seven years of HIV-1-2 screening, there have only been two HIV-2 infected donors identified in the U.S., both of whom were detectible by the HIV-1 test that existed. So, in truth, there has been zero yield in seven years of screening with these combi-tests, which are substantially more expensive than HIV-1 tests.

Similar concerns were raised with the outlier group, O viruses. There was a paper in Lancet about, now, three or four years ago, documenting that some people infected with subtype O test negative on the standard HIV-1 and HIV-1-2 screening tests. This in turn led to surveillance in the U.S. Studies collaborative between CDC, Red Cross. REDS looked at thousands of people who were infected with HIV-1, in determinants, etcetera, and failed to find any HIV group O infected donors.

In a separate study that is ongoing, we have been doing subtyping on a fairly large number of donors over time, both going back to early hemophiliacs infected, early donors during the first six months of screening, and then more recent, donors about -- now, actually, this is a number about 500 totally. And in that whole group, we have only found two variant viruses, a Group A and a Group B, both arising from recent immigrants from Africa who were clearly infected heterosexually in their endemic countries.

The subtype surveillance is important, but if we step back and we ask what the potential contribution of risk is from something like a Group O -- and I do not have time to walk through this -- but the bottom line is, that the risk is extraordinary -- if you take the worst case scenario, the risk is extraordinarily rare that a group O infection will slip through.

These viruses are extremely rare, even in Africa. Their presence in the U.S. is extraordinarily low because people are not -- in fact, now these people are ineligible. The FDA issued guidelines which resulted in permanent deferral, or indefinite deferral of persons coming from group O endemic countries, so even though they were rare in the first place in the U.S. donors from these countries, are now excluded.

And then the tests that we are talking about actually have only about a 20% failure rate, so even if a group O infected donor does slip through, most of the time the tests pick them up. So we end up with an extremely low contribution to these variants.

If we similarly run the same issues for the other viruses, the story is pretty much the same. For all the viruses, the risk due to variants is extremely small, the possible exception being HGLV-2, we are still using HGLV-1 based assays. There's pending licensure of bona fide HGLV-1-2 combi-test, but there may be a rare miss of HGLV-2 by the current HGLV-1 based tests.

There was a significant problem with the first generation HCV test, which has very limited antigen representation from genotype I missing the other genotypes, and that was a major reason for the failure of that test was that is missing non-genotype I infections. So, that has been fixed, too. So, the issue of variant viruses is one of constantly chasing and surveillance for these variants; fixing the tests to detect them, but one that really contributes very little to current risk, because these variants are very, very rare.

The third problem is the non-seroconverter, and there had been reports -- there was an MMWR now almost over a year ago, and just recently published, the paper that describes this case in detail of a Utah plasma donor who was donating and not seroconverting. There have been other cases like that -- this is another one -- a person who developed very high titer(?) viremia, and the antibody tests over the course of a year have just remained borderline positive and negative. But these cases are extraordinarily rare, as evidenced by the fact that antigen screening has not detected any such cases in a year and a half of regular, prospective screening. So, again, these atypical seroconverters are rare for HIV, now this is an important point, because I think this is going to be a major issue for this committee, in the setting of HCV, there is significant evidence that there may be a moderate frequency of people who are PCR-positive, and not seroconverting.

We talked a little bit about this yesterday, the English experience with doing PCR on their recipients who screen negative. They found those four or five cases, which does not surprise me. Actually, Miriam Alter, who is still here, reported five years ago -- to a lot of criticism -- that she was finding in her studies, cases that were serologically negative; people presented with community-acquired non-A, non-B hepatitis, who were testing PCR-positive. And she reported that at a rate of about 10% of all the infections did not seem to be seroconverting, which everybody said, you are crazy.

It turns out, she may be right, and this has become known, because people have begun to do this large-scale PCR screening, and there have been several large studies now in Europe and a recent study in the U.S., both in the source and the Red Cross whole blood plasma collection setting, where people have begun to do large-scale pooled PCR, and there have been no surprises with HIV, nobody is picking up any, but in contrast with HCV, there have been substantial pick-ups of PCR-positive, sero-negative infections.

In the German experience, which is by far the largest to date, they are doing large pools of 500-plus samples and then they are breaking those down and finding the source of the positivity, and then actually appropriately following the donor, and they are also tracing the recipients. And in their studies, they have tested over 700,000 samples, this is as of about four months ago, and they have had problems with false positivity, and samples that test positive initially and then negative, and they cannot explain it all, but they were able to actually identify and firm up and confirm seven positive HBV pools that track down to single donations and yielded a risk of one in 100,000, and almost 70 in this period, one in 10,000 HCV infections.

Impressibly, they were able to get follow-up on a reasonable proportion of these, and three of the five HBV infections they picked up seroconverted to remain viremic and did not seroconvert, even to surface antigen, so this was a very low level PRC positivity undetectable by surface antigen or anti-corp(?).

With HCV, only two of the 36 seroconverted and 33 remained persistently viremic and seronegative for up to six months, so very disturbing data, and the question really is, whether these people were infectious.

In their own look-back studies which were limited, they could not from prior donations document any infected recipients from prior donations, from these donors. But there is data from the U.S., from one of the earlier NHLBI studies, the TTVS study, where Jim Moseley -- we have gone back now and tested these samples by PCR. This was a study in the late seventies and early eighties, where in this study there were 138 seropositive units that were transfused, and 110 of those units caused recipient seroconversion by the second gen test, the third gen test picked up an additional three units.

There were 113 transmissions that were tracked from seropositive blood that caused seroconversion in the recipients, but in addition, there were seven recipients who seroconverted, but there were no positive donors. And we have recently gone back and begun to extensively study the donor units that went into these seven recipients.

One case was found were the third gen test picked up an implicated donations that probably caused one of these infections. There were three other cases, though, that were RNA-positive, and serologically negative that appear to represent these PCR-positive, antibody-negative infections, and appear to be associated now with transmission. This needs to be confirmed and is in the process of being confirmed in multiple labs with sequencing to confirm that these were transmissions.

If you take these three transmissions from PCR-positive, antibody-negative blood, out of a total, then, of 116 transmissions, that is about one in 40 transmissions tracked to these PCR-positive, sero-negative units. And that turns out to fit very well with this one in 10,000 frequency. If you multiply one in 40 times the current prevalence of HCV in the donor pool, which is about, you know, .2, .3%, you end up about one in 10,000.

This is consistent with the observation of one in 10,000 PCR-positive, sero-negative, and suggests that these may be transmitting infection. So, this is an area that I am sure will be a major focus. It has now become the major driver behind expedited introduction of PCR screening in the context of pool-testing before we have standard assays.

The last risk source is test error. I do not have time to go into it, but in a study, we have measured the rate of test error through tracking subsequent donations from confirmed positive donors, and figuring out how frequently do those units test negative when they clearly were infected, and that turns out to be .067%.

For a test error to be significant, it has to occur on a positive unit. So, to measure the contribution of test error, we multiply the error rate times the prevalence, or the frequency of positive units, and that gives us the frequency of risk of missing a positive unit due to error. So, these numbers are small. These are expressed as per million donations, so the worst case for HCV, which is a very prevalent virus, is about one in a million units may sneak through because an error in testing occurs on a straightforward, positive unit, which is, as we talk, begins to be a reasonable risk. We cannot just continue to ignore this, and clearly improvement in methods of testing automation should be pursued to eliminate test error to the extent possible.

This is a summary slide that puts together each of these sources of risk, and again, I have a review that comes to the same table that I can distribute through Paul. Pre- seroconversion window period estimates we have just talked about, immunovariant viruses. These non-seroconverting carriers, where HCV is the big controversy and if this is real, maybe one in 10,000 transmissions may be due to these variants, then you can see that the risk of HCV is substantially higher, perhaps as high as one in 10,000 or greater, and that is really the major issue that is the focus of most blood bankers today, is that residual HCV risk.

You can see, then, we can derive total risk estimates. This is nicely summarized, again. Now, I am going to use four or five slides from the GAO Report compiled from that, and I really urge you to read just that first four or five pages of that. I am not going to go through in detail, I just want to juxtapose these risks with noninfectious risks and then with other risks.

These are pretty much the same numbers I just walked through for the major viruses. These are conservative estimates. They purposely focused on the most conservative estimates, risks of one in 5,000 for HCV, which is even higher than I was talking about. One in 450,000 for HIV, again, a relatively high estimate. One in 50,000 for HGLV-2, because of those missed infections. So, these are really high estimates of risk. They also incorporate a risk of one in a million for HIV. A residual potential risk of non-ABC, some of which may be HGV, but it is probably a very nonpathogenic infection. CMV has a low level risk.

We have a total infectious viral risk of about four per 10,000, which is moderately alarming. For nonviral risks, there are particular focus on bacterial contamination, which risk estimates are as high as one in 10,000 for platelet components appear to exist, of low level bacterial growth in these products that can cause significant morbidity and rarely mortality in recipients. T-cruzi(?) is another concern, one in 40,000 potential risks, although transmissions have yet to be documented in any prospective U.S. studies. There will be a paper coming out in JAMA soon with Red Cross experience on this.

If we sum these nonviral and viral risks, we get five per 10,000. You know, one in 2,000 units cause risk. But if we translate this out to -- as they do nicely -- to the number of recipients who survive long enough to acquire an infection, that translates per year in the whole U.S. in about seven million recipients, to about 7,000 infected recipients, and they also go on to evaluate how much chronic disease burden will this eventuate in? And that is in the range of 12, 1300 per, again, seven, eight million recipients. So, fairly low frequency of disease outcome, but clearly still there are infections transmitted and diseases resulting.

These are the noninfectious risks, which again the GAO Report summarizes; ABO incompatibilities, circulatory overload, things of this nature. And they sum out to a rate per transfusion of these things happening of three per 10,000 units, with a moderate number of recipients being exposed and living with these complications, at least briefly. Many of these are transient. Only about 200 or so develop significant chronic disease or death, though, as a consequence of these noninfectious risks.

Then if we sum this up, we have the infectious risk estimate of five per 10,000, the risk of transfusion reactions, three per 10,000, for a total risk per unit of eight per 10,000, nearly one in 1,000 units is associated with some kind of a risk of an adverse effect. But, again, if we run the numbers through in terms of number of recipients, or frequency of recipients who are affected, it is 2.8 per thousand overall, and who really develop significant chronic morbidity or potential mortality, they estimate four per 10,000, or about 1500 per year among the seven million recipients develop a serious consequence.

Then they go on to nicely -- this is that same number, we had four per 10,000, which is forty per 100,000. This is your risk if you are a recipient of a standard four-unit transfusion of developing significant morbidity from a blood transfusion.

They then go on to say, now, if you are not going to get blood for sure. If you are just going into surgery, then based on the probability that a surgery patient will get blood, they reduce the risk to five per 100,000, and then they look at, what is the probability if you are just somebody currently walking on the street with no plans for surgery, no chronic bleeding disorder, what is the probability that you are going to go to surgery or develop one of these bleeding requiring transfusion problems, and they get a five per million. This is the risk to the general person in the population of chronic consequences, were they to need blood and suffer a consequence of that.

Then they go -- this is that five per hundred thousand. And then they go to contrast that with other risks, five per 100,000, if you are a general surgery patient; 40 per 100,000 if you actually need a blood transfusion. And then they contrast these blood-related risks with other health outcome risks. Being in the hospital and if you get in the hospital, your probability of dying or disability is 600 per 100,000. If you go through surgery, just the consequences of surgery is likely to cause death at a rate of 1300 per 100,000. Injury from a drug complication during a hospital, 6500 per 100,000. So, we are dealing with risks a hundredfold that of blood transfusion for most of these other well-known complications of being in the hospital.

Then they go further to contrast this risk, again, of transfusion-related death with other risks of non-hospitalized, or other problems; malignant tumors, CVA, etcetera. Again, just to juxtapose this relatively low risk of transfusion with the much higher risks associated with many of these other complications, or illnesses.

Then they go on, again, referring you to this document, because it really does a nice job of this, to contrast this risk with lots of other just general life risks. Electrocution, you know, drowning, nine -- so, tenfold, greater than tenfold higher risk of drowning, in general, than from transfusion. Stroke, heart disease, etcetera. So, a nice job of juxtaposing all of these risks.

This is sort of the summary of where we stand relative to those other risks, and I am sure we will hear in the next talk about how we should try to be comfortable or not comfortable with these considerations.

I just want to briefly touch on other approaches and just highlight a few examples of how we can try to make things better, but we often are trying to make things better where the risk is so low that we actually perturb the fabric of safety and make things worse, or could make things worse. So, approaches to reducing risks, these low level risks, involve trying to improve screening, education, enhance screening through the blood tests themselves, increased regulatory oversight, pathogenic inactivation, which is on the forefront, and I am sure, again, another issue will come to this committee but brings with it low level exposures to potential carcinogens, etcetera. And then obviously, reducing allogeneic exposures to autologous, or reducing blood usage, but that begins, as this again, GOA document shows, the benefit to most patients who need blood of getting blood is so enormous, that as you begin to consider not giving these people blood, you really can potentially harm these people much, much more than exposing them to these low risks.

Just one example, a paper that is in press now, where we have looked at the impact of excluding donors -- this has to do with CJD, and because CJD has mostly been documented in older people, there have been proposals and in fact in some plasma centers they do not draw blood from people over 50. And this has led to consideration in the whole blood setting, of stopping collecting blood, or selectively only drawing certain types of blood units, from younger donors. And the truth is, younger donors have much higher incidence of all of these major viral transmissions, so with this study, what we did was to evaluate what the impact of exclusion of older donors was on the net rate of new infections for the known viruses.

It turns out, if you exclude donors over 50, you would increase the risk of the major viruses by 10 to 20%, because you would perturb the blood supply mix, you would move it more towards younger, riskier, donors, and your older, safer donors with respect to these sexually-transmitted, drug-transmitted risks, begin to be pulled out of your blood supply. So, this is just an example of how the mix of donors in the blood supply can be perturbed toward increasing, quote, safety, for an agent that may not even be transmitted by blood, with the negative effect of increasing the risk, for agents that we know are being transmitted by blood.

Another example is moving to new tests, and we can add tests, like P-24 antigen or RNA. There is concern that adding these tests will recruit new donors who are at risk to be tested because they cannot get these tests anywhere else. Fortunately, that has not happened, but in fact, also, these are very expensive tests, and Jim AuBuchon and I did an analysis of the cost-effectiveness of adding P-24 antigen, or RNA PCR, and the bottom line is, as I think Jim showed yesterday, is the P-24 antigen estimate was a little over $2 million, RNA PCR over a $1 million per qualle(?), so very high numbers. This was on the assumption that we would predict -- that we would pick up as predicted in the range of five to ten infections per year.

The truth is, in the first year of screening well over ten million -- this is the Red Cross experience -- but, there has only been one infected donor found in the entire national screening program. In fact, in the first year and a half, in the whole blood sector, probably 18 million units have been tested of whole blood, and only one infected, true infected donor.

These other numbers represent all the horrendous problems with false positivity that we have had to deal with. None of these people are infected. We have had to do large PCR studies on all the people who were negative on the confirmatory tests, etcetera, but the bottom line is, that the yield was much, much lower than we even projected, and there was extensive criticism that these projections were underestimates. So we can put a lot of resources to try to improve safety, with very little bang for the buck.

This is, again, a slide in a review that Jim has developed and I participated in that is coming out in Annals of Internal Medicine very soon, of the overall risk and cost-effectiveness of blood initiatives, and again, contrasting many of these blood initiatives, such as antigen testing, anti-corp(?) testing here, P-24 testing, ALT screening of donors. Contrasting these, solvent detergent plasma, viral inactivation method, with the other commonly accepted interventions that are going on, and just illustrating that we are really beginning to get out of the envelope of reality in some of the things that people are doing to try to reduce risk.

Finally, just a list of the new agents. I do not have time to go into these, but a lot of things down the pipeline of concern, and everybody is out there to discover the next agent and to try to prove that it is a blood safety risk, because that is the political hot button. So, everybody who discovers a new virus immediately tries to document a transmission of this or that.

There was just a recent case of HHGV-8, where an in vitro transmission of the virus from a healthy donor -- this is a very common virus -- to cell culture, resulted in the inference that the blood supply is at risk for this, and now we are having to do large studies on HHV-8. So, most of these agents, you know, through surveillance studies, are, one, extremely rare. Transfusion transmission has never been documented. And even if they are infectious, the risk of disease is very low. So, we have to begin to try to put the same numbers on these things. And you can imagine the risks being so low that the studies must be quite large and intricate to be able to try to estimate the consequences of these things.

But we do have good strategies. Over the last year, there has been good efforts to develop strategies to measure the consequences of new agents through monitoring donor recipient populations, understanding the transmission routes in the non-transfusion settings, and implementing national surveillance of cases to understand whether transmission is from blood. So, I will conclude there, thank you.

MR. CAPLAN: Questions for Mike before he runs off?

PARTICIPANT: Mike, if I could just reiterate one point that you showed up there about the ABO errors and mistransfusions, just so that the committee is aware of this and the relative magnitude, because we are often talking about, as Mike has talked about, what more we can do, or what fancy new technology we can bring in to make the blood supply just a little bit safer. And every year in this country, we kill two dozen patients because they get the wrong unit of blood due to an entirely preventable human error. We do not need a PCR test, we do not need any new fancy approach to that, we just need good practice and traditional checking and make sure things are done right the first time to prevent a very preventable and disastrous consequence.

PARTICIPANT: Mike, have you calculated, or has Jim calculated, dollars per qualle for HCV RNA testing?

DR. BUSCH: Not as yet. I think -- one thing I did not mention, it is impressive to actually benchmark each of these new efforts against the current standard testing. HIV antibody testing is extraordinarily cost-effective. HCV antibody testing actually more than pays for itself. It has a very negative dollars per qualle. It is preventing so many transmissions that the antibody test is extremely effective.

The issue is, you talk about adding a new test, almost always it will never replace the old test, because you are basically trying to pick up window phase or occult infections, but some people who are sero-positive may test negative on a virus detection assay, so you have to keep the antibody test in place. So the cost-effectiveness. The cost is borne by the infrequent, incremental detections, the rare pick-up.

It would be a good one to do. We have the outcome measures, and only in the last six months have people begun to recognize and acknowledge that there may be this pick-up rate of one in 10,000. So, I think we could probably do a good number on that now.

MR. CAPLAN: Okay. I am going to ask Dr. Maibach to come down. While he is approaching the podium, one of the things that I wanted to alert the committee to is, we will spend a little bit of time trying to think about the structure of the report next, and I think the discussion I have had with some of the committee members makes me think that it might be reasonable -- and I would like you to be thinking about this a little bit -- to use the first draft that Dr. Hoots put together, sort of the template, not that Dr. Kenner's insights will not go down in human history as valued, but it may be more useful just to go with that as the document we are kind of working on and editing and building around. So, think about that a little bit before we get back to the discussion of recommendations.

Dr. Maibach is going to talk to us about communication of risk.

Agenda Item: Dealing With and Communicating Risk

MR. MAIBACH: I really enjoyed the last presentation for a number of reasons; one, I am totally uninformed on most of these issues, and secondly, it was a very nice presentation of a substantial amount of quantitative risk information. And, in many ways, I am sure that my presentation will be the only one in two days that is completely devoid of numbers. And actually, I am going to go -- rather than start with my first slide, I am going to briefly reflect on my last slide for a moment, my final point on my final slide, and that is that, people simplify, and our job, in communicating complex risk information is to help them simplify appropriately.

It is a basic truism of human cognition that we -- that -- well, A, people are phenomenally adept at processing a lot of information, but by virtue of the overload phenomenon, we automatically invoke biases, heuristics and biases that allow us to deal with complex information in a rapid fashion, so, i.e., we simplify. And the big challenge in doing risk communication effectively is understanding those heuristics and biases that we apply, so we can understand how to help people simplify appropriately.

One of the fundamental observations in risk perception is that we have a tendency to overestimate uncommon risks; i.e., in this example, the risks of dread disorders associated with blood transfusions, and we underestimate risks that are actually quite common to us; i.e., for hypertensive and perhaps the risk of stroke, or for a smoker, the risk of lung cancer or other dread disorders. And that is sort of a fundamental paradox of risk communication. We overestimate the unlikely, we underestimate the likely.

That paradox is a direct consequence of the fact that risks are not unidimensional; namely, they do not go from the remote to the highly likely; that we do not perceive risks on a single dimension, but rather we perceive risks along a variety of different dimensions, or we respond to risks along a variety of different dimensions. Depending on which cognitive researcher you are paying attention to, that number of dimensions might be seven, or five, or nine, or what have you, but the basic point is that risks are not unidimensional. The way in which we respond to risk is really a direct function of where the risks fall along those various dimensions.

Risks that are perceived to be greater tend to have these attributes. Risks that are involuntary; i.e., imposed on us by external forces, as opposed to imposed on us by our own decisions or lifestyle choices tend to be perceived as greater, and therefore, we are more likely to respond to them with what risk perception researchers call outrage.

Risks that are uncontrollable, i.e., beyond our control, again, we are more likely to see those as greater. Risks that are unfair, namely, where I, as the individual, do not accrue the benefit of taking the risk, but the benefit is accrued elsewhere, the whole issue of environmental equity in environmental health is a nice example of that, namely, communities where there is risk imposed on them by virtue of where they live, but the benefit of those risks, for example, the corporate profits are going elsewhere.

Risks that are unfamiliar tend to be more dreaded, simply because we do not understand what the risk is really implying. Risks that lead to a dreaded condition are seen as living larger in our minds. Risks where there is uncertainty about the risks, i.e., when the experts cannot really explain to us what the consequences -- with any certainty -- what the consequences of those risks really are. Those loom larger in our minds.

Similarly, risks where there are potentially catastrophic consequences, i.e., when many people will have the potential to be affected simultaneously, as opposed to a single individual. And finally, risks that are memorable. If you think of the great catastrophes of our times, Three Mile Island is a single instance, that invokes -- it very readily invokes some mental heuristics that are quite frightening and therefore lead to greater perceptions of risk.

Coming back to the point that our job is to help people simplify appropriately, we might be persuaded inappropriately to use qualitative expressions of risk, as opposed to quantitative; such as, likely, very likely, highly probable. The problem with this approach is that there are very good studies documenting that people interpret those qualitative expressions of risk very differently. So, the confidence interval around the quantitative expression associated with the term, likely, for example, is as low as 30 to 40% and as high as 90 to 95%, depending on the individual who is responding. So, it is certainly not a good way to communicate with any precision, what we are really implying as a magnitude of a risk.

Conversely, is the fact that most people have a very difficult time understanding quantitative expressions of risk; namely, all of those numbers that we just saw, the average viewer, the average listener, myself included, has a really difficult time wrestling with those and putting them in context to what the risk means to us as individuals.

Fortunately, there is some literature indicating that visual representations of risk, believe it or not, the kind of thing that you see in U.S.A. Today, are actually perhaps the best strategy for conveying complicated risk information. And moreover, when visual representations have corresponding qualitative and quantitative expressions of risk, i.e., when the text matches the visual and numbers are included in both the visual and the text, that allows people to get the best sense -- if, by best I mean, most accurate sense -- of what is attempting to be conveyed about the risk.

People value a given piece of risk information more when they perceive the source of that information to be credible; i.e., disinterested, personally or professionally, disinterested in the outcome. Industry is often greatly perplexed by the fact that, even when they attempt to be forthcoming with risk information, they are typically viewed as being suspect in the whole process. The information, even if it is generate by external scientists, is perceived to be not credible.

Government sources, on the other hand, are generally perceived to be credible, although as you all know, as time goes on, even government information is increasingly seen as suspect. But credibility is clearly the key to taking risk information on its face value.

Somewhat associated with that is the fact that people value information that is perceived to be certain. No one likes uncertainty. And when we present risk information with uncertainty, which unfortunately, as you all know, is almost always the case, we are rarely certain about the risk information we have -- was struck in the last presentation by the degree of certainty that was being implied, and perhaps that is because you-all have been on this for so long -- but typically, when there is full disclosure, you have to disclose the fact that you are not fully certain about your risk estimates. And when you do that, people tend to either be suspicious of the source of the information; they feel you are withholding knowledge that you genuinely have, but do not want to be forthcoming about, simply because the audience will not like the implications; i.e., why don't they just tell me, what are they hiding?

In addition, if the recipient of the information is suffering from a risk that is not self-imposed; i.e., externally imposed, to the degree that the information is presented as being uncertain, they will tend to lay blame on the communicator or on whatever source they perceive to be causing the risk, and they will lay blame and want to do finger-pointing as opposed to do problem-solving.

Conversely, if the recipient is the cause of the risk, they will tend to discount the risk; i.e., again, the smoker situation. It is not really as risky as they are telling me it is.

Perhaps the most important point I want to make today is that, people are generally motivated to reduce risks that they face, but that is, I say, generally, because simply because you perceive a risk, and even if you are motivated to reduce that risk, does not necessarily mean an individual will change their behavior to actually mitigate risk. And there is an enormous literature on behavior change that really speaks to this point. There are numerous different reasons why people who appreciate their own risk, who are motivated to reduce their risk, simply do not, or are incapable of taking action to reduce the risk. And therefore, it is risk communication -- the efficacy of risk communication should not be judged, based on its ability to change people's behavior. That is an unfair evaluation criteria.

Instead, it should be judged on its ability to impart an accurate understanding of the risk, that is really the ultimate criterion against which risk communication should be judged. Beyond that, risk communication can be judged by its ability to facilitate reaching a good decision. What constitutes a good decision for an individual, is simply beyond appreciation of the communicator, or the communicating organization. Individuals have many different contingencies in their lives, and are going to have to consider all of those contingencies when deciding how to react to risk information. And therefore, again, the point is, do not judge the consequence of risk communication based on the action, but based on how well-informed people are, and the degree to which the information has facilitated them to make a good decision, based on what they value.

One common, very common technique, in attempting to communicate risk information is by making comparisons to other risks. The irony in this is that it can either be one of the most helpful risk communication strategies, or one of the most harmful risk communication strategies, depending on how you actually go about making comparisons.

The seven dimensions of risk that I introduced earlier, when we make risk comparisons among two risks, or between two risks, that have different dimensional profiles, people will reject those comparisons, and therefore, not only gain absolutely nothing from the risk comparison, but will essentially discredit the communicator.

Conversely, if we are able to identify risks, known risks, risks that we understand because we have a lot of experience with them, that have a similar dimensional profile to the risk that we are trying to communicate about, that can be a very, very useful risk communication tool, but it is -- given the fact that there are at least seven different dimensions that we have to consider -- not at all a straightforward process. Namely, when we make these comparisons, we have to be very careful, we have to be very measured in the comparisons we make.

A number of other ways of conducting comparisons, however, are more helpful, and perhaps more straightforward; namely, comparing the risk with its benefits, and an example commonly in the lay literature right now, or the popular press, is hormone replacement therapy, for post-menopausal women. And, yes, there are undoubtedly risks associated with hormone replacement therapy. Conversely, there are also benefits associated with hormone replacement therapy. The best way to communicate about those risks is to communicate them in juxtaposition to the potential benefits. It is a much more balanced way of communicating, because it helps people understand the information more thoroughly.

Similarly, comparing the risk in question to risks associated with the alternatives; again, back to the hormone replacement therapy example. There is clearly a risk of not taking hormone replacement therapy by post-menopausal women, and understanding the risks of taking action, versus the risks of not taking action, is a very nice way to help people really place the decision that is in front of them in context.

Finally, in some situations, you can compare the risk to natural background levels of the disorder in question. So, for example, cancer clusters, I mean, there will always be a background level of leukemia in any community, and the concern is exposure from an old plant that has been shut down, and there is fear of a cancer cluster. It is very helpful to compare the level of expected leukemia in a given cancer, compared to that actually found.

To sort of bring this to closure at the most simple level, there are a number of ways we know about that are sort of tried and true in terms of improving risk communication. One of those is to provide the information from a variety of perspectives, including both relative and absolute risk, humanizing the estimates of risk; i.e., do not simply present numbers, but present how people figure into those numbers. One in 1,000 people are expected to come down with the disorder, or one in 10,000 people, or one in 100,000 people, and this is equivalent to. Definitely, you need to include people in the equation, because that is how we think. That is the basis on how we make decisions, because we, after all, are people.

As I have already made the point, that it is very helpful to compare the risks we are trying to communicate about with similar risks. Risks, i.e., with similar risk profiles.

Be clear about the degree of certainty in your risk estimates. I have already said that uncertain information can be more volatile. On the other hand, if risk information is presented with certainty, and it is in fact uncertain, that is the most volatile potential situation, because that is where lots of suspicion tends to be brought in, and a lot of finger-pointing and blaming.

A very important point is, use language that is appropriate for your audience. Rarely is scientific jargon appropriate for communicating risk information. One of my own -- sort of a related point, and an area in which I had the opportunity to do some research was, understanding of consent forms as an example of language that is typically inappropriate for the audience. Most patients are incapable of understanding consent forms the way they are typically written. The risk information is absolutely presented there in excruciatingly painful detail. Unfortunately, it is presented in such a way that the person who we are attempting to inform, is incapable of understanding. We have to go to great pains to use language that is appropriate to our audience, even if that entails research to understand what language is appropriate for a given audience.

I come back, again, in conclusion, to my opening point, and that is, that people simplify, and our job is to help them simplify appropriately. It is a basic human process to simplify information so that we can get through our day and get the things done that we need to do. If we really want people to be informed, we, as the communicators, have to go through all of the machinations required to understand how to help people simplify appropriately.

Are there any questions?

PARTICIPANT: That was an excellent presentation. I really enjoyed it because we are trying to communicate this type of information to the lay public, and politicians all the time, and we find it very difficult, not to mention, reporters.

When you said that we should not measure, or we should not judge our communication by its ability to change behavior, but then you said that, you should look at whether or not a good decision is made, as to whether we communicated effectively. Isn't that, then, the same thing? In other words, in the current context, we are attempting to get individual -- we want to notify individuals in one way or another, that they may have been exposed to an infection by nature of the fact that they had a blood transfusion. And whether or not we can evaluate the types of notification that we use is questionable, but if we attempted to do that, one way we would evaluate it is by whether or not they went to see a physician, or got testing. Well, that is a behavior, in essence, so what is the best way, or can you put that in a different context for me, or give me an example?

MR. MAIBACH: The question was, if we are not to evaluate the efficacy of risk communication based on the resultant behavior change, isn't it comparable to say that we are evaluating based on imparting an accurate understanding of the risk? I think I misstated the question, but I understand the nature, and I will address it.

If the objective of the communication is to convey to people who may have been exposed to infected blood, that they may have been exposed to infected blood there, and they should seek medical evaluation, the evaluation criterion for that should be, did they understand what you were intending to communicate? It should not be, did they seek medical consultation? The reason I say that is, because your job as the communicator is simply to ensure that they got the message.

The reasons that they might not seek medical evaluation is, they may be fearful of doctors. They might not be able to afford medical care. There are a hundred different reasons why they might not seek a medical consult. All of those have to be somebody's concern, but they are not your concern as the communicator of the risk information. And if your communication is deemed as faulty for not getting people into a medical consult, well, then that just simply makes you the scapegoat for why the situation is not improving.

My point is, that if you have done your job; i.e., the person is truly informed that they are at risk and they have a balanced appreciation of their risk, then you are off the hook and somebody else is now on the hook, if they do not move into the healthcare setting. And my whole point is that that is an appropriate way to judge sort of a risk management scenario.

PARTICIPANT: So, how do you evaluate whether or not an individual has perceived accurately the message that you intended to get to them, to communicate?

MR. MAIBACH: There are ways. Ideally, you create a risk communication approach, and you test it with a limited number of individuals. You really get involved. You interview them, you ask questions from every possible angle to ascertain the fact that they really understand the risk. That they are using that risk information to make a judgment that makes sense in the context of their lives, and then you go public with your risk communication campaign. And then you may want to do the subsequent step of evaluation, which is, you know, on some population basis, make sure that people actually got the risk communication that you intended they would receive, and just as your test subjects responded appropriately to it, your population -- some sample of the population is responding appropriately to it. If you have gone through all of that, then I would maintain that your risk communication program was an unqualified success.

MR. CAPLAN: Let's go to the next (inaudible).

PARTICIPANT: I think there are those times when, however well we try to communicate risk to populations, there are those other competing forces that are generated by the media that sometimes tends to sensationalize or misinform or confuse the issue, and coupled with that is what seems to be at least in some communities, and it relates to some of the points that you were making about some of the environmental concerns, in particular, that there is a growing cadre of attorneys that are very quick to move into communities, and to take their own information and to find their own list of experts, that come and also try to take a message that is counter to what one is trying to communicate. How does one try to reconcile some of those tensions and confusions?

MR. MAIBACH: That is the art of communication rather than the science of communication. What I have attempted to present this morning really is the science, based on cognitive psychology, and an understanding of how people process risk information.

The art of communication, on the other hand, really involves how to work, how to do media outreach, so that journalists understand what you are trying to communicate, and are able to faithfully reproduce at least the core message that you as the scientist feel must be communicated, which goes a long way towards diffusing opportunists and moreover, an environment, a community environment, in which opportunists can take advantage of alienation and suspicion.

I have the privilege of working with the CDC on a number of HIV communication issues, and we have been particularly concerned -- this is just by way of example -- we have been particularly concerned over the past year, about a potential misunderstanding of the development of combination therapies for HIV infection, and specifically, we were worried that there might be members in certain high risk communities who would sort of backslide behaviorally because they interpret this as being a cure for AIDS, and therefore, they do not have to be quite as vigilant anymore.

We talked both to individuals at risk about their behavior in focus group settings, and we did an analysis of the media, content analysis. Much to our delight, the message as transmitted through the media, was transmitted with almost perfect fidelity; namely, these are not cures, these are wonderful breakthroughs. They are very hopeful. But more is not known about them than is known about them. And they should never be misunderstood to imply the fact that you need not be vigilant anymore. And, moreover, we were extremely heartened to find out that members of -- in this case, we were focus-group interviewing gay and bisexual men, virtually without exception, they understood that to be the case. They did not see combination therapy as a cure, and they claimed, at least, that not in their wildest dreams would they abandon their safe sex practices because of the development of these combination therapies.

MR. CAPLAN: To Jim?

DR. AUBUCHON: This committee is faced with several different approaches, potential approaches to be used to patients who may have been transfused with units that may have transmitted, or may have exposed them to a particular viral agent in the past.

We have some choices including, probably, a direct letter saying, you received a transfusion and we have reason to suspect the unit that you personally got, versus more general public service announcements, just saying that, if you were transfused in the past, you are at some risk. Do you have any comments about the relative efficacy of creating the desired type of behavior in the recipient of that message?

MR. MAIBACH: I am under the impression that there are studies that have actually looked at this question. The National Center for Health Statistics, as you know, conducts a number of national sort of assessments of our health status, and in the process of doing that, the results from physical examination and blood work-ups and what have you, often they detect disorders, and have the ethical obligation of informing the participant that, in fact, they may have this disorder. And I know that they have been involved in some tests of a variety of different strategies. I honestly cannot tell you the outcome. I can, however, give you my judgment as a person who does communication on health issues full-time, and that is, the same message through multiple channels inevitably has a more positive outcome. So, if there were, to the extent to which it is made -- the information is made available to the general public in a broad fashion, namely, a campaign fashion, you will have an audience who are better able to understand, when they receive a letter, indicating that they in fact may have been exposed. But, perhaps even more importantly, is the fact that the way that letter is written, I would say that the specific language used is going to be all-important in determining how people respond to it. Language, a letter that is written in medical jargon will certainly confuse, alienate, cause a lot of unnecessary fear, and may consequently, ironically, wind up producing less desirable action as a result.

DR. AUBUCHON: But just in general, if you had a choice for a health-related concern, would a direct letter be more effective, or an overall campaign that would raise public awareness about a potential risk be more effective?

MR. MAIBACH: If I could only choose one, I would go for the direct letter, but if I could change the scenario, I would say, public communication, direct letter, as well as professional education, because often, that is the person -- health professionals will be the first group turned to for additional information, and when you have synergy among the message in all three, you are going to get the best outcome.

MR. CAPLAN: Let's do one more over here, Bill.

DR. HOOTS: I am concerned about how you translate very low or infinitesimal risk to individuals. Do you -- I mean, would you advise saying, alright, your chances of -- because obviously, part of a physician's and a healthcare provider's role is reassurance, when it is appropriate to reassure. Do you put it in a context of something that you think you understand mathematically, but maybe infer differently by different people, like you said? Well, your chances of this occurring are about like your chances of winning the Lotto today. Or, do you take it more toward the absurd, and something that probably is far beyond their expectation of occurring, to get the point across that it is exceedingly low, even if the numbers do not quite match up?

MR. MAIBACH: The example -- the risk comparison you just made, your chances of a disorder from a blood transfusion is equivalent to your chance of winning the Lotto today, would be an awful risk comparison, for a variety of reasons. One is that, they are fundamentally different behaviors. One is a risk imposed upon you, literally, while you are probably anesthetized and out on the table, and the other is something that you choose, a risk you choose to take yourself, moreover, with great hope that it is going to be your ticket to the big time, and quitting that job, and moving to Florida. So, that is a risk communication that is almost certain to be rejected.

Conversely, or sort of alternatively, if you could find a risk that was well understood but equally improbable -- and it really does not matter if it is one in a million or one in ten million, or even one in a trillion, I do not know, that might be too remote, but for example, the risk of being struck by lightening -- this is totally off the top of my head, it might be that the risk for being struck by lightening might be seen as comparable to the risk of a transfusion-related disorder, simply because those are both things that are imposed on you, from externally. I do not know, I would have to -- clearly, there are many different dimensions we have to consider.

DR. HOOTS: Right. And the problem I think that you run into is, if you use an analogy like that, it may actually, because of their own cognition, or having just read stories of three people who got hit by lightening, exaggerate the risk to them. If you go more towards the relative absurd, something you know they have not heard to have occurred, because it has not occurred within aeons, is that -- even though it is mathematically less accurate, is it both appropriate --

MR. MAIBACH: I would say that, as long as they really understand the risk you are comparing it to. Understand it, and feel comfortable with the fact that it is an extremely improbable event. But you can reiterate that in a number of ways, namely, using appropriate visuals that represent the minuscule nature of the risk, using language that is consistent with the visual, and using the numbers as well. I mean, people are not totally incapable of understanding one in a million, but they are much more likely to misunderstand one in a million, if all you are telling them is, one in a million.

MR. CAPLAN: Okay, thank you very much.

MR. MAIBACH: Thank you.

Agenda Item: Open Committee Discussion

MR. CAPLAN: Let me ask Paul if he could put Dr. Hoots' original draft back up on the overhead. I asked Steve Nightingale if he would help us out, too, in working the overhead. I hope you have a pen and some blank sheets there.

Let me say a few words about our work for the next couple of hours. You remember, we have been tasked, asked by the agencies and the White House, and indeed by Congress, to do what we can to make sure that people who are exposed to hepatitis-C, or face the risk of being exposed to hepatitis-C, know about these risks, understand them, and that we are charged to come up with recommendations to the agencies, that they can really put into practice as policy. I mean, if you were trying to think of, what is the guiding theme for us, it is to come up with recommendations that they can put into practice.

There are a range of questions that have come up, some of which we have been asked about, some of which we have not, and that is why I am not putting the original questions back up on the overhead, because we have kind of come a distance in the two meetings. I have a proposal about the structure of this.

I think the Hoots draft, as we started to formulate it, is very good. And I think there is much to commend sticking with many of the things that are said in there, although I suspect we are going to want to add and debate a bit about some of these points, but I think we might think about a structure that went as follows.

Something like, in front of what is up here, a little bit of what I am going to call, for want of a better term, a preamble, and that gets to the issue of, why are we doing this, some of which is captured in recommendations here. See that recommendation two, it says the advisory committee is dedicated to expanding the public trust about safety. I might move that more to the preamble. That is not a recommendation, it is a reason why we are meeting, why we are trying to make some recommendations, why we are trying to talk about things. And I have actually drafted up a little bit of language of some other things that might go in this brief preamble, but the first part might be, why now? Why are we meeting? What are we trying to achieve? We want to cement the public trust. We want to make sure that people are informed to the extent they can be and so forth.

A second section then might be, facts. When I was thinking about this last night, taking my own assignment seriously, something I am not sure how many of the rest of you did, but when I was thinking about this last night, I thought, well, one way to capture some of the reasons why we are trying to struggle with a policy change at the present time is because there are certain facts that are mercurial, and I could mention a couple of them.

There is a new accurate test in 1992. There is new information about lifestyle that might be relevant to prevention. There is new information about therapies. There is new information, in fact, about risk. There may be new information about new tests coming down the road, PCR-based testing and so on. I've got five or six facts that would make, I think, some sense to people as to why we are saying, well, we did not do this, this, or this in the past, but maybe we are going to do this now. They are not principles, they are basic facts that have shifted about where we are with respect to hepatitis-C. Then I think we could move to some principles, maybe two of those will survive. We could add one or two others if appropriate, and then into the recommendations.

The first thing I would like to have you comment on -- I know it is a little bit hard to wrestle with this as a structural feature -- is, whether that seems like a reasonable structure to raise the recommendations? I am not proposing a GAO-length report here, I think for policy purposes, we have to keep ourselves succinct. I take seriously the advice from the previous speaker that we have to simplify in order to be heard. I do not know what language we should learn in order to be heard, but we have to simplify anyway. So, maybe a preamble-type statement, a couple of paragraphs, two or three. Six or seven facts about where we are at with hepatitis-C. Things that have changed that have made it important to think about what we want to do now. Some principles to follow. And then the recommendations. Does that seem like a structure that people are comfortable with? So, it would run, basically, preamble, facts, principles, recommendations, is what the thing looks like.

We could put that up there, Steve, as a -- what I would like to do, then, if it is alright with you, is just maybe start, not with the preamble, but I actually wanted to start with the facts.

One of the little lessons I have learned over the years about values in policy is, if you can get agreement on the facts, you can sometimes move toward consensus faster than if you try and argue about principles first. So, in the spirit of respect for the empirical, here are just six facts, or six areas of empirical information that I thought were important in thinking about why we might, as a nation, want to change our practices, or what we do with respect to hepatitis-C.

The first is, the emergence of a reliable test to detect hepatitis-C in 1992. It seems as if we have the second generation hepatitis-C test and that shifts the nature of the ability to detect this problem in the blood supply of hepatitis-C.

The second fact I would say, is the availability of new knowledge about risk and transmission. And I noted, one of the things that seemed to me to be important is that as we have heard presentations about the differences that might occur in blood transfusion versus parenteral-acquired hepatitis-C, there is new information coming out, it may not stand the test of time, but there are new suggestions about course of disease, risk of transmission, risk to health of someone who has hepatitis-C, how long the disease may lie dormant and then suddenly activate. These kind of facts are relevant. They may not have been known in 1987. They are known in 1997. So, facts about probability of harm to the individual who is infected, information about transmission.

Third fact. So, fact one, Steve, was just the emergence of a reliable test in 1992. Availability of new knowledge about risk in transmission. The third fact is, we know that there seems to be much greater risk to people who received transfusions prior to 1992. We kept saying that yesterday as something we needed to grapple with, and decide how to deal with it, but there it is, it is a fact, and it is something we know now, partly due to the emergence of that test, that we have a much safer blood supply in terms of hepatitis-C than we did.

A fourth fact, new knowledge about prevention through lifestyle change. I sometimes think of this as the don't drink fact, but there may be other lifestyle things. We have had a little bit discussion about sexual practices, and other things that people might do, but certainly, in the area of alcohol use, based on some of the testimony we heard at the last meeting, there is a suggestion that certain lifestyle changes might be important.

PARTICIPANT: The effect of outcomes on (inaudible).

MR. CAPLAN: Effective outcome, yes.

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Fifth fact, certainly one that is very relevant here, the availability of new therapies. So, this is our interferon, maybe other things as well, but this is different. It is certainly a material fact that makes us think perhaps it is time to revisit the policy about hepatitis-C.

The sixth that I could come up with was the emergence of new techniques for assessing hepatitis-C. This is the new PCR technology that looms. It is part of the reason why, even though Mike has pointed out a number of times that it may not make obvious sense why we are doing, or talking about prospective look-back right now, maybe it does if we are going to pick up more cases than we did not see, and some of the information you actually gave us, Mike, this morning, kind of makes me think that there may be new technologies that are leading us to detect hepatitis-C in ways that we could not do even a year or two ago. Or, will be able to do a year or two from now which might change our record keeping and look-back, and practices.

So, those seem to me, those are the facts that I generated in the privacy of my own slumber. Some of you may want to add to those, but they seem to me to make -- if I was trying to explain in simple language to an agency or a congressman or a person from the general public, well, why change the policy now? Why are we talking look-back, or announcements or something? Some of these facts seem to me to be the driving force behind that. It is not that we just woke up one day and said, let's change the policy. It is because we have this information that is making us want to rethink, perhaps, how we have done things to date.

I am certainly open to altering those facts or adding to them, or changing them. But, anyway, I thought that might be a way for us to get going in terms of the next, let's say, before we get to the break.

PARTICIPANT: [Comment off microphone.] The Army's therapy (inaudible) is inadequate.

MR. CAPLAN: Yes.

PARTICIPANT: The epidemiology, as we discussed yesterday, is not entirely worked out, it is a debate, how many individuals are going to move towards end stage in a relatively 10 to 20 year period, versus 40 years? So, there -- I think it is important to also state in the fact situation, that there are still some unknowns.

MR. CAPLAN: How would you want to -- I understand what you are asking, I am just trying to think how to put that. There is -- I mean, in one sense we could say, I mentioned interferon. I should not neglect the fact, by the way, we did have a ton of testimony last time. This is not what you are asking, but I remind you, we did hear about liver transplant last time as a therapeutic modality, from Dr. Hoofnagle(?), and the prospect is there, one might say that, new therapies and new knowledge about epidemiology may be changing. I mean, that it is emerging, put it that way.

PARTICIPANT: There is rapid development of new knowledge -- recognition of rapid development of new knowledge in this field.

MR. CAPLAN: And you were talking about the trials that you are running on different types of interferon and so on, so it seems as if there is a shift in thinking epidemiologically, and the prospects for even more improved therapies may be in the offing, which will lead to some policy changes. We might try a rapidly advancing knowledge base, leading to the prospect for new intervention, and better understanding of the epidemiology of the disease. Dr. Guerra.

DR. GUERRA: I think a couple of other facts, one is that it seems to be more prevalent in men than women, and two, that it leads to serious liver disease at a relatively young age.

MR. CAPLAN: Do you think those are new facts relative to what was understood about hepatitis-C, let's say, pre-19 -- I mean, is this part of this new epidemiological understanding?

DR. GUERRA: It seems to me that, as we gather more information, we are finding that the median age is really quite young.

PARTICIPANT: I think one of the things that is really important here is, not only do we recognize our charge here with respect to transfusion, but how great this disease is in terms of the number of people infected totally in the country. We need to put that in relationship. People need to know that there are in excess of four million people infected. Transfusions represent a certain part of that, but we do not want to lose sight of all the other people and what we can do in terms of having them recognize the need for them to be tested.

MR. CAPLAN: The way to put that is, if you are amenable to this is, we did say yesterday that we have now recognized that hepatitis-C is at epidemic levels, and it is clear that the nation is going to move to try and grapple with this health challenge, and this becomes a part of that, and that is another reason why we are rethinking what we are doing here, because we are not treating this as somehow a transfusion-related problem only. It is unfolding against the backdrop of the hepatitis-C challenge that is just out there generally, and this has to become part of that.

I think that is true, that is a fact that is absolutely moving this along, so we might put it, it is four million people affected, Steve, but it is also, with efforts being made to grapple with that epidemic, and so this, the fact here is that we are trying to grapple with this problem as part of the overall attempt to deal with this -- recognize -- now recognized epidemic of hepatitis-C.

We keep using that four million, I keep wanting to say, at least four million. All those numbers of the institutionalized and so on yesterday made me think that is the conservative four million. Yes, Tricia.

MS. O'CONNOR: Just a small point here, but -- and I am not saying it is a new fact, but in terms of this word, epidemic, I would say, as opposed to HIV, HCV is something that has been endemic in the population for a long period of time, and it is only with the availability of these new tests that we have the -- you know, can detect it, and measure it in some way and follow its natural history in a much more precise way than we would previously. Because HIV and HCV are very different because of HIV being recently introduced and truly taking on an epidemic course.

MR. CAPLAN: The recognition of a problem as opposed to the emergence of a problem. Jim?

DR. AUBUCHON: Would the committee be comfortable in adding to this list of facts that, the recognition that traditional look-back methodologies have limited effectiveness and/or efficiency?

PARTICIPANT: I would not agree to that.

MR. CAPLAN: I was tempted to put into this list of facts, something about a recognition that record-keeping and the American health system is less than optimal, leaves something to be desired, could be better? I did not quite know how to phrase it. I know what --

PARTICIPANT: (Inaudible) limitations of record-keeping --

MR. CAPLAN: There are severe limitations on --

PARTICIPANT: (Inaudible) recognition of severe limitations in record-keeping.

MR. CAPLAN: Yes. By the way, I was not sure, and just as a matter of fact clarification for me, did I hear yesterday that there was a seven-year legal requirement to keep records, or that is just convention, that a lot of places throw things out after about seven years? Does anybody know what the real --

PARTICIPANT: (Inaudible) real element there of --

MR. CAPLAN: Yes. Jay, do you know?

PARTICIPANT: A lot of establishments are required to keep records for five years, however what was at issue yesterday was the length of record-keeping by hospitals, and that is much more variable.

MR. CAPLAN: Right. But is it -- there is no seven-year statute or --

PARTICIPANT: There are hospital records that -- not to my knowledge.

MR. CAPLAN: Yes, it is not like the IRS or something, we have to keep tax records for years and so forth. Yes. Well, at any -- be that as it may, record-keeping certainly does leave something to be desired, or has limits that have to be recognized, and that may be a way to bridge this issue. Ron.

DR. GILCHER: I want to come back to what Mary said. I think it is really important that hepatitis-C has been endemic in the population, but we did not have a way to recognize that until just recently. And I am going to shift gears for a second and go to HTLV in Japan, exactly the same thing. What was recognized for a long time was the leukemia that occurred from it, and ultimately that led to the discovery of the retrovirus that caused it. But what I am really leading up to is, the Japanese have in fact found a way to intervene and in a matter of two generations they expect to eliminate this virus from their population, simply by preventing the vertical transmission, that is, breast feeding.

Really, what I am saying is, it is important, what Mary said, but also that we now have a way of detecting it, and potentially interventional therapy could eliminate this from the population. So it goes, again, beyond the transfusion-transmission portion.

MR. CAPLAN: I think we do have to get that phrasing right. It is that we have a situation in which we can now detect the fact that four million people are affected, as opposed to the emergence, the burst of some sort of epidemic.

I would say the second part of what you are saying, Ron, I am trying to capture with this phrasing of evolution of new knowledge and the prospect for new cures. It is basically, there is a hope, not just a fear, a hope that maybe we could intervene to do things that would reduce viral load, or allow people to clear the virus and so on, either with some current therapies, but more likely with some emerging therapies.

DR. GILCHER: But, what I am saying are initially two things. Not only the chance to cure the disease with the new therapies, but also to prevent further transmission of this disease within the population, because we did not know that in the past, that is why it is endemic. And now we have a way to prevent further transmission, not only cure those potentially who have it.

MR. CAPLAN: Mike.

DR. BUSCH: Yes I would suggest perhaps two different areas of facts, bullets if you will, one HCV general fact, most of these are that, the level of endemicity, the ability to now detect the virus, all the issues about available treatments, understanding of natural history.

Then a second which is the issues related to transfusion and HCV, which could talk in a little bit more detail about how the risk has actually been reduced over the last several decades, since 1992, with the availability of the current generation, test risk is exceedingly small, get into the issues about ability to track recipients is problematic due to issues such as availability of records, etcetera. And perhaps get into some of the issues in a comfortable fashion that Jim raised, about mechanisms to communicate historical risks are problematic, and the options are --

MR. CAPLAN: Certainly could sort some of these facts along those lines. Let me just suggest this, so that we do not proliferate facts, always feared by philosophers. Proliferate opinions, not facts. What I would like to see us do maybe a little bit is steer ourselves by the facts that are new or materially relevant to explain to someone, not part of this group, sitting on the outside, why did you take a look now? Why were you revisiting this? I am certain there are a bunch of facts we could get into about the tests and so on and their accuracy and what could be done. I am not sure that is what I set up the problem is with. It is sort of, drive it with the facts that tell somebody, why are you taking a look at this now? What shifted, what changed, what is it that is making you rethink this situation?

I think if we do that, the public, the agencies, congress will be in a position to understand, oh, well, they did not have these practices in place before, was that because they were venal, is that because they were stupid, is that because they did not care? I suspect not, I think it is partly -- well, a lot because, we've got a factual situation that is evolving, and that is what I would use to edit what we have in this section, sort of driving it by, here is what made us think that it was time to rethink our public policy, rethink the challenges that PHS faces and so on. Other things that you might want to get up there. Yes, Bill.

DR. HOOTS: But just in terms of kind of constricting some of these to put up, up higher, up on the list, really, the ability to test accurately that allowed to actually classify and clarify and characterize the endemic. And a lot of the things that are below the black dash mark up there, are things that actually come under that context, and could actually be listed at --

MR. CAPLAN: Under there, or part of that.

DR. HOOTS: As in number seven, just in terms of, because that really -- I think understanding just how big an endemic this really is, is clearly driving -- one of the things that is driving our decision-making.

MR. CAPLAN: Yes. And it is clear -- I think that is right. I think part of the push to grapple with this now-detected problem, generally speaking, of hepatitis-C is fueling attention back to, so what about the transfusion-related causes, but without the former, I would not have a look so much at the latter, and so I think that is right. We have to capture that as relevant here. Others?

PARTICIPANT: Just a question about order. If you put 1992 first, my first reaction would be, well, what have you been doing for the last five years? So, while everything there is right, perhaps not focusing on 1992 as the very first thing we put out.

MR. CAPLAN: Yes, I thought of that. It did make me think sometimes, what are we doing with it? On the other hand, I am certainly willing to drop it down. I have no love of that being the number one fact, but the -- I think what happens in part with these new tests is, you get them, they disseminate, people understand them, they trust -- it sort of, it appears in 1992, but it kind of absorbs or what I guess the technology calls, diffusion. It takes awhile, but I understand what you are saying.

PARTICIPANT: We know that, but the public does not, necessarily.

PARTICIPANT: [Comment off microphone] by the new test.

MR. CAPLAN: Starting in 1992, something like that. Yes, Dr. Guerra.

DR. GUERRA: Part of the epidemiology discussion also relates to, what do we know about the distribution of this virus around the world? And we heard some presentations from some developed countries, but what is the rest of the information that is out there?

MR. CAPLAN: Yes, Jim?

DR. AUBUCHON: If I could suggest that there is an element of timeliness related to the fact that the NIH consensus recommendations, only just issued in March of 1997, providing a firmer underpinning to directing people toward testing and possible therapy.

MR. CAPLAN: That is certainly true, and we could list it as an issue there, that we have had the emergence of a consensus report, that has made specific recommendations, and that is driving us to reexamine. I think I could -- I think that is probably true, too. That is part of that general recognition of the epidemic and where this fits in. But the appearance of a major consensus report is certainly a relevant fact.

PARTICIPANT: I think it is important to recognize that no therapy was available until 1995(?), that is when it was licensed, and the fact is that, at that point in time it was licensed for the treatment of non-A, non-BC, because we had retrospectively, basically, looked at those patients, and that was with the first gen test, and they looked to be C patients.

I think the intervening time between 1992 and 1997 has been a time of data-gathering, to see, indeed, if this therapy, although not what we would like it to be, had real efficacy, I mean, in the sense that the original trials did demonstrate efficacy, but as we began to be able to look at the virus itself, to see whether we could eradicate the virus, which is the gold standard.

So, there is a period of time in there where we had a therapy that was being used, and undoubtedly benefitted people, but now we can actually look at the virus and say, yes, the virus is gone, and based on this premise, we hope, as Dr. Gordon said yesterday, that you will not have any more sequelae in this disease. So, I think that, you know, you can explain that five-year period between the NIH consensus conference and 1992, and I think therapy has really been used since about 1995 when we could look at the virus, in that era when we were really able to do PCR and some other things, and see that we could eradicate the virus, and that there was a long-term benefit for this therapy.

That somewhat explains why the --

MR. CAPLAN: Hence, the time.

PARTICIPANT: -- tests, the therapy, and the virus, do not all fit together.

MR. CAPLAN: Yes. I am not ignoring that, I am just trying to -- I am mulling over a way to get that clearly put in there. One of the things we will do is, as we come up toward the break here, I am going to ask Steve if he could just meet during the break with maybe Mike and me, and we could see if we could recast some of these -- get the language sorted out a little bit in there. Yes.

PARTICIPANT: Yes, I think we also heard yesterday that there seems to be perhaps a natural shedding of the virus in the very young individuals. I wonder if that is an acceptable fact. That the younger the individual is, that they seem to perhaps have a natural resolution of the infection and seem fine. Wasn't there about a 10 or 15% I think --

MR. CAPLAN: Self-clear?

PARTICIPANT: Self-clear.

MR. CAPLAN: What I would say there is, I think we may want to capture this by, new understanding of the course and incidence of the disease, somewhere in there. That we understand more about its natural history, basically, that sometimes it goes into remission, sometimes it goes away, none of this understood before.

Let me see if I can sort of shift gears off this fact discussion. Steve, if you would, why don't you put the Hoots draft back up there for a second? Having now attained clarity about the way the world is, perhaps we are in a better position to look at some of the principles and recommendations that we were talking about at the last meeting.

I have one suggestion, which is that we are going to move statement two out to a more preamble, or introductory-type statement. It seems to me -- I have one other suggestion about the draft as it is there now. See recommendation three? Or, excuse me, principle three. It says, all recommendations made by the advisory committee should be ever-mindful of a wise stewardship about the public monies and energies.

I know what we are talking about here. I think what we need is language in those principles, that actually are principles directed to the agencies, not to us. This is a binding principle forever, to make the committee behave in a certain way. That is nice, but my suggestion is, we now want to get all those principles in the form that says, all recommendations, or all actions taken by government agencies should be ever-mindful of the wise stewardship requirement, we are not obligating ourselves, we want to tell them what to do in terms of policy, so that is an obvious change I thought of that we could stick in there. I do not think the principle is bad.

I had one other one, that we are basically following these principles out there, two of them there now, if we do that, which is that there is an old ethics principle that says, treat like cases alike, and it has come up a couple of times. Dr. Penner addressed it yesterday when he said, well, we cannot discriminate against some people just because they are unlucky, they have records, they do not have records. They are in a situation where they could be traced, they are not.

I think we might want to go with a principle that says -- do not write this yet, Steve -- but it is to the extent possible, people who have -- who are known to have received, or been exposed to, hepatitis-C through transfusion, should expect fair and equitable treatment. In other words, what we are going for is, some affirmation that, if they are not able to pay, if they cannot afford --

Well, remember we were yelling yesterday, but we have to say something that says, geez, you cannot just say to everybody, well, you know, maybe you were infected. Sorry you cannot pay for tests, sorry you cannot get treatment. Sorry the health system is not what it should be, see you later, but we told you. We want to make sure that we have a principle in here that says, every measure must be taken not only to inform people, but to treat them fairly and equitably and to get the access to the services they need, or something like that. I am struggling for a principle that commits us to, not just leaving people hanging out there with a warning and then -- not that this does not happen in the healthcare system, but, yes, Jim?

DR. AUBUCHON: Relating to your redirection of principle number three toward the agencies that would be implementing these recommendations, can I toss the hot potato to our colleagues from the FDA? Because the word from the FDA has been that, they are not allowed to regard cost when making recommendations. They will focus only on the safety of the blood supply, or safety of recipients. So the recommendation, if just issued in that form from us to them, would not be able to be implemented by them. Am I reading your philosophy correctly?

PARTICIPANT: I will comment on that. I think you are, the answer to your question is, yes. And in fact, one of the founding reasons for this particular advisory committee was to get a broader scope, other than the Blood Products Advisory Committee, which had in the past considered some of the issues, and then was redirected to strictly deal with the scientific facts, and not to look at the economics of the situation. It does put us in a bit of a Catch-22, to be quite honest.

PARTICIPANT: Also, with regard to redirecting it back to the government agencies, I am not sure that it would totally leave the committee off the hook, in the sense that, no, the committee does not have direct command of a certain dollar amount of money that can be dispersed in a way that we might, the committee might see fit, but I think the public is looking to us to come up with recommendations that are viewed as fair, reasonable, practical, prudent, and I think there is some -- in considering that, there is some element of cost, you know, because you have to consider, I think the committee has to consider -- as Kathy said, I mean, we have that duty, I guess, or responsibility to -- it is not going to be, certainly, the driving force, but it is a consideration, because if you recommend -- the committee recommends something that could translate into billions of dollars; you know, let's send a letter to every person who got a transfusion from -- I am going to say something really unrealistic -- 1970, you know, when the risk was truly quite risky. Then that could ultimately translate into a lot of dollars. I see the committee -- you know, I am sure the committee, can we get entirely off the hook?

MR. CAPLAN: No, but a thought on --

PARTICIPANT: Yes, well, just kind of where this came from is, out of our committee title. I mean, we are not only safety, but also availability, and tied up in the concept of availability are resources and energies. And that is kind of why that is in there, just to make sure that we have to balance, always, but always err on the side of safety, but still balance the completeness of availability of these life-saving therapies. And products.

If we go to the extreme on safety, there is always (inaudible) this, because we as a government, or we -- well, not, we, but, well, actually, I guess we are government employees in this context -- but, government agencies go to the extreme, even the FDA, and I think, certainly in the context of BPAC(?) and everything else, FDA has always said, yes, safety, but there has always been the concept of availability, and availability, I do not think you can separate from energies and resources.

MR. CAPLAN: Mike, then Eric.

DR. BUSCH: Yes, I think all of these issues here in one, two, three, should really be a part of the preamble, the mission of this committee. And in fact, the preamble, I would suggest, might be something that we would develop which would be a common preamble to all of the issues this committee addresses in the future, and then the facts related to specific issues that we are dealing with, and then the principles, how our mission uses these facts to develop principles related to the specific question at hand, and then the recommendations.

I do not know that any of these up here, perhaps other than this (inaudible) obligation to try to get to everybody, but maybe basically, we try to translate the principles, using the facts into the principles specific to this issue. But I think all of these could nicely fold into the preamble, in that sense that, you know, the requirement or the need to deal with stewardship, etcetera, is very appropriate for our committee to acknowledge it as our responsibility.

MR. CAPLAN: Eric?

DR. GOOSBY: I was just going to say a similar thing, so I will just abbreviate it. The committee is charged to come up with what the collective consensus is on these issues, with the commitment to being clear in articulating, once that decision has been made, I think the risk that is still afforded the individual around whatever, you know, blood element you are considering. That is such a fundamental prerequisite to the activity or work of the committee, that you are committed to a clear articulation, delineation, of the risk, once you have decided what is the reasonable activity, okay, in terms of the look-back, how far, what you are going to look at.

Hand-in-hand with that is that there be a clear communication of, at that point, the afforded risk. So, a commitment to what Dr. Maibach was referring to in his discussion, to communicate once that final decision has been realized by the committee, what the afforded risk is, I think is what you are being asked to do. And that could be a generic thing that could go over in a preamble, as Michael was saying.

MR. CAPLAN: Alright, I have a suggestion just for fun. We are getting out to the break, and that would be a time for great minds to meet. Let's gamble on something. Let's take a look at the actual, substantive, recommendations that are on the rest of this sheet, just to get us to be able to talk a little bit about something when we get out there, if we haven't got enough to say about facts or principles, or what might be in the preamble.

Let's do them one at a time. There is a recommendation there about perspective recipient notification, being implemented forthwith. I am just curious to see, just on that recommendation one, does that hold up, are people still comfortable with that? Are there any refinements of language that come to mind about that particular proposal? Sorry, times up.

PARTICIPANT: Well, I think the concept of prospective look-back to seroconverting donors, the truth is, if you do the numbers and you run the experience, it is extremely low yield, but it is a safety valve to unsafety, and perception is good, so I think we have to do this.

PARTICIPANT: I agree. My only comment was one of wording, the second-to-the-last line, be notified to facilitate HCV testing, and potential treatment of the recipient. It is not so much the units that we are worried about at that point, we want to test and possibly treat the recipient, exposed to -- potentially exposed recipient.

MR. CAPLAN: Carolyn(?), I see your hand up there?

DR. JONES: I had a comment, and it was not confined to item one (inaudible). If we want to have a general preamble that sort of is the committee's (inaudible) or what our position is going to be, or policies, or what our structure is going to be for addressing blood issues. I think it would be more appropriate to move number four, your additional one, down to the recommendations section and say that, if we have some perspective notification of recipients, or we have some sort of look-back, that there is some sort of requirement, moral obligation, to allow these people, or to provide people who do not have access to medical care, some sort of treatment or follow-up testing. I think that should be a part of the recommendation, rather than in the upper section.

MR. CAPLAN: Yes, that makes sense to me.

PARTICIPANT: We have not talked about the recipients here, and this is in the prospective aspect. And at one time we discussed the possibility of having an educational program for patients who are hospitalized who may -- or who have received blood products. In other words, the onus should be placed on the hospitals to provide some instruction for individuals as to risk factors, when they are going to receive the blood, or, if they have received the blood. That would be one element that I had considered adding in there.

MR. CAPLAN: Did you mean to add that to one?

PARTICIPANT: It would be at one or two, could even be at two, because that is part of the educational program.

MR. CAPLAN: Why don't you scrawl on the bottom down there, Steve, see under three, you can put three-B or something. Let's see, I better phrase that. The importance of enhancing consent on the part of blood recipients. Knowledge or -- and while that is moving up there, let me ask if --

PARTICIPANT: May I add one more on that, before we get off of it, and that was, the possibility of obtaining a blood sample on that patient who does receive blood for testing for hepatitis-C. I would like the group to think about that. It would be, we always get the blood sample for type and cross-match. We could set aside that sample, and the HCV assay could be obtained. So that we would have documentation, then, of whether the patient was positive to begin with. Then that would give us the alternative, or the option, of obtaining a sample of six months or wherever, to determine whether the patient had contracted hepatitis-C.

PARTICIPANT: [Comment off microphone.]

PARTICIPANT: Pardon? It would be a baseline blood sample.

PARTICIPANT: How long would the hospital keep that?

PARTICIPANT: They would not have to keep it, they would run it.

PARTICIPANT: They just test it.

PARTICIPANT: Anybody who gets blood should have an HCV test done.

PARTICIPANT: Why limit it to HCV?

PARTICIPANT: You wouldn't have to, you could get it broader than that, but I think, specifically, we are trying to address this problem, that is why I bring it up. But the testing that would be done, at least would document whether a patient is coming in with HCV. If they had received the blood, or even beforehand. Then one has the option of going back, if you find that the donor was positive, at a subsequent testing, of going back and saying, well, this individual was negative to begin with, and we know when the patient received the blood, therefore, if the patient is positive now, six months later it must have been due to the blood -- or, is more likely due to the blood product, that is all.

MR. CAPLAN: Carolyn, and then Mike?

DR. JONES: It sounds good in the abstract, but I think there are some privacy issues related to doing that, and I do not know that we want that information just --

DR. BUSCH: It would be the same as when we are doing the biochemical profiles. It would be information that would be available to the patient, and it is obvious the hepatitis-C business would be typically what we obtain.

PARTICIPANT: You know, that speaks to a much broader public health case-finding program. The prevalence of HCV in hospitalized patients, we saw some data yesterday that was just startling, it was like 20-some percent. So -- in your transfused population. Pre-transfusion, you are going to have anywhere from a 2 to 20% prevalence, just a background prevalence of the infection. And the incidence in the community, hepatitis, in the general population, let alone in hospitalized non-transfused patients, is probably one or two logs greater than the risk of blood.

So, even if you had a seroconversion that occurred, you know, it is not to my mind presumptive evidence that a transfusion of current screened blood caused that. And just the logistics, I mean, basically what you are talking about doing is implementing on a routine basis, the Hopkins Houston Study that I alluded to, which focused two or three hospitals, cost $20 million over the course of two or three years to conduct that kind of activity. It is a massive undertaking. You have to then manage those infections.

MR. CAPLAN: One way to handle this, if we are not ready to get agreement on whether it is a good idea to seek baselines, is that we certainly could ask the agencies to explore the desirability of having a baseline. I mean, we can come back to that one, that may not be -- if it is up for both privacy and cost concerns, we can certainly ask that it be explored.

PARTICIPANT: Which agency would have jurisdiction over the samples from a hospital patient?

MR. CAPLAN: Over exploring that one? I have no idea.

PARTICIPANT: I submit that if I were the surgeon, I would want to know whether my patient was hepatitis-C-positive to begin with, before I take him into surgery.

PARTICIPANT: This is getting into whether or not all hospitalized patients should be tested for HIV, as well.

PARTICIPANT: Right, right.

PARTICIPANT: And a whole list of other infections, which has not generally been the public health perspective that the country has used, so I am very antsy about -- and anxious about moving in that direction. Clearly, hepatitis-C infection in the community needs to be recognized, there needs to be increased awareness on the public's and provider's, healthcare provider's parts to pick up hepatitis-C, but I am not sure that we want to recommend essentially a nationwide screening program for hepatitis-C. The CDC might want to do that, but I think that is beyond our purview.

PARTICIPANT: [Comment off microphone.] Recommendation number one up there, one of the situations that is not outlined is, what do you do if there is not a prior negative test?

PARTICIPANT: [Comment off microphone.]

PARTICIPANT: What if it was a donor that last gave before testing was implemented, and so on. I mean, there's --

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Comment on that, what if we have a donor who has come back after six years, or a long time ago, or never?

PARTICIPANT: In the context of HIV look-back, the same issue was encountered, and we recommended a five-year retrospective search, based on the five-year requirement for record-keeping. But I think that the practical limit for how far back to go is availability of records, and a choice can be made whether to arbitrarily limit it to some number of years, or whether to recommend a minimum number of years, or as long as records permit.

MR. CAPLAN: One of the things I think we are going to run into when we get down to the next page of our recommendations, which I would like us to take one look at, just before we go out of here, if you could flip that recommendation forward.

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: No, I am going to come back to that, but it is in the spirit of this, how far back should we look? One of the things that is up there is, we do not recommend targeted look-backs (inaudible) transfused prior to 1990. I expect us to have a big, knock-down, drag-out about whether that sentence stays there, but one way to be consistent is, just as you look at this number four language that was up there before, if we are going to use terms like, as records permit, or as documents permit, that may -- if we do that here, I want you to think about, that maybe the answer to what to do under number one, that is what I was getting at.

If we go with a standard that says, well, we are just going to look back according to some time limit, five years, seven years, ten minutes, whatever it is, as a way to think about possible look-back obligations, to be consistent, one possibility is that we say, you have to look back to the extent to which records are available, or records make that possible. So if that is how we are going to go, we may want to be consistent about that. Because I am leading toward thinking, the language I would like to see personally in one is, to the extent records permit, without getting into the five-year -- so, it may be different, depending on who it is and where it came from. Military may be different from the average hospital.

PARTICIPANT: As a hospital, that is blessed with an excellent record-keeping system, that language could conceivably cause me to go back into the 1960s.

MR. CAPLAN: Before.

PARTICIPANT: Because we have the capability of doing that, and I am not sure that is really worthwhile, but if that is the recommendation of this committee, I will probably feel compelled legally to do that. So, I would prefer that the committee take its responsibilities of stewardship, and set some reasonable time limit, but specific time limit, as to how far to go back.

MR. CAPLAN: You are going to be in that category that we saw from those slides of Britain, where there were those outliers marching back into the late sixties and so on? Yes, Paul.

PARTICIPANT: I guess somewhat in the same tone, I hoped this would not happen, but I would not like to create the perverse incentive, which would then say, do not keep records. So, I have added, the availability of records permit, within a period of seven years, or -- we can say whatever years we are willing to accept, but I think something like that -- would that give you a limitation that would allow you not to go back to 1944?

MR. CAPLAN: Mike?

DR. BUSCH: Yes, I think all these times -- there is a whole separate issue besides record availability, which is, the relative risk, and the risk in the pre-1990 and potentially, 1990 to 1992 period, was so much substantially greater than today, and the look-back I see as kind of the contemporary way to deal with rare risk, whereas the major effort that we have been talking about is to focus, to try to find the majority of people who were exposed in that earlier period, and even if records were available prior to that date, the yield of that process, in terms of the proportionate number of infected people and absolute number per dollars is going to be so much lower.

I think drawing that line at about five years or seven years, is consistent with that date of 1990 to 1992, when the risk during that early period was substantially greater, and a different message needs to be, a different strategy needs to be pursued to detect those people.

PARTICIPANT: And that gets to number three, and actually, what I was going to propose in three is, make it much more aggressive, to see if the committee would go along, in context of the dialogue we had yesterday about this very issue, which is that, every effort -- first of all, number three, I would change to begin with. Identifying all significantly at-risk individuals, or individuals at significant risk, and not exclude -- I mean, include -- being all-inclusive, not just people who have been transfused.

But, for the transfused individuals, that efforts in number two, or number three, that three-B which John was talking about, be undertaken to identify by whatever educational means necessary, individuals within the United States who have been transfused, for advisement to come in for testing for HCV. Which would cover all the ones --

MR. CAPLAN: Let's hold on that for a second now. I know we will get some discussion of that.

PARTICIPANT: But it is directed in -- I mean, obviously, you do not have -- if you are going to do that, you have thrown a much wider net, and you do not have to go to esoteric details about the record-keeping, how far back it goes.

MR. CAPLAN: Let me ask this. And I will drive you forward with the prospect of letting us break. Would we agree just for now on recommendation one, for the twelve months prior to a situation we do not have it, that people ought to trace records back, at least for five years? I mean, is that a cut-off that you can all -- it will get us towards this notion of, as records permit, for up to five years, is that a reasonable --

PARTICIPANT: No.

MR. CAPLAN: No? What would you want to see there?

PARTICIPANT: I think the records do go back beyond that. I think blood centers have records that will go back to 'way beyond that, certainly since the inception of our blood center. On the other hand, the limiting factor is the hospital.

PARTICIPANT: [Comment off microphone.]

PARTICIPANT: That is going to vary, then, somewhere probably around the seven-year -- five- to seven-year.

MR. CAPLAN: Want to make it seven?

PARTICIPANT: Seven?

PARTICIPANT: At the hospital level.

PARTICIPANT: Probably be more reasonable. My hospital keeps it about seven.

MR. CAPLAN: So, let's put -- at least we will answer it in the spirit of that. We have also opened the door to this notion, which I was hoping we would get to a little bit of linking look-backs to some notion of record availability. So, why don't we take our break? What we will do is, during the break, Steve and maybe Paul, Mike and anyone else who wants to, can peer at that first sheet of the facts. We will probably try and rewrite that overhead for you to take a look at when I come back, then we will return back to the recommendations. Work those over a bit more, and oddly enough, following in some logical sequence, we will decide what we want to do and then come up with a preamble to justify it. A little humor there for you to take away. Alright.

[Brief recess.]

MR. CAPLAN: -- get back underway. If I can get the committee back. What I asked an elite corp of wordsmiths to do during the break, was to try and compact some information in the interest, again, having taken some of the admonitions we heard about -- be simple -- seriously.

What you see up here is an attempt to take some of our fact discussion from the morning, put it into both a preamble and fact statement. This would come in front of the recommendations. It is carrying a lot of weight in a relatively small number of words.

It may not have all the details of your particular fact that you were enamored of before the break up there, however, the driving purpose at this point in the document is to both explain why now, to somebody who might not be so certain about why we are thinking about changing public policy with respect to hepatitis-C, and narrow it down to those facts that people can manage, in a relatively simple dose. Steve, do you want to read that thing? We recruited you for your penmanship and --

PARTICIPANT: I didn't have to have penmanship to take this job. The introduction of a new [comment off microphone] improved blood test for hepatitis-C virus in 1992, has substantially reduced the risk of acquiring this disease from transfusion. Subsequent experience with this test has revealed the following.

Number one, about four million in the U.S. are infected.

Number two, many are unaware of their infection because of [comment off microphone.]

Number three, [comment off microphone] causes cirrhosis, liver failure, and hepatocellular carcinoma.

Four, the risk of HCV (inaudible) prior to the introduction of the improved HCV test.

The availability of new therapies and knowledge reinforces the concept that blood recipients have the right to know the risks associated with transfusion of blood and blood products and, specifically, any unique risks related to the (inaudible) quality of blood products they received. That (inaudible).

MR. CAPLAN: And so, what we are trying to do here is say, in an, again, compact form, here is the situation that has shifted? This is a factual summary about why we are rethinking what our duty is to those who have received blood. We are trying to also put it in the context of the general awareness of hepatitis-C, which did not exist prior to the availability of tests, and maybe we have said enough here, too, I hope, about the availability of new therapies and new knowledge, without detailing it, to let people know that that is also fueling our interest in making these recommendations.

The point is, I am not saying we cannot wordsmith or ask questions about whether this is right or whether you liked something else better, but please keep in mind, as we try to do this, that I think succinct is going to be important, and I think the reason for picking out facts, is not that we just know information, but it is to present the facts that have made us offer the recommendations. That is what we are trying to do is, link the one to the other. Let's go to Trish and then we will just start coming around.

DR. O'CONNOR: One concern of the statement of the blood test in 1992 reducing the possibility of transfusion, and then about four million people in the U.S. are infected, to me, implies that they all got it from the transfusion, and that is a concern.

PARTICIPANT: -- from a variety of cause (inaudible) --

MR. CAPLAN: Yes, go ahead, Keith.

PARTICIPANT: One way to go would be to go from generic, in terms of this preamble to specific (inaudible).

MR. CAPLAN: Could do it that way.

PARTICIPANT: Because and then, the other thing that we did not put in here that we had (inaudible) list(?), was to the benefits of changing lifestyles and the potential benefits of (inaudible) should be included as well.

MR. CAPLAN: That is what I was hoping -- the availability of new therapies and knowledge --

PARTICIPANT: Oh, okay.

MR. CAPLAN: I am fishing in there about lifestyles, I do not know how much we have to spell it out there. Maybe we do. I am actually editing myself here, in the interests of fairness. Let's see, Mary?

PARTICIPANT: Just a couple of thoughts, one is, about the statement that is number one. I think you could clarify -- add a clarification for Trish's point, which is a good one, by simply stating that, about four million persons in the U.S. are infected, of whom 7% acquire their infection -- are estimated to have acquired their infection through a transfusion. So, that would provide some information and provide the clarification --

MR. CAPLAN: I see a lot of heads nodding on that, that seems --

PARTICIPANT: One other comment in that sort of introductory phrase, from which these one, two, three, four statements follow. I am not sure if it is a bit of an overstatement to say, the introduction of this test substantially reduced the risk of acquiring disease, because I think as Mike Busch's presentation showed us, there are a lot of other things that happened in the seventies and eighties that really impacted on risk reduction for HCV, through transfusion.

The test allowed us to identify with more precision, who was infected. So, it -- I think -- and that may not be the best way to state that, but that is what I think the test allowed us to do.

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Has contributed substantially to.

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Yes, it is certainly causally there as a -- why don't we try, Steve, has contributed substantially to a reduction? It probably did do that, no?

PARTICIPANT: I am not sure it did(?). Actually, the test in 1992 did not have much of an impact itself on the reduction in the risk --

MR. CAPLAN: You think it was still the behavior before?

PARTICIPANT: -- from transfusion. The major reduction really even took place prior to 1990.

MR. CAPLAN: You mean, from the big [simultaneous discussion] --

PARTICIPANT: That is right. So, by the time 1990 and the test came in, the risk was already below 1%. And we are talking about risks in 1980, of 10 to 20%, so, you know, that is just not really the issue.

MR. CAPLAN: Alright. Let's do it then, has contributed substantially to the ability to detect.

PARTICIPANT: Exactly.

MR. CAPLAN: Contributed substantially to the ability to detect.

PARTICIPANT: One option would be to have a first sentence relating to the fact that, you know, over the last several decades, the number of measures that have reduced the risk, and then the second sentence could be, in 1992, the development of a very accurate test, both reduced the risk substantially, and allowed us to address many of these other questions.

PARTICIPANT: That takes it away from what you are trying to do here, which is, to define the reasons for proceeding at this point. And I think you want to make a strong impact at this level, don't you?

MR. CAPLAN: I do not want to assign magical powers to the test that it cured people, I understand that. What I am trying to say, I think is, that because the ability is there to detect the disease much more accurately, that has led us to want to take a look again.

It is not -- I mean, yes, true the disease incidence was dropping. I remember that big dive in the curve prior to 1990, but I think it is this ability to detect that has us in a corner(?), moving here right into this group(?).

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Not only that, but that is one of the factors. Yes, (inaudible).

PARTICIPANT: I personally like Mike's introductory statement at the beginning of the fact, I think that would show where we have been going over the last several decades, and that this has been a focus of blood bankers for some time.

I would also suggest that at the end of these facts, to state something such as, therefore, it is entirely appropriate that the first action of this committee, the first consideration of this committee, has been to look at hepatitis-C risks in transfusion recipients.

In order to clarify that we have not -- as several of us discussed at the break -- we have not been around for 20 years, waiting for this to happen. We are a new committee, this was the first thing on our agenda, and we are dealing with it.

MR. CAPLAN: Yes. Alright, I may actually ask Mike if he could take his pen out right now and try and craft that first two sentences. He can slip it quietly to Jim, then we can get it up on the board. I love this editing on the fly type thing, wonderful thing. I will have other assignments for more of you later in a minute. Write the Magna Carta.

Any other comments on other aspects of what is up there, does that seem to capture enough to get us set up for the recommendations? What we are looking for here is something that would be, again, pretty simple statement.

Here is why we convened, here is what we are doing, we are not sort of laying on the tasks, per se, but these are the facts, and a little bit of history that led us to make these recommendations. So, we are moving that way.

We have kind of excised principles, moving them more toward the recommendations, to the extent we have them, and that -- we are producing a document that is certainly under ten pages at this point, which I think is good, maybe under five, depending on how things go.

The idea now is to preamble it, set it out with the facts, and then move to the recommendations. That is the structure you have, just so you have it in your head here. Mary?

PARTICIPANT: Fact number three, HCV is a major cause of cirrhosis, liver failure, and hepatocellular carcinoma. I believe that is a bit of an overstatement, and I would ask others who perhaps have more technical expertise to suggest a rewording, but, I think that somewhat overstates.

PARTICIPANT: One of the major causes for death --

PARTICIPANT: It could be -- you could just say it is a major cause of chronic liver disease, or a major cause of liver disease -- well, actually, it is a major cause of chronic liver disease, is what it is. And the problem here probably is more with the hepatocellular carcinoma, it is not a major cause of liver cancer in the United States. It may be in other countries, but not -- it is a major cause of other liver-related problems, but not liver cancer. So -- and it is certainly the leading reason for liver transplant. So, you might say, chronic liver disease and liver failure.

PARTICIPANT: I think what Marian is saying, I agree with Marian in that what she is saying, it is a major cause of chronic liver disease, and I think the wording in that could be used, which can result in liver failure, in cirrhosis, liver failure, hepatocellular carcinoma, and certainly, it is in this country, if you want to use that word, major cause of liver transplantation. So, I think that chronic liver disease is what we want to get at, that is its major cause, or, it is a major cause of, and then say, which can result in the following things that you have listed.

MR. CAPLAN: And that is chronic liver disease.

PARTICIPANT: Chronic liver disease, right. And I think what Leonard was telling us yesterday, is that he is beginning to agree, and the data is saying, that approximately 20% of these patients will, over the course of their disease, develop cirrhosis, which then can result in these other sequelae.

MR. CAPLAN: -- cirrhosis, liver failure, if we can keep the hepatocellular carcinoma -- and I think that will do it, that is fine. Others?

PARTICIPANT: Yes, possibly, the second [comment off microphone] -- some (inaudible) certain (inaudible).

MR. CAPLAN: The question here is, do you want to say something about spontaneous remission, sudden -- no, okay.

PARTICIPANT: I think that dilutes the message.

PARTICIPANT: It is also a small fact. I mean, you are talking about -- firstly, you are not saying in there, how many people develop chronic infection to begin with, you are just saying, this is the number of chronically infected people, and the number who actually resolve, the percentage could be 15%, you know, it could be as high as 30%, as Leonard pointed out, in one population.

The confidence interval around that, whatever it is, it is a minority of those infected and again, it is too much information, I think, for what you want to get across here.

MR. CAPLAN: In writing these kinds of things, in a group, one of the things that you want to do is you want to pull this off when it looks like exhaustion is leading people to a certain kind of agreement. So, after Ron, I am going to ask Steve to take this away as a preamble that might work, and I want to go back to those recommendations, which is the nuts and bolts of what we have to do in the next hour. And bang those around for awhile.

PARTICIPANT: Art, I would propose a new first sentence to be added in front of the first sentence that is up there, something to this effect. The recognition of hepatitis-C, previously called non-A non-B hepatitis, as a major cause of chronic liver disease in the United States, has been facilitated by the development of a specific test for hepatitis-C infections.

Then we can say the introduction, so that we are putting the focus on the recognition of hepatitis-C, and then we go to the introduction. So, the sentence I am proposing would precede the introduction.

PARTICIPANT: That is a good point. (Inaudible) infections. I also rewrote the first couple of sentences.

MR. CAPLAN: Still busy? I was going to ask you about that assignment.

PARTICIPANT: Maybe I will read it. It is, progressive measures over the past two decades have dramatically reduced the risk of HCV infection from transfusions. The introduction of the improved HCV test in 1992 has almost eliminated the risk of acquiring HCV from transfusions, and has allowed for accurate detection of infection in exposed patients.

MR. CAPLAN: Comments? Attempts to merge? I like those two sentences that Mike just generated. Steve, while you are copying that over, why don't we follow my plan of putting the recommendations back up that we started to evolve? I was going to suggest that two facts -- or two, sort of facts come out here.

One is, we were talking at the break, some of us, and somebody said, well, to what extent do we want to get specific about our recommendations? Do we want to say that it is PSAs and they should be broadcast between midnight and six, or you know, make sure that it is in five major languages, and six other linguistic groups? And I think the answer to that is, no.

We could ask for something like the following, if you wanted to. I mean, we are giving advice to the Public Health Service and the agencies. We could say, look, we would like to see a report about how you are going to implement this by time X to take these recommendations forward.

The committee would be very interested in having some ideas about what you think is a -- based upon your opportunity to do it more than in two days, a good education campaign for healthcare workers, or a good education campaign to reach out to donors.

I am not saying we cannot be somewhat specific, but some of it we can say, look, work on it, and we want to take a look at it. That is fair game to do. So that is one issue. How specific are we getting? How fine-tuned? My inclination is to be broad, but it is fair to ask agencies to work together to come up with a statement back about what they propose to do.

PARTICIPANT: I agree with that, but I would want to add in, maybe also, some plan for evaluation of effectiveness. That should be in the front end.

MR. CAPLAN: Then some of us were kicking around at the break one other issue, which I do think is important. It is not part of our recommendations up here yet. It has to do with look-back and records which we are going to, obviously, talk a little bit more about, about contacting people and to what extent we do or do not want that to happen as part of looking backwards.

I think it is important to understand, at least my understanding is -- and people who know much more about this should correct me -- but, what I believe to be true is, records are often destroyed or disappear between a blood bank and the hospital, coming from blood bank distribution to the hospital, about where units went, after five, seven, eight years.

Individuals, patients, have records about their healthcare that are kept more than five or seven years in most cases, in fact, I believe they are probably kept forever in a lot of places. If you wanted to know whether you had gotten a transfusion, because you were not certain, or you were young, you could ask to see your records and come, if you will, upstream.

You could go back to your original patient records and say, did I get a blood transfusion when I was a neonate, or when I was in this car accident, that I did not know about? And we could ask people, or tell people, if you are concerned about this, you might want to find out from your records or talking with your doctors about whether you ever had a blood transfusion, and it in theory could be done.

It would probably be expensive and a complete pain in the neck for the hospital to go drag it out, but you could get it. It is not true to say, there is no means of getting information, the way I understand it, for an individual who wanted, and was concerned about blood transfusion in the 1980s or 1970s, to find out the answer to that.

It is true that if we are going to use tracking blood from the center to the hospital, there is a disconnect in the ability to do that after a period of time. But it is not true that there is no information anywhere that somebody could, you know, oh, let me make up a hypothetical person, say, a lawyer, could -- track back from records back to receiving blood from a person who turned out to be infected with hepatitis-C.

We do have to grapple with that fact. I do not think we want to tell the American people, we are being honest and being fair with you and then say, well, I am sorry, if it is more than seven years, your records are gone and that is too darned bad. That is not quite true. It is just that one type of information moving from blood distribution out to the hospital is ragged, after a certain period of time, but there is information that they could use to build facts, I believe, coming in the other direction. Does that seem factually true?

PARTICIPANT: Yes, that is correct. Yes, the -- the disconnect, or the raggedness, as you used the term, is probably in the transfusion service records going far back, because the blood center would be able to say where they sent the blood.

MR. CAPLAN: Sent the units, yes.

PARTICIPANT: Right. But knowing which patient received a particular unit will fall apart, the further back you go, yes.

MR. CAPLAN: Mike?

DR. BUSCH: Right, and to enumerate another scenario, as our experience in San Francisco and I think generally, the transfusion services destroy the linkage records with respect to the blood component that came from a particular unit, but they usually have records as to which people were transfused in their hospital.

The other strategy that has been pursued with limited effectiveness is the transfusion service sending letters to all transfused patients, not enumerating, discriminating, which have gotten blood from a particular patient, but focusing with letters, the campaign on people who indeed did get transfused. The truth is, that program had a very low response and very low yield of tested people, and of those tested, you know, only 3 or 4% were infected, but that is another -- there are multiple options.

PARTICIPANT: I think this relates to the worried patient that we talked about yesterday, with Dr. Robinson, and one of the things that we might keep in mind is, rather than tell a patient to go track his transfusion records, if he has reason to believe he was transfused and is worried, the NI(?) HCV testing is really pretty simple and you go, find out if you are NI(?) HCV-positive, and that pretty much ends your worry, if you will. And that is a reasonably easy way to do it, rather than trying to track back through a hospital, where the records are in a storehouse someplace that you cannot get at them.

We might want to consider that in the counselling of patients, that maybe sometimes it is easier just to go get checked, if you are worried, if you are a worried patient.

MR. CAPLAN: Well, one of the things that I think we can anticipate happening after today's meeting, is that announcements are going to go out to the United States saying, this group met and said, there was a problem of exposure prior to 1991, 1992, and that people should talk to their doctor, people might want to know if they were transfused. That could lead to a burst of phone calls prior to the CDC's massive educational teleconference experience, which I believe is impending, but has not happened yet.

In other words, people are paying attention to what we are doing right now and may want to go out and ask questions, so it is important, I think, that we do not stir people to action tomorrow morning, in terms of firing off questions, since we are talking publicly here now.

What we are saying is, there is an issue. It has to be talked about with your doctor, but I would suggest that one of the first recommendations we have to put up there, if I was going to re-prioritize these is, this business about educating medicine and the healthcare field, to be ready to field these questions, like the one you are saying, I mean, the appropriate thing to the worried, well may be for the doctor to say, well, then, we will do the test, if you are worried about this. And then, if you are not infected, we are done. If you are, then maybe you do want to drag your hospital records out, or see what happened, or something like that.

If we do not have that loop closed, we are going to be in big trouble. So, my next comment is, it seems to me the number one recommendation that we ought to move to -- and I am sort of speaking to the gallery here as well -- is that, we have to make a very strong -- it is recommendation two, I guess, becomes recommendation one.

That we have to implement, as soon as possible, a systematic program to educate -- I was going to say, healthcare providers, not just medical care -- this is fighting words -- but, healthcare providers, to the appropriate assessment of individuals at risk, as defined in the consensus conference.

If that does not happen, there is going to be trouble. I mean, we will make trouble just by having our little meeting finished today, because people are going to turn on the TV tonight and say, oh, well, I better go call my doctor and -- that is going to be trouble.

That is the first step, the first thing I think we ought to emphasize is that we have to get a person on the other end who can answer that. If we do that, that may solve some of the worried issues about, was I transfused and so on. An appropriately informed physician, nurse, should be able to do that. I would argue that, let's make that number one, and just for that reason.

PARTICIPANT: Art, do you want to add in anything about counseling in that statement number one, or --

MR. CAPLAN: Yes.

PARTICIPANT: -- leave it separate?

MR. CAPLAN: Where does that go?

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: Assessment and counselling of individuals, yes. I think we solved the problem of number two, in terms of that time limit we were talking about of looking back the seven years. I think that does become relevant, relative to those kinds of records. Number three, we did not talk about much. It is time to say a few words about it.

PARTICIPANT: I just will reiterate what I said before the break, which is, I would like to rephrase, identifying all or every possible individual at significant risk for hep-C. And add a caveat that says that individuals determined to be at risk, should be tested.

MR. CAPLAN: Where do you want to put that? I lost -- it is other strategies?

PARTICIPANT: [Simultaneous discussion] -- sentence.

MR. CAPLAN: Individuals determined to be at risk should be urged to seek testing?

PARTICIPANT: Right. Should be tested, either one. It could be the provider who determines the risk, or it could be the patient that determines it.

MR. CAPLAN: Comments?

PARTICIPANT: Should be provided an opportunity for testing. Then they could decide whether they wished to or not. That also puts the mandate on providing some means of getting the test done(?).

MR. CAPLAN: Alright, I can see a sentence that is at the end of three, that individuals be provided with the opportunity for testing. That individuals at risk be provided with the opportunity for testing.

PARTICIPANT: Is that meant to imply that the testing is provided by the government? I think you need to be pretty specific about what you are recommending here.

MR. CAPLAN: I -- that issue, I am waiting to get to with baited breath. When we had our little moral principle about fair access, which we have to come back to in a second here. But before staggering toward that, I heard absolutely no sentiment for what is up there as three and three-quarters. It seemed to fall aside, am I correct on that?

PARTICIPANT: I am too close to read it, unfortunately.

MR. CAPLAN: Can you read that three and three-fourths again?

PARTICIPANT: Three and three-fourths is a recommendation to obtain specimens from recipients of transfusion for documentation of HCV status, or perhaps other infections [comment off microphone.]

MR. CAPLAN: This foundered on the rocks of making a public health recommendation that might be too broad for us right now. It might be a good idea, maybe, but it seemed to fall apart a little bit on worries about privacy and worries about cost, and worries about whether that was just worth doing as a baseline thing right now.

The obligation of providers to provide adequate informed consent for transfusion recipients seems to me to be fine, do we want to keep that, or is that apple pie-ish, and -- Yes.

PARTICIPANT: It would have been a good idea ten years ago, but today, I think most hospitals do use some type of informed consent for transfusion, and frankly, the risks for hepatitis-C now are incredibly small, so there is not a whole lot to inform about, and it is unlikely that the current risk is going to dissuade anyone who really needs a transfusion, or should not dissuade anyone who really needs a transfusion.

MR. CAPLAN: So, what do you think about three and a half there?

PARTICIPANT: Yes, I do not recall whether that is listed, though, on most of the documents that are handed the individuals who sign off on this. Do you recall, Jim --

DR. AUBUCHON: Every hospital handles informed consent for transfusion differently. NHLBI did put out a booklet a number of years ago that listed risks and alternatives, and hep-C was listed as one of the risks, actually, in much greater numerical predominance than we now have.

MR. CAPLAN: Alright, I guess I will take the Chair's prerogative on this one, as I don't hear enough -- it is going to take more enthusiasm than that to keep that one up there.

Alright, let's slide to four.

PARTICIPANT: Art?

MR. CAPLAN: Yes.

PARTICIPANT: Is there any -- future deliberations of this advisory committee, I am not particularly comfortable with that. Can't we make a more definitive statement right now -- [comment off microphone.] It's the second line.

PARTICIPANT: [comment off microphone] makes it this committee's responsibility.

PARTICIPANT: How about [comment off microphone]?

MR. CAPLAN: I think that was just procrastinating until this meeting, is what that was.

PARTICIPANT: Yes, yes. Actually, there is probably a period there -- [simultaneous discussion].

PARTICIPANT: -- recommendation. That is how I thought.

PARTICIPANT: But we are coming up with language that [comment off microphone].

PARTICIPANT: This might be an ideal place to put -- when that was written, obviously, we were in a great state of ambivalence. But this might be a good place to put developed, period, or developed by, with, the follow-up -- you know, the tracking and reporting mechanism (inaudible) of what the outcomes of these strategies are.

MR. CAPLAN: The(?) developed -- [simultaneous discussion]. So, in other words, identifying.

PARTICIPANT: Well, why not -- maybe we should acknowledge that these other strategies for identifying individuals have been examined (inaudible). I mean, that is what we have done, we have really considered the options that we have toward finding these people, and maybe that should be a clear statement.

We have examined the historical experience with approaches for identifying individuals at risk, in collaboration with PHS. And this has led to, now, what we should come forward with as these following specific recommendations.

PARTICIPANT: Could I ask a question? When you say, other strategies for identifying individuals, I am actually -- I read that -- I actually interpreted that one way when I first read it, and now I am interpreting it a second way, and now I am not sure what you meant by it.

Do you mean, other strategies, or strategies other than public awareness campaigns, and look-back, for persons exposed to transfusions, or are you referring to identification of other individuals at risk? Because the vast majority of individuals at risk in the United States, are at risk not because of transfusions, but because of other exposures.

PARTICIPANT: Right, and that was the original intent(?).

PARTICIPANT: And it seems to me that, in a public health context, our responsibility is to -- if we are going to have a public awareness campaign -- is to include all of those at substantial risk, including transfusion recipients.

PARTICIPANT: Well, that was the original intent, actually.

PARTICIPANT: And so, now, we are getting away from that aspect, by virtue of three and four, whereas I thought perhaps maybe three and four need to be combined and include, you know, this, we recommend a public awareness campaign, you know, that --

MR. CAPLAN: Well, you know, what I think is becoming clear to me is that three might want to be revisited in a second, after we talk about four, because what I understood three to mean was, other strategies for identifying people who might have been exposed through blood transfusion, and somehow integrated with efforts on hepatitis-C more generally, looking -- remember that talk about the push to take on the challenge of hepatitis-C and make sure it was integrated with what we were trying to do with blood transfusions. So, I understood it in the first sense that you were talking about, Marian, in A.

PARTICIPANT: Oh, okay. Yes, I see.

MR. CAPLAN: But maybe -- maybe let's just hold off on three for a second, and let's wrestle with four again, if we can pull that first sheet off entirely, Steve. We can come back to three in a little bit, and see if we still need it at all. Let's see what we do with this. We have really been tottering up to it, and it is time to just go full bore.

Right now it says, no targeted look-back prior to 1990. That is certainly an interesting phrase. I would like just some discussion to open this up, about whether we still want to stand with the policy that is articulated there.

PARTICIPANT: Whether we stay with that or not, I think that there has to be some provision provided for these individuals who feel that they may have been transfused, and may have this HCV, and make sure that they understand that they can still come in and be tested. So that they do not feel excluded from that. And make sure that it is communicated in some kind of way that --

We have not discussed who is going to pay for these tests, but I think that is going to be an issue, also. There are going to be many individuals who will not be able to pay for this test, and unless they know prior to this, that they can come and get this test without worrying about paying for it, then we are dealing with another issue here (inaudible).

MR. CAPLAN: Yes, Carolyn.

DR. JONES: [comment off microphone] continue to address that?

MR. CAPLAN: Yes, in fact it was -- it was ambiguous enough so that it both did that and also did this integrated thing. We also had a principle drop out from the earlier slide which said, there is an obligation to make sure that people have fair access. Some bright person has said, to who, who is obliged? But, yes, the issue of who is paying and who is ensuring that this happens is out there.

DR. JONES: I am not so sure from the discussions we have had here that (inaudible) completely (inaudible) belong there. I think (inaudible) recommend (inaudible). Correct me if I am wrong.

PARTICIPANT: I think that is the line in the sand here, and we may as well get at it. I mean, I think everybody should be clear now as to the yield and the reality of doing that kind of targeted look-back, and the process -- you know, a lot of work will go into just the blood center records review, and then that will dead end at the hospitals, because there's no records, and it will dead end at patients on downstream. And I think you are hearing the blood bankers who are -- I would recommend that we not do that. That we not proceed to any targeted notification of recipients who got blood prior to 1992, prior to effective screening.

Other people are echoing the issue that we cannot withhold what might be of value to some subset of these patients who would eventually, at the end of that process, get that notification. I do not know whether a vote of the committee -- but that to me is a very clear distinction that we have very divided opinions of.

PARTICIPANT: At the moment, number four and the new number two leave me somewhat confused as to what it even is that is on paper and that we are considering, because number two talks about a prospective look-back, which to me as a blood banker, if I were implementing it, would mean that only today, and henceforth, when we find someone who for the first time is HCV-antibody-positive, would we initiate a look-back.

Then, in number four, we talk about for persons transfused prior to 1992, they would not be part of the look-back, but certainly, prior to 1990, they would not be part of the look-back. So, I would not -- if I were charged with implementing it right now as it stands on paper, I would be confused as to what I should be doing. And I --

MR. CAPLAN: The only group that would get a direct look-back, the way I understand what we have written so far is, right now. Prospective look-back. That seems to me to be what number two says. So, if you did not get a blood transfusion in 1998, or 1997, as far as I can see, there is no recommendation in here at all saying, look-back to anybody.

We did talk about whether we wanted to cover that period, 1992 to here and now, with look-back. And there is this issue about, what do we want to do prior to 1992, 1991, or 1990, depending on how you define it. But the way I read what is in there right now, just so we are all clear, we have not told anybody to do directed look-back, except contemporaneously.

PARTICIPANT: [comment off microphone] -- if someone came in today, a donor, who is HCV-positive for the first time, and they have had a negative donation some time ago, number two does not say that the recipients of that previous intervening donation [comment off microphone.]

MR. CAPLAN: Excuse me, with that -- yes, with the rider, yes, it does. Number two does say that, up until five years --

PARTICIPANT: But if -- yes, but the problem is, that if a donor came in in 1995, or 1993, and tested positive and had given in 1992, we are not covering that.

PARTICIPANT: No, it does not say that.

PARTICIPANT: I think I would recommend the distinction be, what kind of blood was transfused? That if you have blood that had been screened with second gen, that you would not do look-back -- I mean, sorry.

You would do look-back on recipients, seroconverters, as far back as second gen screening was in place. Prior to that point, prior to March of 1992, these other modalities of tracing recipients are the approaches. So, look-back would stop at the point of introduction of second gen screening.

MR. CAPLAN: It does not say that yet.

PARTICIPANT: It does not say that yet. Right.

PARTICIPANT: Well, I agree with Mike, with the addition that I would like to see, in keeping with what we said -- or what we were just debating about in number three, is that anyone transfused prior to gen II, be tested for HCV, if they have not been previously tested. Be offered testing, again, you know, I mean, not mandated. Obviously, we cannot do that, anyway, but I mean, be offered testing.

So, look-back for anyone who had generation II testing, of the donor that [comment off microphone.] And then, anyone else be offered testing.

MR. CAPLAN: John.

DR. PENNER: I think if we get back to that 1988 transfusion recipient, and how we are going to identify him. And I guess I have not heard specifically why aren't we going to go and track that -- let that individual of 1988 know? And I am hearing two things. One, it is inconvenient, low yield. And low yield to me means cost. So it is saying, it is too costly to let these people know.

MR. CAPLAN: But I do not think low yield means cost, it means that only a very small fraction of the recipients in 1988 who got HCV from blood transfusions will be traceable. It is the proportion of infected people from blood that we can find through that program. That is what I mean by low yield.

PARTICIPANT: But then those who you are able to trace, are going to be missed, right?

MR. CAPLAN: But then there is the misunderstanding that we are doing this program and that this is the way we are going to find you. So we actually dilute out the public health message, because they think, oh, well, if I really have it, they are going to come at me with this other thing, but the truth is, from our analyses, only perhaps 5% of all the infected people in 1988 will be found through this tracking mechanism.

PARTICIPANT: But we have heard from our publicists, that the best way to get at this is by several messages. To get a direct letter, as well as the public. So, if we rely on his recommendation, we would say, we should try to notify these individuals at least by look-back, as well as by public announcement.

MR. CAPLAN: I would recommend doing it from anybody who got transfused, try to get letters out to those people. I think there at least you are going to get everybody who got transfused with a letter. And then you have most people getting multiple messages. You do not have an extremely small fraction getting this message that in truth is a very costly and low yield program.

PARTICIPANT: But then you are withholding information from one individual who has been exposed.

MR. CAPLAN: Correct.

PARTICIPANT: Specifically. And you are saying, you are never -- as an agency, don't really -- we will make the decision for you. We are not going to let you have this information, because we do not think it is good(?) enough(?).

MR. CAPLAN: You are not going to embark on that program that will identify that recipient.

PARTICIPANT: That's right. But that recipient has information that has some meaning to him or her, and would have meaning to their lifestyle and their future management. And you are not allowing them access to the information that is there.

PARTICIPANT: I believe this can be handled a couple of ways, because it is so difficult to get the people further back. If the overall health message to the community-at-large, and the training of the doctors is good enough, you will pick up a lot of the people that way, because they will see the general message. They may say, oh, I did have a transfusion, or I better check with the doctor. Will we miss some people? Yes. But at least you are going at it from two perspectives, rather than just trying to use a letter, or just a limited public service.

MR. CAPLAN: Yes, Jim.

DR. AUBUCHON: I think we should recognize that any of the avenues that we try will have holes in it. Clearly, public service announcements are not going to stimulate everyone to check their medical records, or to have an HCV test. Look-back, as a program, has many inherent flaws in it, primary of which, that the donor has to return on a subsequent donation, and be tested and fall positive for HCV.

We have continually improved the donor historical screening questions over the years, as Mike showed the effect of, and that has resulted in the culling out of many donors who otherwise would have been tested and found to be HCV-positive. We will never be able to identify their previous donations to write the fabled look-back letter.

I agree with Mike that we should not leave the public the impression from a public service campaign that, if they do not get a letter from their hospital, that they are off the hook, and they do not need to be concerned about it. If the patient had been transfused at a time when there was substantial risk of HCV transmission, whether you want to say before 1990 or 1992, we can debate. But if their transfusion occurred during that time, they should check out whether or not they are HCV-positive.

MR. CAPLAN: One of the things we talked about doing, and even though I was yelling about not being too specific, is to put in here something about all physicians should ask their patients if they had a transfusion prior to -- or physicians and nurses. We could add that. That is specific, but it is another way to reinforce that that be routinely part of practice.

PARTICIPANT: Then you could add in there, and then offer them testing, and that would be part of that.

MR. CAPLAN: More discussion. Let's let this go, since it is important, a little while longer in the discussion, about the look-back. I think the case is there to try to make the strongest case about notifying to say, well, you knew that I was exposed from a donation. You knew who the donor was.

It is material to me to know that I got this infection, not because I used drugs or because I engaged in sexual practices, but because of blood. That would make me feel better about myself. I am not that interested in how much it costs and how hard it is, if you know it, you should go back and tell me. That is about as strong a case as I can make. Why didn't you?

You have the records, you know where it is, it is relevant to me, the origin of transmission does matter to me. It may not matter to my treatment, it may not matter to my testing, but it matters to me to know how this happened to me, says somebody. That is about the best case I can see for doing look-back that says, it happened this way, and you got it that way.

PARTICIPANT: I have been writing. For individuals transfused prior to 1992, we recommend that testing for HCV-antibody be undertaken. Identification of these individuals through a targeted look-back is not recommended as the primary method of look-back.

Educational campaigns targeted to the public-at-large, selected high risk communities, and providers, is recommended as the primary form of look-back for these individuals, and that really encompasses, then, everything that we have been trying to say and does not leave any -- it allows both to occur, but we make a recommendation as a primary and in a sense as a secondary.

PARTICIPANT: I would suggest you be very careful on the use of your term, look-back. Look-back refers to a very specific mechanism, okay? So, what you mean when you use that term, look-back and all those others, is really your primary or secondary method of identification, not look-back, actually.

PARTICIPANT: Correct.

PARTICIPANT: Only because in the end, we should be clear about our terms.

MR. CAPLAN: If I am a recipient of blood from 1986, and I say, well, I have had a test, and I want to know how I came to get hepatitis-C. I did not know I had it and now I see I have it. Can I stride off to my blood bank, or hospital, and demand a look-back in the sense in which you are talking about it, now? In other words, could I activate -- what you are talking about, Ron.

Could I say, look, I want you to track this back, because I am going to prove to my spouse, or I am going to prove to my mother, that I got this through a blood transfusion, and that is what I want to do, and I am telling you, look back. Tell me how I got this.

PARTICIPANT: I would not call that look-back, either. That is a case investigation. You get yourself a lawyer and we will do that for you, no problem. That is the mechanism where that linkage will occur downstream. I mean, this, why don't we do this or that, has no implication on whether there will be (inaudible), because those linkages can be obtained.

That is very simple. If you come in and say, I was transfused this date -- it is a very simple track to see whether we have ever subsequently identified a donor who has given a unit of positive blood. That takes ten minutes.

PARTICIPANT: I would just like to point out that there is an inconsistency between recommendation three(?) and recommendation number four, either as written or as rephrased by Dr. Gilcher.

If it is 1997, and you do seven years of look-back, you are back to 1990. If, on the other hand, you are recommending that you do not do a directed look-back prior to 1992, you have two years of inconsistency that you need to address.

The situation gets even more confusing with the period that the committee has not yet addressed, which is what to do between 1992 and 1997. I assume that has not been forgotten.

PARTICIPANT: I think Dr. Busch addressed that, but we have been throwing lots of things about here. I believe the concept was that, we would perform a look-back, based on all donors found to be hepatitis-C antibody-positive, with second generation, or subsequent testing. And that once the donor had been identified as second generation or later positive, we would then look backwards in our records and identify recipients back to some date. And we could say, 1992, we could say, 1990, we could say, 1980. But, we would use as the trigger mechanism for the look-back, a second generation confirmed test -- I am sorry, supplementally supported test. And then look backwards to some date.

PARTICIPANT: I agree that that was as stated by Dr. Busch, that directed look-backs be triggered by generation II tests. I would assert that it has not been clarified whether those are confirmed or unconfirmed.

I would further assert that this language in number four, or Dr. Gilcher's language, focuses on when the recipient was transfused. That is independent of when and by which test the donor was screened.

That changes things, because you are saying not to do the look-back if the date of transfusion antecedes 1992. That is a second, completely independent concept. And it does render inconsistent the effect of recommendation two.

In recommendation two, a seven-year look-back is triggered by a generation II reactive donor. That could go back to 1990. However, it is being stated in number four, that if the transfusion occurred in 1992, do not do the directed look-back. So, they are not in fact consistent, although I agree with what you stated as the criterion. I would suggest that the committee needs to decide whether the criterion is a test in 1990, a test in 1992, or a confirmed test, and we have not talked about how and when. That ambiguity is still unresolved.

PARTICIPANT: If I am hearing right, then, we are disregarding all of the testing that was done between 1990 and 1992, because it is only 50% positivity. And we are doing that because we are going to lose 50% or, as I still hear it, it is going to be more costly for each individual that we identify.

PARTICIPANT: The other distinction, I think, is again, the risk to recipient prior to 1992 was substantial. The risk since then is two orders of magnitude lower. And it is that message that I think -- I think dovetails nicely with the decision to only do look-back to units that were transfused prior to the introduction of the second gen test. Because those people prior to that point, irrespective of our look-back, a lot of first time donors were coming in who never came back and were transmitting. Those people had substantial risk that we will never be able to get to them -- that a substantial proportion will never be traceable through targeted look-back.

So, those people prior to 1992 deserve a message that catches everybody and says, everybody was at risk and everybody should be tested. Whereas, subsequent to that point, we are not going to give those people that message. Post-1992, mid-1992, the blood is relatively safe. You do not need to get tested routinely. And what we have during that period of people who received second gen screened blood, we have this safety belt of trying to get the rare transmissions through this look-back.

PARTICIPANT: Why not do both look-back, though, and the general, as was suggested, for those individuals prior to 1992?

PARTICIPANT: Well, if you do it for the people who come in, in let's say, 1993, test positive, and then they gave units prior to 1992, that is basically the same thing as when we first introduced the first gen test in 1991, and then you do it to the ones prior to that. And then you go, why not just --

PARTICIPANT: So long as you have a test that is 50% positive, there is a high risk, then, that that individual will have been infecting.

PARTICIPANT: I do not disagree in the slightest. The risk of a recipient of unscreened blood, getting blood from somebody who is subsequently confirmed positive, is very high. And there is no question about that.

PARTICIPANT: So, why aren't we going to notify those individuals?

PARTICIPANT: Because it is extraordinarily ineffective and low yield from the perspective of the proportion of exposed people found.

PARTICIPANT: But those individuals were found.

PARTICIPANT: They have not been found, they are not known, and the question is, do we begin the process --

PARTICIPANT: But if they are found, and you expect that they will be found, then you have to say, it was important to them. But it still comes back to me as you are saying, it is too costly to do this. And that is the only reason that we are not going to look back further.

PARTICIPANT: I am the one that is -- well. I am very interested in cost-effectiveness, I am very interested in cost, but I am not focusing on cost here. I am trying to focus on yield, and how we can identify the people who are really at risk, because a number of individuals will never be told that they got a unit, that they were at risk for getting HCV. We depend on the look-back program, and I think -- I agree with Mike. We do need to be clear as to where the risk changed, and have a specific program targeted for specific years of transfusion.

If we try to -- whether we use 1990 or 1992, we can argue about, but before a specific date, we need to be very clear, we are not doing look-back before this date. You cannot depend on us to tell you that you are at risk. Listen to this public service announcement. You are at risk, go do something about it. After that date, the risk is lower. We have a system where we can manage a targeted look-back, and in that time period, we can provide some information for patients. And you can depend on us to tell you.

PARTICIPANT: One case of data that we saw yesterday that is worthy of comment. The relative prevalence of HCV in first time or one-time donors versus repeat donors I think was 13-fold. So there is 13 times as many infections in the people who never come back, as in those who are these repeat donors who we are screening again. Those people were never going to trigger the look -- those people were giving blood. All along they have been giving blood, and never coming back. And they are never going to trigger a look-back.

PARTICIPANT: Well, I am talking about the individuals who actually were infected, and that you are going to pick up by going -- by doing a look-back. Those individuals are there, and you are denying them information -- [simultaneous discussion] --

PARTICIPANT: There are some there, but they represent [simultaneous discussion] --

PARTICIPANT: -- that has meaning to them, by saying, it is not important for us to give that information.

MR. CAPLAN: Just for my clarification here, it seems to me, two issues up on the table. One is, if we say that a second generation test is where we do look-back, and that is the trigger, does that handle the consistency question we were asked about by Dr. Epstein?

PARTICIPANT: The clarification is that, one, the second generation test triggers a look-back; only people who are confirmed positive. And two, is that the look-back will only proceed back to the recipients of second gen screened blood. So, it will only go back as far as the people who got that kind of blood. Because prior to that point, one, is we are talking about seroconverters, but prior to that point, the risk was so high that this whole other strategy is playing in. So, I think those two points make it consistent.

MR. CAPLAN: And let me ask then the second question, just for Keith or anybody who wants to jump in on this, if we were to say, our reasons for not doing look-back prior to 1992, are that we feel we are going to miss too many people. Not that it costs too much, or not that it is a pain in the neck, but because there are a lot of single donors that are lost forever to -- could have exposed folks.

What we want is a very aggressive campaign to let everybody know, prior to 1992, who got a transfusion, that they may have been at risk, and that would include, not just the public service announcements, but change in medical practice to make sure that every physician and nurse asked about this, as part of a routine question or physical, maybe also, putting in some sort of notices at hospitals. I mean, up on the wall.

What we are talking about is a serious push to let folks know the best way to let the most people know who might have been exposed is simply to really undertake a huge outreach campaign through providers and through public service and general media announcement. It does not get to my issue about how I got this, and my right to know that. It does trade it for, there are so many people that might be at risk that we could not pick up by tracking back, that we better spend the resources doing general physician education and general public health outreach. Is that roughly where we are at, with that?

Somebody mentioned, just to toss it up and I was curious. We could have letters written to every person who had a blood transfusion in the United States, prior to 1992. Yes? That is another kind of notification, it is a lot of stamps, but --

PARTICIPANT: Actually, that was kind of the compromise that I was going to propose, which is, make first of all, semantic distinction between look-back and risk notification. Look-back, define look-back precisely as Mike and Jim defined it. And then say, we are advocating a risk notification program for people who received transfusions. That would include a letter notifying them that they received blood, a blood transfusion, during a time of risk. Of significant risk. Maybe not high risk, but of significant risk. And that we recommend that they be tested.

PARTICIPANT: That program, I should give you the numbers, because this -- one issue is, this was HIV, which obviously should have triggered a lot more concern in patients' minds than HCV. The other is, the time period of risk was much more discrete, and the time when this effort was undertaken was relatively soon after the time period of risk. But at UCSF, 17,000 people received such a letter. So the whole, you know, transfusion service, history of transfusions was scanned and they identified 17,300 people who -- they were discharged from the hospital after getting transfused.

Only 808 people responded to those letters, and of those -- which is 4.4%. And only 15% of the estimated 12,000 recipients. And of those, they only identified, I think, about 15 infected people with HIV. This was a major area of focus in the late eighties, and the American Hospital Association recommended against broad scale implementation of that program. This was at the same time as the national recipient notification effort in the late eighties was undertaken vis-a-vis risk of HIV from transfusion.

So, the program, the recipient programs, are not high yield and, from many people's perspective, cost-effective, either.

PARTICIPANT: I do not think anything you do with respect to look-back or letters are going to be the kinds of yields that we would like to see, but either you are going to have to do, or think about doing a targeted look-back, or a general letter to those folks who have been transfused.

It is easier to send a letter informing people of a risk. And if they do not take advantage of it, you are still going to have the public service announcements. You are still going to have several modes of attack to these people, and you can take a horse to water, but you cannot make it drink, but our obligation is to inform these people that there is a risk associated with them being transfused prior to the time that reasonable testing had been implemented. And that is the only obligation that we have.

We cannot force these people to go in and be tested. Even if the yield is low, we have addressed our obligation, and if folks choose not to be tested, that is their choice. But we have to do something. Either we do the targeted look-back, which will be a problem for blood banks, for hospitals, and so on. I mean, either way, we are going to end up spending money that is going to have, not the kind of yield we would like, but something is better than nothing.

MR. CAPLAN: Well one other way to -- I mean, I do not want to get into the details of the UCSF thing. The idea here is, of course, obviously, we have learned to do these things in tandem. Not just the letters but the public service announcement and the medical question being a routine part of a physical and a properly prepared health establishment to answer these questions. I will be moving to this country whenever it exists.

The fact is, too, that if we look at the outreach effort, the situation is such that with the potential for cures, and with the potential for lifestyle information, that might justify the expenditure. The other side of it is, you have something to offer. It is not just a letter saying, hey, bad disease, may have it, do not want to communicate it. It is, bad disease, something might be done, and you might want to alter your lifestyle. So, the momentum that comes behind a letter campaign could be different, potentially.

PARTICIPANT: Well, and it also gets to the issue of shifted risk, because unlike look-back which takes energies, significant energies, away from personnel within the blood banking community, with computers, I mean, it is pretty easy to put out a standardized form letter, so that it really comes to a dollar issue, not an energy issue.

I think that that is a major advantage, and probably still has about -- still will -- I do not know this, but I would guess it has the likelihood of uncovering people for follow-up and testing, to a degree that is every bit as effective as look-back during the period prior to 1992. And probably more.

MR. CAPLAN: Mary, then over to Paul.

PARTICIPANT: I guess I would like to seek some clarification, because I think the committee should be -- at least I do not have as clear an understanding of what the impact of that kind of a recommendation would be, you know, a more general letter notification. Are you --

Will we be recommending that hospital transfusion services search their records, and identify people who have a blood transfusion during some period of time, and as opposed to HIV, where we have kind of a sense as to when that was introduced into the blood supply?

We do not have that luxury, if you will, for HCV. It has been endemic for decades. So, transfusion services, then, would have to compile lists of all people who had a transfusion dating back some period in time, and then, use addresses that they might have had, you know, X-number of years ago --

I am trying to sort of operationally translate this recommendation into steps so at least it is clear to me, because right now, I apologize, but it is not entirely clear as to how that would be done.

MR. CAPLAN: Actually, you are going to laugh when I give you this answer. I think what the committee will do is say, the agencies have to work together -- if it goes with this recommendation -- to find a means to notify all recipients of blood, to the extent possible, prior to 1988, that they are at risk and need to go talk to their doctor. Then we will be waiting to find out what course of operation -- how that goes.

In other words, I do not think we are going to go today and say, here is this database. Find that mailing list. Go out and -- the charge will be, we think there is an obligation to let those at risk know. It looks like the yield of trying to do it specifically from infected donors to those who got their units prior to 1988, is just going to give low yield, but a general notification followed up with a comprehensive campaign involving docs and PSA, and whatever else we can think of, that is the way to go do it, in tandem. Plus the prospect of some benefit for you. Lifestyle or maybe therapeutic intervention.

That is how I see the package coming. It is moving more towards saying, that has to be then operationalized, not here, over at CDC, NIH, FDA.

PARTICIPANT: You said something about, the agencies would have to develop a plan? A plan --

MR. CAPLAN: To make that happen. Or a set of recommendations that they would send out. In other words, I do not think we are going to get, today, out of this group, the information about how often have the mailing lists been updated, what has the Oklahoma blood bank got that, I don't know, the transfusion service at Irwin does not, I do not know.

We will not know, but there are going to be some general recommendations about, try to make this happen. That will have to be put into a policy or a plan at that point, is how I see that going. Although I may be alone in this vision.

PARTICIPANT: I think, probably the experience with this letter(?) activity is what I just told you. That is the total world experience with trying to send letters to transfusion recipients in a broad fashion. I think it is a worthwhile effort to try to pursue that strategy as a pilot study, but I would recommend that we recommend that they explore this option in addition, as a potential feasible option, to find these patients.

PARTICIPANT: My thought(?) was going on the exact same vein, and that, if I am following what John is saying, we are really trying to get to the most people, and I think there is some significant question as to whether this type of letter is (inaudible). I go give blood, or receive blood today, and I move tomorrow, and now we are talking about people who had blood decades ago, and we are assuming we are going to find them.

I think that maybe the word, pilot, that Mike just mentioned -- what is the effective yield of the direct letter, and if that yield is crappy, then John's point becomes more and more relevant all the time of, we have to try to get to as many people out there as we can, and it is easy for us here to say, let's send this letter, but who the heck is going to get it?

PARTICIPANT: [comment off microphone.] -- and healthcare (inaudible) period of time, over years, most likely, as the knowledge evolves. It has to be part of it, it has to be a cornerstone of this whole approach and I think we are more or less agreed that a targeted look-back from the point of view of the introduction of the second generation screening (inaudible), does make sense, so what we are debating is, prior to that, and education has to be part of it.

PARTICIPANT: It seems to me we are talking about perhaps something similar to notify all of the troops in the Gulf War that you may have been exposed to nerve gas, and withholding the information that those within five miles of it were exposed critically, and we are not going to notify them, specifically.

We are withholding information from people that you already have the information for, you are not going to allow them to have. And they have a right to have it. We say that in our premise, to begin with, that we are going to try to give all the information that we have, and this is where it differs from, I think, what Ed is saying, is that we have specific information that belongs to that individual, and we are unwilling to provide that information because it is too costly.

MR. CAPLAN: Let's go over to Ron.

DR. GUERRA: Some points that I think need to be clarified. First of all, as one goes into the past, risk increases. In other words, the risk of hepatitis-C was actually far greater in 1970, than it was in 1990. And I think we need to put that into perspective.

The second point is, and I want to clarify this. The hospital blood banks, most of them, do not have the records in the blood bank of who was transfused. Most of the blood banks in the hospitals have kept information on who the problem patients were, because I have been an inspector for the ABB in the past, and clearly, that is what they retain, if that is even retained. So, there is no way, going to the blood bank, that we could get a list of transfusion recipients that goes into the distant past.

The last point that I want to make is that one method here should not preclude the other, and the statement that I have given to Steve and I have modified it with what Marian had said, so to speak, encompasses this, and I would just like to put it up at this point. It is not the final statement, if you could read that, or if you would like me to read it.

PARTICIPANT: For individuals transfused prior to 1992, we recommend the testing for HCV antibodies be undertaken. Identification of these individuals through a targeted look-back (inaudible) is not recommended as the primary method of identification. Educational campaigns targeted to the public at large and selected high risk communities and survivors is recommended as the primary form of identification of these individuals.

DR. GUERRA: That really encompasses everything that is being said and it does not preclude one form over the other in terms of looking back, where records do exist.

PARTICIPANT: Then you are going to disregard two years' worth of information from the laboratory tests that (inaudible) provided, that would identify more selectively a small group of individuals who were exposed.

DR. GUERRA: No, I am not precluding -- I am not saying that one should not do the other. I am just saying what we would recommend as a primary.

PARTICIPANT: Yes, but they are not going to do the targeted look-back.

DR. GUERRA: Not recommending it. It does not say, we are not doing it, it is saying, not recommending it as the primary form of identification. Are you --

PARTICIPANT: That almost reads as though, if you are not recommending it, then it is not going to be (inaudible).

DR. GUERRA: No, that is not the intent of what I have written, and -- I mean, are you concerned, John, with the 1990 to 1992 time frame?

PARTICIPANT: Yes, the assay when it was first installed in 1990, we have that -- that is the data, I think we are looking at. So, those tests that were positive during that time, we now have a look-back that we are not proceeding with. You are saying, we will disregard that testing.

DR. GUERRA: No, I am definitely not saying that we should disregard that. That is not intended or implied by what I have written.

MR. CAPLAN: Just so I am clear, could we -- is there a position about this, what if we changed that date up there until 1990, I mean, just erased 1992 and made it 1990, and took the look-back requirement, targeted look-back requirement for that interval to include both the one and two tests?

I mean, we are sort of rattling around on this -- one of the reasons I am inclined to think about this is, it seems to me that is where the record-keeping is still going to make it feasible. Beyond that, we start to get into this antsy area of, what is the yield and what have we got, even if we are trying hard to do it, it starts to -- then we are back, more than seven years, and things get -- so would that --

PARTICIPANT: Again, you have two issues, as Jay enumerated. You have the donations triggering it, and then you have the recipients of those kind of screened products. In terms of donations triggering it, the downside of going prior to second gen, is the test was less sensitive, less specific, you did not have a supplemental to discriminate, so you will be triggering look-backs from the first gen test that will -- half of them will be nonspecific, and you will be creating anxiety, etcetera, those issues.

The other is that the people who got blood between 1990 and 1992 had virtually a zero risk, and -- I am sorry, had a substantial risk, and this look-back will only find a proportion of them, so there is the issue that those people do need to get the public health campaign, because they did have risk, and we are not going to find most of them through targeted look-backs.

MR. CAPLAN: I understand that, but in a way, if we went to the 1990-1992 and included them in, wouldn't that give us what some of you are talking about, which is the ability to see how that look-back went? In other words, that could, in a sense, function as the pilot.

If we said, today, well, we have this inaccurate test and we do not want people to just rely on the notion of, they looked back, did not look back. We are going to give them the public health message, anyway, that prior to 1992 you should get a test.

We will send that message, anyway. But, in that 1990 to 1992 window, we will make every effort to do the targeted look-back, and we will see how it goes. De facto, we have a pilot there. I mean, we will know pretty quickly, or relatively quickly, if there is any point in doing this. It may be a position that lets us at least not saddle the system of trying to either find people from 1945, and so forth, and at the same time, gives us some idea of whether there is any point at all.

I mean, is that -- I understand what the pluses and minuses are in terms of looking to the false positives and so on, but it may still give us a means by which we are exploring in a reasonable way, with reasonable records still around, what people might want to know.

PARTICIPANT: -- 1990 or 1992 as the date of transfusion.

MR. CAPLAN: It is the 1990 date of donation. Donors would be in before --

PARTICIPANT: -- hepatitis-C --

PARTICIPANT: -- say someone came in in --

PARTICIPANT: In 1990.

PARTICIPANT: -- 1990.

PARTICIPANT: That was the first generation test.

PARTICIPANT: Okay, if someone tested -- say, someone came in in 1992, or 1991 --

PARTICIPANT: In 1991?

PARTICIPANT: And tested positive.

PARTICIPANT: Okay, there is a 50% chance that that test --

PARTICIPANT: And they had --

PARTICIPANT: -- would be truly positive, and not a false negative.

PARTICIPANT: And so you decide to do look-back.

PARTICIPANT: And we look back.

PARTICIPANT: And say, the last time that person gave was 1988. Would you go back to those 1988 recipients, or do you cut it off at 1990, being --

PARTICIPANT: We would go back -- [simultaneous discussion].

PARTICIPANT: So, what if he had given in 1988?

PARTICIPANT: That is the whole game of look-back, that is not a pilot at all, that is the whole game. Yes, but we would go back.

PARTICIPANT: So, it is just standard look-back.

PARTICIPANT: That is it.

PARTICIPANT: We can always do more. We will never be able to do everything for all patients. If we want to consider including in look-back those donations that were found positive in first generation testing, where there was about a 50% probability that the donor was truly infected, well then, why don't we also do look-back on the surrogate testing we were doing in the last half of the 1980s? Because about 30% of those individuals who were core-positive and/or ALT-positive, were truly infected with non-A, non-B hepatitis. They were not great tests, they were the best that we had available at that time. But, we have gone from a 50% probability to a 30%, well, that is not that different.

So, where do we draw the line? And I do not think anyone is thinking that it is reasonable to use surrogate testing, and I think we have to bite the bullet and say, we feel this is the appropriate point.

I would be willing to -- I mean, philosophically, I guess, since I typically could see some benefit to extending the look-back to transfusions to 1990, based on second generation testing. In other words, a donor donating in 1992 or 1993 might have donated in 1990 to 1992, and we would look back on those transfusions during that time. But to go further back, probably has low yield. We are dealing with now records that are a decade old, difficult finding patients, and relatively very little return for a lot of effort.

PARTICIPANT: Practically speaking, if we agreed on a date today, how quickly could this actually begin? Are we talking 1998 or 1999 before we actually have a broad reach out to the public, which then really takes us out of that seven-year record-keeping?

MR. CAPLAN: I have no idea, except to say, if anything happened before 1998, of any nature, I would be very surprised about it. So, I will say there is going to be at least some time lag of that nature.

PARTICIPANT: Well, it just seems from what Jim has said, and also just the practicalities of getting this implemented, that 1992 probably makes better sense.

PARTICIPANT: The ALT test is not specific though, Jim, and you know(?) it(?). And there is specificity in the hepatitis-C first generation, (inaudible) 50%. The ALT is nonspecific for just about everything that comes up, including walking up the stairs fast. So I do not think one can really use that as a reasonable look-back procedure, whereas I think we can take a look at a specific test, for 50% specificity, and consider that it would be useful.

PARTICIPANT: What concerns me, John, about the first generation tests, is if there is a substantial number of individuals tested with that procedure, who are actually infected, that were not picked up, so, the potential there is, a recipient of the unit who is actually infected, might be aware of, well, look-back has taken place during this particular period of time, and I am fine.

I really think that the 1990-1992 period is a -- it is a problematic one, and I particularly like Ron's terminology here, from the point of view of not the primary method of identification, because it is certainly a useful tool, but it is not the only tool, or the key tool here.

PARTICIPANT: I still am looking at this from a personal view of an individual, that you are withholding information --

PARTICIPANT: But you could get false information.

PARTICIPANT: That is alright, but that is up to that individual to decide, when you give them the information, that it may be false, it may be true. But you have to give them at least the option of making a decision. If they do not have the information, they do not get the decision.

That hearkens back to the syphilis trials, as you remember. We just will not tell you that you are positive for syphilis, and we will just kind of watch you. This is the same sort of approach that we have here (inaudible). We have a situation where you do not have any knowledge of what has gone on that would indicate that you are at higher risk than someone else.

MR. CAPLAN: Carolyn.

DR. JONES: I disagree, because if we had some other sort of programs(?) here [comment off microphone], you are not denying them information, you are only denying them information about a particular unit that may have infected them, and that is not -- you are not going to give them a donor's name or anything.

What you are doing, is informing them that there was a risk associated with the transfusion. You cannot even say for sure if they were transfused with a positive unit, whether they actually came down with HCV. So, I mean, what information are you denying them?

We are saying, there was a risk associated with transfusion at that time. Go get tested. I do not see what you are denying them.

PARTICIPANT: Yes, because it is a general information, whereas you have specific information on a test(?) (inaudible) for that individual. And I think that is going to give rise to some real problems, legally, and I can see individuals saying that, I now am tested positive, you gave me a general statement in the community newsletter that came through that said, anybody who has received blood and so on and so forth should be tested, but you also had knowledge that I was at a much higher risk than somebody else, than my neighbor, who got a blood transfusion. And I just overlooked that, because I did not get that information that I was maybe at a 50% higher risk for getting it than anybody else.

MR. CAPLAN: That is the point I was going to stress. The question is, if you know that you have a positive unit, and you have some test indication, plus records, it is not just the test, but it is the records, the ability to find it.

Do we feel some obligation to say, not only should you get tested, because you were at risk prior to the availability of a very reliable test, but we know, in our records here, that Mr. X was positive, and we tested and found out that there was some higher chance that you could have been infected; we have to tell you that.

My interest in feeling obligated to get that path closed down is beginning to diminish, the further back we go in time, because I am becoming persuaded that that is just not the route to do it. The records, the moving, the phone numbers, the mailing addresses, things start to fall apart.

I am not sure in this 1990, 1992 window area, that we shouldn't or couldn't, if we wanted to, and I am not sure whether it is donation date or transfusion date, still, I hear that, but nonetheless the general question is, if I know that you have been exposed to a transfusion that had hepatitis-C in it, because I had some means of testing that, not ALT, but the first generation test, and it is in my records, would I want someone to call me up and let me know or try to find me?

To me, that is the bottom line of that push, and when we cannot call me up or find me anymore, then I think we have to be very aggressive about just letting you know, hey, hepatitis was a big problem in blood transfusions prior to this testing. We do not who is at special risk, you are all at equal risk, go get tested. It is when there is more information that somebody might have that says, you are at risk because of what I know that is in the record, that triggers some duty to disclose. So, that is why I am still hung up.

If we could push past or resolve the 1990, 1991, 1992 issue, in terms of how that goes, I suspect, if we get consensus there, we might agree that all we can expect people to reasonably do before that date, is not really look back, and maybe not mail -- I am not sure we haven't knocked the mailing to the transfusion recipients out. But that massive notification of, folks have to know that before we could test for this, there was a big hepatitis problem out in that blood supply, and you really have to get in there and talk to your doc and your doc has things that they can tell you about.

By the way, Trish, that campaign to educate the doctors, if we really say that has to come first, that takes us 'way into 1998, by the way, just as a -- if you are doing that first so they can answer the questions that they get, and what are my options, and so forth, we are into 1998, somewhere. Mike?

DR. BUSCH: Yes, just to pursue your sort of spirit of compromise in exploring this interval period. One could make the argument that there was HCV testing beginning in 1990, and there may have been a perception put out in the community that blood was safer then. So for that purpose, to consider overlapping time periods where the general campaign really extends to 1992, because we know there was substantial risk through to 1992, but that indeed the look-back program extend back to first gen screened blood.

That both potentially, the people who were identified obviously by second gen once implemented, but even during the first gen screening period, the seroconverting repeat donors, you would go back to their prior, first gen screened blood, blood transfused since implementation in 1990, and track records for those. The records are probably mostly there. You will get a higher yield, as Jim showed, from that more recent period, but that we abort the process at unscreened blood.

That unscreened blood, even though recognized right up front that unscreened blood has a substantial risk, probably a higher risk than the screened blood, but the process just falls apart, as you said --

MR. CAPLAN: The moral difference is -- if I can talk that way for a second -- weirdly, we know there is more risk before the test, but the converse is, there is more obligation to say something because of the test. I mean, it comes -- the more you know, the more you have some duty to tell somebody what you know.

When you do not know anything except that the risk is worse, then I think you are diminished and your obligation -- you cannot sort of go find anybody and tell them anymore than, gosh, there was a big huge risk that has been going down like this on a curve, but if you look at the fifties and sixties and seventies, 20% -- whatever it was -- it was really risky out there and you want to pay attention to that.

But when we tested, even in a crude way, we got information. We have some of that. We have some duty to let you know or try to find it. That is what I am fishing for here. It is not that I do not understand very well that folks are not off the hook because they did not get a good test or even a half-baked test. They were probably the most risky group of all, but when we have test results, that is what you were fishing for with that analogy about withholding information.

PARTICIPANT: Yes [comment off microphone.] If the information is there, you are obligated to provide it to the public. And if the information is not there, you are not. So then you are safe.

PARTICIPANT: I guess the issues that I am hearing here are, if -- well, number one, it is -- I think we need to decide upon a date, and depending upon that date, I am hearing the issues of, there may not be records available for persons. I am hearing also, financial impact that it would have, and also, ethical responsibilities that we may have. And I think we have to go back to ourselves and ask ourselves the question, if we choose one date over the other, are we being ethically responsible to the American public?

What if we choose the 1992 date over the 1990 date, if the American public understands that there was at least a test, the generation one test in 1990, what is going to be our answer to them to say, we have chosen one over the other?

PARTICIPANT: As a parent with two children who get a lot of transfusions, what I would like to see is, when it is all said and done and the American public realizes why we recommended certain procedures and not others, I think that we have to consider how we answer them in that regard.

I also think that when we start talking about the low yield, as a parent, that does not have much meaning to me, and I understand your end of it, and the cost effects of it, but also, the people that we do identify, will at least have the knowledge available to them, and if we agree that we are not going to reach everyone, do we not have an obligation to at least notify the ones that we can?

Part of what I am sensing out of this is that, you know, it is like if we had a vaccination for a disorder, but it is so costly that, unless we are able to identify enough people to make it worthwhile making this vaccination, then we are not going to put it out to the public. And I mean, the public's perceptions since the HIV problem has changed, and I think if we educate them properly, and give them the tools, whether they use all those tools or not is not our part of this.

Our part of it is making sure that if the information is available, we do our very best to get it to as many people as we can, whether the yield is low or not, I think the quality of those people that we do reach, the quality of life that we are helping them with is substantial, and I do not think that they are going to be upset with you because -- you know, it is just a matter of the quality of life of the people we can reach.

When we start talking about numbers, then as a parent, I have a problem with that. I understand that, sitting here on this committee, but as someone who, if I were not here and I heard about this years and years later, I would have a definite problem with all of this.

I just think that we have to reach the people we can, and not worry so much about yield, because the people we do reach will be the ones that will thank you, and the ones that we cannot reach, no matter what we come up with at this point, I do not see us finding a way to reach everyone. So, I think we do have to reach whomever we can, however we can.

MR. CAPLAN: Well, let me put on my stovepipe Abe Lincoln hat and say, how about if we tried a statement that we recommend for individuals transfused prior to 1992, that they be tested? I think that is a very radical recommendation, but maybe one you want to go with.

That we say, in the interval, once the appearance of a hepatitis-C-specific test appeared, we recommend look-back that would be generation I- and II-driven -- is what is going on here, with the notion that the obligation is, if you have the results, you are going to try and track them back as far as you can, so I think we are in the 1990 to 1997 period there.

That we recommend prior to 1990, almost what the language is here, the targeted look-back is not the primary way to go, but that a massive public health outreach and healthcare educator program is the way to go, to let those people know what is going on. Is that a -- that is not quite English, but are those the elements?

PARTICIPANT: How could you do look-back prior to 1990, if there was no test -- [simultaneous discussion] --

PARTICIPANT: You have to understand that the blood centers have records going back to the sixties, fifties. So, as soon as we decide to do what you just said, we would go to all the records of all the HCV-reactive donors, starting with first gen testing, and then the confirmed positive with second gen.

We would go back and determine which of those donors had prior donations. We would go back to all the records of prior donations, dating back to the sixties, fifties. We would then have to disseminate to the hospitals the list of all the components that were issued at those hospitals over the past few decades.

It is up to the hospitals to individually determine which records they can then trace. So, they have to go through a process that is a huge undertaking. The scope of this, what we are talking about now, is probably a hundredfold that of the entire HIV look-back program that we undertook in the mid-eighties that tipped the boat to near sinking in many transfusion services.

What we are describing is a massive program that we are arguing is a very low yield program, and you are saying, do it anyway. Some of you are saying, do it anyway. It is going to get a few people on the back end and it is worth getting those few people. That is the question at hand. Is it worth that?

MR. CAPLAN: More comments.

PARTICIPANT: (Inaudible) with the obligation. The testing was done, and therefore you have to follow up. And information must be made available. I think you could construct some limits on what the hospitals have to do, and I think the hospitals actually will construct limits on what they will do. But also, I have seen what happens in the hospitals when they do the look-backs, and also, I have seen what happens when we do that in our Red Cross centers, and I think it is doable, but it is a pain.

PARTICIPANT: Your comment yesterday -- well, two responses. One is, your comment yesterday that the cost will be passed on. That is no longer the case. We are into managed care. The price of blood used to be $100, it is now $80 and driving -- due to the HMOs driving, so there is no pass-down of costs, there is contractions of service that(?) we(?) see(?).

If you juxtapose -- again, assuming that there are limited resources, to me, if you juxtapose -- if I were a transfuser and my child was transfused in the mid-1980s, and I were to weigh the probabilities that that child, if infected, would be found and notified and tested through a broad public health message versus look-back, there is only at best a 5% chance, if he got the infection from a transfusion in 1985, that that will be eventually found through a targeted look-back. If you run the numbers, and you can walk through that.

I think there is a much better probability that he will be found as a consequence of an effective, broad public health notification to recipients, the things that -- the other issues we have talked about, much more probability that he will be tested. And I think that is the way to go, and I think we are wasting our resources with this double effort, where one is so much poorer than the other.

PARTICIPANT: As a blood product consumer every week, I am conflicted here. But I honestly have to say that, if I had a real specific, directed message, that anybody that is a blood product user, or anybody that had a transfusion, needs to go get tested.

I would be more comfortable with that, than I am after hearing all the data, depending on some blood bank, or some hospital, or infusion service, to tell me that I had a specific lot or product that potentially was contaminated.

Again, I am conflicted, but I would feel much more comfortable -- if you have any question at all that you had a transfusion, for whatever period, you need to get tested. And that we spend our resources letting the public know that, because the other side of the coin is, only 5% of those that have HCV, as we have heard in testimony, or 8%, are by transfusion. And we have an obligation to let those that are affected otherwise -- if you will, that they have to be aware of this condition as well.

So, as a blood product consumer, I am -- again, I would rather just go down and get tested and make sure everybody in my community who has (inaudible) anti-trypsin deficiency, gets tested, and we probably all have.

PARTICIPANT: But you are very knowledgeable about this, John, so I think you are coming at it from a different --

PARTICIPANT: That is true.

PARTICIPANT: As we all are here. But about the cost factor, maybe just a perspective on cost, and we talked about this a little bit at the break. The cost of blood has gone down, as we said, to $80, and there is a lot of pressure on blood. This is a lifesaving commodity. What does it cost us to give an antibiotic program to an individual? Some of our antibiotics are running $1,000 a day.

I ordered antithrombyn-free concentrate for a patient the day before I left, $2,000 worth, as a lifesaving precaution, one shot. Our hemophilic population are getting $800 to $1000 worth of product for an acute bleed. And blood is very inexpensive when it comes down to the fact that it is $90, or whether it is $100. And I was very strongly in favor of a $40 -- and keeping the price down when I was overseeing my Red Cross responsibilities. But I have come the other way around and realize how inexpensive this product is, in comparison to everything else that is lifesaving, that we are providing our patients in the hospital.

It is a negligible cost when we consider it, and the fact that we can provide a very clean -- with respect to reducing some of the infectivity -- and it is going to cost us another $10, I think is something we need to get out of our minds. Most of us were brought up in the era of turning the lights off when we left a room, and I think maybe we still have that kind of concept. But I have been willing to extend my thoughts on this to the fact that we do need to perhaps price blood much higher. It is more important to have it as safe as we can get it, and safety means that we are going to have to add more cost.

PARTICIPANT: And John's concern about depending on a transfusion service to notify him that he might have received a unit that could have transmitted an agent, I think is an appropriate one. Not that blood bankers are cavalier or that they are not trying to do the best job possible, but there are limits and constraints put on their operation.

I just recently completed a survey through the College of American Pathologists which has not yet been published, that inquired of hospital transfusion services what their budget had looked like over the last several years, and whether they had had to make changes in their operation.

Of the responding laboratories, 15% noted that their -- I am sorry; 21% noted that their budget had been restricted, reduced, out of proportion to any changes in their workload, over the last two to three years. So, 21% have seen a contraction in the amount of resources they have available, and in fact, 8% noted that because of those restrictions, they have had to make changes in their operations, that in the opinion of the pathologist or the laboratory supervisor, reduced the safety of transfusion at their institution.

Now, that is worrisome, if just with ongoing operations, these laboratories are having to cut corners to the point where they are having a riskier operation, now, on top of that, we are going to be asking them to perform some type of look-back procedure. Many of these hospitals, that is going to be a manual procedure; in any case, it is going to very labor-intense.

The hospital administration is not going to provide extra staff to do that, and the patients in the operating room and the emergency room are still going to need transfusing. Someone is going to be doing double duty, trying to do too many things at a time that could lead to an avoidable error that has been caused by our good intentions of trying to get information out to patients.

MR. CAPLAN: Let me try this. Sense(?) of the meeting, nonbinding. Just on the issue of, should we ask, or look-back, from donors, who are positive by, let's call it, hepatitis-specific test, 1990 to the present. As part of our recommendation.

I think we are going to have no problem saying we recommend that everybody get the test. It seems to me, we probably are not going to have a problem saying, prior to 1990, it is going to have to be a massive campaign and well-structured thing. I think we have the two ends of that recommendation running.

What I want to get is a sense by hand -- I will not hold you to it, you can change -- would you favor a recommendation right now that said, in the period 1990 to 1998, given what we know about cost, burden, yield, obligation to tell when you have these results, that we include a recommendation to look back for any donor who tested positive in that time?

That is not exclusive of the public health campaign. We know we are going to miss -- they are going to be subject to that, anyway, but I am asking about, as a supplementary strategy in that 1990 to the present strategy. Well, we have test results that indicate there is a unit that is positive, do we have to look back and find out who else might have been exposed to the donor of that blood?

Not all can vote who are sitting around here, but of those who are on the committee proper, how many would favor including that now, that provision? You have to put them up 'way high. We won't hold you; you can even close your eyes. Against? Uncertain, indifferent, no -- Okay.

PARTICIPANT: Art, let me make sure that I understood what you said. The period that is creating the problem for us is obviously the 1990 to 1992, that is where we are hung up. Are you recommending that both targeted look-back and educational campaigns be utilized as primary methods for that group?

In other words, we say that we are going to put equal --

MR. CAPLAN: Equal emphasis?

PARTICIPANT: Equal emphasis in that time period. Is that what I hear you saying?

MR. CAPLAN: Yes. That we would do both. We had test results. We could probably get someplace with the record. And we would do the same public education -- they would be told to get a test, anyway. They would be told that there is risk in the blood supply. They would be told that it was 50% accurate. You are not safe because you did not test positive in this interval, blah, blah, blah.

The question is, do we include that along -- I have to say, by the way, the sense that I am getting here is that there is still not strong sentiment from the committee to include that as a co-equal way to do this right now, just for the sense of the meeting, so far.

PARTICIPANT: Kind of gets back to what I said before, since this is obviously the impasse area. I would be very much in favor of being very intensive in identifying any type of risk to transfusion group, during this time. Because I really believe that, if we -- you know, that the yield is suspect enough that any resources we can to get everyone identified -- and that is the reason I originally proposed the mailing, who was transfused at all, is more likely to get those people into identified HCV-positive status, if they truly are, and into therapy, than the targeted look-back.

I really think that, from a physician's perspective, I would rather see more people get more care out of this, than if we have to make tough choices, rather than saying, well, we got a little bit closer to identifying the actual ones who some test said was accurate, when it may not have been.

The reason I am so leaning towards really -- and I think maybe one of the ways we could get past the impasse is to say, because this is a problematic era, during which we had something that may or may not -- I mean, it may have been better than nothing, but it certainly was not what it was to become.

That we could apply more resources during this time, along with the blanket educational programs, to identify those individuals, across the board, who are at risk. And then, ease backward, knowing that it does not make any sense to do that for people who were transfused in 1981, because we will never get their addresses. But we probably do have addresses, as we have already said, during that period, and if we took a lot of intensive effort to identify everyone who was transfused and get them tested, we could come closer to solving this impasse, I think.

PARTICIPANT: I guess I am moving much in the same direction. I am very sympathetic to the points that John has been trying to make all morning long, but we -- I guess, my economics training tells me that no matter how badly we want two primary strategies, that in reality, that is not going to happen.

Then I listen to the types of things like Jim and others have said, that if we devote a whole lot of resource to something that is not going to work, will that then diffuse the effectiveness of the type of thing that Keith has just described? And I guess my leaning -- you notice my hand did not go 'way up when you asked -- my leaning towards what Keith was saying, let's really get big barrels, and this educational campaign that we are talking about in this area(?), and as I mentioned earlier, my concern about whether we are going to find people.

Yes, I think we will find more people in the 1990 to 1992 than in the eighties, and maybe we can be more effective there, and it is in that spirit that I would tend to say, let's not push back to 1990, but stick with 1992.

PARTICIPANT: That section [comment off microphone] --

MR. CAPLAN: Just their accuracy, or lack thereof. Alright. Let me try --

PARTICIPANT: I will just state again something that Jim reiterated, the suggestion that we do look back from second gen screened blood, which would include the prospective look-back beginning back -- you know, to seroconversion beginning back from 1992, but would also include tracing back to recipients of first gen screened blood, who got blood that was detected as infected after the second gen test, which is about one in 1,000 units.

That would basically give us an overlap for that period of 1990 to 1992, when there was transmissions going on, some people may not get the full message because they will think that was screened blood, I do not need to worry. So, for that phase, we do do a look-back to first gen screened blood that was negative, that was picked up when the second gen test was introduced.

It then gets us also to the issue that the people who we are doing look-back on, are people who have a supplemental test to target that, to the specific period. Some level of overlap coverage for that problematic period. But would not go back -- I would propose it not trigger from first gen screened donations for which there is no supplemental.

In addition, it would not extend back to recipients of first gen unscreened blood. It would only go as far back as 1990, when first gen screening was introduced, and the major thrust of our education would be to the pre-1990, but bring along also the 1990 to 1992 people for general testing recommendations.

MR. CAPLAN: Go ahead. Keith.

PARTICIPANT: One of the questions that Carolyn raised yesterday after we met, there is another group here that we have kind of ignored in a way, in terms of, if we put them in or do not put them in in terms of look-back, but suppose we decide to go with the 1992 rigid cut-off?

There are still the donors who are identified. Whether you do look-back or not, if they truly are positive, those people are accessible a lot of times in a way that even the recipient may be much more difficult to achieve. Could we also go after that group, in that gray range we are talking about? Because -- in other words, they are probably positive. They need the same counseling, they need the same further testing, if they have not already had it. Do blood banks have the names of those donors? Could we identify --

PARTICIPANT: Those people, we have been obligated and have notified those people all along. So they are always told they are deferred from donating, they are told they are positive, they are told the data we have and referred to physicians.

PARTICIPANT: Oh, okay. That's good. So they are always told why they were deferred?

PARTICIPANT: Yes.

PARTICIPANT: Okay, alright, that is good.

MR. CAPLAN: Mike, can you put that proposal forward again, that we would only do look-back with a confirmed test on the gen I group, is that what you were saying, basically?

DR. BUSCH: Right, right, that we would -- look-back would be -- would commence with the introduction of second gen screening donations, and the second gen confirmed positive donations would trigger a look-back, be it to prior, second gen negative donations, which would be seroconverters, or to the recipients of first gen screened units for when the test was first introduced. Over that first year or two there were large numbers of donors who had been previously giving regularly, who had been negative on the first gen, who were picked up and confirmed positive. Rate of about one in 1,000.

What I am proposing is that, during that period, we would notify the recipient, trigger the look-back to attempt to notify the recipients of first gen screened blood from those second gen pick-ups. But that that would stop at the point of unscreened blood. So it would only go back to recipients who got blood since May of 1990, when the first gen test was introduced.

MR. CAPLAN: Let me try to tax Steve's ability. New slide up. The reason I am going to ask you for a new overhead here is, I want to see if I can get this into a -- Mike might even want to write it himself if he is so inclined up there, but the idea is, if you take the bookends, of what we had up on Ron's proposal, and agree that we are going to call(?) for the testing before and that we have a strategy about outreach for the pre-1990s.

What I want to do is get a vote, again, sense of the meeting, about this proposal, because if that becomes the middle package for how look-back goes, then I think we have a -- we can pull these apart or list these separately, but we may be somewhere. If it is not, we will have to go back and debate a bit more, but let me tell you this. I am not going to call for a final vote, I am going to let us take a break. But if we get close to agreement on this, then we can reconsider it and finalize it when we come back.

I apologize, by the way, to those in the group who do not think that the look-back issue is the nub of what is going on here. There may be some of you saying, yes, that is the least of our worries, but we will take the meeting and just continue with it, the other issues, other things that you may want to get on there as recommendations when we come back, so we will just plod on.

If we do not get to the post-lunch topic and speakers, if we get consensus, and get this written up in a form that we can understand, that will be fine. We will pick up what to do next the next time we meet. Alright, so one last time of trying to see if --

PARTICIPANT: Okay, actually, I would recommend outreach extend through 1992, that that gives us the overlap, so that continues -- the targeted look-back be undertaken for donations confirmed positive with second -- who are detected by the second gen EIA, and confirmed by appropriate supplemental testing since 1992. So, it triggers on the donations that are confirmed since introduction of second gen in mid-1992, that this targeted look-back extend to, as far as first gen screened blood only. So, back only as far as May 1990.

MR. CAPLAN: Do I understand what we are going to be asked to see if we agree or not? All clear on this? Okay, how many people would be willing to go with --

PARTICIPANT: -- (inaudible) for an example. So, there are two relevant examples. The first is, we have been screening with first gen for several years, we have just introduced the second gen test, and a couple of months later a donor comes in and is screened positive and confirmed positive. And that donor had been giving first gen blood, and the test was missing them. That was happening at a rate of about one in 1,000.

So, because we are concerned that people who got first gen screened blood, one, we think we have a relative potential of getting to those people; and two, we are concerned that they would have a mixed message. We were giving them screened blood, they would think they were relatively safe, so for that period, we would go back to those screened units. But if that donor had also given into the eighties, what I am proposing is that we not continue that look-back trigger to the hospitals, prior to May 1990, when unscreened blood was given.

The other issue is, the donors who have continued to give and were second gen negative on several occasions, and now they seroconvert in 1985 or in 19 -- sorry, 1995 or 2,000. Those will routinely trigger the look-back per FDA's recommendation that you would go back, as far back as one year prior to the most recent second gen screened negative blood.

PARTICIPANT: If they do not have a second generation screened blood?

PARTICIPANT: They don't trigger --

PARTICIPANT: If they go back -- if they gave a unit in 1985 or in 1988?

PARTICIPANT: My proposal would be, no look-back to units given prior to -- to units given prior to first gen screening.

PARTICIPANT: But, how far do you (inaudible) go back --

PARTICIPANT: May 1990. No recipients will be triggered --

PARTICIPANT: So, the donor that is positive in 1992, you are not going to go check out individuals who received blood that he gave in 1988.

PARTICIPANT: That is the proposal, right. That we would have the overlap period where we would have both targeted look-back and the education campaign for that period of 1990 through 1992, but that it would not extend back. Otherwise, if you are not going to -- if you are going to say, just trigger by second -- and trigger by first as well -- not trigger by first, 90-plus percent of the donors who were infecting recipients prior to 1990, were picked up by the first generation test which, if you believe that is necessary, then you have to trigger it from first gen donors, as well.

PARTICIPANT: Yes, but you have gaps. You have people who came in in 1992, and their last donation was 1987. What are you going to do with those?

PARTICIPANT: Those people will be tackled through the broad education campaign, because the records are not there --

PARTICIPANT: [Simultaneous discussion] -- specific information for them.

PARTICIPANT: We are drawing a line, that is right.

PARTICIPANT: You are drawing a line, and -- you are drawing a line because it is going to cost more --

MR. CAPLAN: Well, let's see how many --

PARTICIPANT: And so it is a cost basis that you are dealing with right here.

MR. CAPLAN: Let's see how many folks favor drawing a line in this place -- sense of the meeting again. How many can go with the look-back for the second generation test, plus the look-back to confirm first gen tests by a second generation test, as the overlap period for where the look-back period is? That particular line.

PARTICIPANT: I need to state that again. I think you just misstated it. What is stated here is correct, that we would trigger targeted look-back for donors detected by the second gen test, that that would only extend back as far as transfusions from donations that were screened with the first gen test.

MR. CAPLAN: Agree? Hands? Disagree? Hands? Looking for a sense. Okay. Alright, let's revisit, we may be staggering toward a destination that no one can remember where we are going anymore, so what I would like you to do is be back in this room in 45 minutes. Short lunch. Got to come back, though.

[Whereupon, at 1:00 p.m., recess was taken until 1:45 p.m. that same day.] A F T E R N O O N S E S S I O N (1:45 p.m.)

MR. CAPLAN: Take your chairs. Somebody said to me during the lunch break, they know we are building toward a proposal to look at, again, about narrowing the time period at which look-back might be done, but somebody else did say to me, they had lost a little bit of track about what some of the other recommendations were, and in particular, we did have two versions of a preamble put up on the table. I think Ron had some language, and then Mike Busch had some language, that were opening sentences. And I do want to make sure we get back to those.

What I have in mind for our last hour -- and then I am also told, since I did a little surveying around lunch -- is that everybody is leaving. So, we will get to consensus in the next hour. I warned our speaker for the afternoon that we would not be able to get to them. We will take the hour and march toward this modicum of agreement of recommendations, but what I will do is, Steve and I -- did Steve ever come back? No, he left. That's alright.

We will work with Paul to get typed up the recommendations that we have here. We will send them out to you for a final look-see, and then finalize them that way. I am not going to try and commit you to what gets put on the overhead. They are hard to read. The penmanship is fuzzy and things get changed. So, we will try to edit them back into a usable form.

Having said all that, why don't we see -- I know, Mike, you must have been trying at lunch to polish that language a little bit on the issue. Why did I know this was true? And you could actually -- why don't we read it? We are still on recommendation four, it is what I call the middle sentences.

You will recall we said, HCV testing ought to be recommended for everybody as a beginner, and we were trying to do the major public health outreach and educational campaign to get people prior to 1990 as points of consensus, I think, and we are still in the sandwich here of recommendation four.

PARTICIPANT: Right. One here is basically a restatement of what you just said, although again, I think that should continue to include recipients through the second gen test introduction in March 1992. The PHS education campaign.

Then the second issue which, through some discussion of sort of compromise and I think appropriate rationalization of, can you or can't you draw lines at certain places? We have made a modification to what was on the table before. Basically, it is proposing to implement a targeted look-back program, triggered by donors detected as HCV-confirmed positive by second gen screening supplemental testing, which was introduced in March 1992.

There is a minor detail here that the supplemental test was not approved until early 1993, but most blood banks did have access to that, so most of your second gen screen donations were confirmed. One probable scenario would be if you did not have confirmatory, you perhaps would have to notify, based on the screening. But optimally, you would have confirmatory tests.

Then, who would be notified is the three bullets below that. Clearly, you would notify per, you know, the earlier discussion, recipients who had gotten prior second gen negative units. And this would be the second gen seroconverters, and this would be what would continue henceforth.

The second group we discussed was actually going back and notifying recipients who got first gen EIA-negative units. That would be the recipients between 5/90 and the introduction of second gen testing.

And then the third group of recipients that could be triggered from these confirmed positive donors would be recipients who got unscreened blood prior to 5/90. And in discussion right after we broke, it was felt that stopping at 5/90, when you have already initiated the identification of the confirmed positive donor, you have looked at his donation history records, and then you have a history of his having donated in 1990, 1991, but also back in 1989 and 1988.

To draw that line is somewhat arbitrary when you have that information in front of you, so perhaps an appropriate compromise would be to extend this second gen triggered look-back to recipients of unscreened blood.

Now, just a general opening the door to all prior donations creates somewhat of the uncertainty as to how many of these products are going to be traceable, etcetera, so actually in discussion with Dr. Epstein at lunch, he felt that perhaps, per FDA's usual recommendations, that extending back to some reasonable time period, but not indefinitely, would be appropriate, and putting a front-end bound on that of 1985 or 1987 might be reasonable.

MR. CAPLAN: Comments? Again, remember, we are trying to balance very important, two sets of principles here, which you have heard again and again, that what we are trying to do is decide if we have knowledge from tests, that someone may have been exposed to a donor who is infected, to what extent do we want to push for a tracing -- I do not want to call it a look-back anymore -- tracing back strategy, for that exposure, when the two types of tests existed.

I think, you know, the political, the moral question, the PR question is, if you knew something, why didn't you go back and find the people that had been exposed? We have to be prepared to answer that question. This is an attempt to lay out at least a strategy that says, in some areas, we are going to look, in others, maybe drawing a line that is somewhat arbitrary, we are going to go with a bigger outreach campaign. So that is what we are struggling with.

I think that just to put the heat on us, yet again, we will be asked by some, why didn't you just look at anyone who might have been exposed when you knew it? That is a moral question, and we have to be prepared to answer that question, if that is not what we do. And, at the same time, we also have to be prepared to answer the question, why are you making us look, if the yield is low, if the costs are high, if the records stink, why don't you do it some other way?

We are trying to balance in there, don't ask us to do what we cannot do at one end. Why didn't you do what you should have done if people knew, at the other end? Somewhere in between is what this is working to answer, is the way I would put it.

Any other -- aside from that brilliant comment? I do not have a sense, and I am really looking to the blood bankers here about the time limit issue, which has become issue C about looking back to the unscreened thing, we are going to draw that. I know there is not a right answer here, it is just an answer, but --

PARTICIPANT: Considering this, I think, from the blood bankers' perspective, one, this is a fairly enormous task. We have to go back to all of the donors found. Probably in an average blood center of 100,000, we are probably talking about tracing back and finding the records on about 5 to 10,000 donors who have been detected as HCV-confirmed positive, identifying the subset of that group who were repeat donors prior to that, through the computers, generating lists of their prior donations, tracking back through archived records to determine where those products went, and compiling notification lists to the hospitals, of all of the products that went to their hospitals from these donors who have been now found as confirmed positive.

It is a large undertaking. The issue of whether you draw the line and stop that process at 1990 versus 1987, versus 1985, versus indefinitely, for the blood centers, it is modest additional work the further back you go, but I think the issue really becomes at the hospitals at that point.

Once we have started to track these records, we can basically go back and find these archived -- I need to go back to my own center and find out what the impact would be, but I really think the burden here, if we are going to move through, and we are talking specifically about how far to extend it back, is going to be on the hospital's -- particularly concerned that a hospital has a blanket understanding that they do not have records prior to some date, but they are getting these notifications from us about transfusions prior to that date. They have certain liabilities and efforts that they consider they might have to undertake to go beyond any reasonable effort, to ask the hospital folks to comment on.

PARTICIPANT: Well, to put a number on this, we are probably looking at, if we talk, let's say, 1987 as a cut-off date, I am working off the top of my head here. But we are probably dealing with about a quarter of a million tracings that hospitals will have to do. That is, they will be notified of about a quarter of a million components, that they will have to go back through their records, a large share of them being manual records, to identify the recipients, and then initiate the notification by letter or however they will do it.

PARTICIPANT: That is nationally.

PARTICIPANT: Nationally. And if the FDA, in providing recommendations, guidelines, or regulations as to how this is to be done, implements it in the way that HIV notification, look-back notification, is to occur, it is a substantial process, because we are required to make at least three attempts at notification, if I remember correctly, within a defined amount of time. And this could be -- if we were to be required to do this, or if that were the standard of practice to do this for HCV, it could be a substantial undertaking for hospitals. Not one that is not worthwhile, understand.

PARTICIPANT: Clearly, Jim, the link is broken -- the three blood bankers are speaking here. The link is broken at the hospital transfusion service level, and they will not be able to track back that 250,000 components.

The blood centers will be able to notify the hospital that the link is broken, unless they actually went to the patient files, which are on microfilm -- and that truly would be an impossible job, and I would never recommend that -- so, I do not think we are going to see the hospitals being able to track back this number of components.

MR. CAPLAN: One thing we can do, if we want to draw a line, somewhat arbitrarily, about 1985, 1987, with this look-back, coming back again, the trigger being still, the gen II test confirm, which still leaves some window of people that we know something about, maybe not well. I just want to remind you of all that, that we have not -- that we have decided we are not hunting for it all. But, we could say -- I hate to be temporizing about this -- but, our recommendation is now that we launch this undertaking, that we watch what happens here, and we are going to revisit this and see if the entire hospital, blood bank, United States health system, collapsed, under the duress of trying to meet this requirement.

If it does not, we could then extend that look-back to the other population we have been talking about, or we could say, go back to 1906, and continue on. This has been a wonderful demonstration of your efficacy -- in other words, I have to say in all honesty as I look at this, some people have said, can't we pilot, can't we check? Couldn't we see?

Well, I think the American people will probably want to know that, if we knew that some of them were exposed to infected blood, that we did what we could to tell them. I personally am going to find it hard to say, I knew you were exposed, even in a probablistic way, and I did not tell you. That makes me uncomfortable. But, I am much more comfortable if I say, well, I have a two-pronged strategy. I am going to make a big push on the public health side, and I am going to take this on as my trace-back, and we will see what we get.

I mean, we could say, this is our attempt to sort of reach a balance, and if it is getting us where we need to be in terms of letting folks know that hepatitis-C is something they have to worry about, okay. If it turns out to be something that is implementable before the year 2,000 and carries forward, okay. But it may still leave us a little bit of room to come back and then add to or follow-up on it, depending upon the cost and the precision and the real impact of those public health outreach efforts.

If they do not work; if no one spends any money on the PSAs, if there is no budget put forward, despite what we have to say about doctor education that really dents(?) in there, we could be back here saying, well we have to do more tracing. The other is a failure.

In other words, I am hopeful that some of what we are hoping is going to happen will happen, but if it did not, we could come back and say, well, we went halfway down a road and we are ready to finish the trip now, because we did not get where we were supposed to be. So, we may ponder that as something to bring into this discussion.

PARTICIPANT: If you raise the issue about who is going to allocate money for these various efforts, who is going to pay for the look-back effort? There is a huge amount of expense there. I would have to calculate it out, but it is in excess of several hundred million dollars.

Are hospitals going to be expected to bear this? Do blood centers have to just eat what it cost them? Obviously, if you are tracing back a quarter of a million donations, that is a lot of person hours, a lot of time and effort, not to mention just stamps and stationery. Is this something that the hospitals just have to swallow?

MR. CAPLAN: Mike(?), you know, I will tell you what my answer is to that, too, is the cost issues have come up sometimes about access or getting testing or who is carrying the load for either the public health outreach campaign that we call upon the agencies to implement and coordinate, or these kinds of trace-backs.

It seems to me that what we have to do is decide what we think ought to happen. At that point, we will let the agencies, the White House and Congress, the lobbyists, whoever, battle it out. And I mean that in all sincerity. We will not do it as soon as we leave this room with a set of recommendations, phones will go off and people will start to say, well, what can we do to avoid getting this cost stuck on our plate? And I wish all of them well, but my attitude is, we should decide what we want done, and then the chips will fall.

I think we should send a clear message about that hepatitis-C testing, that that -- to put that up there and say, and you know, we hope it happens, is not enough. That we have to send that out as a major public health challenge that ought to be addressed by this country, and we think that that testing ought to be available.

I am perfectly willing to say that that ought to happen, and that we urge the White House, the agencies, Congress, to make it happen, but in all honesty, we will not sort out the, who is going to pay for this, and we should not. It is not even our -- yes, Trish?

DR. O'CONNOR: I had a question. Were you suggesting with this, I guess sort of a huge pilot, that we just do A and B and [simultaneous discussion] --

MR. CAPLAN: No, no, the whole thing, but watch it. That we are monitoring and evaluating to see what is really happening here, so that we know, if we have put our eggs in the basket of the pre-1990 look-back, and no look-back is happening, that is going to change my thinking about the duty to inform.

I mean, if it just does not happen, I am going to be very nervous that people are not getting the information they want, because no one made a sincere effort that we are going to go back and urge tracing to eternity. I mean, I would, not we would, I would. John?

PARTICIPANT: We have already heard from the British and the Canadians, that it did not sink their ship, despite some of the critical nature of some of their budgeting problems, so it does not seem to me that it is that overwhelming in comparison, and the fact that they have already accomplished this on their own, and that we are kind of just following along, seems to me that if we do not do a pretty good job and a thorough job, it is going to look very sad that we could not keep up with the other members of the world.

PARTICIPANT: Could I clarify what we mean by test, when we talk about test recipients? Does this mean that, for those recipients that do not have private healthcare, that there will be some mechanism that they will be able to get a test performed? Is that what we are proposing?

MR. CAPLAN: We are not proposing the mechanism, but we are saying there is a -- there should be an effort made to make sure that that testing can be done.

PARTICIPANT: What we are basically saying is, that the private healthcare, and/or some other mechanism, should allow every patient who is identified, or who was part of this group that was contaminated, post the people we are looking back at, should be able to get a test. That is what we are saying.

PARTICIPANT: Okay. I agree. I just wanted to clarify it.

MR. CAPLAN: I think that is in a word, aggressive.

PARTICIPANT: One question I have had throughout the day is whether -- we have heard the CDC's sort of plans for this national hepatitis testing or information strategy, and one option that is attractive would be to fold this recipient piece into that. It is an existing mechanism, presumably funded, presumably there is testing on the heels of that.

The caution or concern about that is that the recipient piece of that should be small, because it is a small piece of the total HCV risk. So, the other option would be to argue that we need a separate recipient-specific public health campaign that is perhaps piggybacked or on the heels of that, but I think that is an issue that perhaps PHS can deal with, but I think our group would recommend that it be a focused recipient piece, and not simply a small piece in the bundle of HCV planned already.

MR. CAPLAN: Well, let me see if I can be parliamentarian for a second here. Are people interested in voting on that at this point, or would you like to have a little more discussion?

PARTICIPANT: [Comment off microphone.] -- very serious consequences and (inaudible) pretty much part of (inaudible). I just wonder if you couldn't in some way try to connect this part of (inaudible) campaign (inaudible) so that when we take the story about the hepatitis-C virus (inaudible) right to conclusion and we have recognized across all sectors of the communities, this is now another virus (inaudible), because otherwise, we effectively will have had a national (inaudible) hepatitis-C (inaudible) campaign, my kids are protected against it.

It just occurred to me and I am a little bit concerned that maybe we need to add one other(?) dimension(?) to this(?).

MR. CAPLAN: Probably not bad for preamble language, too, in terms of saying, again, we know there has been a lot of attention to this, but this is a special area. Hepatitis-C, here we go. Again, once we say that, we can certainly expect that the agencies, as they try to implement the public health dimension of this, are going to wrestle with that very problem.

That is going to be there for them as a major challenge. Because I do think a lot of people hear, hepatitis, hepatitis, and assuming they have had one negative hepatitis test for some hepatitis virus, will assume they are done with all of this, and we are going to have to explain that in clear language. Not us, I mean, as this gets implemented.

PARTICIPANT: But I think it is incumbent on us to inform them (inaudible) statement(?) (inaudible).

MR. CAPLAN: I am seeing some, put it to a vote sentiment here by expressions. Alright. What you are being asked to do is approve maybe with some minor wordsmithing, as things go along, or dissent from, this proposal, that we seek out aggressively to know the test recipients transfused prior to March 1992, through PHS physician and public education campaigns.

We may actually -- you can make a note of this, Steve -- put in something like that the financial means be made available to allow people to get that testing. We can add that so that we are clear that that is what we were talking about. Or that, whatever. The financial obstacles be eliminated or reduced, to the extent they exist.

Implement the targeted look-back program triggered by donors detected as HCV-confirmed positive by second gen screening and supplemental testing. And -- I will not read the rest of it, but with these qualifications, and I think what I heard was -- although I am open to this -- the vote will be actually on drawing the line on 1987, unless someone wants to pipe up. And the justification for drawing it at 1987 is partly due to where other record look-backs of this type have been, but acknowledging that it is somewhat arbitrary.

Alright. In favor. Opposed. Abstain. Is that an opposed or abstain, I jumped fast.

PARTICIPANT: Abstain.

MR. CAPLAN: Abstain. Alright. Let's do it again by count. I think I am actually going to record this vote. All in favor. Opposed. Abstain. One, alright. What did you get?

PARTICIPANT: Eleven(?), zero and one.

[Whereupon, the motion proposed by the Chairman was passed unanimously by a vote of hands.]

MR. CAPLAN: I am in a state of shock here, since I had not expected to get that thing off the page. We will go back and fold that into the language that Ron gave us, too, Steve, if you can remember this, on this recommendation. We will try and wordsmith that to make them come together. And again, on recommendation four, which is what we are still talking about, we will make this part of something that we will send out to you to just make sure you knew exactly what the language was.

If we could, then, let's go back to that first sheet, recommendations one, two, three, and also, Steve, keep handy a blank one, because I have a feeling we may hear a couple of other recommendations.

Let's revisit for a second, recommendation three, which we did clearly establish before the lunch break that no one knew what it meant anymore. It is a valued recommendation, it is an important one. We do not know what it means, and I guess I want to put the question, are we getting rid of recommendation three completely? Is there something there that we were trying to say that we have not said very well? Mike?

DR. BUSCH: Well, my thought was that perhaps this could be a preamble point to what other recommendations we just voted on, stating that we have reviewed the range of strategies available for identifying individuals at risk and propose -- in collaboration with all experts in Public Health Service, and based on deliberations, propose the following, and then that leads into the recommendations.

MR. CAPLAN: Yes, Keith?

PARTICIPANT: I would like to include testing of people who for some reason -- and there will be a lot -- who do not come through the look-back. That we recommend they be tested -- (inaudible).

MR. CAPLAN: That is up in -- that is in the first part of that next statement. Yes, it just fell over to the --

PARTICIPANT: Okay, sorry.

MR. CAPLAN: Alright, we can use three, then, as a little bit of boilerplate to set up what will now become recommendation three, as this is now numbered. Do you want to say -- let me just ask you again to revisit, briefly recommendation one. Implementing a program to educate providers of medical care in the appropriate assessment and counseling of individuals at risk.

We talked a little bit before about that this should proceed, I believe, or that this has to happen very, very quickly. If we are going to be having people coming in for a testing and look-back and so forth, and asking questions -- I mean, it is going to happen so quickly.

It is going to happen tonight when these news things get on the air about people calling their doctor but, do we need to say anything more about the speed? Do we want to see this plan presented to us at a future meeting? Would that be appropriate? And we have not said anything here about evaluation or audit of the activity. Well, in fact, I will propose that we add that this educational program be monitored and assessed in a timely fashion.

PARTICIPANT: With regard to our education program, one of the things that I see in the patients is that they may have dual problems. They have been transfused, and they may be remote IV drug abusers.

I think it is very important as part of this, when we are reaching transfused patients, to also advise them, there are other high risk behaviors, that even though they may not have been infected by transfusion, they could have been infected by some other lifestyle that they have indulged in in the last 20 years. So, I think it is important to get that in there, as well, because more of these patients, I think, will have been infected by their lifestyle, like IV drug abuse, remotely, than would probably have gotten it from transfusion.

PARTICIPANT: There is also the group -- we might want to say something about, you may not be completely -- you may not know exactly if you were transfused or not. If you were in any of the following high risk groups for transfusion, please consult with your physician about possible testing. Like, prematurity, etcetera. I mean, I don't know if you have an exhaustive list, but --

MR. CAPLAN: Do you think it necessary that we do this on a -- do you want to get this specific in the recommendation, or do we want to see what they bring us back, in terms of how they decided to implement here?

PARTICIPANT: Well, I don't know. Maybe singling out prematurity as special, or a low birth rate, infants as a special issue is not appropriate, but my personal experience is that, there are a lot of babies, particularly who were born peri- or pre-HIV, who got (inaudible) transfusion, may not have been aware of. They may know they were born premature, because obviously, their parents --

MR. CAPLAN: Well, how about this? We -- Steve, one recommendation that we did not put up there yet, which we had talked about is, the obligation -- the need to inform people that they may have received a transfusion, and that they may want to talk to their doctor about finding out how they could establish whether they did or not.

PARTICIPANT: This sounds good. You know, what is currently number one here, could perhaps just be expanded. I think it clearly should be the first statement coming after our statement of facts, that we recommend that there be implemented a broad program to educate about appropriate assessment, counselling, and then perhaps a second sentence, this program should include specific information about the risks of transfusion over time for HCV, and subgroups of patients who are at risk of having been transfused.

Then, what we are doing with the next piece that was on the other page, is we are going on to the specific additional focused public campaign that should target recipients, above and beyond this general implementation planned already, I think.

MR. CAPLAN: I don't know if you followed that, but, Steve, that was a -- emendation. The issue of alerting people to the possibility that they may have undergone a transfusion and the need to counsel about that. And the need to make sure they understand what other behaviors might put them at risk, if I was capturing your point, too. Those are -- we will add those in under --

PARTICIPANT: The other point is, what is currently number two here, I could argue is actually subsumed under what we already have in that other piece, but I also think it could and maybe should stand alone, as a third bullet under that other piece of our recommendations. It would come after the piece about, this committee has considered strategies for identifying -- and(?) we recommend. This would be one of perhaps three pieces there.

MR. CAPLAN: Yes, I understand what you are saying. All that would mean is just moving that particular strategy for individuals at risk under, to go along with the others. We can do that, too, that is fine. It will be so simple. We will have one giant recommendation in a second, just with many subpoints.

Are there other recommendations -- oh, no. Actually, before I do that. Are people comfortable -- I am going to ask for a vote -- on recommendation, what is now up there as number one, about the implementation of the program, the counseling -- oh, and the assessment. I forgot. Assessment, a timely assessment of this education program should be on there, too.

Here, we are asking you for a little bit of a vote on some vague language, but -- I mean, we have notes up there, but the spirit of it is, to get this education program going forthwith, but the agencies work to establish this. That it be both there to assess and counsel. Let people know about ways in which they could have been put at risk in the past; that is, transfusions they did not know about. Ways in which they could have been put at risk right now or in the past by lifestyle, and then to assess this effort to make sure that it really is doing what it is supposed to be doing.

In favor? Opposed? Alright, I actually got a unanimous one there.

[Whereupon, the motion proposed by the Chairman was passed unanimously by a vote of hands.]

MR. CAPLAN: Having tested our penmanship beyond belief, why don't we put a blank one up there and see if there is anybody who wants to add any other proposed recommendations? Maybe I shouldn't have forced a blank one up there --

PARTICIPANT: Art, it occurs to me that maybe -- and I am sure that probably the agencies, especially the CDC, will ultimately be coming out with some recommendations, but the first one that gets out to the public is the one that is going to probably raise a lot of questions.

One of the questions that will invariably come up, that we are going to need to be prepared to field is, how do you deal with the circumstances that affect those individuals that are very close to someone that fits into the risk categories?

The newborn infant of a mother who is HCV-positive and, you know, fits into the category, well, we have done the look-back, find her to be HCV-infected. Can she breastfeed? So, I mean, there has to -- it seems to me to be that one other section that at least opens up the opportunity to build in those other considerations.

MR. CAPLAN: Yes --

PARTICIPANT: Not included under assessment and (inaudible).

MR. CAPLAN: There is a tendency here to want to make sure that we -- well, let me put it this way. We will not necessarily have to spell out some of these points in the recommendation, but what we will do is send the transcript along with the recommendations, so that the kind of specific points that were originally thought about or reflected about, will be on the record.

If we try to do each one of these, we will wind up -- the recommendations will get lost, but we have taped and will transcribe everything, so when we pass the recommendations along, a lot of this is going to be captured in there. I got asked -- I keep reporting to you, my big empirical studies --

The other thing I did not tell you was, the three cases that I found where the doctors did not know what to counsel, one of the things that was of concern to the people was, could they maintain sexual relationships with their partners, I mean, knowing that they were -- so, and I think that was a question that the doctors were not prepared to answer for them.

There is breastfeeding and might I have a problem if I go to the health club and all these other kinds of questions are going to come flying out. They have to be addressed, that is exactly the sort of thing that people are going to ask, and that is what I am a little nervous that our physicians are not yet in a position to give good advice about right now. Dana?

MR. KUHN: I do not know if this has anything to do with the hepatitis, but I have heard a lot in these discussions that we were having problems with the record-keeping. I do not know, maybe this is going to be something in the future that we would want to look at, to include, or encourage blood establishments, hospitals, however, through whatever mechanisms, to keep longer -- records for a longer period of time, considering that other epidemics or endemics are possible. And if that would even help future look-backs, or notifications.

MR. CAPLAN: Comments?

PARTICIPANT: [Comment off microphone.]

MR. CAPLAN: I am just giving Steve a minute to catch up there. On the issue of, do we want to say something about records? There is certainly no reason why this body cannot make recommendations based on hepatitis-C about record-keeping, or the need to set up a system that is as good as it can be in terms of finding out about infectious diseases.

I do not think we have to link it to hepatitis-C, we could. We had a little discussion, you will recall, about CJD at the last meeting, and there are other bacterial things and so on, where people have raised issues about record-keeping and information.

The only reason I am hesitating to put it in here is, I guess I see us today as just sort of trying to address, to the extent we can, the hep-C issues, but I do not mind at all coming back at a future meeting, to the issues of records and information.

PARTICIPANT: I agree with that. I think we should come back, because I think there are a number of issues that we have not even touched on that need to be.

MR. CAPLAN: Any other potential recommendations about hepatitis-C, specific? One of the things that we did say was that, we would be concerned about making sure that obstacles to testing, to counseling, were minimized, and that is in our recommendation in terms of aggressive efforts.

Do we want to say something that stands out as a recommendation on that? One of the things that will jog your politicians is if you make a special point of recognizing that there are obstacles out there in the current healthcare system, both to testing, to counseling, to tracing, to many things.

I mean, my personal inclination is to try and bring that out, that every effort be made to minimize obstacles, and everyone will stand up and salute that and say, yes, every effort should be made to minimize obstacles. The question is, do we want to put it in there so we can hold people to account, to find out what they did? Maybe we can have a few congressmen in for testimony. Keith?

PARTICIPANT: Do we want to say that this is -- rather than lose everything by saying it belongs to everybody, actually say, it does belong to everybody, but there has to be a concerted effort by the government, the private sector, the whole healthcare delivery system to [comment off microphone] stating that everybody and therefore nobody is going to take responsibility. But at least it singles out that it is not likely to occur until every one of those components is in agreement, and is cooperative.

MR. CAPLAN: Yes.

PARTICIPANT: We have now identified these people. What are we going to do with them? The NIH consensus makes some statements around offering therapy, and I think it is unreasonable to identify people, if we are not going to do something about them and determine whether they are appropriate for therapy. And this, then, gets into our healthcare system again.

MR. CAPLAN: Certainly, an appropriate therapy up to that list of, minimize obstacles to testing, counseling, and appropriate therapy. And even, I might add -- you do not have to say this, but I can -- even to access to clinical trials, which is, I know from my own healthcare center, getting to be an increasing pain in the neck about finding out who is going to pay for the clinical trial type thing, that is getting to be a big issue.

One way to do that is to say, we want to minimize obstacles to testing, counseling, access to clinical trials, and appropriate treatment, and that all sectors must work together to make this happen, or have an obligation to work toward this end, private and public. Does that kind of language seem to -- I mean, I am keenly alert to the idea that I am mouthing extreme platitudes here, so I do not mind dropping it if it does not seem to go anywhere. Good after dinner speech rhetoric, but maybe not worth it -- paper.

PARTICIPANT: We are talking about public health issues. Public health issue puts it in the public sector. If we are serious about it, why don't we say that the public sector needs to step forward and make sure this gets done?

If we are not going to be explicit about it, given the anti-tax and all that stuff going on, it is going to be shoved back into the private sector, and without throwing stones, the private sector has to protect its own self(?) -- and there are going to be holes. So, if we are serious about it, let's say it is public sector responsibility.

PARTICIPANT: I think it may be the public sector responsibility to perhaps do the counseling and (inaudible) the testing, but I think when you get into the treatment, I think that has to certainly be a shared responsibility, I think especially today, when it is very blurred out on the playing field, in terms of what is public and what is private, and the way that the systems of care have emerged. So, I think the suggestion that you offered, Art, seems to be a very reasonable one, to build into it, the opportunity for therapy and clinical trials, as well.

MR. CAPLAN: Paul, can you just scribble that up there under three, up there? The access to the clinical trials and therapy (inaudible) -- it is just in penmanship. And actually, you know what, Steve, that is actually just a continuation of three, not even a new four there. Access to clinical trials and appropriate treatment. You can put that appropriate treatment thing in there, too.

What I am going to do is try and cast some language that says the public and private sector must work together. I do not know what to do with Paul's suggestion about it, if it is a public health responsibility, move it on the testing and counselling (inaudible), are we prepared to endorse that?

PARTICIPANT: I think that is impractical. You are not going to get a public health mechanism to do all this. Most of it is going to fall on public -- on private or general public insurance, etcetera.

PARTICIPANT: This is essentially the cost of doing business.

PARTICIPANT: The other thing is, if this is going to be -- the identified infected people are going to be out of this broad catchment, and only a small proportion will actually be transfused recipient infections. Some of them may have mixed risk factors, including recipients. Others, you know, basically, we will not be able to prove or disprove whether who is found infected was transfused or not. For the same reason as we cannot do effective look-backs.

MR. CAPLAN: Let me tell you one reason to keep this statement in here, and then I will get off my hobby horse about this. If I was trying to get coverage out of my HMO for a test. Or if I was trying to get coverage for an innovative form of therapy that might come down the road, I might, if I was very lucky, know somebody who knew that this group had said something like this, and use it. I mean, I might bring it forward.

What happens from groups like ours is, they have the function to some extent of a consensus panel. If we say, well, this is very important, and this is a kind of public health challenge, it is a standard of care. You can sometimes wedge resources out of the private sector, invoking that kind of thing. I have done that myself on occasion.

It may have minimal power. I am not sure we could probably command the public health budget, I think you are right. It is just not going to happen. It probably is overhead to somebody's business costs, but if you are sitting there saying, are they going to put hep-C testing into my HMO plan, that kind of statement sometimes works, and say, well, this group said that they should, so, you ought to. That happens.

Alright. Those who want to include -- let's do this backwards. Recommendation three? In favor of that, that we vote to include language that will minimize these obstacles to these things? Opposed. That is a unanimous.

[Whereupon, the motion proposed by the Chairman was passed unanimously by a vote of hands.]

MR. CAPLAN: Should we move votes on one or two? We want to keep the language in at the end, as knowledge of hep-C accumulates, we want to revisit, does that seem reasonable to include? Favor? Okay. Opposed? That is a unanimous.

[Whereupon, the motion proposed by the Chairman was passed unanimously by a vote of hands.]

MR. CAPLAN: I cannot remember what the first one was.

PARTICIPANT: Medical records (inaudible).

MR. CAPLAN: Oh, the medical records one. That one we agreed we were going to hold over and revisit. Okay, we do not have to vote that. I am getting the feeling that we may have reached the end of the road of today's meeting. We have not reached the end of the road of penmanship and calligraphy, but what Steve and I and Paul will do is try to reconstruct the language that we have been doing. This is tough on overheads that can be wiped out. We will have to put them in amber or something, as soon as we leave the room.

We will send these out to you. For those in the audience who are interested, we will keep copies available through Paul's office, too, if you want to see what the final language looks like. This is not to edit in any major way, we are just getting it back to you so that you can make sure that the commas and the periods and the textual coherence is what you thought it should be, and at that point, what I will do after that is, I will be back in touch with you after Steve and Paul and I and maybe Eric, have some further discussions about what we might do next.

I could be cruel and open that discussion up now, but I am not going to, because I think we have come pretty far today, that is a little mean to ask you to think about other things, having gotten this far.

There are other topics we can revisit. I would love to hear your ideas about what some of those might be, whether it is plasma pool size. Availability of the blood supply. Medical record-keeping. I do not much -- let me know. E-mail me, fax me, phone me. Steve can be reached through Eric's numbers and fax machine. You can let him know if you have any ideas. That is going to become the basis of our agenda.

We have not been asked formally to comment on the CJD issue. It is there. We did not finish it, we just heard about it. We could add in some of your infectious agents. As somebody told me, I think it was Keith, this is a committee that is supposed to look at availability, as well as safety, so availability issues are out there. Whether you think we ought to look at something like paid donation or whether blood costs too much. That was bandied about today a little bit.

We actually heard a speech which I never thought I would hear in my life by someone in the blood system, calling for a higher price. Charge me more. And all those are fair game for us to move to.

We have no scheduled next meeting yet, but I will probably see some dates and presume that there are things that this group wants to talk about. So, I am not going to wait until I get the topic, you are going to hear something about next dates. Probably, some time in December, January, which is when I think we will reassemble. And we can fill in the agenda, if you will do that.

So, watch to make sure that you check the text as it comes to you, by mail or fax. If you do not get it, within, let's say, the next three weeks, call Steve and say, I did not get what you said you were going to send. Just call him or contact him. And send us your ideas about what you would like to turn to next, and we will be getting back to you with some dates for the next meeting, okay?

MR. CAPLAN: Art, I would like to get a sense of the committee on recommendation number two, in the draft submitted by Mike Busch. It is clear enough what is recommended if there is a positive, supplemental test. However, that leaves unstated and therefore unclear, what is to be done if a supplemental test was not performed. And there will be a significant percentage of generation II EIA-screened positive, or repeat reactive, as we call it, for which a screening test was not performed.

Is it the sense of the committee that the directed look-back should be recommended, if supplemental testing was not performed, or should it only be recommended if supplemental testing was performed and was found positive?

MR. CAPLAN: I have a comment about this, but I will open it up.

PARTICIPANT: I agree with Jay. There was a period of time between March of 1992 and early 1993, before FDA had licensed the original second gen REBA(?) test, which was an appropriate, complementary, supplemental test to the second gen screen.

Most blood banks gained access to second generation REBA(?) at that period, ongoing, for online confirmatory testing through a reference lab service that FDA allowed to transpire. But there may be some modest number of screen-reactive donations that the blood center did not go through the hoops to get those things tested.

Now, to my mind, I do not think we can let those off the hook. Those do need to trigger a look-back. You either have the confirmatory data, or if you do not have it, you get it now, if you have the sample, or you have to notify, and that is part of FDA's language.

The same problem exists today, unfortunately, because the third gen test was licensed now about a year ago, and has been in place and regularly screening donors without a third gen REBA(?) test to prove, and again, Red Cross and a number of other blood centers have developed mechanisms to gain access to that third gen, pre-licensure test, and have those data coming on online, but other blood centers have not done that. So there is again, now, a period of time when there are people who are getting screen-reactive results, without confirmation. And I think it points to another issue that is dear to my heart, which is the need for these companies and FDA to bring up supplementals in concert with the screened, because otherwise we are out of linkage and we have all these problems.

I think, there again, if you do not have the supplemental, I think you have to proceed with a look-back.

PARTICIPANT: That is what I would have said.

PARTICIPANT: Yes, I (inaudible). So that is the spirit there, looking back, if you do not have it.

MR. CAPLAN: Yes. Mary.

PARTICIPANT: I guess I also wanted to hear(?) the agencies, I guess, wanting the final clarification. CDC, as you know from Marian and health presentations, has really had the lead within the PHS right now for the educational campaign. And I just wanted to make sure that it was clear to me, the messages about education of providers and the public, who include people at risk, among those, transfusion recipients.

As I understand now, the structure of this document, there is going to be something as a preamble that kind of endorses the need for, and speedy implementation of, a campaign, an education campaign, for providers and the public about modes of acquisition, diagnosis, treatment, prevention.

Then, within the specific recommendations, which was up there on the transparency, number one. Aggressive campaign to notify and test recipients transfused prior to 1992. And what is not clear to me is, is that the same? Is it different?

MR. CAPLAN: That campaign -- the campaign discussed up there is to tell people that they ought to go out and get tested.

PARTICIPANT: Right. And I mean, if you review -- and (inaudible) passed out, I mean, admittedly, it is the bones, it is the structure of, a working document, something in progress. But, the messages that were originally crafted were, transfusion before 1990, and this committee is recommending that that message be changed to 1992.

Okay, fine, but you know, for each of these bodies of -- for each of these target audiences, be it the public or providers, you know, these are the messages, that recipients are surely among those who are considered at risk for infection.

What I am trying to distill out of this is, this is the plan, the bones of the plan as you have it now. Is this not, as best you can surmise, adequate to accomplish one? Because -- it is confusing to me, because it is mentioned in two places. It is in the preamble. We endorse, there is this education campaign, dah, dah, dah, dah. And that is going to happen, you know, as Hal and Marian had spoken to earlier.

MR. CAPLAN: The way I would answer it is this, and other committee members can pipe in if they want. I see that one as a specific, get tested campaign, aimed mainly at the public, with doctors being told that that is something they should do.

I see the -- what you are calling the preamble and set of recommendations there as, supplying doctors with the information they need to answer questions, when people come in and get results and say, can I have a baby? Can I breastfeed? Can I have sexual relations? Should I ever donate blood again? Can I -- am I safe forever from hepatitis if I tested negative today? What about my lifestyle, blah, blah, blah. How can I get in a clinical trial if I am infected? Would you take interferon? That is the list. I will send you my 20 failed questions of the three physicians.

PARTICIPANT: No, we concur. We concur that there is -- many providers are not informed to the level, anywhere near we would hope for at this point. So there clearly is a need to get information out to providers about all these things.

MR. CAPLAN: That is the preamble part, it is that (inaudible).

PARTICIPANT: Right. And because bear in mind, the PHS campaign is directed towards providers, as well as to the public-at-large, and the messages do, as I understand it, include that if you were a transfusion recipient before, now, 1992, yes, you are among the people then that need to be tested.

PARTICIPANT: That is the concern I raised earlier, that if the only public health campaign is what is currently planned, and transfusion recipients are being obviously targeted in this program, but it is a very small piece of a much bigger risk pie, and what I envisioned was, that this would transpire, but that this would be supplemented by an additional notification campaign, targeted at the recipient population. Practically, that is probably saying a whole lot more money.

It is somehow enhancing the recipient piece to what you already have planned, which may be more practical, but --

PARTICIPANT: Again, I do not -- my understanding is that the messages were going to be directed to an audience, listing, these are the risks for HCV infection. And even though we know that transfusion is a small part of that pie, I do not think the message is going to be minimized in any way. But I really cannot speak to that in as much detail as the hepatitis (inaudible).

PARTICIPANT: You know, perhaps it is not the patients that need the extra message related to transfusion recipients, maybe it is the hospitals and the physicians. It is the point you raised about the physicians need to be educated to ask about transfusion histories. Perhaps hospitals need to put messages up around, were you transfused --

MR. CAPLAN: Signs.

PARTICIPANT: Enhancing(?) it through that method, rather than necessarily changing the message to the public.

MR. CAPLAN: And Mary, I will talk to you if you want, subsequent to the meeting, about those -- I have a pretty good idea, just based upon the discussion of concrete things that people had -- whether it is the PSA that says, were you transfused, or making sure that the doctors do it.

Remember, we also said yesterday, putting it into medical training so that when you are taught to take a history, you begin to ask about this. So, I captured those pretty well and I do not mind -- I will bring those back to you. John?

DR. PENNER: A question to Mike. If I am (inaudible), do you think that generation I reduced the numbers of positive generation II tests?

DR. BUSCH: We are back to this. I think the generation I test detected probably four-fifths of the infected donors that the generation II detected -- would have detected.

DR. PENNER: And therefore, excluded them from --

DR. BUSCH: That is correct.

DR. PENNER: From the donor pool?

DR. BUSCH: That is correct.

DR. PENNER: So that made the generation II 1992 a much more manageable group.

DR. BUSCH: That is true.

DR. PENNER: Had generation II been available in 1990, had been implemented as the first set, then all of those individuals, minus say, 50%, would have to be followed up, as well as --

DR. BUSCH: I think -- if we had had an excellent test like generation II with the supplemental, with the knowledge we had today, and we were introducing that test, I think I would recommend that we do look-backs routinely to prior donations, period, across the board. It is the context in all those issues, though, the supplemental being online, which did not exist for the duration of 1.0 screening. The knowledge we have today. The time that has elapsed, ten years, since that test.

DR. PENNER: And essentially, the difference would be about 50% would have been excluded because they were false positive -- [simultaneous discussion].

DR. BUSCH: I think that is (inaudible), I think that is one of multiple reasons why that first gen look-back is very problematic.

DR. PENNER: As the only one that has abstained, I would just like to have my statement that I agree with what we have done so far. I do not think it has gone far enough. But I think our recommendations are inconsistent and not really ethically justified in excluding generation I donor look-backs. And hopefully, that might be revisited.

MR. CAPLAN: And what -- I think I went on record as saying, I think we have to revisit, and I think we owe it to the people who received blood, and for whom these tests were done, to come back to this and see how we are doing. So, at least I will try to get this back up on the agenda, after some reasonable amount of time has passed, so we can see what has been done and what is going on. Because I did not vote on that one. I was trying not to influence this group, and it made no difference to anybody, I lean in the direction that John is articulating.

I do think that we have to go back there and take a look, but if this does the job, or if it turns out to be manageable in certain ways, it will give us certain evidence, so I am willing to go with where we -- as far as we have gone.

Okay. Thank you.

[Whereupon, at 3:00 p.m., the meeting was concluded.]