Department of Health and Human Services Meeting of the NATIONAL HUMAN RESEARCH PROTECTIONS ADVISORY COMMITTEE (NHRPAC) Tuesday, October 30, 2001 The DoubleTree Hotel 1750 Rockville Pike Rockville, Maryland 20852 A G E N D A Tuesday, October 30, 2001 8:30-9:00 Welcome--Overview of Meeting Mary Faith Marshall, Ph.D. Chairperson, NHRPAC 9:00-9:30 Lesson's Learned from ACHRE; The Evolving System of Human Research Protections Ruth Faden, Ph.D. Executive Director The Phoebe R. Berman Bioethics Institute and the Philip Franklin Wagley Professor of Biomedical Ethics John Hopkins University 9:30-12:00 Update and Discussion: Social Science Felice Levine, Ph.D. Co-chair Working Group, Executive Officer American Sociological Association Jeffrey Cohen, Ph.D. Co-chair, Working Group Director, Education, OHRP/DHHS 1:00-3:00 Update and Discussion: Third Parties: Research Subjects? Mary Kay Pelias, J.D., Ph.D. Chair, Working Group Professor of Genetics, Louisiana State University, HSC 3:15-5:00 Research Ethics Occasioned by the September 11, 2001 Tragedy; Children & Research Alan Fleischman, M.D. Chair, Children's Workgroup Senior Vice President, New York Academy of Medicine 5:00-5:15 Wrap-Up Mary Faith Marshall ROSTER OF MEMBERS MARY FAITH MARSHALL, Ph.D., Chairperson, Director of Program in Bioethics, University of Kansas Medical Center GREG KOSKI, Executive Secretary, Ph.D., M.D., Director, Office of Human Research Protections, Office of Public Health and Science, OS MARK BARNES, J.D., LL.M., Partner, Proskauer Rose LLP MS. MARGARET BORWHAT, President, Women's Cancer Advocacy Network SANFORD CHODOSH, M.D. ELLIOT N. DORFF, Ph.D., Rector, Distinguished Professor of Philosophy, University of Judaism ALAN R. FLEISCHMAN, M.D., Senior Vice President, The New York Academy of Medicine JENNIE R. JOE, Ph.D., M.Ph.H., R.N. Professor, Family and Community Medicine Director, Native American Research and Training Center, University of Arizona SUSAN Z. KORNETSKY, M.P.H., C.I.P., Director, Clinical Research Compliance, Department of Clinical Investigation FELICE J. LEVINE, Ph.D., Executive Officer, American Sociological Association ROBERT LEVINE, M.D., Professor of Medicine, Yale University School of Medicine ABBEY S. MEYERS, President, National Organization for Rare Disorders JONATHAN D. MORENO, Ph.D., Emily Davie and Joseph S. Kornfeld Professor of Biomedical Ethics, Director, Center for Biomedical Ethics, University of Virginia Health System Roster of Members (Continued) MARY KAY PELIAS, Ph.D., J.D., Professor, Department of Genetics, Louisiana State University Health Sciences Center ROBERT R. RICH, M.D., Executive Associate Dean of Research, Emory University School of Medicine ADIL E. SHAMOO, Ph.D., Professor, Department of Biochemistry and Molecular Biology, University of Maryland School of Medicine JUDITH L. SIEGEL, Ph.D., Vice President, Head U.S. Clinical Operations, Hoffman-La Roche, Inc. DENYSE THORNLEY-BROWN, M.D., Assistant Professor, Division of Nephrology, University of Alabama at Birmingham KATE-LOUISE GOTTFRIED, J.D., M.S.P.H., Executive Director, National Human Research Protections Advisory Committee EX-OFFICIO MEMBERS Dr. James Shelton, USAID Dr. Maryann Danello, CPSC Dr. Joseph Spence, USDA Ms. Linda Beth Schilling, USDOC Mr. Tryn Stimart, Alternate Ms. Blanca Rosa Rodriguez, DOED Ms. Patty Boll, DOD P R O C E E D I N G S [Time noted: 8:40 a.m.] WELCOME--OVERVIEW OF MEETING CHAIRPERSON MARSHALL: -- made their way into policy or into guidance. And I hope that this committee can look back at the most important work that they did and put some of those things -- there is Ruth as we speak -- put some of those things into place. It was most wonderful work. Then this morning we will be spending a fair segment of our time on discussion of social and behavioral sciences, and Felice Levine and Jeff Cohen will lead that discussion. We will then segue into an update on third parties as research subjects and Mary Kay Pelias will lead that discussion. And both of these groups have been doing just a phenomenal amount of work in the last couple of months, and I thank you and commend you for everything that you have done. I am not sure that there's a social scientist in the country who hasn't been involved in one way, shape or form in this work. So I am delighted I want to tell you. Really, it is difficult for me to say how much I admire Ruth. If we had a queen in the bioethics world then she would be it and I say long live the queen. DR. FADEN: Thank you. What a way to start. CHAIRPERSON MARSHALL: Ruth is the Executive Director of the Phoebe R. Berman Bioethics Institute at Johns Hopkins University and is the Philip Franklin Wagley Professor of Biomedical Ethics. She has written a book on informed consent and -- DR. FADEN: I had a little help. CHAIRPERSON MARSHALL: She had a little help. Yes, she did have help along the way. She wrote half the book on informed consent. And as I was saying before you got here, Ruth, we are going to revisit ACHRE and the work that that commission did is so important, and I said that in many ways it was very prescient. It is certainly apropos today. It will be apropos in this afternoon's discussion of the response to September 11th. As the chair of that presidentially appointed advisory committee, Ruth had what I think is the genius to demand as a condition of accepting that appointment that the commission not only look back at the wrongs that were done by the Federal Government and others in areas of human radiation experiments but that they also take a look at the present; at what informed consent to human subjects research means; what is understood by participants in human subjects research. And that work really is the foundation of the work that we are about today. She provided us with a numerator that goes over the denominator. And we have said -- I think we have lamented too many times in the course of our work that we do not have the data that we should have in order to make public policy and to do good work. And the foundation of the data that we have, a good deal of it, was provided by the scholarship of those involved in the ACHRE group. And they also did look forward in terms of their recommendations, and I would hope that, as a committee, that many of the things that have not been implemented in terms of public policy guidance can be implemented because stellar and superb work was done by that group. So, Ruth, we're delighted to have you. Thank you so much for being here. LESSON'S LEARNED FROM ACHRE: THE EVOLVING SYSTEM OF HUMAN RESEARCH PROTECTIONS DR. FADEN: Now I am going to blather and stumble. It is going to be so awkward. [Laughter.] DR. FADEN: We have a little mutual admiration society going, Mary Faith and I, and so you'll have to forgive us. Thank you, Mary Faith. Thank you. I also want to acknowledge, and everybody I am sure know this, Jonathan Moreno was my colleague at ACHRE. And you know what is funny, we never called it ACHRE then. We called it the advisory committee. It became ACHRE after the fact and, I guess, in history it is ACHRE and I find myself using it, too. Although, did we ever use it at the time? I don't think it was ever an acronym that was in business at the time that we were actually functioning. It also strikes me that it is in a way truly history now. It is almost six years since our report came out, which is hard for me to integrate. I am delighted that Mary Faith thinks that our recommendations and our work still resonate. I certainly believe that it does. I have to say that Kate, in talking with my office, said that in 15 minutes I should address an overall view of human subjects research, the lessons learned from ACHRE, and where human subjects research is going or ought to go, including a discussion of the Hopkins situation. [Laughter.] DR. FADEN: Which I thought, you know, was a good charge. I am happy to talk about all or any of those things. With respect to the Advisory Committee on Human Radiation Experiments, it would take me much more than the 15 minutes, and there's to be discussion as I understand it, to tell you all about its history and how we operated and what we did. I know that you're living in the midst of a commission so there might be other times or other opportunities, or certainly from Jonathan directly, you can both learn from our mistakes along the way as a commission and from where we did exercise some good judgment. And I think on balance the commission was extremely successful. I thought, though, that now is not the time to go over what we did and how we did it. I just want to highlight a couple of the key recommendations of the Advisory Committee on Human Radiation Experiments that I think still resonate. Some I think resonate all the more given the current situation. Then maybe supplement with a few thoughts that I have generally about the ethics of human subject research. Greg has probably heard me say a couple of these things before. I have certainly heard Greg on these issues. And then if people want to talk about the Hopkins cases, I am happy to do so in general. That is to say I have some very specific thoughts about what those cases at Hopkins should cause us to think about and I am happy to talk about that as well. One of the key recommendations of the Advisory Committee on Human Radiation Experiments that I think is critically important today and really has not been fulfilled is the one that Mary Faith has already made reference to. We recommended that there needs to be in place periodic, and I'll use the term "outcomes research" of the system of institutional review boards. Al and I are working together on a group that is thinking about exactly that question but the message here is simple. What we really believe is critical, we really believed is critical, I still believe is critical, is that we need information on how well the system is doing what it is supposed to do, which is protect the rights and interest of human subjects, and that means looking to outcomes. That means looking to data collected on a periodic basis with the full force of law, and I'll explain what I mean by that in a minute. Basically I'll explain it now. It means that institutions are compelled to participate. That is to say we don't have to go through the process of pleading our way into the system but rather that it is an obligation of a general assurance arrangement, for example, that you will cooperate with regular monitoring but not really monitoring in the sense of auditing of the sort that is going on now but real outcomes based research as we think about it, for example, in medical outcome studies. We have to figure out what the endpoints are and then we have to see whether the endpoints, in fact, are being achieved or not. This is a very simple idea. It is just incredibly complicated and almost impossible to implement. And, in fact, if you look at the history of research involving human subjects, the only time in which there has been a successful attempt to gather systematic data about what is actually happening, okay, rather than audits of after the fact alarming situations or paper audits of whether particular aspects of the regulations are being complied with in a technical sense has been when a commission, a national commission has done the work. The President's Commission collected data. Our commission collected data from a national sort of context and arena but so far it has taken the sort of authority and stature and status of a national commission to be able to pull it off. To get in there to do it and to pull it off. As an aside, the importance of compelling this. When we started to do our work, initially our contemporary -- the advisory committee started to do its contemporary research projects, and Mary Faith is right, it took some wrangling to persuade anybody that we had any reason -- there was one clause in our executive order that created the advisory committee. I was a presidential commission. And that was the line that said, "And ensure that it never happens again." Everything else was retrospective and then there was this line that said, and included in the charge to the executive committee from President Clinton was, "And ensure that it will never happen again." We took those four words and said, "Okay, we can't assure that it will never happen again unless we look at the current state," and that became the wedge into our authority to look at the current status of research involving human subjects. We had some considerable difficulty persuading universities initially that it was appropriate for them to allow us to have access to their records and to some extent to their patients. We didn't have subpoena power. We only had a complete authority to demand records from the Federal Government. And so research records that were in the hands of the Department of Defense or HHS or NIH we could compel and we received. Although initially the response was, "Wait a minute, you're only supposed to be looking at the past so we don't have to give you any of today's records." And we said, "Yes, you do. It says all records. It doesn't say records before 1974. It says all and any records." And so we were able to obtain anything that was in the hands of the Federal Government. Private universities or state universities, for that matter, however, were outside our legal scope of authority. We did not have subpoena power. We couldn't compel the universities to participate. We ultimately got participation by -- and this is not a secret -- by basically saying to the universities, "We are going to at our public meetings report the universities who are cooperating with the study and report the universities that are not cooperating with the study." And it became very clear that no university wanted to be in the list of universities that were not cooperating with the national commission's attempt to investigate the state of ethics of research involving human subjects. It took us a long time to figure out that we had that leverage, however, and for what it is worth, it would have been a whole lot better if there would have been a very plain structure that said, "You have got to do this." Now this issue of outcomes-based, evidence-based structures for looking at whether we are adequately protecting human subjects, that can -- the importance of that, I think, cannot be overstated. The fact that historically we've only been able to collect such data systematically when we've had the force of the national commission is very lamentable but also there's a lesson there in terms of the way things should be done. And this ties into another very important recommendation that the Advisory Committee on Human Radiation Experiments made, and that was that there needed to be a standing body that would be able, in effect, to continuously interpret and specify and expand the regulatory guidance provided to institutional review boards and to investigators. And, moreover, that standing body presumably would be the location for the ongoing outcomes based continuous evaluation of how well we are doing in protecting human subjects. NBAC, the National Bioethics Advisory Commission, was created the day that President Clinton obtained -- we turned in our report and, in his remarks in which he accepted our report, he announced that he was creating NBAC in response to that recommendation but it is important to note that NBAC was not a standing body and now we have NHRPAC, which is a standing body or not depending on what happens. And the point is what we don't have is any assurance that there's going to be a continuity of both public examination of ongoing unresolved and new issues in the ethics of research involving human subjects and also ongoing oversight. So there's sort of two functions that we envisioned for this standing body. One was a vehicle for continuing to reinterpret, to specify, to expand on the ethics of research involving human subjects as new issues, as new cases, or as new dissatisfactions with old standards and practices emerge. Some place where there can be public discussion of these issues and guidance provided and at the same time some place where there can be oversight that continues across administrations, across personalities, and which is focused on obtaining regular periodic outcomes based data. So those were two of our most important recommendations that, I think, have not changed at all. We also had some considerable recommendations with respect to research in the military. This may resonate even more given the current situation. Some of the recommendations there is -- there is a report by the Clinton Administration that you all may have seen called Building Public Trust, which is that administration's account of the extent to which they accepted our recommendations and how they acted on them. It is kind of unusual that there is such a thing and it is nice to have, and in there it is detailed the extent to which they accepted our recommendations with respect to the military. We also had recommendations with respect to the conduct of secret human experimentation in the United States. Our recommendations there were only, in part, accepted. Again something that may resonate very importantly now in the current situation. We made two pleas that every commission has made in the history of national commissions looking at human subjects research. Maybe not every but at least recent ones. One was extend the regulatory structure to the private sector. Not news. Okay. And the other was come up with some of a compensation scheme for research injuries. Not news. So those two recommendations were made by the commissions before us. You may make the same recommendations or not. The real question is not whether those are good recommendations. The question is whether there will ever be a time in which either one of those things will happen. Okay. We had some very specific recommendations that came from our examination of then currently conducted and currently approved research involving human subjects that were more particular kinds of recommendations, and I am just going to wrap up with those. They turned, for example, on recognizing that we have to confront a continuing tendency on the part of investigators to overstate the benefits of participation, particularly in clinical research, and kind of understate the harms and risks and discomforts of research; that there has to be some way, and I'll talk about this a little bit more in a minute, there has to be some way to help institutional review boards learn how to allocate their time better and differentiate between research projects that are pretty much un-problematic and research projects that are morally problematic and need more time and attention and consultation. And, also, a very poignant plea, and I'll stop there and go to some observations about the current state of affairs that extends our work, that when you look at history, people tended to think initially that the reason why things were not good in the '50s, '40s, '50s and '60s, into the '70s, is that there were no codes and there were no regulations and there was no guidance, and that is why these unethical and bad things happened. Well, in fact, what we learned was there were codes and there were regulations, and there were guidances, and they were largely ineffectual. And while I have, you know, the highest regard for the need for adequate regulations and for regulations that are responsive and useful, the obvious truth of human nature is that you can't regulate ethical conduct. You can't regulate people into behaving properly. And so we spent a lot of time talking about the culture of the '40s, '50s and '60s, into the '70s, that gave way to the kinds of attitudes and practices and values that caused people to behave the way they did or sort of reinforced people for behaving the way they did and sent a kind of message out. And we still think or at least I now speak for me, I still think that these cultural issues are hugely important. If we don't shift how people think about the importance of ethics in the conduct of research involving human subjects, we can regulate from today till tomorrow and it will make relatively little difference. And the line that I like to use is we want to make ethics as central to science as the science itself; that is to say when people are doing their work they should be as concerned with how good the ethics are as they are concerned with how good the science is. That is a hard thing to kind of get your arms around. It is not as easy as saying we have to change the consent forms or we have to change the composition of the IRBs or maybe we should regionalize IRBs or get them out of their local institutions. A lot of those kinds of ideas that are percolating and being discussed but I think it is really at the core. Okay. Two or three observations if I might or do you just want to go to questions. CHAIRPERSON MARSHALL: Go ahead. DR. FADEN: On the current state of affairs. And here, if we want, we can relate that to some of the experience of what is going on at -- or has gone on at Hopkins but it seems to me that we have a couple of things that we really, in particular, need to focus on. One is the sort of status, and Tom will talk about this at some point tomorrow, I am not sure when exactly, Tom Beauchamp, of informed consent in research involving human subjects. And here we've got a kind of double edged sword as far as I am concerned. On the one hand, increasingly, I think, and rightly, some of us for a while have been saying we have to recommend -- recognize the moral limits of informed consent, that there's only so much moral work that informed consent can do for us with respect to protecting the rights and issues of human subjects, and this is quite correct. If you go to the specific case of the death of Ellen Roach at Hopkins, there is no question that the consent form was unacceptable. There is a lot of discussion, however, that it wouldn't have made any difference, that if the consent form had been straight forward and appropriate, it is very likely that the people who participated in that experiment would still have participated. And I am willing to grant that counter-factual. It is a counter-factual. The response I get then from medical students, in particular, would be, "Well, then why are we bothering if it doesn't matter and people would still participate no matter what you say. Then what is the point of the consent form?" And the message is more complicated and complicated messages are more difficult to persuade people of their correctness. The message is it may be in certain instances that what is disclosed to research subjects won't make a difference. People are going to agree or not agree for a complex of reasons that have almost nothing to do with what is disclosed to them in at least some instances. That may well be the case but that doesn't mean we don't have a moral obligation to be honest, okay, in the presentation of what is at stake in participating in research. What it does mean is that we can't put too much of the burden of assuring that a research project is ethical on what is disclosed to the research subject, that there's an awesome responsibility that comes before you can even make the proposal to a research subject that the proposal itself is ethical. You can only put, is the argument, ethically acceptable offers in front of people and then you go on and then you proceed. Another big issue that I think we need to focus on is characterization of risk. In the advisory committee experience, we kept seeing over and over and over again disclosures in consent forms that seemed to the committee members to understate the risks and discomforts. This was probably the single most disturbing finding in our review of contemporary clinically oriented research projects. The problems in the system that we have currently have to do with the extent to which the review mechanisms rely on the characterization of the risk by the investigator. Usually it is the investigator that provides not only to the potential subject but also to the IRB the definitive sort of description of just what the risks are and the discomforts of participating in the research. And if you're lucky, there's somebody on the IRB who might have relevant expertise to perhaps raise a question but this is not something -- the possibility that there might be somebody in the room who knows enough about this area of research and clinical practice if it is a clinical protocol to be able to challenge the characterization is -- you know, it is a hit or miss thing. The odds are there probably isn't anybody in the room at the time and this is, I think, at the core of a central problem in terms of providing adequate protection for human subjects. Some institutions have gone to a system whereby, for example, there is departmental review in an oncology center or an HIV/AIDS group will vet their research projects before they go forward to an IRB around these issues where presumably there's a depth of expertise on this question of how you manage melanoma or how you manage a particular complication of HIV/AIDS and, therefore, whether this description of what is at issue with respect to risks and benefits and alternatives is adequate. Some people find comfort in this. I think it is a good mechanism but, of course, you have the situation of its all being colleagues and sometimes, of course, there is only one expert at an institution. Depending on how deep the institution is or how rarified the research problem is, there may only be one person who understands or knows, in fact, at that institution about that particular line of research and there may be 10 or 15 or 20 people in the world but there's only one, you know, at this one institution. So that is to me a very central issue that I hope that you will take on, which is how best to help the system figure out this problem of adequately and accurately characterizing the risk. This relates to the third thing, which I already introduced, and that is this issue of somehow triaging research projects into those that are problematic or at least requiring a greater level of oversight and those that are not. A characterization of risk is obviously a critical part of a process that says let's red flag or tag certain research projects for special handling but there probably are many others that need to be identified and a way in which we can structure those so that we know which projects we should be spending the most time on. There are whole sets of issues that I don't have time to go into now that have to do with considerations of roughly characterized justice. This is all part of making sure that the proposal is fair and ethically acceptable before it ever goes to a potential research participant before the issues of whether the consent is adequate even come up. The consent process is appropriate or adequate. Maybe I should just stop there and say how delighted I am that there's a body that is continuing to pursue questions of this sort under Mary Faith's leadership. CHAIRPERSON MARSHALL: Thank you, Ruth. Generally just explain to you our procedure is that we give the committee a certain amount of time -- we're equitable here in terms of access to our speakers. We will give the committee some time, the ex officio members some time, and members of the public the same amount of time. So, Alan, I see your hand up. Adil, okay, and then Abbey. DR. FLEISCHMAN: Ruth, thank you very much. As always, brilliant. A quick question. DR. FADEN: Yes. DR. FLEISCHMAN: Each of the groups who have come to chat to NHRPAC coming from major commissions of various kinds, we've asked to tell us the numbers of staff and the numbers of dollars, who helped to make the work possible and so important. DR. FADEN: And you want to know how much money we had? Well, let me tell you, just so you will feel really deprived, at one point we had a staff of over 60. Now it has to be remembered that the Advisory Committee on Human Radiation Experiments did have a huge historical discovery charge, that is to say we were asked essentially to scrutinize the past in the way in which the historians on our staff thought of it as the world's largest history project and the lawyers on our staff thought of as the world's biggest discovery process. So we had two different disciplinary perspectives both trying to uncover -- am I getting this right roughly, Jonathan? So we had this huge number of the staff were actually allocated to this extraordinary historical discovery process. Most of the time there were about 40 staff. Dollars, the truth of the matter is I don't know and the answer to that seems bizarre but the fact was we were not given a budget. We were given a job and what we did got paid for by the Executive Branch. It was a rather remarkable moment. A couple of times -- I think there have been efforts to characterize or try to calculate the budget and different figures have come in. They're all grossly inadequate because a large part of the work that was done in terms of discovery was actually done by small armies of people who were working, and I see people in the audience who have been part of -- who are part of that history at the various agencies. So there was a small army of people at HHS and at DOD and certainly at Energy, at NASA, at the Veterans Administration. I mean, it was a rather remarkable thing. So the staff numbers -- I am giving you the staff that were on ACHRE's -- under ACHRE's aegis. And there were all these people that were in the federal agencies that were working on this. Some at very high levels and it is impossible to calculate but Greg's characterization is right. We were given a job and we were told to do what we needed to do. We were given a time period. We were a sunset commission. We were originally given 12 months and it was extended to 18 months. And I think that actually for our mission was very advantageous. The fact that we knew we were going out of business and that nobody was going to be able to make a career out of this meant that we were politically really feisty, you know. What could they do? Shut us down? We were going to go out of business in 12 months anyway. I mean, what is the worst that could happen? Right? It was a very freeing sort of thing and so when I say that we need a standing body, I say it with some anxiety because standing bodies are much more vulnerable to political whims and considerations. You know, are we going to exist next year? Are they going to shut us down if we are too controversial and maybe raise too many issues? Or if the administration changes or if there's a crisis in the world that makes us irrelevant, and on and on and on. But there are models for standing bodies that are pretty politically insulated and we can look at those. So you're welcome. DR. MORENO: Can I just add, as Ruth indicated, the comparison is a little misleading because of the discovery process, which took about six or eight months, and we just cut it off at a certain point. DR. FADEN: Right. DR. MORENO: Because we had started getting into ethics commission mode as Ruth called it. And presumably we will never be in that kind of position and other presidential commissions have never been. I heard Jeff Cohen report, who is the staff director, that the GAO estimated about $25 million worth of staff time, not only the advisory committee directly but also all of the folks in the agencies who were amazing. Including, for example, the Pentagon's Radiation Experiments Command Center, which I always thought was pretty cute, and they did an incredible job because -- not to bore you with the archival details but we couldn't go through the millions of pages ourselves. The technique was they told us roughly where they thought the paper was. They would do their best to find it and we would -- then we would ask them at briefings how they -- what their methods were. And that turned out to be pretty successful. We got a lot of stuff. DR. FADEN: I am not confident of that figure but it gives you an order of magnitude and we had 18 months to do it and I think it could have been higher. It could have been lower. It could have been higher. It could have been higher or it could have been lower. It is just hard to say. CHAIRPERSON MARSHALL: Adil? DR. SHAMOO: Thank you, Mary Faith. I am sure Mary Faith would take $10 million. It would be okay with her. There are those of us who have recommended for a long time the enactment of National Human Research Protection Act and actually did make some comment in a paragraph about human research protection should be encompassing of all research. NBAC came very clearly and recommended NOHRO, that is National Office of Human Research Office or something, and to be outside HHS. The example I can come up with is the Nuclear Regulatory Agency and the Securities and Exchange Commission, which are truly now trusted as independent agencies even though they are governmental agencies. They have a great deal of independence and there is a great deal of public confidence in their activities. And especially the Nuclear Regulatory Agency, issues of how you regulate radiation in universities. And those of us who are old enough to remember there was a lot of resentment, there are none now. It is a routine function of the university to make sure that we comply with radiation issues. I just wanted to see whether we -- you come on that subject. DR. FADEN: Actually I sort of incline to the Securities and Exchange Commission's model a little bit more but, yes, I think that my thinking goes in that same direction. CHAIRPERSON MARSHALL: Abbey? MS. MEYERS: First, in light of what is happening today, we've been requiring our soldiers to take an anthrax vaccine that is still experimental. It has never been approved by the FDA. And they're not given informed consent. They're told they have to take it and if they say no, they have to face some kind of consequences. I understand some people have been discharged from the military for not taking that anthrax vaccine. So the first question is I'd like your comments on that. The second one is your first recommendation that the regulations should be extended to the private sector, which is a recommendation of many, many commissions, nothing has been done about it. What are your thoughts on that? DR. FADEN: Abbey, everybody has recommended it. I don't think anybody has a coherent argument why there shouldn't be an extension to privates, a coherent argument that I find persuasive. They may be coherent, just not persuasive. A persuasive argument why it should not extend to private sector activities. It is more complicated obviously, and you know this better than I do by the sort of, you know, political and private sector interests that can sometimes make it difficult to get what some people would perceive as progress but it is -- I don't know. I don't know. I haven't given up and I think that part of the reason, too, if we go that route we not only need to extend the language to private sector context but also we have to be more thoughtful about what ought to be covered, that is to say the definitions of what constitutes research need to be thought through in such a way as to catch some of the more troubling arenas that are under regulated or outside of regulations. Like advances in new reproductive technologies, for example, which would fall out if we didn't change certain things more than just moving to the private sector. On the anthrax situation, I am not a sort of expert on what was going on. The anthrax issues were sort of co-contemporaneous with the Advisory Committee work and we didn't look at it officially but there was another group that did so I am not now speaking specifically about what was going on in the first round of movement with respect to the debate about whether anthrax was experimental and, therefore, that it was research involving human subjects. There is a provision in the military, which we did investigate at great length, which has to do with compelling medical intervention for the health and safety of the troops. And this is an understanding that says military personnel cannot say no if the purpose of the medical intervention is essentially just that, to protect the health and safety of the troops in a context where it could undermine military preparedness to have people not treated medically. I don't want to comment on the past. I certainly think that frankly, Abbey, this is not an issue to worry about now. The issue to worry about now is the fact that we don't have enough of the anthrax vaccine. I would be thrilled from working with my colleagues in our center for Civilian Biodefense if we had enough anthrax vaccine for every person in the United States. We don't have it. I am perfectly happy to take that vaccine. I would take it tomorrow if it were made available to me. Frankly, right now, it is not as -- and again I am not an anthrax -- there probably are anthrax experts in the world -- I mean, in the room and certainly there are many in the world but I think that right now, in particular, for people who have been exposed to anthrax the idea of not -- of backing up anthrax antibiotic treatment with an anthrax vaccine makes incredibly good clinical sense, speaking as a non-physician but talking again to colleagues who know what they're talking about. And certainly from a public health point of view, I wouldn't worry right now about anthrax being experimental -- anthrax vaccine being experimental. I would worry about having enough of it. CHAIRPERSON MARSHALL: Greg has a question. DR. KOSKI: Just a couple of comments on point there. Actually the anthrax vaccine, although I am not sure what the formal FDA approval status is, is probably the most studied vaccine that we've ever had. It has been extensively studied for a number of years and I think the -- I would venture to say that the days of surreptitiously doing dastardly experiments on the military today as guinea pigs for the sake of doing dastardly experiment on the military are gone. I don't believe that those days are with us any longer and so I think, indeed, the effort here, particularly with respect to anthrax vaccine, is clearly for the benefit of not only the military but perhaps others as well but I don't want to dwell on that issue. MS. MEYERS: Let me just say one thing. It is still experimental technically so technically it should fall under all the regulations for experimental drugs, which means informed consent. And, if not, then we need some regulation that gives everybody a safety valve there. DR. FLEISCHMAN: Again, Abbey, I am not an expert on this and I can't -- I am looking at Jonathan to remember but there are provisions for military personnel which takes things out. Also, moreover, there's a public heath emergency kick out clause that could well apply with respect to anthrax right now. Right now. So I didn't mean to sound so impassioned a minute ago but I am working on, you know, right now worrying about how to allocate anthrax and smallpox vaccine, and my only concern is getting more of the stuff and not worrying about getting -- and I don't think the worry is going to be people, you know, not wanting to be vaccinated for anthrax and not being compelled to have it because anthrax is not an infectious disease. So it is not a situation where -- I have to be vaccinated so you'll be protected. If I am going to be vaccinated, it is to protect me, I think, right now. DR. KOSKI: To just finish -- DR. FLEISCHMAN: But I am sorry. That was off a tangent of the working group. DR. KOSKI: What I really wanted to comment, though, was on the issue of private research. I think it is important to note that already the vast majority of all privately funded supported research is already subject to regulation under the Federal Government's existing structure through the FDA, EPA and many others, as well as on the federally supported side so that most of that research is already there. There is a certain part of the research, though, that is privately conducted and supported that is not subject to the regulatory agencies and that is an area of some concern. The bills that are currently pending in Congress, both in the House and the Senate, or to be introduced, we've had -- in the light of recent events, there has been some delay there but both of those bills would bring all human research under the federal oversight mechanism and the mechanism for oversight would be transferred to an agency -- although it would not be a new agency, okay, it would actually be a group set up with sufficient autonomy to achieve functionally that goal so that there is, I think, progress that is being made in that direction. And certainly even under the existing oversight mechanisms we're working very aggressively towards creating a level playing field that extends across -- and a uniform system that extends across the entire spectrum of currently regulated research. CHAIRPERSON MARSHALL: Just procedurally, Ruth has given us very graciously ten more minutes so I have Jonathan on point and I have Bob, Mark, Felice and Elliot, and then I am going to open questions to the floor. DR. MORENO: Briefly, Abbey, on the anthrax business, Ruth is quite right that the Uniform Code of Military Justice does set a different framework for people in uniform and Reserves. However, having said that, the FDA was made a promise by the Pentagon that people who were asked to take anthrax during Desert Storm and Desert Shield would at least get an information sheet about it and that apparently was not the case. Now there have been congressional hearings on that and the FDA recognizes that that did not happen. The Pentagon recognizes that did not happen. There has been a whole argument about that. I am not going to go into that. Secondly, the concern, I think, about the anthrax vaccine that was used 11-12 years ago -- 10 years ago by the veterans, including those who believed that they had contracted Gulf War illness or Gulf War syndrome was about the quality of the vaccine. We all know about the bio-port problems and I gather they are being worked out very aggressively now. Thirdly, my understanding is, and this goes to Greg's point, that the primate model for anthrax vaccine is thought to be an adequate one so that to the extent that there is a human research component to widespread distribution and use of an anthrax vaccine, it would be more like a Phase 4 study. That is to say we would want to follow people for their reactions once the stuff has been widely distributed. CHAIRPERSON MARSHALL: And this discussion probably we can carry on this afternoon as well so I am going to ask you please to make your questions concise so we can allocate time fairly. Bob? DR. RICH: I had three brief comments. The first is with regard to your call for a standing committee. I would -- I believe that you raise a question as to whether this was -- whether this was a standing committee? And I would say it is a standing committee but with a sunset clause and I think that that is appropriate for federal advisory committees that they ought not to be permanently authorized. There is a role definitely for a distinction between the federal employee based regulatory process and then an external advisory committee. I think that is what NHRPAC is. The idea that it needs to be looked at from time-to-time to find out whether it is doing is job I think is a good idea. So I think this is a standing committee. The second comment just as part of that, however, you said that the role of such a standing committee ought to be -- and I don't want to put words in your mouth but is to continuously interpret, analyze and expand regulations for human research protection. I would quibble with the issue of expand in the sense that I think that part of the role of this committee is not necessary to expand but to reassess and evaluate. We will find sometimes things that have been done in the past, in fact, are not working and are not effective, and that sometimes we need to retrench from things that we are demanding because they were not useful. So that the philosophical principle that this is a continuously expanding balloon, I think, is something that I would not want to embrace. And then, finally, in your introductory comments you asked for a mechanism for mandatory outcomes analysis and I would just caution about that with respect -- it almost sounds as though to the extent that outcomes analysis is, in fact, a kind of human subjects research, it sounds to me like a carve out for bioethicists to get around the issue of IRB review of what they do and analysis and careful scrutiny because, in fact, I believe that outcomes analysis of human subjects protections could intrude directly on human subject protections. And I think that the mandatory part of it, which means that you don't have to jump through all the same hoops that the rest of us do, could be problematic. DR. FADEN: Let me respond to all three. First of all, I think your first point, I think exactly what needs to be debated is whether what we need is FACA committee after FACA committee, federal advisory committees, act type committees, or we want a regulatory committee that is a regulatory agency or body. So that is a key question. We've had FACA committee after FACA committee. Those committees that are chartered as federal advisory committees and they're very important. I liked our's a lot. I think we did a lot of good stuff. I think all of the commission, NBAC has done some wonderful reports and had some important impact. And if you go back in time, certainly the national commission got the whole thing going in the 1970's. The question on the table and continues to be on the table is whether the model of FACA committee after FACA committee is the model we should have or whether we want a standing regulatory commission. So that is my response to your first comment. The second one, I did not mean and you did put words in my mouth or maybe I misspoke, that by expansion it is expansion of regulations. The first function that I see is we want to continue to specify, to examine, to make particular ethical analysis of issues and also, where appropriate, to determine that regulations need to be changed in either direction. What I see as a continual problem is that regulations, no matter how good they are, only provide limited guidance. Issues keep coming up. People struggle either because they don't understand what the regulations require or the regulations provide inconsistent guidance or is silent on the issue. Take the whole set of controversy around international research in the developing world, the continuing debate about what constitutes minimal risk in research with children, the continuing debate about what is appropriate to do with people who are incompetent or of questionable competence who are adults. We can go down a whole long list of grey areas, under examined areas. What are we going to do with the first stem cell human trials -- research therapy based on stem cell derived products? There's going to be a whole string of these, genetics and so on, and we need some place where these issues continuously get revised and expanded. The mandatory part, to go to your third comment, I meant for the institutions, not the individuals. And certainly not in any respect to suggest that there ought not to be IRB review or other appropriate review maybe at some federal level of the ethics of any outcomes research process that is undertaken. The compliance part I want directed at the people, the institutions that do research. I don't want a situation in which any private or public actor can say, "Sorry, you can't -- the University of Pennsylvania, Hopkins, Harvard, Stanford, Genentech, you name it, we're not participating so we're out of here. We're walking, we don't like the study." And I just picked those names because I went to all of -- went to some of those institutions and they came to mind but that is my concern. Not the concern at the level of human subjects. Obviously human subjects can always -- should always be free to refuse. DR. MORENO: Can I just add to that? The -- we're the only industrialized country that doesn't have something like a national advisory commission on bioethics. Those groups continue to meet together. They continue to set policy. They have a voice with the World Medical Association. We get frozen out. We show up in Durban with our objections to the draft of Helsinki and we look like bioethical imperialists. I mean, we have rediscovered coalitions in the last six or eight weeks and this is a coalition that I think we need to be part of so I heartily disagree that we don't need some kind of standing national advisory committee. I don't care if it is regulatory or not but I think we need to have a nonpolitical mechanism, as nonpolitical as it can be, to have a voice on these issues internationally. We operate in a global pharmaceutical marketplace and this is part of the ball game that we're operating in. I'd also say, as a bioethicist, who has done outcomes research, I have had to go to an IRB. It is a whole lot of fun. We did get expedited review. [Laughter.] CHAIRPERSON MARSHALL: I am going to ask that when you ask your questions, committee members, you ask one question. No smuggling in of other questions or I am going to cut you off. And I have Mark, Felice and Elliot, and then we will open up to the audience. So please be mindful of other's need for time. MR. BARNES: Ruth, hi. The -- just a quick question and it is perhaps a naive question but in terms of outcome research -- this is maybe to follow-up on what Bob was talking about. What would it look like? I mean, what would you do that would be different than simply a paper audit of the medical records and a paper audit of the informed consent process? And then in the course of outcome research, then, I mean, do you really think that every -- should it be a sampling methodology or should it be that every study has outcomes? DR. FADEN: No, it obviously has to be a sampling methodology. I mean, that -- and that -- and we -- this -- I don't really think this is full employment for bioethicist people. I think this might be full employment for epidemiologists and health services researchers. But really what we're talking about is -- what I am talking about would be a very thoughtful, I hope, enterprise in which first there would be a careful examination of what the appropriate endpoints should be. So let's take -- and then very creative development of methodologies to try to operationalize what the endpoints -- how we might measure whether the endpoints have been satisfied or not. So let's take the consent process and let's use that as an example. Instead of looking to see whether the consent forms have been signed and dated and logged in, okay, instead of even looking -- although this would be an improvement -- at whether the content in the consent form can be reconciled with the description in the original protocol and grant proposal, that would be moving up one, right. Instead of doing even that, that you begin a process either of observing consent solicitations in selected projects, interviewing people after they've been asked to participate in research and seeing the extent to -- what they think they have just agreed or refused to participate in. Going to people and getting a sense -- this is what we did in the advisory committee. We looked at research and medical records of patients to see whether they had ever participated in research and then we asked them had they ever participated in research to see the extent to which people had a memory. Now that was a flawed methodology in this respect: If people didn't remember, it didn't mean that they didn't understand what they were consenting to at the time but it certainly didn't leave a lingering trace, right. So we had to be careful. That was a flawed methodology. There would have been better methods that we could have employed to get closer to the endpoint of interest, which is whether people understood and felt like they could say no at the time. There are qualitative methodologies that can be used more like those that anthropologists employ. There are quantitative methods that can be employed. There's a lot of role for -- Mary Faith was talking about this need for really creative social science to marry with good regulatory oversight. CHAIRPERSON MARSHALL: Felice? DR. LEVINE: I think, Ruth, in your last answer you're moving close to my question but it wasn't specifically directed to the social and behavioral sciences. It seemed as if -- it is a two-part question. CHAIRPERSON MARSHALL: One part questions only. DR. LEVINE: One sentence. -- that we have a real need for capacity building and a range of forms of research in this area, and as you have observed in the context of each of the FACA commissions, committees over generations, there's always a call. I even think it is part of the IOM mandates to address the issue of future research. It is one of those areas that ends up being on the kind of cutting room R&D floor and so the question is from your range of experience do you have any counsel to us so that when we take up this issue it might have some sustained impact? DR. FADEN: One of the -- one of the, I think, documentable impacts of effects of the work of the Advisory Committee on Human Radiation Experiments was an increase in federal funding for empirical research on informed consent. A series of NIH solicitations which now have continued over time specifically originally referencing our committee report and so here's a little piece of success where now there's a small stream of income -- of resource research funding for largely social sciences but not exclusively social scientists who are investigating questions about informed consent in the context of research involving human subjects empirically. I obviously am a big advocate of work of that sort. I think it is an important part of the field of bioethics. I think it is obviously critical to this issue of understanding how well we're doing and also we're coming up with some creative problem solving. At the same time I think it is no substitute for philosophical and legal scholarship. They're kind of all complimentary ways of trying to move the ball forward. So if we get to, you know, continually perplexing questions, for example, about what constitutes minimal risk, there are empirical methodologies that should be employed to help illuminate that question but there's certainly good philosophical work that needs to be done there as well. CHAIRPERSON MARSHALL: Elliot? And let me ask if there are those in the audience who would like to ask questions, if you would just come on up to the microphone. Thank you. DR. DORFF: Any thoughts about how to create the ethical culture that you're talking about? DR. FADEN: Now that you ask, I have lots of thoughts and probably many more than Mary Faith wants me to talk about now. I think, for example, that an awful lot has to do with leadership in academic medicine and in academic social science, behavioral science, biomedical sciences. What we need to see are the leaders in those areas, and we're seeing this increasingly, making it a point to emphasize the importance of ethics in their work. This has to translate into real modeling opportunities so let me give you some specific examples. I have discussed with some of the leading foundations who give very prestigious awards for senior investigators and for young scientists that when they interview and vet potential candidates that they vet those candidates not only for how good their science is, how promising their work is but they vet them on whether they have thought about the ethics of their work and how well they understand the ethical issues that their line of science proposes. I have suggested that every doctoral student in the sciences and in psychology and in the social sciences be held accountable in their doctoral defenses for examination on questions of ethics. They ought to be able to answer pointed questions about the ethics of research or animal, whatever area they work in. And, also, that they defend the ethics of their research project in particular. I think that all appointments to universities in areas involving where the person's work will take them into research involving human subjects, and I would extend this to animal subjects and to lab research as well, should be vetting candidates for appointments along their competencies with respect to ethics as well as their competencies with respect to science, and I can go on and on and on. These are very practical things that can be introduced into the training and into the culture of the academic community. It can be extended, I think, into the private sector where I believe there's great willingness and interest in ensuring that the corporate culture is sensitive to these issues. They may use different language, language that maybe we're not as comfortable with but I think you can make the argument, as has been made in business ethics for years, that good ethics is good business. Good ethics is good science. It is just very elemental but I have a whole -- and I could go on, Elliot, of like lots of really practical specific things and don't think that a computer course that you take in order to get your whatever so you can be -- I mean, it serves a certain value but it doesn't go to cultural change. It goes to reminding people that there's yet one more hurdle. You know, I have got to plug in. I have got to sign in. I have got to finish this stuff and I have got to hang up. This is not cultural change. This is more the equivalent of auditing consent forms to make sure that they were signed and dated. CHAIRPERSON MARSHALL: Vera, you're looking especially lovely today. That is a wonderful hat. DR. HASSNER: Vera Hassner, Alliance for Human Research Protection, AHRP. Ruth, to paraphrase you, I think the crux what made the ACHRE report so credible and so important was that you took it as we're investigating in order that this may not happen again. And I think that part of the problem is that subsequent commissions, and I think that NBAC's report, for example, which has some very good recommendations and went beyond you, however the report itself does not provide that evidence-based backing for the recommendations. And when you talk about outcome studies, that that was so important to see what the -- that is actually a form of accountability and that is what everyone seems to be very leery of and yet I think we in the public want that more than anything. No one wants -- a taxpayer doesn't want 100 commissions standing or rotating that do not, in fact, bring about a change in the culture and in enforcement and in compliance. And I would suggest -- and, please, don't take it personally but Johns Hopkins, with the public revelations all around, not a single case -- we're talking about whether a literature search is even being done before human subjects are put at risk. We're talking about putting -- exposing children to lead paint poisoning. We're talking about doing research in India without even IRB approval. I think that the Johns Hopkins, overall case now, is a wonderful opportunity as a case examination so that this will not -- find out what went wrong, what things need to be fixed so that these kind of things will not happen not just in Johns Hopkins but elsewhere. I am very worried whether NHRPAC is even more removed from -- CHAIRPERSON MARSHALL: Concise, Vera. DR. HASSNER: -- doing just that. DR. FADEN: I think we're agreeing about the importance both of evidence and outcomes-based work both in terms of its role in research involving human subjects in terms of scholarship but more importantly with respect to protecting human subjects in a meaningful way. In terms of the opportunity at Hopkins, I also agree with you that this is -- as it has been I think at all institutions where there have been crises -- it is a wonderful opportunity and I am hopeful that the institution will seize it. In fact, there's every indication that we will. I want to make a sharp distinction, though. I disagree with the characterization on the lead paint study. That would be a whole other discussion for a whole other opportunity but I think that case is extremely problematic, extremely confusing, and not in any way a simple sort of slam dunk. In fact, in terms of -- that was, you know, a bad, ethically wrong research project. I don't think, in fact, that it was but we will have to wait and see as more people come and look at the lead paint study. With respect to the tragic situation in which Ellen Roach died, that was extremely awful and I am hoping that some of the comments I made earlier that some of the most poignant changes will come or are already coming in response to what we learned from what happened in that situation. DR. HASSNER: I am talking about an outsider coming -- CHAIRPERSON MARSHALL: I am sorry, Vera. There are two other people and Ruth is being very generous with her time. Thank you. Could you please tell us who you are for the transcript? DR. MAKHENE: My name is Makhene. I am a medical officer at the Food and Drug Administration and I am not here in any capacity representing the FDA. I am just here out of my own interest. CHAIRPERSON MARSHALL: Good for you. [Laughter.] DR. MAKHENE: My question actually has -- it is actually more I wanted to just get some comments from Dr. Faden and maybe other members of the committee. It goes specifically to vulnerable populations. DR. FADEN: Yes. DR. MAKHENE: And vulnerable populations and probably maybe outside of the way we think about them, and probably this will be addressed, you know, during the two day session but beyond children, beyond the disabled, but particularly vulnerable populations with respect to the economically disadvantaged. DR. FADEN: Yes, poverty. Right. DR. MAKHENE: And, also, in terms of informed consent and -- DR. FADEN: Right. DR. MAKHENE: -- those populations who maybe are not able to understand consents very well. And then, also, that intersection of someone who has been given a consent and at the same time is economically disadvantaged and how that sort of meets and -- DR. FADEN: Right. DR. MAKHENE: -- impacts on how they really view the study. And I say this in a capacity of, one, having been a co-investigator in clinical trials and having sort of watched the process and, two, being, you know, as I said at the FDA and reviewing clinical trials. So I'd like to just get some comments on that. DR. FADEN: This is one of my favorite subjects actually. Mine as well as a whole bunch of people. But I think that probably the biggest moral questions that we face in research involving human subjects today turn on questions of justice and not around questions of informed consent. And in that I mean that when we scratch the surface almost always -- I shouldn't say always -- often what is troubling us has to do with whether we are taking advantage of people, whether we are exploiting the unjust or unfair background conditions that potential research subjects come from and whether those background conditions are ones of, you know, economic poverty, of relatively political powerlessness, or profound illness, you know, the sort of situation where someone is powerfully and desperately ill and in a circumstance where they're likely to grasp at anything. And this is why I keep emphasizing that the real -- that a critical, essential, necessary but not sufficient condition of ethical research involving human subjects is a vetting, and we will leave open for the moment by whom and under what conditions, that the research proposal is ethically acceptable to put to someone. That is to say before you -- I just cannot emphasize this enough -- before you go and ask anybody whether they want to participate in research there should be such a level of confidence that that research project is ethical to go forward, that it is ethically acceptable to even ask somebody. And what that does is relieves some of the burden around concerns with respect to whether people really feel they can say no, whether they really understand what is being presented to them and so on. Again, when I say that, I don't want to suggest in any respect that we don't want to diminish -- that we want to diminish the importance we attach to respecting the individual research subject and his or her capacity to protect her own interests. We want to honor and respect that for lots of reasons but because there are so many people who are not positioned well to exercise that kind of self-protection. I don't mean that in a demeaning way because the world is unfair and people are not always situated in ways that put them in a good bargaining position. We have to either conclude that all those people are off the table for research, which may not be at all in the interests of those people, or we have to make sure that would not put before them a proposal that it would be unethical for them to accept. DR. MAKHENE: Well, and I have been there, as I said, having been a clinical investigator. CHAIRPERSON MARSHALL: I am sorry. Than you very much. This will be our last question. DR. MINER: Yes. My name is Caroline Minor and I am not officially representing anybody. I do work for the Bureau of Prisons in the Department of Justice. And I thought it was very interesting that in your discussion of outcome measures or possible outcome measures that two of the main ones you mentioned were interviews immediately after the consent process and follow-up interviews after the experiment was concluded. And we had actually in my bureau had discussed doing this as a regular operations type procedure but the reason we kind of hadn't been able to decide that that was a good thing is because in most of our research our primary concern is with a breach of confidentiality. And so we were very concerned that by implementing some type of follow-up like this that we're just increasing the risk to our subjects. DR. FADEN: You're right. DR. MINER: So how would you address that? DR. FADEN: It is a huge issue. Research on research, which is really what we're talking about, right, so we're talking about doing research about the process of research, and you end up with levels of consent. I have done this so I know it is very awkward. The person says, "Well, I just consented to that study." "No, no, you're consenting to a study to study their study." And people go, "Well, wait a minute, you know, how could -- wait a second, what am I -- what --" and so it is a very difficult thing ethically to sort of pull off. And not the least of which is the fact that in many studies and many particularly more problematic studies protecting the identity of the participants is the single biggest human subjects concern. So then how are you supposed to study the study? You know, how do you study people when what you really want to do is make sure that no one knows that they're in the research? I don't have a simple answer. I think this is where you get real contextual. You look at particular cases. I think, too, if you were trying to get a national pulse beat, you know, on what is going on to exclude studies, the whole class of studies in which confidentiality is the principle moral concern would be problematic. It may be, however, that with respect to those studies, the critical endpoint is was confidentiality adequately protected? So maybe in those studies we don't look at whether the consent form was as the person -- as the research subject -- as the investigator said it should be but rather whether the confidentiality procedures really were as they said it should be and was there any evidence of any breaches of confidentiality? And maybe that is where the outcome goes because that is the principle outcome of concern in that type of research. CHAIRPERSON MARSHALL: I am sorry. I think Ruth has given us an extra -- we don't want to tell her how long -- 25 minutes of her time. Ruth, thank you so much. DR. FADEN: Thank you very much and good luck with your work. I am delighted that you are in business. [Applause.] CHAIRPERSON MARSHALL: I gave Greg time to make a comment. DR. KOSKI: That is right. I reserved that time. Before Ruth leaves, I just wanted to also thank her very much for coming but I would want to remind everyone that the goal of moving towards a performance and prevention oriented evidence-based system for human research protections has certainly from the outset of the creation of OHRP been one of the department's primary goals. And, indeed, from the very beginning we had asked the Institute of Medicine as part of its ongoing work now to develop the objective measures for assessing the effectiveness of the system as part of their work so we are eagerly anticipating the second phase of their report. Accreditation was phase one and the outcomes initiative is phase two so we should get some valuable input there. Finally, I'd just say that the shining light of public interest into the human research process is something that serves us very well. The road for human subjects is just littered with one recommendation after another from federal advisory committees of one sort of things that have never been implemented. And, of course, one of the things that we're trying to do in OHRP is to implement those things but I think that there would be perhaps real value in having some of this process for assessment of outcomes and modifications to remain outside the government. The government is one of the most vested bodies in terms of both the conduct and support of research. And so I think that we have a track record of seeing that the government doesn't always, you know, hold itself to the highest standards here. So that having part of this process remain in an independently publicly based body, I think, is something that again would serve us very well rather than relying exclusively on the government. CHAIRPERSON MARSHALL: You're welcome, Greg. Thank you. I think what I am going to do now procedurally is give us a break so that we can have our next session not be interrupted and I notice people are getting up and making necessary runs at any rate. So why don't we take 15 minutes. We will resume at quarter past 10:00. [Brief recess taken at 10:00 a.m.] CHAIRPERSON MARSHALL: Susan Roe is here and I'd like for her to make an announcement about an upcoming meeting which looks wonderful. DR. ROE: Good, I need that kind of reassurance. For two years in a row we did -- not two years in a row, one site visit of one of our sites and then again four years later, and the community member that we interviewed did not know why she was there, had not -- couldn't figure out any project she had ever reviewed, and the site -- when we said don't -- you know, don't give her cement shoes and throw her overboard, just find some other community members. It is a Bay Area institution and they said they could not find any other community members. So rather than getting into a fist fight with them, I decided to try and figure out if I could find community members so the result is this meeting, which I am thrilled you think is going to be wonderful. I think NBAC has called for an increased membership. We truly need folks who (a) care and (b) who are savvy. So they have needs that we haven't had time or money to address. So one of the focuses of this meeting is going to be to listen to community members. We found them all over the country, all sorts of size institutions. We are going to do one session with IRB chairs saying their favorite things about community members and also the things they hate. The guys who never show up. We are going to kind of do the reverse with community members. What do I like about doing this? What is difficult? So the first day is going to talk about community members and the second day is going to be by community members. So I think it is going to be provocative and I am thrilled that you thought it was important enough to mention. The only problem, yesterday at the end of the day someone from NIST called me, the most imperious phone call I have ever had, saying how dare we use their auditorium without their approving of everything I am doing. So I think we still have to duke out the room. I am still aghast in the spirit of human subjects that I am going to stay here forever and not go back to work because I don't want to fight that fight but there's no registration cost. In the next couple of weeks I am going to have an agenda to send out and I would be delighted if folks would put that on their agenda. We have one of the groups who's represented at the table who is going to be a participant. The National Organization of Rare Diseases. We wanted to bring big membership groups into this so we could kind of talk to them about educating their folks to be members of ethics boards and community members. Thank you. CHAIRPERSON MARSHALL: Thank you, Susan. So this is a -- DR. ROE: I am sorry. I thought it was mailed. I don't -- no one is getting mail unfortunately. CHAIRPERSON MARSHALL: Yes. So the flier, which I have a copy of up here, this is a preliminary conference announcement from the Department of Energy, from Susan, "the Community IRB Member, Neighbor and Partner," and the date is April 8th and 9th of 2002. DR. ROE: And in case we don't mail for a while, it'll be on our web site and then when the final agenda -- just for an enticer, I just -- I am a ham so I get one more. DR. __________: Where is it? DR. ROE: Pardon? DR. __________: Where is it? DR. ROE: It is going to be in the parking lot at NIST unless things -- [Laughter.] DR. ROE: I got another idea for the concept of community -- and I don't know if you know this woman but she's going to be one of my keynote speakers, and she is wonderful -- Sister Carol Taylor from the Bioethics Program at Georgetown is going to talk about what is community. That even shocked her but she said, "Yes." CHAIRPERSON MARSHALL: Wonderful. Thank you, Susan, so much. It is a great idea. So let me turn the floor over to Felice Levine and Jeff Cohen, who have been working very hard with a large number of people, and all of those people have been working very hard. They're going to give us an update on their working group on social and behavioral science. Thank you. UPDATE AND DISCUSSION: SOCIAL SCIENCE DR. LEVINE: Thank you, Mary Faith, and we appreciate the opportunity of having prime time here to kick off the NHRPAC discussions of a number of issues that have been ongoing on the Advisory Committee's agenda for the last -- well, really since we began and that is, I think, important both pleasure and opportunity. One of the many user friendly parts of NHRPAC is not only having tents that work one way but they also work two ways so for those of us who can't remember who we are, I thought this was terrific. So at my right is Jeff Cohen, who is my co-chair, and what we want to do is give you a little bit of an overview of what has been going on for the working group to get everyone up to speed. Hopefully, the cover memoranda, though it arrived late, for which I apologize, gives you kind of a touch and feel of what we have been doing. And then we want to this morning focus on two issues. One, the public use data files where we felt in July our discussions were quite far along where we could, we hoped, move to the kinds of recommendations that could potentially lead NHRPAC to move forward. In the risk and harm area, we really had only the most -- not exactly even an abbreviated discussion so we hoped that before the noon and lunch break we could have much more discussion and input from the group about risk and harm so that our working group could continue its work. The third issue on our working group's agenda over the past six months relates to third parties and we will be contributing our input such as where we are at this point as part of the third party discussion this afternoon. In many respects, it almost seems like two years ago and in other respects yesterday since July 30-31, which is when we last met, because, of course, as we know, the time since early September has really been so eclipsed. I attended really an exceptional conference planned well before the events of September 11th on humanitarian emergencies and one of the features in both dealing with humanitarian emergencies and studying humanitarian emergencies is how fast everything travels and we were discussing how you could address human subjects issues in that context but in a way many working groups I know plan to either meet, have conference calls or engage in much more intensive interaction in the period what I would call post-Labor Day between then and the end of October. And while our working group had a scheduled meeting on the 20th, we managed actually September 20th to have both a place based meeting and a conference call that was really a substantial turnout of the full working group of 16 but it was -- instead of it being a meeting we had scheduled as a full day, it was a several hour meeting, and I think a productive one. And I know that other working groups have felt the similar fragility of having had trouble scheduling either meetings or even conference calls because time has been really so accelerated. It is an interesting subject for study. But over the period since we met, I think one of the hallmarks of what we have been trying to do, both NHRPAC really and the working group in similar spirit, is to maximize input and communication from almost every community at various levels of experience both with research and human subjects. So that as the cover memo states, we have really been involved -- Jeff and I, and really every member of the working group -- in a very extensive outreach process of attending meetings, posting. We have -- the working group has a web site portion of the American Sociological Association's home page just so that it is readily accessible. And we have been trying to maximize our capacity for input and in a way, picking up on where the conversation ended with Ruth's presentation this morning, that I think one of the real pluses of what, I think, NHRPAC, in fact, has already achieved, and I know our working group wants to similarly contribute to that is that in ratcheting the profile and the openness with which NHRPAC wants to reexamine the human subjects protection system and make sure it is working as effectively as possible for the promotion of the best ethical practices and the best science that conversations are occurring in lots of places where they haven't occurred before. Greg was saying at the break, groups that haven't before had ethics codes or guidelines within their own scientific communities are beginning to do that. Chairs of departments are convening and discussing these kinds of issues more actively. And many of us who went through the process of either developing ethics codes or in the case of the American Sociological Association, revising our's in 1997, very much at the heart of that activity was not just to revise the ethics code but really to focus more on the educational aspects of having that kind of guidance and how that really needs to be part of the doing of quality research. So I think in terms of -- if we were studying our own outcomes and impacts, I think we are already beginning to see some of those serendipitous effects. One of the highlights for me was actually Greg coming to the annual meeting of the American Sociological Association and not just doing a wonderful town meeting that was very well attended, that was by attendees at the meeting, and he did a terrific job in the context of a Ted Koppel type format. But really also it was kind of a two-for because he came to the chair conference which starts the day before our annual meeting and then runs through day one. And the chairs had all of our information in advance and he really did a very good job in having a very good and open dialogue with 80 or so chairs about how chairs need to be thinking about their roles in relationship both in leading and being scientific leaders in this particular field, and it is applicable certainly to any area. There is nothing unique to sociology about that other than Greg's presence. But as a consequence of that, we almost had another event similarly for the deans of schools of education that was scheduled, I think, for September 24th or 25th and obviously was called off, and the historians are seeking. So it gives everybody more work. And I must say when Ruth talked about a staff of 40 to 60, and what was that, $25 million, I thought 10 percent of that would look pretty good for all of us, let alone -- so we know that not only is this heavy on the volunteer persons but in a way, you know, the recommendations of NHRPAC are going to sort of trickle down to our quality colleagues like Jeff and Greg and others who are our full-time eyes and ears in trying to put these recommendations into practice. Jeff really has participated in two or three different kinds of meetings, including last week at Oral History, and I thought just by way of introduction he might give kind of a little bit of a sense of that meeting because it is sort of prototypical of the kinds of things we have been doing. DR. COHEN: Yes. I have attended three of these programs of outreach programs. One at the American Psychological Society. Another at the American Psychological Association. And the third at the Oral History Association. And there are two sort of themes or two general conclusions that I can draw from all three of them, which all three had the same sort of general tenor and theme. Initially at all three of them there was considerable adversarial interactions. Let's put it that way. And it struck me very soundly that there is considerable resentment out there among social and behavioral scientists toward IRBs, toward the regulations, towards the whole system. I mean, resentment, not just this is irritating, this is bothersome but clear resentment. In each -- all three of the meetings, as the discussions continued, all three groups came to sort of the same conclusion that the problem was not the regs. The problem was the implementation of the regs by IRBs. And there were serious issues raised about how IRBs are interpreting the regulations when they are applying them to social and behavioral sciences and also procedural things, logistical things. How often IRBs are meeting? The delays. The unreasonable paperwork demands, et cetera, et cetera. Nothing new but again this degree of resentment towards all that is clear but all three groups had -- came to the same conclusion that there was an important role to be played by the professional societies. This is, I think, the most promising, most positive thing to evolve out of this. Yes, OHRP has a role and NHRPAC has a role, et cetera, et cetera, but the professional societies are starting to see that they have a role to play in this as well. And all three groups have committed to, number one, doing a lot more at their next annual meetings with regard to human subjects, having pre-conference workshops, having more sessions on human subjects, and doing a lot more at their conferences a lot more visibly. We at OHRP have committed to work with them to help that move forward. Also, some of the groups, in particular I think the Oral History Society -- Association because that was I think the most interesting, traditionally APS and APA have had major efforts in this and Oral History Association has not. They have concluded -- they have decided to form a working group from their board and to come up with actually their own sort of white paper and guidance for their members that their members can take to IRBs to talk about the appropriate review of oral history research. One of the interesting things most people don't realize about the Oral History Association is their ethical codes are probably more stringent than most IRBs would impose and that has been a bone of contention because IRBs have been ignorant of and unwilling to address or respond to the Association's ethical code. So this dimension of the professional societies taking a more active role, an active role in training and education, and in turn this need for IRBs to look to the professional societies and their codes of ethics to help move this effort forward and to protect subjects, I think, is a very important issue and one that we and the working group are resonating with and trying to move forward. DR. LEVINE: What I am hoping we can do is -- and I think our working group and I believe NHRPAC is well situated to do -- is to move forward in discussing and ideally addressing the recommendations in what is one of the core areas that has been of substantial concern in the social and behavioral science community, and that has been the issue of IRB practices over the past several years, in particular, in terms of reviewing public use data files. You have received by attachment in our revised recommendations still entitled -- well, I think they may be entitled "Draft Recommendations Regarding Public Use Data Files." Indeed, we are commending them to NHRPAC for their consideration and action. They have been -- we discussed those at some length in July as you remember. We did get some e-mail interaction from members of NHRPAC. It seems that in July across our group there was support for the recommendations and, as I have said, as I have spoken about the original drafts, which we had posted and circulated already so that we didn't revise them at that point, no pun intended, but some of the language was infelicitous and that it really needed tightening to avoid things that we weren't intending to mean or imply but that the essence of the recommendations seemed to be well received in this group. I don't want to quite say that passive consent is consent but every posting we got since from this group certainly suggested or reaffirmed what we thought we were hearing in July and so has the input we've gotten from others in -- I will say including from IRB participants. It is true that the IRB participants that members of IRBs that are most likely to be at the range of meetings that we've been going to or maybe even accessing this particular web site or even caring what the Social and Behavioral Science Working Group is doing are persons who perhaps more readily encounter social and behavioral research. And so I start with the caveat that those persons who have served on IRBs are, indeed, our IRBs administrators who have given us feedback at a range of meetings are more likely to be persons who are more cognizant and familiar with public use data files. But the general sense is really what Ruth was saying earlier, and that is that in not knowing kind of what to do in the past couple of years of ambiguity, they have been in some places, in some institutions re-reviewing the use of public use data files and it is not only burying their activities. They are kind of wondering why, indeed, they are doing that and are, you know, kind of concerned about moving to putting their attention on significant issues that come before them. So we have sensed a real support for these recommendations but they are obviously brought to NHRPAC for your consideration and discussion. I won't repeat them. I think there was a handout outside. We did a transparency that is really not particularly -- it is 12 point so I am not sure how visible it will be. So if someone doesn't have it at your place, we will try not to use recommendation numbers but maybe we just want to start with Recommendation 1. Which is that the working group is recommending to NHRPAC that OHRP should clarify that a public use data file, that is non-identifiable data in a publicly available file, falls outside of the code of federal regulations for the protection of human subjects as set forth in 46.101(B)4. Needless to say, the section near and dear to everyone's memory, which I have the code here if we need to be more specific about that. OHRP should further clarify that under 46.102(F) research using public use data files is not private information because any private information -- according to that definition -- because any private information in such files is not identifiable or readily identifiable. MS. KORNETSKY: Can I ask a question? DR. LEVINE: Sure, absolutely. I just wanted to kick it off by that basic recommendation. Others we can go through. MS. KORNETSKY: My question is, is this saying that the public use data file is exempt from review or is this saying that it is not even -- it doesn't even get to the point of exemption? DR. LEVINE: It is under -- there are sort of two uses of the word "exempt." 46.101(B)4 really talks about what is exempt from the policy. MS. KORNETSKY: Exactly. Okay. DR. LEVINE: Exempt from the policy meaning exempt from the federal policy for the protection of human subjects under the Common Rule. MS. KORNETSKY: So is that what this is saying? DR. LEVINE: That certainly is what 46.101(B)4 states. MS. KORNETSKY: Okay. I would highly recommend for the IRB community that you use the word "exempt" in this recommendation. I was confused by what you were trying to say here and the reason that I am bringing that up is because when we get through the other recommendations, to determine that something is exempt, someone other than the person who is using the data has to make that determination. And so that needs to be somehow also, you know, addressed within here that the person who is going to be using that information, if you're saying this is exempt outside of the policy, that individual investigator can't say, "I am making that determination, therefore I need to do -- that I can't -- that I don't need to go to anyone." DR. LEVINE: There are two places where both the prior set of recommendations and this, I will say, fused recommendation 1 because it was the integration of, I think, recommendation 1 and 2 before. I think that is -- and that is that for a type of work that is exempt under the policy -- it doesn't fall under the policy and that is the 46.101(B)4 language. Indeed, we indeed altered the use of the language exempt there to say it is -- it falls outside of the policy because if something falls outside of the policy, it is not under the process under the Common Rule of the human subjects protection system. Those areas that are outside of the policy. Now the -- MS. KORNETSKY: That is where I am very confused here. You know, having worked on an IRB -- I mean, I fully agree that this type of -- I am trying to determine whether this is falling under, you know, under one of the exemption categories and, therefore, someone has to make that determination that it is exempt or whether you're saying we don't even have to deal with this at all. DR. LEVINE: Well, we are -- if you look at the -- of course, you have looked. The intersection of the set of recommendations is that a review has occurred for it to have become a public use file. And it has. It has gone through an IRB process in order to make that as part of the determination of the producer or the supplier of that data. Once that has occurred, we are saying it doesn't need to be re-reviewed by another IRB either because -- either because it falls under the 46.101(B)4 and also because it is de-identified. It is not identified -- MS. KORNETSKY: If it falls under the 46.101(B)4 someone else has to make that determination. I mean, someone has to determine that it doesn't need to be reviewed. Am I totally confuse? DR. LEVINE: The question is -- MS. KORNETSKY: Am I confused or am I just not getting it? DR. COHEN: Can I -- this is one of the areas and this is something that again is an issue that NHRPAC, they want to address. There is some confusion in the regulations because there is a -- I guess to use the word -- a conflation (sic) between activities that don't meet the definition of human subjects research and activities that are exempt. The public use data files are probably the clearest example of that because on the one hand if they are appropriately de-identified then they don't meet the definition of human subjects. It is not identifiable private information under the definition of human subjects in 102(F). On the other hand, 101(B)4 talks about publicly available information. And I think where it gets confusing is that public use data files that are de-linked are both publicly available and unidentifiable so they sort of fall in both categories and that is where it starts getting confusing. I think one of the issues that hasn't been clear, OHRP has been very clear that to determine whether a research project is exempt under Section 101(B) that someone other than the researcher must make that determination. We have not, I don't believe, been as clear as to who makes a determination that an activity doesn't meet the definition of human subjects, and that is, I think, another area of ambiguity. We have focused primarily on the exempt research and not on activities that don't meet the definition. Who decides that? So we need clarification on that and I think the only reason we haven't been as clear is it hasn't come up as often. DR. FLEISCHMAN: I think, Jeff, that is very helpful and for Susan's comment if you don't come to the dance then I don't know whether I want to dance with you. Susan asks everybody to come to the dance and then she says, "You're either exempt from dancing or you have to dance." But what we're really saying is that there are investigators who do not wish or do not have to get that assessment based on this information. So I have two questions. First, will we be in any way as institutions being able to recognize that such research is going on in any kind of listing or knowledge apropos to Ruth Faden's comments about accountability and outcomes and important issues that I think the research community may wish to have known. Will there be any such accountability listing? That is a question in this rubric. And the second question, I still don't quite understand where the criteria are written that constitutes a public use data file. Is it written somewhere because nowhere have you told us that. You have told us what they are but is there somewhere where everyone has agreed that X is a public use? These are the criteria and how will I, as the president of the university, know that my social scientist is using a bona fide public use data file? What will be the stamp on the file so that, if I am criticized for my university, I'll know that she used the right stuff? DR. LEVINE: Yes. In a way what one -- in crafting these recommendations -- starts with one because it is a logical way to begin, meaning trying to classify what these public use data files are. But two and three really sort of speaks to the process because, indeed, the public use data file has met the test of scrutiny when a supplier or producer takes that two and it has gone through an IRB review, indicating that these data meet that descriptor. The emphasis on this is really on whether other researchers can then use those files once they have essentially been certified by the supplier or the producer having that intent and it having already been examined to meet that ambition through an IRB process or in the particular case of some federal statistical data collections where other, like the census data where there are other regulations and review processes in place to establish that it is a public use file. I feel like I want to stop there, though there are also instances, and I have some descriptions of them, of what archives do that disseminate data to ensure that a file is a public use file and what the criteria are for establishing that. But the real thrust of this is that there is an affirmative responsibility to the producer and supplier to have that reviewed through the IRB process or, I am going to say, or the equivalent with some federal statistical collections. And that once those data are identified as public use files through that process, it is reproducible to deans, chairs or other IRBs. What we are essentially saying is that it doesn't then require additional re-review because it is not additional data collection for the construction of that file so the thrust of this is really recommendations 2 and 3. DR. COHEN: Let me add to that a little bit. First of all, I think this goes back to some of the heart of what the discussion was in July and what we think some of the confusion was. It is important to remember that at one point this was identifiable data. At some point the data was gathered from human beings. At some point it was identifiable data and the gatherer of that data went through a process of de-linking, whatever you want to call it, to make it appropriate for public use. What we are saying particularly in Recommendation 2 is that should go through a process -- an IRB review process. In other words, the IRB at the source institution should review what procedures this source of the data will go through to protect -- to de-link and then deem it, so to speak, a public use data file. What our recommendations are trying to deal with is the secondary use of that data by investigators after that has been done. And the issue that the social scientists are concerned about is that in the current environment that investigator that wants to use that data, his or her IRB is wanting to re-review that again even though it has already been re-reviewed. So what we are trying to clarify is that once this has been done, it falls outside of 45 CFR 46. It doesn't need to be re-reviewed again. And Recommendation 2 and Recommendation 3 focus on that pre -- how the data gets to that point. DR. FLEISCHMAN: I get it. But what you still haven't done is you still haven't crisply and clearly identified those criteria by which an IRB would so designate this? DR. COHEN: What we're saying is that OHRP should do that. Now the working group isn't going to do that. There should be clear criteria. There should be guidelines. DR. LEVINE: I have Bob, Adil and Elliot. DR. RICH: I think this is a very important discussion but I think it is broader than social and behavioral sciences, quite honestly. And I very much hope that the discussion gets extended beyond that to the biomedical sciences as well. And the best example that I can give is the Human Genome Project after all. Those are clearly huge public data files and they should fall outside of the definition of 45 CFR. One should not need to go to an IRB to ask whether or not you can go into the Human Genome Databank to query that information. That should not be something exempt, that should be outside. My one question is in your definition of a public data bank, you say available to anyone who wishes to use it. And I might -- I would prefer to see anyone who in a sense qualifies to use it and part of the reason is that it doesn't necessarily -- and perhaps the issue also is to do -- it means what is really public. There are really two genome databanks that are out there right now. One is "public," which means it is free. The other is proprietary, which means you have to pay to use it. That is the SOLARA databank. I don't see that there's an ethical distinction between whether or not you paid for access to the data or whether you can get it free. It is the quality of the data itself. So in a sense qualifying in that sense means you paid the fee. So I really congratulate the working group but I hope that it will take the broader view that these concerns are not just concerns of the social and behavioral scientists. DR. COHEN: First of all, I hate to say it, that is not the -- the purview of this working group is the social and behavioral sciences. If NHRPAC wants to broaden that, we would strongly support that and encourage that but our charter is the social and behavioral sciences. DR. RICH: I understand but as we are talking about what NHRPAC does, I hope we can take the broader view. DR. COHEN: Yes. Let me also bring up another point, going back to what Alan raised. One of the other issues about the regulations is that there are minimum standards, a set of minimum standards, and that institutions are free to adopt additional requirements. It would be fully appropriate once you have said, okay, these activities are outside the purview of the regulations, for an institution to say that it is this institution's policy that any researcher using such records has to in some way, hopefully minimally burdensome way, report that to the institution and to include the basis on which they're doing that. That this database was certified or meets certain criteria. That is an institutional responsibility. The regs are silent on that. The regs, you know, again to establish what the minimum is. Institutions set their own policies beyond that. DR. LEVINE: I think for this to be implementable there are already procedures being used by IRBs to ensure either that data files are -- if not anonymous, that they are appropriately de-identified. I think the IRBs that have been doing this in the context of publicly -- I'll say publicly available datasets, have their own procedures. I did reproduce a set of procedures, not anticipating, not only because I anticipated the question but also because I thought it would really be helpful general background and education more broadly for our group. Indeed, the first set is with three hole punches. This comes from the -- this can go to the audience -- from the Inter-University Consortium for Clinical and Social Research. Other IRBs that more routinely are reviewing large scale datasets that involve public use files have, I'll say, the equivalent practices. I reproduced the Inter-University Consortium's guidelines selectively and they're quite -- there's quite a book and a workbook both for the supplier of the data and the user of the data. And I reproduced really an excerpt of this so that you see that those -- even an archive charged with doing this with a dataset that is presented to them as public use -- re-scrutinizes and re-examines, and holds both the supplier accountable for the fact that it is de-identified not only directly but also in terms of indirect measures that might unobtrusively reveal the identify of the subject. But also holds the user accountable for not trying to in a way re-redact, re-identify. So this will give you, I think, a good and, hopefully -- I really only brought 50 so I apologize for not having enough for everyone. We will post these on the web site as an illustration. But the indirect -- you see the list of direct identifiers. The indirect identifiers, which is often in terms of ways that rare populations could be identified, either a small locale or an unusual demographic, earning over half a million dollars, you know, female lawyer earning over half a million dollars in Dubuque. I mean, that may, you know, be a small set. And what happens -- and the processes in place for handling them and this handout kind of outlines some of the processes. Removal, bracketing, top coding, collapsing, that is how to handle indirect identifiers, scrambling, impugning data, aggregating and disaggregating. These are the steps that the original IRB addresses and these are the procedures that the original IRB is addressing with these datasets when they are approved as public use data files. So the thrust of our recommendation is once approved by this form of procedure, and this is only one, that the second user of those data should not have to go through a re-review. I wanted to speak just briefly to the issue that Bob raised. The reason why we stated that these are data files that are available for anyone to use, it is operationally the case that, of course, the most active users of any form of dataset are researchers and scientists using these datasets. But there are many public use files that are used by persons in state and local agencies, by private corporations, by the public, the census data, who are indeed inquiring about things that may interest them and they're not necessarily just doing it for research purposes. Those data are already out there and, indeed, when the census data are produced, those are produced, for example, with a very explicit congressional intent of making that available to anyone in the public who wants to understand the nature of our population at any point in time. So that is why it is a wide language. DR. SHAMOO: Thank you. My concerns are sort of complimentary to Bob's. It is too bad he is not here because we really probably compliment each other. His concern is, of course, which is legitimate in not impeding research of your kind of work, which is legitimate, but my concern is those producers of public use data that we are inhibiting really and infringing on First Amendment. Let me give you an example, and that concerns me. My attitude is, and I have expressed it a couple of meetings, once it is public, my attitude is it is none of our business. It is public. Anybody can have access to it, do whatever they want with it. Let me give you -- public interest groups now publish name of researchers, name and address and phone number and e-mail, and how much money so they can delineate who has and does not have conflict of interest, and how much grants they have. They publish which doctors have run afoul of the law, et cetera, et cetera. I am saying it is those data, if a social scientist wants to do research on those data, which are available on the web site or newspapers or wherever they're published, why should an IRB even review it? These are publicly available to anybody. I am saying aren't we encroaching and almost using censorship to intimidate someone from using those kind of data? DR. COHEN: Obviously this is a very difficult issue and it borders on the whole issue of the distinction between journalism and research. DR. SHAMOO: That is right. DR. COHEN: And that is a really murky, murky area. As a matter of fact, it is one of the conversations I had at the Oral History Association because the oral historians are in that same sort of issue. I think that clearly there is a need for more guidance on distinguishing between journalism and research. I think the courts are doing that. I mean, the courts are addressing what constitutes journalism and the extent and scope of the First Amendment rights, especially in the context of the internet. Publishing something on the internet, does that make it journalism and so forth? And so I think that the human protections movement should look to the courts for guidance on some of that. There's also a distinction, though, between -- in a sort of common sense way -- between journalism and research. Journalism is done for the public knowledge and for the public good in the sense of providing information the public needs to know. Research has a sort of different context and that is, you know, further -- the development and furthering of generalizable knowledge, which is a somewhat different thing than the public's right to know, although they're blurred. So I think that it is very clear in practice that the government, institutions and IRBs do have sort of a right or a responsibility particularly when it is focusing not on censorship but on protecting the rights and welfare of the subjects of research to review that, and I think the courts have upheld that. Particularly, I think, was the University of Minnesota case, which wasn't particularly about human subjects but it was on research integrity. The courts upheld that right as opposed to journalism where infringing on that would be censorship. There is a murky area in between. There's also the sort of traditional knowledge that your right ends at the tip of my nose. And so that you can't yell fire in a crowded theater. And so there are things that even though we have constitutional rights, we also have the right to protect subjects and that there's a balancing there that needs to be done. It is not that there's an easy answer to that. DR. SHAMOO: Your answer really answers. Subsequent to my concern is that those who are producers of public use data, you want them to go through some mechanisms before -- unless there is a bylaw like the census -- or they go through some mechanism, may include IRB before they may their data public. And my attitude is no. There should not be -- if it is public use data and it is out there, and they are going to collect it themselves, if I am in an advocacy group and I want to publish the name of every research scientist, I don't have to go through an IRB to publish that kind of information on the internet. And it doesn't have to be research, that is within our First Amendment, within journalism, and I am not a university professor. If I am a university professor and I want to do subsequent research on it, it is maybe that is okay within your definition to have it reviewed by an IRB. DR. LEVINE: There may be another issue of the one that you raised that we should discuss at some point more generally in NHRPAC but for us here in this area what we're seeking to establish is that when researchers are producing datasets that other investigators can use, that once established as meeting the criteria of public use through the IRB process, that that should be sufficient and not have to go under those subsequent reviews. DR. DORFF: I am just playing philosopher here. Okay. I think that in order to get around some of the confusion, what you may want to do is in terms of restructuring these recommendations, is just simply start with a definition, as you do, right. In other words, that a public use data file is this. Okay. And then ask -- you know, ask Greg's office to come up with criteria if you want to. In other words, as to exactly what meets that definition. Or come up with the criteria yourself. Either way, okay. So you first define it together with the criteria for making it fit the fact that this is not -- there's no longer a file that deals with human subjects and, therefore, not -- right, it is no longer a file that meets with human subjects. Then the second thing is after the definition which meets this criteria, talk about the process for determining that data is -- does, indeed, meet this definition of being outside human subjects. So that process -- that is two and three that you have here. Two could be either that the government decides this or that a private agency decides this. And you now have them built in, beforehand, the criteria by which it is going to decide it. So you have a definition. You have the process for deciding -- you know, for determining who does this and, you know, who can be respected for having done that. That is Susan's issue. And then, finally, third, the result of this process is that individual researchers don't have to get independent IRB approval. But, see, in that way what you're doing is making it clear that the point here is that these public files, once having been declared public files in recognizable ways, in a credited recognizable way, right, are then no longer the business of individual IRBs. DR. LEVINE: I think I would maybe recommend a different process because I think what may be clearer to some of us as being here, you know, what a public use file is, is not necessarily clear to everyone on NHRPAC. And so -- and yet this is an area where we all seem to know it and thus want to move forward. So I would -- I'd like to put in place some kind of process where we could move forward. I'd call it -- indeed, I anticipated it. It was my first note at the top of the page. I call it the Mark Barnes' strategy. Meaning it would if we all think we know what we're stating, it would be -- I am thinking -- and certainly the working group knows what it is stating. It is not seeking to revise and resubmit if I can put it in those terms. We may be at the point where it would be more functional for those of us on the working group who are on NHRPAC with some of us on NHRPAC not on the working group to try to re-craft this in a way that meets the needs of NHRPAC, frankly. That is what we're talking about. What will permit NHRPAC to make a recommendation? So whether it is Mark or Alan or someone, or you, to -- where we think this is clear where we can really move this from this is what the working group is recommending, this is of the spirit. We're not really trying to have this published. This is of the spirit of the kind of guidance that we think NHRPAC needs to put in place to take a terrible stumbling block in the system and free the system so that analysts can use these public use files. And thus the issue for me as a NHRPAC member is what is the next steps to permit NHRPAC to be in a position to do that? Not that I am -- you know, I am not opposed to the revise and resubmit notion but I am thinking if we all kind of are agreeing in the spirit of what we want to achieve. The question is what is the next process when NHRPAC has this guidance and there isn't a disagreement on the facts. There's a disagreement on how to best explicate it that we need kind of a process that will get us there, I am going to say somewhat quickly, ideally. CHAIRPERSON MARSHALL: I am going to say that I agree with that and maybe ask Susan if she would sort of head up a small group. I am just going to say Susan and we will call others from the committee to finesse these recommendations. MR. BARNES: Felice, there is a regulatory issue that lurks in what you're talking about in regard to HIPPA and the medical confidentiality regulations that are already in effect but will be enforced in April of 2003. Under those regulations there is a definition of what is and is not de-identified data. Because HIPPA is a medical privacy or medical confidentiality statute then that definition that was invented by or fashioned by the people within HHS last year adheres closely to a clinical de-identification definition, a clinical medical de-identification definition as opposed to things like occupation and where you graduated from college or high school and elementary school, things like that. They didn't think about things like that when they were writing the de-identification definition. But there is a definition in the statute which basically says that if you -- if one, if a hospital or a health care provider or anyone who has this protected health information sheds the data of 18 identified categories of information like medical record number, social security number, zip code for those zip codes that have fewer than 20,000 people in them, et cetera, et cetera. There are 18 data fields. Once that information is shed out then it meets essentially a safe harbor definition and then that information can be freely disclosed by the holder of that information to anyone in the world and can be freely used by anyone in the world. The reason I bring this up is because ultimately -- I mean, the way that the world works is that the definition of de-identification in HIPPA is probably at some in the future going to coalesce with the definition of de-identification of de-identified data under the Common Rule in one way or another, and there is no definition under de-identification of the Common Rule. So this is all a long way of saying that it seems to me that one thing that would be a useful exercise would be for the working group to look at the 18 data fields to compare those to the things that you think would be -- you know, ought to constitute de-identification, to see if there are things where you differ. And ultimately, you know, to make some kind of recommendation, if not to -- you know, I mean, this committee, you know, essentially reports to Tommy Thompson so, you know, I mean, he is the one who also controls the HIPPA regulations. But it would be nice -- I mean, what I am saying, I guess, in the end is that it doesn't make -- it doesn't -- it is not a -- it is a less valuable product unless your product actually refers to and compares what you're talking about to the de-identification standard under HIPPA. DR. LEVINE: I agree. I remember reflecting as that briefing was going on. Indeed, it is the one note I took from the -- in addition to reading the report from the July briefing and that was the 18 criteria. I remember as we were going through the criteria, I thought, oh, well, that report sort of reinvented what the social and behavioral sciences have been doing with public use files for, you know, many generations. And I think that there's a comfortable fit, as I recall the 18, and it probably makes, as Bob was raising, it probably makes more general sense for NHRPAC to reflect on those criteria for de-identification with respect to a large number of datasets. But I think that would be helpful and, whatever the vehicle, we should do that. DR. MORENO: You can always tell when a philosopher has been on an advisory committee too long because he starts acting like an editor. Elliot, you actually did editing, good editing, rather than philosophy. I thought you were going to point out or suggest that perhaps the second use of the word "private" in Recommendation 1 looks like a contradiction and we ought to get rid of it or modify it in some way. That is just a minor editorial point. I am not sure how that will affect the thrust of the recommendation. The other thing I wanted to say is more generally, and this might go to what Bob and Adil were talking about, and maybe more to third parties and genetics, is that in the -- conspicuous to me, at least, from the examples given of indirect identifiers in the Michigan list -- and I don't know if this goes to HIPPA also, Mark -- is relationships. For example, offspring of the current President of the United States or relatives of victims of the World Trade Center disaster and so forth. Somehow we have got to figure those in and this also goes to what Alan is going to be talking about later on. I don't know how they work in HIPPA. MR. BARNES: Under HIPPA there is a -- you know, there is this kind of safe harbor of 18 datasets but there is -- but that is if there is other information about the person that despite the shedding of the 18 data sheets could nevertheless identify the person then it also is not de-identified data under HIPPA. So although it is a safe harbor, it is not a safe harbor if the researcher or the holder of the health information knows that there was other data that would tend to identify. So relationship would be an appropriate example. DR. MORENO: So given that there are fewer than -- let's imagine that there are fewer than 20,000 children orphaned in the World Trade Center. We're getting ahead a little bit. This may be relevant this afternoon. Would that, therefore, not be in the spirit of HIPPA de-identified if you were to do a data -- do a study of the psychological effects of the disaster in that group? MR. BARNES: What HIPPA says is that once you did the study and you had all the information, if you stripped out all these 18 datasets, then you could use them to disclose that aggregate data or even that individualized data. DR. MORENO: Okay. MR. BARNES: But what HIPPA doesn't talk about is whether you could actually conduct that research and the method in which you could conduct the research. DR. MORENO: Okay. Thank you. MR. BARNES: Does that make sense? DR. MORENO: Thank you, yes. DR. LEVINE: I want to add that I think the scrub and analysis in these social science datasets to meet the criteria for a public use file as opposed to, I am going to say, an individual investigator seeking to "de-identify" his or her own file is actually quite a higher bar. This was just the briefest excerpt but it is -- that is examination every which way and IRBs looking at whether investigators are doing it in order to make it more generally accessible, recognizing that what an investigator might think is de-identifiable for one's own research use may with other wider use really require lots of different forms of matching and combinations. So it is quite a high bar process before public use data files are essentially vetted and released by an IRB for that purpose. CHAIRPERSON MARSHALL: A procedural point. I don't think that we are going to have time within the 40 minutes that we have left to perhaps move on to risk and harm. So we should come to closure on this if we can. So if you want to -- that is how we're going to proceed and I want to leave time to open the floor to questions from the audience. I have Sandy, Judy, Greg and Susan. DR. CHODOSH: I guess I have to take a philosophical point of view right now and I think what this does very well is to try to eliminate the risks that are associated with breach of confidentiality because it is gone once you de-identify data. What it does not look at is the benefit and this is, I guess, the part that has always been bothering me and it didn't really become very evident. Is the research that is going to be done with that data appropriate? You know, the research institution, all of us are under special scrutiny and that I think that using resources, whether it be intellectual resources, whether it actually be real monies, et cetera, to do bad research, you know, is not in our interest. Somehow in this whole process if we eliminate the IRB which variably looks at the scientific benefit, who is going to take over that role? Not all researchers are big time NIH, HHS money getters and who looks at this to make sure that this is appropriate use? I grant you that it is independent research as people will sometimes go off and decide that, you know, relatively is a real great idea and eventually it works out but there is -- you know, why -- philosophically it bothers me that we are sort of dividing this in a way without introducing something in there that works and I am not sure that we have it at the present time. DR. LEVINE: I think it is an important issue you raise and I want to answer in two ways. One was implied in Recommendation 4, and that is in the -- in making these data available, the benefit to both the subjects and the scientific enterprise is that these data operate in the finest scientific tradition of permitting with this information testing rival hypotheses, addressing new questions, even maybe falsifying inferences through new forms of analyses. That is sort of on the one hand a real benefit to science. On the other there is, or additionally there is the -- what is the role of the human subjects protection system and then what is the role of other parts of the science review system, promotion and tenure, publishing and non-publishing if you're addressing trivial or less important scientific questions, or being funded or not being funded. And that is not to say that any of those are imperfect but one of the real pluses, really where the social and behavioral science community has taken a lead is in seeking to make data accessible for really meeting the essential fundamental principles of science. And that is whether, indeed, the results are reproducible, whether they were accurately drawn, whether other hypotheses can be tested, and that is why, indeed, we didn't try not to address it. We tried to address it in Recommendation 4. I think the other issues you raise are addressed in other parts of the kind of scientific oversight system, however imperfect that may be, I think would be a big burden for the IRBs to also be in a position of evaluating kind of the scientific meddle of an otherwise scientifically vetted dataset. DR. CHODOSH: I didn't suggest that the IRB should do it. I am saying that if you eliminate that aspect of it, you have got to put something else in its place. And, frankly, that doesn't exist in a lot of places right now. DR. COHEN: Well, one of the problems with that is sort of turf or purview. If we've determined, and I think there's some agreement that once done properly these do not fall under the purview of the human subject regulations, then in some sense our role ends. That is then where -- and I think that is Felice's -- that is where the institutions' responsibility comes into play and the scientific community's responsibility comes into play. But what we've done is established that this is not -- no longer human subjects research. That doesn't mean that there aren't still ethical issues but the venue of the human protections system, it is outside the venue of the human protections system and falls under other responsibilities. DR. CHODOSH: But sending it out into cyber space doesn't exactly absolve us of our societal responsibilities. DR. SIEGEL: First, I want to thank Elliot for preempting and saying it better but I also tend to take a very pragmatic approach so I would actually like to support. The questions that I was having with this document are the set up and what do we need to do, how do we need to evaluate, and I think what maybe hasn't been brought up so far is who needs to know that this has been done. So I think some of the questions are, well, how does one IRB acknowledge, accept and defer to another IRB's judgment? And I think if the criteria upon which IRB's were making these judgments were more -- first of all, in one place and not knowing these, there seem to be several places where these criteria might be available now. Maybe HIPPA will also play a part in this but then how do you then disseminate that information among IRBs so that everybody has the same understanding going in? DR. LEVINE: My sort of -- in the spirit of friendly amendment to Elliot's suggestion to sort of migrate that task to us, even me with my shifting hat, from the working group to the NHRPAC hat, is that we really need to think about how to put this in a form where this group is provided the expertise. How do we now put this in a form that is in a way that NHRPAC wants to then move forward with an agreeable content? And that is kind of a second process that I think a working group of persons, some of whom have agency identification, some of whom are on NHRPAC, and some of whom are in the field really can't as much be the eyes and ears of NHRPAC in terms of how to move the content forward once there's, I think, an agreement on that. DR. KOSKI: Well, maybe this follows on that. There was a physics professor at the University of Wisconsin who once tried to explain to his students the concept of close enough for all practical purposes. And the way he described it was that if you took all of the males at the University of Wisconsin and you lined them up on one side of Main Street and all of the females and lined them up on the other side of Main Street. Okay. If they stepped halfway towards each other and then again halfway towards each other, and then again halfway towards each other, they would actually never really come into direct physical contact but they'd be close enough for all practical purposes. Now I tell that story because I want to -- I have noticed the social and behavioral sciences working group doing an extraordinary job since the beginning. This was one of the first working groups that came together. And we're making great progress moving halfway toward the goal, halfway toward the goal, and halfway toward the goal, and I want to compliment the group for what it is doing. But at the same time, forgive me if my frustration is showing, but one of my greatest sort of challenges in my first year here with the government is that we tend to go halfway toward the goal and halfway toward the goal, and sooner or later we have to take that final step. And so what I am urging you to do, okay, is not to -- we have some very smart and committed people in government who actually have some understanding of these issues, okay, and we don't need to have everything nailed down, okay, to every minute detail in order to get the job done. So what I am urging you to do is to get together and take the next steps as soon as you can because we are anxious and willing and waiting to move forward in a constructive way, recognizing the enormous opportunities for both streamlining, improving the oversight process, as well as helping research in the social and behavioral sciences community, as well as other areas that would rely upon these kinds of things to do it. So that this is just a pep talk but I would like to encourage you to cross the threshold and take the giant step that is necessary to get a final set of recommendations from this group so that we can go ahead and act on them in an intelligent and meaningful way. Thank you. DR. FLEISCHMAN: Gee, I agree completely but I don't understand how to do it. And the reason I don't understand is because I strongly believe that volunteers are necessary but not sufficient. And we have not yet in NHRPAC in each and every one of our outputs, and I have been involved in some of those outputs, we have not sufficiently been supported by good and important government employed staff. No offense to our staff. I believe that if NHRPAC had some staff, we could comfortably then do what Felice is trying to do, which the working group for children is trying to do. And that is to bounce the tennis ball toward Mary Faith and to say, "Chairperson, the next step is get it organized so that we can approve it and send it on to OHRP." But she has got a day job, at least she used to, I think she still does, and I really believe that we need some conversation about that process because I would not be proud as a NHRPAC member of signing off on this as it sits here knowing full well that there are smart people sitting in OHRP that could use this in good ways. So I really have a problem about this process and I wonder how we could -- and maybe other people feel differently and maybe this is a process -- you know, this is good enough and it is approximately close enough, and then we will get OHRP people to do their job but that is not how I thought we as a FACA committee ought to be functioning. So it does speak to this issue and, you know, I don't want to perseverate around having asked everybody how much money and people they had who come here from NBAC and from the advisory committee, and everybody else. But we are being hampered in fulfilling your goals and I think we could move very rapidly on a lot of these things. DR. MORENO: I thought, Greg, that you were going to say the next step between the undergraduates was the one prescribed by Geikergard (ph), namely the leap of faith. [Laughter.] DR. MORENO: And in that spirit perhaps what we could do to resolve this problem is at the conclusion of this conversation to give it to OHRP staff and then in an iterative fashion ask them to come back and report to us their refinements of these -- their recommended refinements of these recommendations, and then we will have another go at it. CHAIRPERSON MARSHALL: I really didn't slip Alan five bucks under the table before the meeting but he makes an excellent point and I think that is something that we just need to keep working on, is trying to get ourselves adequately resourced because you're exactly right. And we have been relying on the committee and on the good graces and volunteer time of others to do excellent work but we could certainly have moved forward more quickly if we had more resources to bring to bear. I guess my sense at this point is that I would like -- I do very much want to bring closure to this and it has got to be something that the committee is happy with and proud of. I would like to -- at least in the phase between the time between this meeting and our next meeting -- see what a group headed by Susan can do in terms of fine tuning and perhaps at the close of our meeting tomorrow when we talk about housekeeping issues and so forth then we can look at resources again or sort of examine our procedural shortcomings and our needs as a committee. I have Susan and then Alan again. MS. KORNETSKY: Now that I have been assigned something. [Laughter.] MS. KORNETSKY: I just want to go back to a point that Bob made. When I read through this document, the only place that the word "social and behavioral science" really shows up isn't until Recommendation 4. And, I guess, I am wondering in taking direction -- I mean, if we are going to limit this, which I don't agree with, to behavioral and social science research, do we have a good definition of behavioral and social science research? Because it is not so clear where one ends and one begins and I think my approach would much rather be to take this and extend this to all public use data files and not even include behavioral and social science research. So I think I need direction -- DR. MORENO: Specifically. MS. KORNETSKY: Specifically, right. I am sorry. Thank you. [Laughter.] MS. KORNETSKY: Thank you. So, I mean, I think I need direction from this group about that as well. DR. LEVINE: I think the more generic approach is certainly one that we should all seek and feel comfortable with. And in a way one of the functions of the social and behavioral science working group is that we are sort of populating issues that have been particularly -- where they have been particularly kind of fault lines or problems with respect to these arenas of science but they are not necessarily unique. Indeed, that was part of my message when I spoke before you not as a NHRPAC member in December. So that, for example, when we're talking about minimal risk it obviously is generic and we need to come to terms with that but in a way our working group is working on them not thinking that they are somehow special but part of a -- but often part of a broader conversation and we would not see these as necessarily being generous to some forms of datasets called social and behavioral science but those that have been kind of the stock and trade of the nature of the work that we do, which frankly increasingly is rather multi-disciplinary, including health history and biomedical science. And so it is not that there is a divide between one and the other and I think that we could re-craft -- you are right that the words weren't there. We're particularly worried about the fragility in the system with respect to very significant data analysts using very significant datasets that would have classically been defined as social science, but I think we can more broadly do that in the same spirit. DR. FLEISCHMAN: Just a quick clarification in Recommendation 7. The two uses of the word "encourage." And why aren't we a little more directive and strong? DR. LEVINE: Sounds good to me. I mean, I think the group wasn't quite sure. In a way they are in a -- I won't quite say a rank order but the top of the book are clearly things that would free use but we certainly want to have better educated IRB reviewers with respect to these forms of data and for those who are essentially the preparers that they have much more savvy so something that is stronger than encourage would be desirable. DR. COHEN: Let me just say again that there is very little authority to go much beyond encourage. I mean, there's nothing in the regulations that requires education and training. We're dealing with this with the FWA within a whole lot of other contexts. So that if NHRPAC wanted to go beyond encourage they would have to find some basis for doing that. Moral suasion is about it at this point. DR. LEVINE: But I think what Alan was saying would be something like instead of OHRP should encourage the development of a training module is that OHRP should develop or support or fund or receive funds for so that it was done and that doesn't reach to whether someone has to take the course. CHAIRPERSON MARSHALL: Let me ask if there are ex officio members, public members, those out there in the audience who have comments or questions? We certainly would like to hear them. DR. CORTAZY: I am Roger Cortazy, EPA. CHAIRPERSON MARSHALL: I am sorry, Roger, this is on the record so you need to use the microphone. DR. CORTAZY: How does this process treat the question of information that is publicly available and identifiable but maybe shouldn't be used? I take from the security example just because you read about the performance characteristics of a new airplane in the New York Times doesn't mean that information isn't still classified and you can get into a lot of trouble for using it. How does this deal with that problem for human subjects? DR. COHEN: I think this blends a little bit into what Adil was raising before. And the regulations are fairly clear that publicly available -- the use of publicly available information is exempt from the human protections regulation regardless of whether that is a good idea or not or whether it should be used or not. Under the regulations it is exempt. So if this advocacy group publishes the financial relationships of the top 100 investigators, clinical research investigators in the country, whether they should have done that or not, whether it should be out there or not, once it is out there under the regulations it is exempt. Now that then gets back into there are other spheres. There are other players in terms of enforcing ethical or responsible conduct. There are institutional policies. There are, you know, scientific publication policies. There are funding agency policies. There are other spheres. The regulations have said this is the extent of the involvement of 45 CFR 46 and so since NHRPAC is geared towards the regulation -- you know, human protections and the regulations sort of defines the sphere of human protections, unless we recommend changes -- we -- you want to recommend changes in the regulations to broaden the sphere, that is the sphere and then we look to other players to enforce other ethical standards. DR. DORFF: Just a question. In Recommendation 3, the second sentence, this recommendation pertains only to those instances where non-identifiable data files were not planned or contemplated at the outset. Why do you say that? In other words, why couldn't it be that what we would recommend is that you could have the government has an accredited agency to determine that a given file is a public data file but you could also have any of these organizations that you just mentioned at the bottom or the University of Michigan, that you could -- what I have in mind here really is the accreditation process for universities, right. Where you have a nongovernment agency that is recognized as providing the criteria for something. In that case accreditation. So why couldn't we recommend that in addition to Federal Government agencies that there would be other known and respected organizations in the various fields, like the ones that you interviewed, that would be given -- that would be accredited by us and by the government policy to determine that public interest files are, indeed -- that certain files are, indeed, public data files? DR. LEVINE: We are in heated agreement. That is what we thought 3 was saying. Not that the government should establish its own but they should -- they should have in place some process by which they would say ICPSR, your processes, if you determine that something that wasn't previously judged through an IRB process to be a public use data file, ICPSR qualifies as one of those organizations that could do that. I mean, that was an example. So that is certain organizations. It wasn't trying to establish more federal apparatus but really a way of OHRP designating certain forms of public -- certain forms of archives ideally that disseminate data, and there are some number of them that do, as qualified to attest to the public use files. DR. DORFF: Okay. So then why -- these things were not planned or contemplated at the outset? DR. LEVINE: It is for when -- for those data that when those -- when the producer of those data undertook that research, the investigator undertaking that research that he or she did not contemplate the creation of a public use data file. So it wasn't reviewed by the original IRB for that purpose. If it is reviewed by the original IRB for that purpose it is Recommendation 2. Recommendation 3 was when it had not been reviewed as a public use data file because it had not been anticipated at the outset of the work. DR. RUBEN: Hi. Phil Ruben from the National Science Foundation but I am speaking as the chair of the HSRS Nonbiomedical Working Group and we would like to strongly encourage and basically second what Greg was saying, and strongly encourage NHRPAC to move as rapidly as possible without compromising anything on compartmentalized things like the public use data file because we think in this case the behavioral and social community but broader communities are reaching out to both NHRPAC and to the Federal Government to give them the kind of guidance. We think that this is a case where it is very close. We do agree. We want a professional product and we think we are very close in our group waiting on this group and so this is one where we really encourage people that this would provide a good example and would also let us determine how the process plays out after that. Who should be doing the guidance, where it goes, and we think this is an excellent compartmentalized example of a case where we can all work together. So we support what you are doing and encourage you to move as rapidly as possible. DR. SUSSMAN: Don Sussman, USDOT. A question. Once one of these public use data files has been developed, what is to stop it from being FOIA'd? Why do we have to worry about it once it has been recognized as a public use data file? FOIA is Freedom of Information Act. DR. LEVINE: While certain federal data collections are public use files, much of this also includes data collections that are not government files. They may be federally funded. They may not. DR. CHODOSH: I hope that in Susan's new little task here that process is worked out whereby there is some method so that the secondary user somehow has documentation to show to somebody that is the process you have to work out. That assures that somewhere along the line it was properly authorized. That, it seems to me, to be a very major key which has to be in the process. DR. LEVINE: Let me say that a number of us sort of discussed that possibility but it was one that wasn't in the original -- it wasn't in the original set of recommendations and what we were trying to do was hone the feedback and commentary we received around these but there was a sense that something that -- where it would -- I mean, in certain contexts it happens at the department level but somewhere where it was reproducible as to who users are would be, in general, useful. Though I say this more with my NHRPAC on because it hadn't really been actively discussed by the working group. CHAIRPERSON MARSHALL: Mary Kay? DR. PELIAS: I have a concern in Recommendation 4 and it is going to spill over into our discussion this afternoon, I think. One of these sentences says, "The general notification investigators provide to human subjects." I think that this phrase presumes that the investigators have a right to notify or essentially to tell human subjects what they are going to do down the line and I don't see any provision in our thinking so far about asking human subjects whether or not they would want to have their data declassified or de-identified down the line for purposes of public use. I am bringing this up just for our consideration because I believe particularly in the world of genetics and the fact that we are getting new data all the time about conditions that people may or may not develop in the future, I think there's a place in our thinking for the idea that human subjects should be asked whether or not they want to have their data de-classified or de-identified and have a choice about possible feedback in the future. I think this is a decision that should rest with the individual human subject and not necessarily be a subject of notification to the human subject by the researchers. Just an observation that will spill over into this afternoon. DR. COHEN: This issue is a very important one and it resonates with an interaction that I had participated in back in January at a conference on data sharing of neuroimaging data. They had a conference on the technology and the ethics of that. And this got into a very big discussion at that group of primarily researchers. What rights does a subject have to the use of his or her data once it has been de-identified? Once under the regulations it is now no longer considered human subjects data. So if you're a researcher and you gather data on me, and the example that was at the session was on my brain function, and I agreed to be in the research for a certain kind of research to be done for a certain purpose. Now you take that data and you de-identify it and then use that data in a study say on racial differences in intelligence, which would offend me, which I would not want to participate in and I would not want to be involved in. But you have de-identified that data. Is it still my data and do I still have some rights over the use of that data? And I will tell you after considerable discussion by some very intelligent people, we were unable to come to a conclusion on that issue. That is a very difficult issue and I can see where it gets into some of the genetics issues and third party issues and so forth. But that is a fundamental issue that the regulations are silent on and not easily resolved. CHAIRPERSON MARSHALL: Other comments, observations, questions, suggestions? [No response.] CHAIRPERSON MARSHALL: No. I would like to thank Felice and Jeff, who really have been doing a yeoman's job and to thank Felice especially because she has been very creative and generous in terms of coming up with resources, using her own resources to get this work done. Felice, we really do appreciate what you have given to us and all of your dedication. We have had a Greenwald fellow, Jim Taggert, who spent the summer with us working with Felice on this project. And maybe just to look forward to the discussion of housekeeping issues and so forth, Kate and Greg and I have talked about the idea of hiring consultants on an interim basis as we are trying to push forward and asking for an increased budget and increased resources and so forth. So we have decided that we will hire writers and other folks on a consultant basis in the interim to help us get our work done but that doesn't mean that we won't continue to try and push forward in requesting the resources that we need to get the job done. Felice and Jeff, thank you very much. [Applause.] CHAIRPERSON MARSHALL: I think we can commit to a finished product by our next meeting moving forward having Susan help us craft something with Felice and with Jeff that looks final enough, penultimate enough for us to bring to this group for final action. I guess, January, Kate, is when we're meeting again. DR. CRAFT: Yes. CHAIRPERSON MARSHALL: So thank you all very much. DR. LEVINE: Thank you. I mean, this is -- I just want to say this is not only a Jeff and Felice Levine collective effort but really it is a very active working group. Phil Spoke, Caroline Miner. I want to introduce Paula Sketchfold, who has joined ASA as a senior science policy analyst, and she is giving a large chunk of her life to human subjects protection. And we are -- you know, we are really interested in recognizing that we need to do a lot of the heavy lifting in order to really bring the good ideas to the table and we are planning on doing so with other issues and I want to have those percolate up and move forward in those ways. Ivar Pritchard is from Education. He is on our working group. There is almost no one listed who really hasn't done a huge amount of work. I want to pass out because risk and harm always ends up being deferred but it won't be next time because we will -- you know that will be one of our next subject. I want to pass out which were the July recommendations but I just tried to do another handout that focused on the July recommendations, and then -- and this is just background reading, and we wouldn't mind the input. This is the current guidance on the expedited review because we do talk about making clearer how expedited review relates to low risk or minimal risk. And then this is a working table that we were going to -- I think I have 50 copies of that. This is a working table that we only really -- we view it to be a continuous work in progress that sort of tries to provide a spacial way of understanding the relationship between different kinds of harm and different levels of risk. Not just understanding the levels of risk per se but also ways of ameliorating that risk and what steps can be taken. So I am passing it out to NHRPAC in a way to engage you early on in what will be our continuing conversation. We are going to be meeting on this on November 28th and our hope is to have something that reaches the highest bar of refinement possible by January. So it will be helpful to have that continuing interaction even though we didn't have yet the substantive discussion that we will in January. DR. KOSKI: I wanted to follow up on the sort of comment, I think, either Alan or Jonathan made. It may be that we can expedite the process of some of this by doing as you suggested. Rather than wait for serial processing, if there are areas in which we already have substantial agreement or close to consensus among the full membership of the committee, then rather than wait, as I said, until every detail is ironed out, we might be able to sort of take what we have got at this point and begin the process of developing either educational materials, guidance or if there are new regulations that needed to be developed, we could actually get those in the works and then bring them back so it might be a more efficient way to move things forward, recognizing that the resource constraints are very real. But it might be a way to try and facilitate the process to move forward in a more timely fashion but clearly we would want the full committee to be agreeable to that approach, recognizing that we would not be trying to preempt the committee decisions but rather -- recommendations or advice or whatever but to actually facilitate translating those into what we have to do on the government's side. DR. LEVINE: Thank you, Greg. I think I can speak for Jonathan and say he is all for consensus. DR. MINER: Since Greg mentioned parallel processing, I just thought I would bring up that are -- CHAIRPERSON MARSHALL: I am sorry but could you just -- DR. MINER: Oh, I am sorry. My name is Caroline Miner and I am also on the HSRS nonbiomedical working group. And I just thought I should mention that we have actually a small subset of our members working on creating the guidance documents that we seem to be hearing from the NHRPAC working group and the public that are so badly needed. So we are hoping to actually have something on that relatively soon. CHAIRPERSON MARSHALL: Well, we are right on time for lunch and we will see you back at 1:00 o'clock to talk about third parties. Thank you. [Whereupon, at 12:00 p.m., the meeting was recessed to be reconvened this same day at 1:00 p.m.] - - - A F T E R N O O N S E S S I O N [Time noted: 1:05 p.m.] CHAIRPERSON MARSHALL: You have been very good, I think, about staying on time today so let us move along in the agenda to a large topic. I am sure this will be an interesting discussion and I am sitting close to Mary Kay so if any objects get hurled they're going to hit me, too, so be careful. DR. PELIAS: I'll duck. CHAIRPERSON MARSHALL: But I hit back. Update and Discussion: Third Parties as Research Subjects. And Mary Kay has chaired the working group in this area. Thank you, Mary Kay, and we look forward to a fruitful discussion. UPDATE AND DISCUSSION: THIRD PARTIES: RESEARCH SUBJECTS? DR. PELIAS: Thank you, Mary Faith. As you might imagine, we have collected a mountain of e-mail printouts over the last six months and I would like to assure everybody that I have read every word of them and what we're going through now will be a distillation of opinions. The first thing I would like to do is to thank the working group for all of their help. Although I do want to say that some members of the working group have been absolutely loyal in participating in all of our telephone conferences and we have had limited to no participation by a few people. So when we talk about our vote and the references to some members of the group and other members of the group, I will tell you that some members of the group means a very, very small majority and other members of the group means a very substantial minority. So we've gotten into a situation where in all of these issues we have not been able to come to a consensus on some very important questions. I would also like to thank Felice at the outset for the help that she has given us in putting all of these recommendations together. Early on the advisory committee asked us to form a working group on third parties. It was supposed to be, I think, initially a working group on genetics but it was expanded to third parties in general and our charge was to review the definition of human subjects as we find it in the federal regulations and then to determine whether the collection of data about third parties requires the informed consent of those parties and what are the implications of trying to go to these third parties for consent. Our working group did for a while join forces with the working group on social and behavioral sciences and we put together a good deal of information for which I thank Felice and her staff. But then towards the end of the summer, the ideas and the job reverted back to the working group on third parties, and that is what we're here now to examine. I'll be over my adrenaline rush in a little while and my voice will quick cracking but I haven't gotten there yet, folks. [Laughter.] DR. PELIAS: I'd like to invite you to look at the handouts, the materials that were given to you. I sent them out about a week ago and I think most of you got them at that time and I am going to work my way through them, and then get down to what we might consider as guidelines for OHRP. The first question is why is third party information important to research and this is divided into considerations that arise in the world of genetics and genetics research and then we will look at social and behavioral sciences. In the world of genetics, which I am most intimately familiar with, we have been taking family histories ever since genetics research began. These days it is particularly important to have good family histories because, as we find out more and more about genes and how they work, we find that this kind of information is helping to point us towards various courses of management or therapy for people who are having trouble with their genes. Family histories have also given us a tremendous amount of information in finding and tracking genes. One of the things that we do with gusto is to construct pedigree charts and find out all about everybody's relatives and we do that with good purpose. A lot of information that we collect when we write down family histories comes from information that is common knowledge within families. I think this is a very important thing to remember. The idea that there is information shared within families that we could say is in the public domain, at least the public domain within the families themselves. Other times we write down information in these family history interviews that qualifies as real hearsay. That is we have situations in which family relationships are strained and we have situations in which families are separated by very large distances. So what we hear from the people we are interviewing really is hearsay. A lot of times the information that we gather in our family history interviews is immensely private and very sensitive. Now that I have brought those two words into the discussion, private and sensitive, I would like to make a point, a very important point, and that is the perception of private and sensitivity varies in genetic studies from disease to disease, from family to family, and from study to study. The bottom of the paragraph says the main message for genetic studies and that is that we would expect all family history information to be treated as maximally sensitive and deserving of the strictest safeguards of confidentiality. This thing will repeat itself over and over again as we go through this, this afternoon. With regard to social and behavioral science, we are talking in terms of social contexts that have to do with the environments and the social networks and the social relationships that people find themselves in. It has to do with perceptions and experiences and feelings and knowledge, and interactions, and I would stress again at the bottom of that paragraph that all of the information that is collected in social and behavioral sciences should be treated with the highest standards of confidentiality. I have included in the handout material a repetition of defining the definition of human subjects that we find in the Code of Federal Regulations, and that is just for our use. There's no use repeating it here. With regard to the status of information that human subjects provide about third parties, I will start out by saying that our working group is divided on this issue. With regard to numbers of who are some people and who are other people, I will tell you that we divided six to five on these issues so some people are six or seven if I include myself and the other people are five voting members. So it is so close that we are bringing these issues to this committee for help in making decisions. Some people in the world group felt that we should make it very clear that third parties are not human subjects. Others said that we should definitely make all third parties human subjects. And the problem arises with regard to the interpretation of the language. There wasn't much problem in deciding that if an investigator has an intervention or an interaction with someone that that individual would be considered a human subject. But the other language which has given us so many headaches is the idea of identifiable private information. This definition is unclear. The definition of human subjects is unclear on what we mean by identifiable private information and I think that we are going to see that this phrase can cover a huge spectrum of information. What is clear to us is that IRBs need some serious guidance about how to evaluate the risks to human subjects and we think that they should be allowed to do this on a protocol by protocol basis. The reason for this is that we don't think that we can establish a dichotomy about defining human subjects or not human subjects. We think that the IRBs are going to have to consider this on a protocol by protocol basis. We noted that the Code of Federal Regulations does not specifically address information that human beings provide -- that human subjects provide about other persons or third parties. Some people note, some of us note the possibility of risks to human subjects themselves, that is the people we interact directly with if third parties are not otherwise enrolled as human subjects. Others of us are more concerned about how these risks to human subjects are managed and again there is a consensus, I think, among the whole group that IRBs must ensure the confidentiality of all data provided by human subjects. Let's move on. In genetics research a third party may consent to be contacted -- we are now at the top of page 4 -- may consent to be contacted either through the family member who is already the subject, that is the person we are talking with face-to-face or through a health care provider, somebody who would urge an individual to enter a study, or maybe even sometimes a third party volunteers to enter a study. But what is important is that that individual becomes a human subject when contact is established at the time that an interview is conducted with an investigator. With regard to the social and behavioral sciences, information that is offered by human subjects about third parties seldom results in personal contact. So this does not become so much of an issue in social and behavioral sciences as it does in the world of genetics. The next section is a short discussion of third parties and waiver of informed consent. Some of the folks in the working group held the view that informed consent and its waiver applied only to human subjects and others were more concerned about issues of consent and the idea that if a subject supplies information, identifiable private information about others, that those individuals would become human subjects, automatically become human subjects as well. And this is going to introduce a little wrinkle in how we make contact with those people. In most instances, consent of third parties would be waived after an examination by the IRB. Now if one assumes that the regulations do or should apply to third parties, we may have a real problem about how to obtain informed consent from those third parties or how to seek a waiver of informed consent, or how to justify to an IRB that the research should be exempt from any IRB. The example that I will give to you is the idea that if we are talking about a -- if I am a subject talking to an investigator about a third party and this information is germane to me as well as to the third party, how are we going to notify that third party without invading his or her privacy that this kind of information has been exchanged and we are recording this information? I think that is a very tough question that we are going to have to grapple with. It can be summarized in this way whether it is possible for an investigator to protect simultaneously the interests of both the human subjects and the third parties as well. The IRB has to honor the ethical responsibilities to the human subjects but then we also have to consider what is happening in relation to the third parties and we have to define and weigh their interests as well. So there then comes a question, and this is a parenthetical that is not included in the write up, about whether or not we should have some sort of new officer or new office or new role defined with IRBs that would be specifically dedicated to protection of third party interests. This is a question that we might get to some time this afternoon. With regard to criteria for waiver of informed consent, the IRB has to evaluate minimal risk across the whole study and give its attention to the safety and interest of the human subjects relative to the interests of the third parties. For an IRB to say that the research carries minimal risk if sensitive information is conveyed about a third party would be difficult unless the data remained confidential. And here again we bring up this idea of confidentiality of all of the information that we collect from third parties or about third parties. That is important. Moving on down. The strong consensus of the working group is that the best mechanism for honoring and protecting the rights and welfare of the research subject while still protecting third parties from an inadvertent disclosure of information is to enhance significantly the protections of all of those data. For the sake of getting on into our discussion, and I again thank Felice for this, we have somewhat modified recommendations here on pages 6 and 7 and 8. But I would like -- since we spent a lot of time talking about recommendations and rationale and some of these ideas or many of these ideas are repetitious of what we have been through with social and behavioral sciences, I am concerned that we are going eventually to get to a position where we have to make recommendations for guidelines to OHRP. So as I went through this and as we went through this in our three conference calls and I tried to distill this information down into so what is the bottom line, where are we going to be with all of these recommendations and rationales, I did go through it one more time and put together some suggested guidelines, and this is what we might work on very seriously. I would like to say this is just a first stab at guidelines since I have never written guidelines before, and I expect that they will be expanded and that there will be explication of each one of these points but these have been put down as a distillation of our conference calls which were many hours of conference calls, and we will see where we go through it from here. On page 9 you see the beginning of the suggested guidelines and I have divided this into three sections. Who are the parties to research projects? What is the business about confidentiality? And what will be the role and responsibility of the IRBs? Parties to research projects, this is number one, will include research investigators or their agents, human subjects who interact personally with research investigators, and third parties about whom researchers obtain information from human subjects but who have no personal interaction with the researchers. Secondly, third parties may become human subject if they are encouraged to participate in research protocol by relatives or by physicians or other professionals and if they voluntarily enter the protocol. Third parties may become human subject if they enter a research protocol on their own initiative. A third party becomes a human subject with respect to the procedures and protections of informed consent when he or she has contact with a research investigator or an agent and enters the protocol. Please note that number two here is divided into the first paragraph that is separated from the second paragraph by capital letters that say "OR ALTERNATIVELY" and this reflects the fact that our working group is so closely divided on this issue so we are going to have to come to some sort of decision about this sooner or later. The "OR ALTERNATIVELY" says third parties who are identified by human subjects become human subjects themselves. The requirement for informed consent or waiver of consent from third parties is a question for IRB judgment. Moving on to confidentiality of research information. There really is only one paragraph here and that is the first one. All research information gathered from human subjects about themselves or others from medical records or from stored tissue samples should be regarded as maximally sensitive and, therefore, deserving of the strictest safeguards of confidentiality. The second paragraph really is more or less of an explication of the first. This was a mistake on my part. It just says we want to encourage candor and truth and all this sort of stuff so that the information that we collect is accurate. And on the last page with regard to role and responsibility of the IRB, first of all, because identifiable private information is increasingly difficult to define, the IRB is charged with determining the nature, sensitivity and relevance of research information that is sought by a research investigator. Because research projects cross a spectrum of information about human subjects and third parties, the IRB should evaluate on a project-by-project basis what information is acceptably identifiable and/or private in the context of the study and the provisions for confidentiality. Secondly, the IRB is charged with evaluating the status of third parties, the protocol for collecting information about third parties, and the procedures for maintaining confidentiality of research information. Relevant considerations include the extent and accuracy of identification of third parties, the relevance of information sought to the goals of the research project, and possibly effects of the information on human subjects or third parties. Thirdly, the IRB should be aware that any information sought about third parties may inhere in both the human subject and the third party. The gathering of such information should be relevant to the study and justified by the scientific rationale of the study. Finally, the research investigator has the responsibility to present the research protocol in clear terms and to justify the collection of information both from human subjects and about third parties. The IRB has the responsibility to approve or disapprove the study and to suggest appropriate modifications of the research protocol and the IRB may enlist expert advice on projects and protocols that are of a highly specialized nature. So we're defining the onus where it should lie. The investigator should make very clear what he or she is planning to do and then the IRB will pass judgment on it and/or enlist expert advice. So that is our summary and now I invite you to have at it. [Laughter.] CHAIRPERSON MARSHALL: Mary Kay, I want to commend you on the work that you have done and how nicely it is organized in terms of being concrete and pragmatic, and something that we can dig our teeth into as a committee. So I see Elliot's hand. Sandy, thank you. Elliot? DR. DORFF: First of all, with absolutely no -- I just can't avoid this. The rabbi of a congregation is told by its president that the congregation's board has voted to congratulate him on his 50th birthday, the vote was 6:5. [Laughter.] DR. DORFF: It is now out. Okay. No offense to the working group. All right. I have four points. They are all small but in any case the first thing is really not a question to you so much, although -- I mean to your group, although you would be the one as a geneticist who is -- would be interested in this. At some point or another -- I am on page 2, just coming from the regs -- human subject means a living individual about whom, so on and so on. At some point or another we have to at least define out of our purview if that is what we're going to do things like genetic, you know, tissues and fetuses and things on that order. I mean, if we are going to do that. But at some point or another no where in any of the things that we have talked about so far at least that I remember have we ever talked about what a living individual mean, and maybe we shouldn't. Okay. Given the politics about all of that. But at some point or another it seems to me that we at least should either decide not to deal with it or to deal with it. The second point I had is on page five. And again it recurs in your summary and also on page seven. The strong consensus at the working group at the bottom of the page is that the best mechanism for honoring and protecting the rights and welfare of the research subject is to enhance significantly the protections of data. My question is -- I mean, after it is all said and done, you mention later on page seven that there are some laws that would impede that in things like required reporting of, you know, abuse cases. But there also are court procedures. I mean, courts can invade data and so the question is, you know, at some point or another, I mean, I think we have to at least inform people that no matter what we do could be invaded. You know, in more -- in general, your data will be protected but you should know that a judge could decree that it will be opened and that that at least has to be -- I don't know what the limits of that are by the way. We can talk to lawyers about that but in any case I think that that has to be at least mentioned. Page eight, the point that you raise here is something that I found absolutely fascinating and I had not thought about before. Namely that asking third parties for consent could actually undermine their confidentiality. And I am just -- this was all new to me and I was just sort of wondering how new it is to IRBs and how we train IRBs to recognize that even if they are going to be zealous about protecting confidentiality -- in other words, they start out with this kind of zeal -- that the very act of asking for consent from third parties may, in fact, undermine the confidentiality of those third parties. That is a thing, I think, that has to be at least -- that people have to become sensitized to. I mean, I had not thought about it before I read your document. DR. PELIAS: Is it okay if I answer that? DR. DORFF: Sure. DR. PELIAS: My own experience in this area comes from collecting family history information for genetic studies and we may find a nuclear family that we talk to and we record a lot of information from and they may tell us that maternal uncle or cousins or paternal, whatever, are -- have a disease or a mental illness and so on and so forth. We may know who those people. We even may know where the live and what their full names are and so on but the way that we go into the family, that is go out into the family to get more information, is to ask the people that we are interviewing face-to-face, the human subjects in all of these definitions, we ask them if they would be willing for us to contact so and so out there to see if we could get more information, corroborating information, more information about diagnoses and so on and so forth. The human subject says, "Sure, you can call up my uncle. I'll be glad to have you talk to him." And then we go one step further and say, "Would you mind asking your uncle if he would let us call him so that we don't go jumping into that individual's life with no introduction whatsoever." This is the kind of invasion of privacy that I am particularly in tune with. If our human subject says, "No, you may not. I don't want you talking to him," then we let it drop. We may get into that subject through another route in the family but the idea is to respect what the subject we are talking -- the people that we are talking to and also to be very careful of not invading the privacy of those relatives out there who don't even know we're talking about them. So, anyway, you had another one? DR. DORFF: Right. It is on page six at the bottom of the page. "Thus if the purpose of collecting information about others is to answer questions about those other persons." Isn't that always the case when you're talking about third parties? I was -- DR. PELIAS: Well, we may be asking them about information about themselves when we get into those third parties but then we may take it another step further and go to other relatives in the family. DR. DORFF: I was reading at the bottom of page six in italics. It says, "The purpose of collecting information about others is to answer questions about those other persons as against -- " see, that is what I was asking in other words. In contrast to asking questions about themselves or about other members of a family. Is that what you have in mind? DR. PELIAS: Well, if I am talking to a human subject and I go out into the family with permission to talk to those others, I certainly want to ask them the same sort of questions to corroborate what we've been told in the first interview. There will be differences of perceptions and ideas of who is sick and who is not sick and how bad it is and so on and so forth. So what we hear from our human subject we would like to have confirmed by the third parties when we get to them but we need to correlate all this information. Sometimes a human subject is completely wrong about the disease status of somebody out there and we need to find out all of this so that our pedigree data will be accurate and we will be able to make a reasonable genetic study about what is going on in the family. DR. CHODOSH: I don't know where to start but I'll try. I, being very naive about this, read the definition in the regulations as not necessarily having to have us deal with third parties. It says that human subject means a living individual about whom an investigator conducting research -- and go to two -- obtains identifiable private information. It doesn't say from whom, how or whatever. It seems to me to be explicitly clear that that automatically means that anyone that you obtain information on is a human subject. There's no explanation of how you get around that. It seems to me that our working group has been working to get around that. Maybe it needs to be gotten around but I think that within what we have here right now it is often easier -- I mean, you can always go through this whole document and substitute human subject for third party and it reads fine. And it fits within the regulation and does not necessarily require change. That also carries over somewhat into page nine in which the parties to research projects -- it says, "human subjects who interact personally with research investigators," that is B, and then C, "third parties about whom researchers obtain information from human subjects." Well, lots of information is obtained even about bees, not directly from that human subject. So that this doesn't cover it all and it needs to cover it all or else we're not really doing our job properly. But it does fit in with the simple reading of the definition of what a human subject is according to the regs. I probably should stop there and I'll be piping in again later on but there's a basic assumption here that we need to consider third parties when it seems to me that if we consider human subjects and let it fall to the IRB, the investigator to explain it, et cetera, to make the determination of how that individual's information is going to be handled, whether it has to do with the informed consent or whether it has to do with confidentiality or what have you. It is no different than when you look at a human subject per se, that is the kind that we're all talking about it seems, and therefore it should not be such a critical issue. DR. PELIAS: I think part of the distinction that you and I are making has to do with what people we can -- what persons we can legitimately obtain consent form from for participation in the study. DR. CHODOSH: It has nothing to do with -- who a human subject is has nothing to do with informed consent. That something that gets determined afterwards. When you look at the protocol and how that individual is going to be -- has got to be treated. We shouldn't be mixing those two issues. DR. PELIAS: Well, I guess some of us think we can't separate those two issues. DR. CHODOSH: We all have to do that in IRB review. We do have to, in fact, look at those issues separately. DR. MORENO: Can I raise a question for Sandy? On your reading of that line in the regulation, does it matter whether the information that is obtained about say my uncle from me is information that is relevant to the study that is being done or not? DR. CHODOSH: I think those are issues that have to be determined on the basis of reviewing that protocol, what it is after, what kind of -- there are situations -- I am sorry Jeff isn't here -- you know, where clearly almost all the information is obtained from somebody else about the real subject, such as children when the histories are taken from the parents. The parents are not the subject or may be but they don't have to be the subject. The subject is the child. Well, who is the human subject here? It is easier to consider them human subjects and move on from there and let the determination be made from the best judgment of what the investigator feels is appropriate and then on the review of the IRB using judgments which are supposed to be within our capabilities to make such decisions about, one, is it a human -- you know, do we have to go after this? Is it exempt? It could be exempt. You know, can it be waived? You know, or is this really information that is so non-identifiable that it doesn't matter? That is up to an IRB it seems to me. And to confuse it by trying to get algorithms of what is a human subject and what is not a human subject as to how they get treated is going to create more problems for IRBs than they currently have. DR. KOSKI: I mean, we separate the human subjects and the consent all the time. I mean, most of the medical records research that is done is without informed consent from individuals but investigators access the information, it is private sensitive information, under IRB approval but they have protections that are appropriate to make sure that the interest of the individuals about whom they are getting the information are protected. So we do that all the time so I don't see any reason to say that they would have to be coupled in such a way that it would, you know, sort of lead to this, you know, road block. CHAIRPERSON MARSHALL: I have Felice, then Susan, then Adil. DR. LEVINE: It is fortunate I am sitting next to Sandy. We are not going to hit each other but we are book ends here on this. I think, if I can just bounce off of Greg for one moment, the reason why they are coupled, you know, even though each can be determined separately, is that once there is the possibility -- once something is identified as a human subject, while consent may be waived, consent may not be waived. And this is -- so there is the conceivable possibility that the interest of the indexed human subject could conflict with the interest of other referenced persons. And this is getting to something that I think Elliot picked up in what has become this draft document, which was -- kind of that add on sentence came -- in a way came from somewhere else originally and this heroic job that M.K. and F.J. were sort of seeking to do. If I can -- the reason why it is 6:5 in the Third Parties Group, and I will say in the Social and Behavioral Sciences Group, while not a head count, where the vast majority of those in the community as well as in the working group have problems with referenced others by a human subject being treated as if they are a human subject for the core reasons that you picked up on. One, it is all about the protection of human subjects. The importance of autonomy of that subject to provide information that has been judged to be relevant to his or her participation in research. The critical importance of protecting the identity of that person and providing them with anonymity in participating in this research and the confidentiality of the information they provide. Now we really don't see a difference between the social science and behavioral science work and the genetics work in any respect but what I will say is that with the vast amount -- there's a huge proportion of research where, indeed, subjects are referring to their social space and in that sense there is almost little work that isn't making reference to others. And for a subject to potentially run the risk as subject, and to be chilled by the possibility of some other person having to be contacted or -- and as that work becomes higher in risk, which is why we use the spouse abuse, but we're not trying to be alarmists -- as the work becomes more sensitive you are running more risk to the human subject when they are making reference to others. An employer making reference to an employee. Or as I said to Mary Faith if there was a study of chairs of FACA committees and she was trying to report on the dynamics of interaction in our group, she would be constrained from perhaps making certain sorts of references that would limit her ability to participate in that work for fear that she was not operating in the autonomous way that we're trying to protect human subjects. So in that sense who a human subject is, is the core issue in this challenge. Now we could take that and engage in big philosophical debates or we could, I think, really focus our work on whether subjects of study can in a way that is protective of the information they provide about themselves or others, and I prefer referenced others to be the language, I think we should focus on that -- on whether human subjects can make reference to others in protected ways as long as all sense of information is held with the highest degree of confidentiality and not take up all of the other ways things become subjects. I think those are, in my view, for another day and another -- maybe another working group. CHAIRPERSON MARSHALL: Greg, you had a comment on point? DR. KOSKI: Well, to just go back quickly to the medical records analogy. If you wanted to, you could easily say that when we do a medical records study, in fact, the patients from whom the information was obtained are third parties because you are not getting the information from them, you are getting the information from the group that holds, okay, the medical record and you're getting this information about those third parties without their consent in those instances. But we do place specific conditions on certain types of information that we have agreed and many state legislatures agree are so highly sensitive that you don't go there without asking consent. Reproductive records, psychiatric records and others. So that somehow -- and I don't think that it is a totally irrelevant analogy because you're getting information about individuals with whom you have no direct relationship through another resource, if you will. So that it seems to me we already have some workable models there and if we again focus on the fact, you know, that these are all sort of human subjects then the goal should be to make sure that we have appropriate protections for them then, you know, we have a framework that would serve us well. CHAIRPERSON MARSHALL: Susan? If you are on point, yes. DR. CHODOSH: I am sorry. Along the lines, though, of the worst case scenario, say spousal abuse. Those are exactly the protocols that need the most careful attention by an IRB. DR. LEVINE: No one is saying that they don't. Of course, they do. DR. CHODOSH: It is the same story. It is not a question of whether you consider those to be subjects or not. It is the consideration that an IRB gives you. For us to devote so much time to trying to figure out who a human subject is divorces ourselves from the issue of these are tough decisions sometimes that have to be made case by case by an informed IRB, hopefully, so that it can manage this whole range of things and some things very simply and some things not. DR. LEVINE: Because of the -- in the shadow of the recent ways in which IRBs have really, I will say at the very minimum, been confused, in studies of this genre, they are discussing whether, indeed, information that a subject is providing can be provided without the consent of those that the subject wishes to speak to. So, for example, in a recent study of grandparents raising their grandchildren, a researcher submitted his protocol including interview questions to the IRB and the IRB -- the protocol included how did you happen to become the primary care taker of your grandchild. And the IRB felt that that was a sensitive question because the answer would probably reflect poorly on the child's parent. So that led to quite an extended discussion of whether the parent should consent. Now this was a study of grandparents rearing their grandchildren. We have studies of -- long-term studies looking at the accumulation of wealth in families where there is references to other persons in the family and what their income and financial capacity is. IRBs are now asking whether all of those other persons need to consent to what is known information to the subject. Now in this case of the grandparent study what the IRB apparently decided, hearsay from my point of view, but the IRB decided that the question -- that the question should not be asked and as the researcher, of course, readily saw that once you have a nice rapport with a subject and you're talking with grandparents about the raising of their children, the topic of how it was that they got to raise their children readily came up. Now for research to operate in the shadow of that did nothing but potentially chill a grandparent from participating or unsystematic information that would skew our intelligent inferences about the knowledge we need to have about the study of grandparents. And in that sense the IRB discussion in a relatively, I will say, erroneous way was focusing on other persons in the family as potential subjects and that is what is going on and I think that is why it is functionally an important issue to address. CHAIRPERSON MARSHALL: Susan? MS. KORNETSKY: I have a couple of points from a very practical standpoint and then from trying to think about the regulations. Elliot, you asked about the issue of invading privacy and Mary Kay gave you sort of the example of genetics but I think Greg has also pointed -- and I work very well with analogies about the medical record analogy. We go in -- the investigators go in and look at medical records all the time. And if you keep or want to maintain identifiable private information they are considered a human subject. And so we have to think about informed consent all the time in those situations. Do we get it or do we not get it? And if we do determine that we need to get it, we have the same issue that we think about with invading privacy. These people who came in for clinical care and now all of a sudden we need to approach them. Are we invading their privacy? So there are many models and in the genetic research we have -- it is almost a nice way to be able to handle that but we think about that and deal with that all the time. You know, my gut reaction to this is I would like to find a way from a very practical standpoint and do -- you know, do I now consider third parties human subjects? I think many IRBs don't and I would love to try and find a way to do it. However, when I really read the regulations and think about it, I think we are just really trying to work around things and I just -- I can't reconcile those differences. I don't see how there's any way when we consider reviewing medical records about someone -- human subjects, how we can get away from thinking about that here. Now some of the examples that are given, if they are considered human subjects, we take that situation, there is a way to think about -- doesn't automatically mean we are going to have to get informed consent from everyone? And some of the examples that are being given would be very good rationale why it would not be practicable to do the research, to get informed consent and do the research. So I think there are other methods that certainly can be used if we do consider them research subjects and there are four criteria that can be met. And some of the examples given about spousal abuse and all these different type things are exact reasons why probably it may not be practical to get informed consent and do that research. So there are other mechanisms of working around it. I am very conflicted because I am sort of speaking what I didn't think I was going to be speaking about but I have -- I really sort of said, no -- when I first heard this, of course, they're not human subjects but really thinking and reading and contemplating it, I just don't see any way around it. CHAIRPERSON MARSHALL: Mark, are you on point or do you -- I have a list here. MR. BARNES: I am on point but if you -- other people may be on point, too. CHAIRPERSON MARSHALL: Well, I have Adil and Abbey so why don't you go ahead. MR. BARNES: Susan, I disagree with you and I disagree with you for a number of reasons. Let me give you one within the Common Rule and then one using the medical record analogy. Within the Common Rule if -- I mean, as Greg pointed out, there is this mechanism. As you pointed out, there is a mechanism to waive informed consent. That is something we do every day in retrospective chart reviews and that is not a problem and you meet the four point test of the minimal risk and not practicable, et cetera. But when you do that, when you -- this is on a practical level -- then you -- what the IRB is considering is the risk to those people whose records are there. They don't know even at that point who else is in the record. So in order to really fulfill the criteria you have to do like -- you know, you'd have to do a review preparatory to research, understand all the information that is in those records about everybody else who may be imbedded in the record, and then you would have to apply the four part test to those people as well as the index subjects. So what I am saying is that it is fine on one level to say that we do have a way around waiving informed consent within the Common Rule but when you really get into it then, you know, becomes much more complicated than simply, you know, waiving the consent if you are applying in an honest way the four part test. Not to imply that you wouldn't apply it honestly. [Laughter.] MR. BARNES: So that is one point. The other point on the medical record issue, though, is that one has to think in this case that although what we're really talking about is within the Common Rule and research regulation that the analog is -- the real basic -- the base issue, getting away from the legalities and -- I mean, because you can change the law if you want to but the principle -- I mean the kind of principle that you extract from it is whether when there is a record about one person and then that -- and then from that person you're discovering other information about lots of other people and their family or their social circle or their sexual context or drug use, you know, contacts or whatever it might be. Then -- or even their employer if they are mad at them and they talk about them in psychotherapy or whatever it is. And once you get -- then if you think that the subject of that record is beyond the index case, himself or herself, and you apply that standard not just to research but you apply it over to the medical record, then that person can't really consent even to the release of their own medical record because the rights of all those other people in here in their medical record. So in other words, what you would be -- you would have an infinite regression of having to tease apart everyone's medical record and no one could ever consent to the release of their medical record if it had information about other people in it. You'd have to go back to those people and get their consent. And so you have an infinite regression and you end up with really an impracticability, I think, you know, both in the research context and also in the -- you know, applying that principle to the extra research, regular medical records context. And none of that is to say that we don't -- we shouldn't, you know, go down the line of trying to protect those people's privacy but I would think that one would have to come up with some kind of middle ground like I think that Mary Kay and others did of trying to respect the third parties privacy rights, while at the same time not really -- you know, not crippling the way that we do things. CHAIRPERSON MARSHALL: Sure, Sandy, go ahead. DR. CHODOSH: Mark, you know, you are not supposed to go into medical records and just go on a fishing expedition. You're supposed to have a protocol which says what information you're going to obtain. And if it does, indeed, include information about your family, that has to be part of the protocol and it has to be considered by the IRB. So it is not as if you're going to stumble on stuff, you know. If you stumble on it and you plan to use it, you shouldn't be using it unless you go back and get a new approval. MR. BARNES: Can I respond to that? DR. CHODOSH: Yes, please do. MR. BARNES: No, I understand that we're not talking about the regular medical records release research. I understand that. But what about if you -- but let's say you're doing a retrospective medical chart review and you come across information that you didn't anticipate but that has some clinical relevance? You go back to the -- you don't -- I mean, one would not go back to the IRB and say, you know, in three of the 100 records that we reviewed, we found some additional information which seemed to be bear on this. I mean, one would not normally do that. DR. CHODOSH: We are talking about information about someone beyond who the medical record is about. We are talking about the third party business, not about the information about that individual. MR. BARNES: What I am saying is that when you -- if those people themselves, those extra three people of the 100 -- let's say you have 100 records in a retrospective chart survey that has been approved under a waiver, under the four criteria, and you have a protocol because you're looking at -- you know, at the incidence of cancer on particular populations or something. And you -- and in reviewing those records your research assistant or you as the PI or your co-PI come across three instances in those 100 where there is a family history of cancer. Now one would not go back to the IRB and say, "Can I report on those when I write up my research findings that three of these 100 people had a family history of cancer?" One would not do that. So I agree with you if what you did from that is you found the names of three more people and then went back to their records, et cetera, then you would have to go to the IRB. But that is not what we're talking about and I would not disagree with you that in that case you would have to go to the IRB because you would be asking for more consent to go back to those people as cases qua cases, subject qua -- as a subject qua subject. But what I am saying is that in the other case where you're just looking at the -- and reporting the generic information, I don't think you have to do that. DR. PELIAS: I'd like to ask whether or not you could anticipate that in your original application to the IRB and say we expect to find a few cases in which there's familial cancer and we'd like to have consent or permission to pursue that when we find them? MR. BARNES: You could do that but obviously, you know, the IRB would only have jurisdiction. You know, the IRB -- the institution would only be able to authorize the release of records within that institution unless you had some IRB that had, you know, some cooperative arrangement or something. DR. KOSKI: When you are going to do a medical records review for a particular project you tell what information you're looking for, you get the information that you have said you're going to get, and you're not supposed to be, you know, using other parts of the record that are not part of what you have said you were going to do. So if you were going to do something different that would be considered a change in your study and you should go back to the IRB for approval. CHAIRPERSON MARSHALL: Abbey? MS. MEYERS: I have to tell you that my entire family has gone through an in-depth genetic study and they ended up finding relatives I didn't know that I had. They went to, I don't know how many different states to find all of these relatives. A lot of the information that they gathered was not just about the living relatives but dead relatives. And when I think about third parties, for example, my grandmother was dead at the time, my husband's grandmother was 99 years old and still alive but, you know, should we say that a dead person is a third party? If we go back to the definition a human subject is a living individual. Does that mean they shouldn't ask any questions about dead people because they can't give their informed consent? I mean, the whole thing just -- it gets more and more complicated as you look at and I think that we looked at this definition and we said this is not great stuff. It really -- really if we had our druthers, we would rewrite this definition of human subjects but it is impossible because if you rewrite it, you have to go through the whole Common Rule thing. You have to get all these agencies to approve of a new definition. It will take eternity to do it. So working with this definition, how do we look at this third subject question? And I think that you don't want to inhibit a lot of the genetic research that is going on now and that will go on in the future that involves, for example, some research subject who is talking to investigators saying that their Uncle Harry really believed that he was Napoleon and not being able to say it because he's a third party. You really want the person who is face-to-face with the investigator to be able to have the freedom to talk about things in their family that knowing -- and this is really -- knowing that whatever they say to that investigator will remain confidential and nobody will ever know where Uncle Harry lives or what his address is and what his social security number is. And I think that is what is really reflected here in what Mary Kay wrote because throughout it is repeated that confidentiality is primary in this. DR. LEVINE: I would like to follow up on this, on what Abbey has said and go to the question that Elliot raised about protection of this kind of information against subpoenas or any other legal procedures. I think that we can do everything we can with certificates of confidentiality and we can guarantee all that is humanly possible with regard to confidentiality but I do believe that we need to be candid with people that we're talking to by saying that while we will do everything we can there is always a possibility, and I think that is part of being forthright with the people that you're talking to about very, very personal and sensitive information. DR. PELIAS: And if I could add, I am sure to retain recommendation from all the various drafts that we viewed that to be the case, whether one was talking about themselves solely or providing information -- or discussing other aspects of their life that happens to involve others that we always needed to inform subjects about the potential limits of confidentiality. And that would obtain whether they were only speaking about themselves or if they were speaking interactionally about referenced others. CHAIRPERSON MARSHALL: Adil and then Judy? DR. SHAMOO: Thank you. Could we -- I am talking in terms of who is a research subject or third party. Could we have a degree of relationship or degree of concordance, you know, the concordance in genetic information is 50, 80, 90 percent, make a difference whether the third party is a human subject or not? Could we make such a gradation if it is only concordance of some kind of genetic illness for third party? It doesn't have to be all social or behavioral or spousal abuse. I am talking only now genetic information. That degree of relatedness have involvement of who should or should not be a human subject as a third party because I don't see it in the list of the third parties. Can that narrow it down? Your face looks like you don't follow what I am talking about. DR. PELIAS: You're right, I am not following. [Laughter.] DR. SHAMOO: Okay. If someone has Huntington's disease and he's my brother, and we're identical twins -- I am just giving you an extreme example, if I give that information, may as well say that to my brother. So my attitude is my brother is really another human subject. I am saying that degree of relationship from between 100 percent concordance to 90, 70, 30, 20, down to zero, is that degree of relationship and sensitivity of risk, has anything -- I don't see it here -- in definition of potentiality of third party being human subject or not? DR. PELIAS: I will try to answer by saying that we take into consideration the fact that information that a human subject has about himself and his family necessarily inheres in other individuals that you may not ever see because it is family information. What I call the limited public domain of the family. It doesn't make other family members necessarily human subjects but we have a question of a body of information that is out there that is my information as a human subject but is also information that inheres in other individuals. And this is one reason why I want to have the IRBs charged with looking at this kind of a study very, very carefully and making a judgment, as I have written here, on a protocol by protocol basis. CHAIRPERSON MARSHALL: Judy? DR. SIEGEL: I think I brought this up not last time but the time before that we spoke about this and, unfortunately, I did not bring the FDA regulations. I think if I remember correctly, if you read the FDA definition of a human subject, we would not be having this conversation. Because I think -- and if there is someone here from the FDA who can bail me out with the facts, please help, because I think the FDA regulations speak more directly to intervention. I mean, they are talking about -- this is usually a drug trial or device trial. So it is -- the person who receives the intervention whom we get information for, which I think again now suddenly a human subject is in a whole different domain. But we are in those kinds of trials actually collecting in some cases lots of information about third parties. CHAIRPERSON MARSHALL: Tell us for the record who you are. DR. LEE: I am Bonnie Lee from the Food and Drug Administration. And the definition of a human subject off the top of my head so it is not absolutely precise but it is basically an individual to whom a test article is administered or who acts as a control in an experiment. DR. SIEGEL: And having said that, where does this discussion and how do we apply this discussion to that situation? MS. MEYERS: Can I say something? The main point here is that the FDA never signed on to the Common Rule and when we make the statement, some of us, that private research, privately funded research really ought to fall under the Common Rule and then others say, well, but private research is controlled by the FDA. Well, it is regulated by the FDA but those regulations are not the same as the Common Rule. So what we have here is probably the most important health agency in the U.S. Government that follows its own rules and doesn't follow the rules that the rest of the government is following. DR. KOSKI: I will speak to this, although Bonnie probably should, but as Bonnie has said many times, FDA, in fact, was adopting the Common Rule but doing so within the constraints that it could under its separate statutory authority because it has to apply to that research that they regulate. So, in fact, the FDA regulations, which I say repeatedly are 99.99 percent the same as the Common Rule, okay, are intended to apply to that specific domain of research which is under FDA's regulatory authority and, you know, clearly there's interest in making this as uniform as we can. I think the fact of the matter is that when FDA came up with their definition for 21 CFR 1556 for human subjects, this issue of the sensitivity of information obtained from one individual about other individuals just wasn't on the radar screen. So it wasn't an attempt to try, you know, sort of simplify things at that time. It is just that is where we were at back in 1991 when all of these regulations were adopted. So we're dealing now with a new set of issues and that is why it is on the table regardless of what the FDA regulations say as a definition for a human subject. CHAIRPERSON MARSHALL: Let me go ahead at this point and just open the floor up to ex officio members and public members, anyone who would like to join the discussion, please, feel free to do so. DR. SHAMOO: I just want to say something for FDA. If anything, Abbey, I think they are much more specific than 45 CFR 46 in some cases. MS. MEYERS: You know, it is interesting because I pulled out the regulations a few weeks ago and looked at them and they are only about three pages long. But then there is about 20 to 30 pages of exceptions and waivers. DR. SHAMOO: This is only a small portion of it. There is a huge number of them. There are at least 30 pages. DR. AUSTIN: Finley Austin from Hoffman La Roche and I serve as a consultant on the Third Party Group. A couple of comments I'd like to make as far as the last discussion. I think I'd like to put forth a little bit more strongly that part of the reason, I believe, there is a distinction is that in the Belmont Report, World Medical Association Declaration of Helsinki, all of the sort of ethical preludes that led up to a lot of regulations have always distinguished between research for pure knowledge generating sake versus research and a health care delivery setting. And all have acknowledged that when delivery of care is involved in the research, family history, all of this, anything that would be standard of care in the medical setting is allowed in the research setting. And I think that distinction is very important and needs to be there and it is my assumption that is part of the reason why the FDA definition is limited in its scope. And I -- you know, of course, this gets into how you see the third parties. I think everyone on the group knows that my leaning is more towards let's view people as subjects and then let's deal with when it is appropriate to waive consent versus it is not appropriate to waive consent. And I think in listening to this discussion I have heard the word "impracticable" brought up numerous, numerous times. Well, the regulation says if it is impracticable then that is one criteria of waiver being met. The one word I have not heard this afternoon is "voluntary" because to me the crux of it comes down to, you know, my sister can go participate in a research study that may involve providing family history and I may not know about it. I probably won't care about it as long as it doesn't come back to haunt me in some way that I didn't expect and I get blind-sided about. But I guess I wouldn't see myself as uninvolved because I could do a study of domineering older sister syndrome and I could just as easily write -- you know, and I could five graduate students working on the project and we could have one paper that says younger sisters have higher rates of hospitalization for depression in families reporting dominating older sister syndrome. You know, I mean, you could -- and then you could say, well, was that research about the younger sister or was it about -- I will use the genetic term -- the proband because that is what I am used to. And so I think -- and everyone knows I don't want to stifle research either but to me I think what we need to look at is when is voluntariness waive-able, when does the competing interest between my sister and I, who should win in what situation, whose rights should supersede the right of the other? Her right to participate in the study? If it is a genetic study and they have to come get bloods from family members, and I say, "No," and she can't find enough cooperative family members, she's out of the study anyway. So I don't -- I think IRBs if we educate them, they can think through these rationally and come to good answers. My question would be is the bar on the waiver if we are looking at changing the regs, is the bar on waivers maybe too high in some instances of having to meet each of the four criteria solidly? And is there any wriggle room or is there any room in the interpretation in that line? I will leave my remarks there. DR. LEE: Bonnie Lee from FDA again. I was also a member of the Third Party Working Group and one of the issues that I raised was even with FDA's different definition of what constitutes a human subject, there would certainly be instances where you could have FDA regulated research involving what we are calling here third parties. FDA's informed consent regulations because of our statute, however, do not set forth the four criteria contained in the Common Rule for waiver of informed consent. Our criteria for waiving informed consent tends to be solely in life-threatening situations. Thus if you were to define third parties as being human subjects and this were to be applied under the existing regulations to FDA regulated research, that would mean that you would be obliged to get informed consent from those third parties. I'd also like to just speak briefly from the point of view of a mother of identical twins as well as I am the daughter of identical twins, so you can figure it out, there are clearly two different sets of individuals. But what I hear both from my mother as well as from my daughters, and her twin sister as well, is that they are their own individuals. They have every right to speak for themselves, to make their own determinations for themselves, and I think that if they were here they would speak very strongly that they have -- would have rights individually as a research subject that should not be impinged upon solely because they happen to be identical twins and what they may say is not something that their sister would otherwise want to reveal but who is not part of that research study either. So there again relying on privacy and confidentiality, I think, would be the safeguard that they would expect in those situations. DR. WATERMAN: I am Paula Waterman and I am standing in for John Mather for the Department of Veterans Affairs but I am actually going to speak sort of on my own. One of the things that interests me about this discussion is the centrality of it and yet to my mind, and I admit my bias, one of the strengths of the IRBs is that they represent local communities and local attitudes. And I don't think there's any one solution that will fit all. There are some communities for whom individual privacy is enormously important and they would be very, very, very offended at the thought that someone else could give out their information. As I remember, part of the problem at VCU was a parent who complained about information revealed by his daughter. There are other communities, perhaps at universities, where there is a lot of research going on where they just shrug it off and I think we can't expect you to supply all the answers because in all honesty I think there may be local communities who have their own ideas. CHAIRPERSON MARSHALL: You're implying that we don't have all the answers? [Laughter.] DR. PELIAS: I want to agree with you that we don't have all of the answers and a lot of our decisions are made within local communities. This is, I think, one reason for the fact that over and over again on the conference calls we've had over the last six months is that it really boils down to the idea that the local IRB is going to have to judge the situation and ask questions or send it back to the investigator for a clarification and then the local IRB is going to have to be the one to make the decision. I would also add that knowing my own IRB in my institution, I don't think they're going to like that but I think that is where the decisions are going to have to be made. DR. LEVINE: I mean, in a way that has a face attraction but on the other hand IRBs operate under the shadow of believing that there are some "as if" policies defining what seems clear to my colleague on my right, is not at all clear to me but that was, indeed, the intent. I think that the intent was probably a little anticipation of what happens when subjects are referencing others. And I think Finley's example is a very good example because it was my example in the reverse and that is the emphasis was, I said, on the autonomy of an individual. Bonnie caught it. Of the autonomy of the individual and that is what the regulations are about. The autonomy of the individual to provide information that is their -- I am going to say that is their information. Mary Kay has a nice way of putting it in terms -- but it is -- once shared, it isn't confidential. Now if you're saying I have in my briefcase my mother's psychiatric record, should I, as a subject, turn it over to a researcher, well, that is, you know, a kind of , well, what am I doing with them. In any event -- and should I turn them over. But what I know and what I -- you know, what I know is really my -- it is my information. It may also belong to someone else and what we are fundamentally talking about is whether we in any way should limit an individual's capacity or put it up for an IRB to possibly determine that we should limit the capacity. We all agree with respect to the confidentiality of this information and issues of privacy but I do think that in the end we will need to provide some guidance with respect to whether subjects of study can meaningfully discuss, meaning meaningfully related to the research, meaningfully discuss information that is their information that happens to involve others. DR. KOSKI: Along those lines we heard earlier Mary Kay and others have tried to put this discussion in the context of the routine history and physical examination because, indeed, the regulations make reference to that type of thing and defining risk levels and so on. In a routine medical history the information that is sought from the patient about family members is, you know, what did your grandmother die of. Well, she died of cancer. Is there any heart disease in your family, diabetes, so on. Those are the kinds of questions that I would think most reasonable people would agree probably, you know, used in that context and subject to appropriate confidentiality protections and all carry little -- relatively little risk to the individuals about whom that information is being provided. Okay. That is the third parties. Those are very different kinds of information, okay, than when you ask the question, you know, does your mother use drugs. Okay. Did your mother abuse you as a child? Okay. Have you had sex with your father? Those are questions that come up in the context of some of the very important kind of research that may be done in other settings but it is entirely different information than what you would ask in a routine history sort of thing. I think somehow again we are trying to, you know, sort of shoehorn all of this stuff into a one size fits all thing. I would think that guidance that would say the kinds of stuff that you would normally ask in a routine physical examination or history would be something that would be useful guidance to IRBs but when you get into these much more pointed questions -- indeed, these were the kinds of questions that raised issues in the VCU case where there was a sense from the offended family member that the researchers were going after information that was not within the rightful province of the research participant to reveal about him. And so that is where we somehow, I think, have to target, you know, our guidance to, you know, help the IRBs that have to make decisions about this and the investigators that have to design their studies to be better sensitive to the issues and not just sort of put it all under one umbrella. CHAIRPERSON MARSHALL: Thank you, Greg, that is a perfect segue for me to remark that we have half an hour left and we need to be thinking in terms of where we are going. MS. MEYERS: In reference, though, to what Greg just said. Greg, there are some families that this simple question of whether or not your mother had cancer, to them it is an awful stigma. To them, they won't even say cancer in their family. And it really depends upon how an individual feels about illness. To some people, illness is shameful and in others they're happy to tell anybody they meet what their diagnosis is. And so I don't see a difference. I say to you that if this committee feels that all third parties should be asked for their permission in order for the primary subject to be able to talk about what they know about them then no doctor should be able to take a family history. It is the same thing. DR. SHAMOO: Research is different than health care for direct medical benefit and that is why the bar is higher than health care, Abbey. CHAIRPERSON MARSHALL: Elliot? DR. DORFF: How high does that go? Let's say that you wanted to -- you were researching ways of stopping the spread of AIDS and so you do these histories of people's sexual history, which inevitable involves talking about whatever number of people they've had sex with in terms of their third parties but you're doing it presumably not for the health care of any one of the individuals that you're talking about but for the public policy of trying to figure out how you're going to stop the spread of AIDS. So the distinction between the research and health care, I think, is -- you know, it is like most distinctions. I mean, there are ends to the spectrum where it is very clear that if I have got bronchitis and I went to the doctor for an antibiotic, that is health care. Whereas, if I am talking about, you know, the other end of the spectrum where there's no report of third parties at all, you're just talking about what drug works on -- you name it -- then that is research. But, I mean, it seems to me that it is precisely in these middle areas where -- you know, especially when you're talking about things that might be sensitive. I mean, you know, the person might have had sex with people that don't want to be known to be gay. So, I mean, I think it is -- I think that what we have to recognize here is that one of the real -- one of the -- that the hard issue is precisely the kind of information which would be found to be either embarrassing or shameful or something by the parties involved. So what you're then dealing with is, well, what constitutes shame in a society. And I think what Abbey is arguing is that it is different for different people and I think that that is right. As a matter of fact, in the Talmud there's a very interesting discussion about this which I think I mentioned in a previous meeting as to whether shame means only that which the person feels or whether it is that which is understood to be shameful by the other members of the society. Whether the person, you know, feels it or not. I mean, the person might be asleep, for example, or just unaware that it is shameful. And I think that when we're talking about these kinds of things we have to be -- I think we need to have some kind of a statement to say that by and large in society these kinds of things are, indeed -- might be embarrassing and, therefore, we have to have more controls over, you know, what gets said about these kinds of things. Other things that normally are not shameful to most people in society still are private and that, therefore, you know, the level of confidentiality has to -- at least a level of confidentiality has to be preserved even though you may want to -- you know, to allow it with less safeguards if it is not normally seen to be shameful, for example, to have -- you know, to be in a family that has a history of breast cancer. Whereas it is -- may be considered to be shameful to be in a family that has a history of schizophrenia or something on that order. CHAIRPERSON MARSHALL: I see Belinda. I want to make just one point, Belinda, before you begin and that is to ask members of the committee, and those of you in the audience who have a document in front of you, to look at the 2 -- under the suggested guidelines, number 2. Neither one of those guidelines currently reflects anything about the nature of the information being provided or being disclosed. That is really what we're talking about now. So I'd like for you to be thinking concretely if we're going to move towards endorsing one of these today. How we would incorporate that in and where? And moving forward. Belinda? MS. MEYERS: Page 7? CHAIRPERSON MARSHALL: I am talking about page 9. Belinda? DR. SEATON: Belinda Seaton, National Institutes of Health. I am a member of the NHRPAC working group on this topic as well as the NIH working group that is co-chaired by Sarah and Jim and they're both here in the audience. We all recognize that we have individually very different measures of what is sensitive, what is shameful information, and in order to issue guidance that would be helpful to both our investigators and IRBs, we take into consideration that there are different measures. That is why in both groups we spend a lot of time discussing protecting the data. Data security not only for the index case but for information collected about a third party. The other point I want to raise is the case by case determination by IRB. IRBs, I sense since the VCU situation, are desperately looking for guidance and the guidance cannot be, well, we will do it locally here and then in this locality it may not be quite so -- I absolutely respect the local -- respect for the local community but they are looking for guidance on how to proceed on the subject matter be they in Wichita or in Boston. DR. CHODOSH: Along the lines of confidentiality, it seems to me that one thing that we could do is offer some better protection for such data. You know, we now have a certificate of confidentiality which pertains to certain groups but it is not that easy to get. You have to go to HHS to get it. Whether, in fact, that could be made an easier avenue so that the data when we promise confidentiality we are more reasonably assured that we can do that because right now it is not that easy. In truth, you have to put a whole bunch of qualifications when you offer -- promise confidentiality. Now two things that did come up in our conference call and really in the report is that we -- it seems to me that we all agreed that the concept of private or non-private, that the truth of the matter was that what is private to one person is not private to another, that all data should be considered private. So that part of it doesn't have to be really considered. It is just there and the other is that it is clear that institutional review boards need to be reinforced about what their responsibilities are in all these regards of human subjects, informed consent, waiver, all this sort of thing, exempt categories which have not been used very well. That this is probably as important as anything else that we do in terms of protecting confidentiality or what we want to call it. CHAIRPERSON MARSHALL: I guess I want to, in a sense, call the question before the committee and ask you to look at on page 9 under number 2 the two alternatives, to think about incorporating, or whether we would incorporate into one or the other of them some language about the nature of information provided. But we need to move forward, committee. I am sorry, Mark. I neglected to recognize you. MR. BARNES: Well, my comment, I think, is sort of a segue into answering your question, and this is not just my comment. Susan had, I guess, contemporaneously the same idea when we were talking about it. I mean, in terms of a concrete recommendation, you know, you don't want to get caught up in the legalities. Just because the law says one thing doesn't mean the law can't be changed. So you shouldn't take the world as it is but the world as it should be. CHAIRPERSON MARSHALL: Hear, hear. MR. BARNES: But it seems to me that if we want to avoid the harshest impracticality of saying that every time someone is mentioned in the course of research that all of a sudden they become a research subject and then the things that that would trigger, then -- if you want to avoid that and thus allow -- and it is not even a question of allowing the research subjects to speak about it. That is not the question. They could say whatever they want to. The question is -- the real question is whether the researcher can actively elicit the information and record it. That is the real issue there but that is a side issue. But it seems to me that everyone -- in terms of the points of agreement, I think everyone agrees that where the -- I am going to try to list those because maybe that is a way to get to consensus on this really is to start down the road. I think everyone agrees that where there is information that is intensely private that could have some significant potential for shame, embarrassment, humiliation or even adverse consequences for the -- for a person who is mentioned, that there is a need for the IRB to exercise when it approves the research and looks at the way the research is conducted, there is a need for the IRB to pay attention to the privacy and confidentiality of that information that is collected, the way in which it is collected, the way in which it is interpreted and disposed of and used, the way in which the research subjects, the index subjects themselves are questioned about. So that the questioning is not viewed either by those people or when they go back home and tell their families, is not viewed as abusive. That when the information that is being elicited about other people who are not the index subjects becomes really not about the index subject but about the other people themselves then that is a point at which the IRB ought to think that they are crossing over the line, the researcher is into actually making those people research subjects themselves. And that finally that when there -- that when the information is really recorded about third subject -- third parties in the course of research that in order -- and this is what Susan and I were talking about -- that in order to kind of get around some of the technical problems and also to protect confidentiality that one should try to record -- the researcher should try to record the information in such a way that it would insofar as possible, given the need to protect the integrity of the research, de-link the information about the third party from being identifiable to that third party. And you might be able to do that through -- I mean, if you really had, you know, complicated questioning about intensely private matters, you may want, you know -- the IRB might want -- just as an example -- to require that the -- even the names of the index subjects from whom the information about their moms or dads are elicited be -- they be coded, that the code be linked only with one thing that is held in a -- you know, at midnight in a safe, which can only be unlocked by a person with three toes. I mean, you know, there are different things that you can -- kinds of protections that you can put on that information in order to kind of heighten the safety -- the protections of privacy or confidentiality for it. So that is just to say that, you know, maybe to get some points of consensus to get us down the road. MS. KORNETSKY: I would just like to add -- I mean, I was thinking of the example in genetics if you said a maternal aunt. I mean, you know, is that identifiable private information? To me that is not identifiable, yet you can collect all of the information about that individual. So that is sort of one way that I am trying to sort of come to that. DR. LEVINE: The situation arises when the referenced person is only one person, however masked, or with others it is not -- it isn't a problem when they're labeled A, B, C, D in any way. All of the modes of privacy and confidentiality, de-identifiability, and all of the discussions, there is absolutely 200 percent agreement. And all those we could craft. One issue that you raise in your middle zone is the tough one or the one where there's a 6:5 split in the Third Party Group or 7:5, depending upon how you count, and not that much of a split in the Social and Behavioral Science Group, a more modest minority, and that is putting IRBs in a position when work is on highly sensitive issues because this is the issue where it comes up. Where it will make a determination that something has crossed over the line and what that will do for chilling a range of important work and the capacity of an individual to talk about that. So for adult subjects to be able to talk about their abusive life histories and to limit science on that, limit people who have been in that situation, to be able to have informed science done on that, for those of us in the 7:5 or the 13:16, however it all falls, we thought at that -- and that is why the example that Finley gave versus the one that I gave is right at that margin. That notion stemming right from the Belmont Report and the Code really has to put the emphasis on the index subject. So it is not so -- we could agree on one and three really easy really. It really always gets back to how we are going to make an informed decision about the place of the index subject in talking -- in being asked about a subject that really is his or her own information even though it happens to also relate to others. MR. BARNES: In other contexts where the IRBs exercise judgment like that, for example, in regard to financial conflicts of interest and all sorts of other things that IRBs look at or at least that research administrators look at, you know, we don't say that they must do one thing or another. We simply say that they must understand it, appreciate it, discuss it and manage it. And I guess I don't -- you know, even though it is a tough issue in the middle there, I think it is entirely within the province of the IRB to say, you know, you can't really do this research without -- you know, this research is valuable and you can't really do it without eliciting those histories and it is not practical or anything else to get the consent of all those extra parties. So go ahead and do it but while you do it, since we recognize that it is sensitive information, then be very careful about confidentiality. And in your protocol or in your revised protocol, describe all of those precautions that you're taking. I don't have a problem with -- you know, with vesting that authority on a case by case basis in the IRB as long as they are -- as the guidance document from OHRP really sets forth these are the kinds of things that the IRB needs to consider when they make that decision. DR. FLEISCHMAN: Does the IRB need to say -- MR. BARNES: I don't think in that situation they would need to say that they are waiving consent because I really don't believe that these people are third parties but that is my -- you know, I mean that those people are -- those third parties are research subjects. But I think that -- but what I would recognize in my scheme is that those third parties, although not research subjects, do have -- need to be protected because they have a vested interest because this information has been judged by the IRB to be sensitive, critical, shameful, embarrassing or potentially harmful to people other than the research subjects themselves. CHAIRPERSON MARSHALL: Margaret? MS. BORWHAT: On our conference call we discussed the idea of the intent of the investigator and found it be lacking because what is really important is the protection of the human subject or the third party and the intent of the investigator is really not a good guideline to go by. CHAIRPERSON MARSHALL: Virginia? DR. __________: I am really very concerned about the idea of identifying what is sensitive and what is not sensitive. There is no way to tell. We talk about reproductive histories. For some people having a child out of wedlock is extremely sensitive. For other people it is part of their every day occurrence and it is how their families are formed. So the notion that somebody else could identify for an individual what is sensitive, it just doesn't seem like it is a really viable approach to take. From everything I have heard, there are sufficient protections within the IRB to handle everything that we've talked about without identifying third parties as human subjects. DR. SHAMOO: I think I will talk of my own center. I hope one day in the future we are going to discuss how we strengthen the IRBs themselves and their structure and function. Somehow I have a feeling this body, some of you at least think that IRBs walk on the water and I think they'll sink. They will not be able to walk on water. Here is what Mark said: Exercising judgment, understand it, appreciate it and manage it. Well, they do all that in three minutes. A lot of IRBs are part-time -- not part-time people. They are doing this as a favor to the research institution. They have a huge amount of load of work. Until now they had poor staffing. And then you have an IRB administrator who has somewhere around 50 protocols per meeting and within two hours he has a sheet of paper. Oh, this is scientifically valid, check off, check off point. They have a two or three minute discussion, maybe even a 10 minute discussion. How in the world are you going to do even in ten minutes, which is five times the normal average because GAO said it is two minutes per protocol, exercise judgment, understand it, appreciate it, and manage whatever all these confidentiality and other risk/benefit analysis and, God help us, what else there is attached to it? We are throwing everything into the IRB as if the IRB is such an excellent system. If you remember my first meeting here I said it is the weakest link in the chain of protection of human subjects. That chain is nearly broken and it needs to be fixed before we go on and pile up more work on them. CHAIRPERSON MARSHALL: Thank you, Adil. Ivor? DR. PRITCHARD: Ivor Pritchard, the U.S. Department of Education. I am not speaking on behalf of the U.S. Department of Education. [Laughter.] DR. SHAMOO: Sounds like a Federal Government employee. [Laughter.] DR. SHAMOO: Statement. DR. PRITCHARD: That is bureaucrat to you. DR. SHAMOO: I didn't say that. [Laughter.] DR. SHAMOO: I didn't say that, a Federal employee. DR. PRITCHARD: I'd like to make three points. The first is that it is fairly common in education research for people such as teachers and parents and students to be asked questions only about other people and not about themselves. In that circumstance if you say that third parties are not subjects, it is not research involving human subjects, it never goes to the IRB because there are no subjects. Neither the third party nor the informant because you haven't collected identifiable, you haven't collected data about the living individual who is the person, right, and if the living subject isn't one, if the third party isn't one then I don't think you can count them as a subject either so it is not research involving human subjects. The second is I fear I am missing something but I think that the problem with having to get informed consent from other people you get information about in medical records is pretty much moot because if you go into medical records without the patient's consent then it seems to me that you have to presume that it is still a confidential record. In which case you would be obliged not to seek the consent of any other person because you have to keep the information confidential and so you have to go for a waiver of informed consent for anyone else about whom there is information in that medical record. The third point I wanted to make is that unless you want to take one of the extreme positions of saying that all identifiable information about a third party makes them -- or any identifiable information about a third party makes them a research subject or no information about a third party who is identifiable makes them a research subject, then you have to make the principle of -- the concept of private a functional concept. You have to have some way -- and I think this is what you were driving at. You have to have some way -- you have to at least provide some general guidance as to how IRBs should make a decision about whether the information is private or not because that will serve as the boundary between whether they are a third party or not. There are lots of circumstances I can imagine, and I am sure everybody else can imagine, too. There are lots of circumstances in which even sensitive information about a third party is clearly not private information. All of the people who were asleep in the last 20 minutes might be sensitive but it is not private information. And so I think it is important to look at privacy rather than sensitivity as some working way of making the discrimination of that boundary. CHAIRPERSON MARSHALL: Well, we are close to time. I have some thoughts but I am happy to entertain ideas or suggestions from any committee members as to coming to a consensus and resolution. DR. LEVINE: I am usually in line for reaching consensus, consensi. I have to say that, rightly or wrongly -- I mean, not that this should be controlling if we had a strong consensus that could emerge. I believe as a consultant to the Third Party Group and as chair of the social and behavioral science working group that ultimate recommendations would flow from this group in January. And that, indeed, part of the third party -- that the Third Party Group was going to have an opportunity for more discussion or maybe we could all agree that we would just come down to the same difficult choice point. But I actually thought more deliberation and counsel was going to take place across our -- at least within the Third Party Group and I think Social and Behavioral Science Group thought similarly. I don't mean to eclipse closure if we thought one was jumping out but I have to say that one hasn't -- as someone who has had many sleepless nights trying to argue every point of view possible, one hasn't jumped out and hit me yet. CHAIRPERSON MARSHALL: I guess I would like for us to move along at least to the point where we can garner some input from others. Not that the discussion is final but I would like between now and January to have things -- something posted on the web for input from others rather than simply sending this back to committee. I think Mark has given us some things to consider in terms of evaluation by an IRB or a privacy board or whomever is making these sorts of determinations, and we know that it is likely in the future that a privacy assessment of the risks -- breaches of privacy are going to be much more thorough in terms of protocol review than they are now. I guess one suggestion that I might have is to incorporate Mark's suggestions for privacy review into perhaps one of the two definitions under No. 2 or under the two alternatives under 2 and go from there in terms of a working draft. My sense of the committee, but I'd like to hear about this from you, is that we seem to have a majority of committee members who endorse the first option as opposed to that third parties -- the second, which is third parties are identified by human subjects become human subjects themselves. Am I correct in what I am hearing? DR. PELIAS: We are going to take a vote? CHAIRPERSON MARSHALL: Yes. Let's -- so I am going to sort of put this before you that -- DR. __________: No abstentions. CHAIRPERSON MARSHALL: You may not abstain. [Laughter.] CHAIRPERSON MARSHALL: I guess that there could be two or three options, although two would be nice, is that we take on page 9, under parties to research subjects No. 2, the first paragraph, the first alternative, and we either take it as it is or that we add language regarding the nature of information that is provided in the assessments that would occur in IRB review. My own thinking is that language about the nature of information and how that would be helpful in terms of IRB review would be helpful to IRBs, research participants, subjects, investigators themselves, sort of the whole gamut of the system for protecting research subjects, that that would be of more assistance to IRBs and others than leaving it out. A, A(1) and B or 2. We're on page 9, parties to research subjects. Under 2, the first alternative we would say is A. A(1) would be the first alternative with language incorporated about the nature of information provided and the responsibility of the IRB for exercising judgment, understanding, appreciation and management. That would also, it seems to me, apply to the investigator and other parts of the system as well. Sandy? DR. CHODOSH: As this is worded right now it excludes a whole area of information obtained by other means than through direct contact that you intend. The way it is worded right now sounds like only if you have contact with the individuals do they become -- CHAIRPERSON MARSHALL: So can you suggest alternative language for me? DR. CHODOSH: I have. It is called the alternative. It gets more and more complicated as you go down this road because the way it is written right now, only if you obtain information from the subject about somebody else is that a third party, number one. And if that information obtained about the third party is not obtained from the human subject, according to this they're not even -- they're not part of the game. DR. LEVINE: I might be able to at least offer a semi-friendly modification there. Third parties referenced by indexed subjects become human subjects. CHAIRPERSON MARSHALL: That is nice language. DR. LEVINE: And if I could make one other little friendly change. DR. PELIAS: But, Felice, that is in the very first sentence. DR. LEVINE: Third parties referenced by index subjects. CHAIRPERSON MARSHALL: Index subjects. DR. LEVINE: May become subjects if they are encouraged to participate in research protocols by relatives, physicians, or others since there's a lot of network analysis where it is not relatives or physicians. [Pause.] MR. BARNES: Can I add something to this alternative? CHAIRPERSON MARSHALL: Please do. Let's draft something. MR. BARNES: It is not simply the -- I mean, it is if they're referenced by the -- if it is third parties referenced by the index subject. It is if the third party's information is derived from the index subject and the third party -- and the focus of the research, either through the analysis by the IRB of the effect of the protocol or an analysis of the intent of the research team, is that that third party, himself or herself, is actually one of the foci of the research. And in that case then what this alternative could say is in that case the IRB needs to judge that, needs to understand both the effect -- the intent of the researcher and the effect of the implementation of the research. And it needs to assess whether the impact on third parties is such that they should be considered as research subjects. And that whether they are considered research subjects themselves or not, it's the responsibility of the IRB to try to assess the sensitivity of the information, the potential for shame, harm, abuse, et cetera, and to exercise appropriate judgment and discretion in -- to exercise appropriate judgment in setting terms or conditions for the research that would be appropriately protective of the information of the third parties. I couldn't re-say that but -- DR. PELIAS: I'd like to point out that my intent in putting these first paragraphs under parties to research was just to define who is a party to the research. Okay. And if you would go to the next page, and we look at role and responsibility of the IRB, these may be our direct charges to the institutional review boards, and I think that paragraph 2 under that pretty much addresses what you were saying, Mark. CHAIRPERSON MARSHALL: Adil and then Elliot? DR. SHAMOO: I think Mark took the words out of my mouth. I wrote down -- because it's written really in this alternative as an absolutist. The third parties who identified by human subjects become human subjects themselves. That is an absolutist statement. And that makes it what I call easy for the fit. The other way one can look at it, the IRB may determine that third parties who are identified as human subjects become human subjects themselves. And then they could decide another alternative of whether they should waive informed consent or not. That makes it more palatable, I think, than absolutist. CHAIRPERSON MARSHALL: Elliot? DR. DORFF: Two points. One is just a little one, which is at the beginning, "parties to research projects include --" there is a D there, which is the sponsor of the research project. You're not talking about them here but that is sometimes very important. On the issue itself, on number two, see -- I mean, what we're weighing is how intrusive should we be in what researchers do. In other words, how much do -- how much do we subject research protocols to public scrutiny. That is really what we're talking about. And I think that the -- what Adil just said was an interesting kind of a compromise in which what you would say is that we're not necessarily defining a third party either in or out of being a human subject but that it will depend. But that definition would depend upon how the IRB evaluates the impact on third parties of a particular research protocol. And I think that is a nice compromise. CHAIRPERSON MARSHALL: Committee? I see some heads nodding. DR. CHODOSH: I agree with that. CHAIRPERSON MARSHALL: All right. Let me ask this procedurally. Can we craft something that doesn't exist on paper right now but has just been suggested? Put it forth to the committee by -- obviously via e-mail, I think? And then go from there to posting something on our web site asking for comment. How does that sound to you all? We may be able to do it by tomorrow. It depends on if you guys are willing to do this overnight. We may very well be able to do it tomorrow. Okay. That sounds like a plan. Yes, Greg? DR. KOSKI: In looking at the language -- you're talking about the alternative, this thing A(2). With reference -- since this goes in context to the -- you know, the parties to research up there, it wouldn't make any difference whether the third party is identified by a human subjects. Just that any third party to research may become a human subject and so that not only putting in the word "may" but removing the "who are identified by human subjects" which is irrelevant to the whole thing. It's just that they are one step removed from the immediate interaction between the investigator and the research subject. CHAIRPERSON MARSHALL: Well, let me ask that prior to beginning tomorrow morning that Mary Kay, Elliot, Adil and Mark work on a revised definition, a revised paragraph that we can then put before the rest of the committee, agree on tomorrow, and then have something that we can post for comment. Great. Thank you. It's time for us to adjourn for our afternoon break. Mary Kay, again thank you very much. [Applause.] [Brief recess taken at 3:00 p.m.] CHAIRPERSON MARSHALL: I am going to give the floor to Kate because we wanted to do this, this morning, but she was hard at work and we would like to introduce a special person to you all and I am going to give the floor to Kate to do that. MS. GOTTFRIED: Well, as many of you know, we have been inadequately staffed and although that continues to be the case, we have a temporary reprieve or I should say an intermittent reprieve because she's not temporary, Keisha Johnson is now staffing NHRPAC. [Applause.] MS. GOTTFRIED: And I want the people on the committee to pay special attention to all the different commands and orders that Keisha sends to you because she is now your lifeline and pipeline for the committee and very earnest in getting everything taken care of. So if you have any questions, concerns, you should send them to Keisha. If she can't answer them, she will find out how to get the answer or who to talk to. So thank you very much and welcome, Keisha. CHAIRPERSON MARSHALL: I would like to say that she is also an artist, that Keisha made the cover for our books. Thank you very much, Keisha. They're gorgeous. [Applause.] CHAIRPERSON MARSHALL: Let us do move along. We have gotten a lot of work done today. It has been an interesting day in the good sense of the word. Alan Fleischman first suggested to Kate and to me that we make some time on our agenda for a response to the research ethics opportunities that have been occasioned by the events of September 11th, and Jonathan Moreno has obviously been thinking about them as well. So I think it's important that we do make time for that. I want to just preface this session by bringing to your attention an article in the November 5th New Yorker, the financial page, "No profit, no cure," in which the observation is made that in 1967 then Surgeon General William Stewart declared that "The time has come to close the book on infectious diseases. We have basically wiped out infection in the United States." Well, that turned out not to be true. And the article goes on to quote Gary Woodnut, who is a vice-president of MicroBiology at Glaxo-SmithKline, who observes that "this year private industry is expected to spend somewhere around $4.1 billion on antibiotic research" and Dr. Woodnut says that "the genomics revolution really was a revolution as far as antibiotics are concerned. We are in a new age of drug discovery." So I think there are going to be all sorts of clinical as well as social and behavioral science research responses to September 11th. Alan and Jonathan, I thank you very much for putting this on the agenda and for helping to -- and for leading the discussion and helping us to think about it. RESEARCH ETHICS OCCASIONED BY THE SEPTEMBER 11, 2001 TRAGEDY; CHILDREN & RESEARCH DR. FLEISCHMAN: I think I am going to start out to say that New York medicine responded admirably to the tragedy that occurred on September 11th but very shortly after the tragedy it was very clear that although there are some data, there are not a lot of data about the consequences of such massive disasters man-made on victims, their families and the populations that surround them. And it was appropriate, both in terms of the need to do the kind of social science work that this tragedy created, as well as to do assessment data collection for the purposes of determining the needs and the intervention needs, it was obvious that there was a need for substantial research. Very quickly after the tragedy, large numbers of social scientists and clinicians amassed in New York. It was surprising to many New Yorkers how many long distance airplane flights were taken by well-intentioned researchers to begin to develop their research protocols and to get access to these populations. I hasten to add that I believe all of these people were well intentioned, all of these people wished to help and to learn and they were not people who were there in order to benefit financially or in any other way from this tragedy. Now when one looks at the data in the literature about research in those who have been traumatized, there's very little guidance for IRBs or for that matter even for researchers, except for the admonition to be sensitive to the needs and interests of the research subjects and that goes almost without saying. Ruth Faden this morning, I think, made a very important point and that is her comment about too much reliance of the research ethics enterprise on informed consent or the respect for persons principle in the Belmont Report as compared for the importance of the beneficence principle in the Belmont Report. I will come back to that. Unique to this tragedy perhaps, and I am not sure whether it's absolutely unique, but unusual in this tragedy was that within two days of the tragedy the names of those missing and presumed dead were known publicly with virtual certainty at about a 98 percent level. Anyone could assess who the victims were and who their families were and because of the proximity of those living in the region you could virtually find their address and phone numbers with no problem. So this created some very serious concerns. We knew the names and addresses the firemen, the policemen, the employees in the World Trade Center who had died or who were missing and we knew the names of a large number of those who were visiting the World Trade Center, and all of the rescue workers. And the epidemiologists in my shop said to me within 48 hours they could give me a complete listing without having access to that through the public officials. Not that I asked for it but that they were quite concerned about this from a research ethics perspective. In New York City both the mayor and the health commissioner, who I believe did superb jobs, were quite concerned about the interests of the victims and their families. I think the whole country and the world saw the mayor's leadership. They gave services to these people and are continuing to provide services that are quite astounding and they were quite concerned about over sampling of that population, an abuse of that population with well-intentioned research. So we laid out a series of questions that seemed to be relevant in this particular episode and might be relevant if, God forbid, there are future episodes. It comes through the language of research ethics about vulnerable subjects. In social science when we think about vulnerable subjects, we think about people at increased relative risks or susceptibilities to adverse health outcomes generally. In research ethics when we think about vulnerable subjects, we really think about those whose ability to make autonomous choices to give informed consent in a voluntary way, those people being compromised by their age or their capacity to make rational choices due to cognitive impairment or mental illness, or the social circumstances which make them vulnerable or subject to nonvoluntary or coerced consent, such as prisoners. However, these potential research subjects seemed to most of us to be quite vulnerable as well because of the acute and severe grief they were experiencing, the possibility of being re-traumatized by a research intervention, and the potential for over sampling and redundant research. So we thought of the following scenario: Let's assume that in Palo Alto a very well intentioned researcher in the social sciences had forwarded to his IRB a quite ethical, acceptable and appropriate research study on all the wives of the firemen. And in Cambridge another investigator did the same. Both of these IRBs carefully and thoughtfully reviewed these protocols and found them to be ethical and appropriate research, yet they were identical or overlapping. Were there ways for those IRBs to deal with this problem? In Oklahoma City I am told that the university there basically developed a process by which no research could be done in Oklahoma City without the approval of their review process. Now it seems that was agreed upon voluntarily and that this came forward. I am also told, and correct me if there are people in the room who know more about this than I, that there was not the same outpouring of interest in the social science community with conflicting potentials. And there was ease of collaboration and, of course, there was a great deal less number of victims. So the questions basically revolved around how ought we without precluding the ability of bona fide social scientists to do bona fide research, how could we protect these, what we considered, vulnerable subjects? And I raise this as a question for discussion, not that we have answered it. But there does appear to be need for guidance to IRBs in general that in such circumstances there needs to be coordinated and collaborative research work that perhaps will require a super IRB or coordinating group to review already approved proposals to assure that there is no redundancy or over-sampling of subjects, to bias the research towards research with potential for direct benefit to subjects, to create appropriate consent monitoring and assurance of voluntary participation, to have special protections, including the training of the researchers, the interviewers, the people who interact with the subjects, assuring that they are able to identify symptoms of distress, the availability immediately of referral for those subjects who reveal symptoms and require services on an immediate basis, and the assurance to subjects that they may learn the results of the studies even if the studies do not have direct therapeutic intent. There was also a discussion in the New York City level about whether studies on the general New York City population, not just people directly impacted upon by the tragedy, ought to be subject to the same kinds of constraints. That is an open question. I will tell you one of the 13 studies that the New York Academy of Medicine embarked on since September 11th is a random phone survey of New Yorkers to learn about the medical, social and health -- medical and mental health impact in order to do a needs assessment for what future services might be needed in New York City. This was a study that we embarked on which will be very important to the commissioner and will be very important to the people of New York but yet our IRB knew that a random survey of New Yorkers might touch family members of victims. Our IRB also knew that a random survey of New Yorkers would likely touch severely and seriously distressed people, and created a very stringent procedural approach towards this simple random survey with a professional random digit dial survey company who we did not allow to do this survey without direct input through psychologists, psychiatrists and others who did training and who were available as the survey unfolded. And, in fact, in the first 200 who were sampled, and this is a 3,000 person random sample, in the first 200 who were sampled there was actually the need for two interventions, immediate interventions. So, in conclusion, it would be -- it seems to me that we could clarify for IRBs and for the community of social scientists and clinicians who would be interested in such research, the need not only for sensitivity but, in fact, special protections that are appropriate in such unusual circumstances, and perhaps create some understanding of this type of vulnerability based in a massive tragedy of this kind. I do not really have an answer for exactly how to do that. Yet I do think that it makes some sense for us to consider this not just today but consider this over the next period of time so that if, in fact, we are faced with such events in the future, we will have at least some guidance for investigators and IRBs. CHAIRPERSON MARSHALL: Thank you, Alan. That was beautifully thought out. You have given us a framework in which to consider moving forward. Jonathan, let me ask you how would you like to proceed? DR. MORENO: I am not on point so I would suggest opening the floor for discussion. CHAIRPERSON MARSHALL: I would like to welcome Denyse. I would be remiss in not welcoming Denyse. We are glad you are here, Denyse. She did observe herself that she had made it in time for the break and dinner but she is going to have to do a little work in between. Adil? DR. SHAMOO: Alan, you made the comparison to the vulnerable groups. Since I was one of those NBAC groupies, they call us, we attended most of the NBAC meetings and took notes as well as testified, as you know in the NBAC they have for vulnerable groups and for this you need a type of research which local IRBs might not be able to take care of them for lack of expertise, whatever. They have what is called SSP, special standing committees or program, whatever it is. Special standing panel. And I think, to be very honest, what you just suggested fits rather well to what NBAC has already proposed for us to consider, and there is a logic behind it. And, again, it was done in the context of what tomorrow we will be talking about, decisionally impaired. And to be very honest, those who have been traumatized at such a horrendous act, they are traumatized and they are vulnerable and they may be decisionally impaired not in terms of cognition and brain function but nevertheless impaired because of the trauma. CHAIRPERSON MARSHALL: Felice? DR. LEVINE: I think this is a critical issue for us to work on and actually when I saw it on the agenda, I was hoping, and when I came in today I said to Mary Faith that it was a subject I was hoping we would raise. I had seen it on the agenda when I looked at the agenda but I assumed -- this was a psychological process on my part -- with Alan leading the Children's Working Group that we were going to focus on sort of research issues involving children as opposed to the more generic issues that you raised, and I think it is an extremely important one for us to get our hands around. I think, though, I will say that this is an area with whatever -- whoever may have aggregated around Oklahoma. There has been a long history of cadres of researchers working in the whole area of disaster research. Of course, working in the area of epidemic research and working in the area that I said I was speak to this committee on -- that was dealing with humanitarian emergencies, which are dealing with persons in refugee status and forced migration, in particular. And, similarly, one of the things I suggested in speaking was the possibility of some kind of in-site/on-site coordination among projects so that you could also have some form of also peer review guidance because you can't go back and say, hey, I want go modify my protocol when you're in the midst of a something. But I have sensed, in part from reading a lot of wise papers from people working in this area, that there has actually been -- that (a) in these communities there actually is a Center for Disaster Research in Colorado and Delaware. They are in close coordination. So I would suspect that there may be more natural affinity of what is actually going on than in some -- I am going to say some other social issues where we all may bump into each other and not exactly realize it. So I would like us to sensibly but kind of intelligently find ways of being briefed about how things are operating in these communities now. And I know from my own recent experience with this Committee on Population there is a lot of wise people thinking about this and a lot of eagerness to really engage with the human subjects community in trying to figure out ways to do it better. So, I think, it would mutually meet with receptivity and I am really pleased it came up. CHAIRPERSON MARSHALL: Thank you, Felice. Margaret? MS. BORWHAT: Alan, I just want to thank you for highlighting, in particular, and would like to support it strongly as possible the availability of immediate intervention by health care professionals in providing psychological counseling. More than 20 years ago I did some research with adults recalling sexual abuse in their childhood and from that experience I know how just a -- what one would think was non-threatening environment and questions that were toned down, if you will, that still it triggered -- it could trigger some powerful emotional reactions. So I cannot support strongly enough always having health care providers available for immediate intervention. CHAIRPERSON MARSHALL: Sandy? DR. CHODOSH: Although not exactly the same thing, I dealt with a situation like this way back in the 1970's in which the institution that I chaired the IRB for was also the institution that had the predominant numbers of patients with sickle cell disease. You cannot imagine when suddenly funds became available to NIH with specific direction for sickle cell, it was just like the crash in a way, everybody wanted these subjects and we had to set up essentially a registry. Because not only from my own institution did people want to utilize such individuals but also they were coming to us from other places because that is where the material was. It was something that had to be done and we did it fairly quickly actually once we realized we were seeing just numerous, numerous protocols with a limited number, it turns out, of subjects who could easily have been enrolled in three or four different studies at once, and it just wasn't practical. So it is good -- I think it is almost like emergency planning in a way. Perhaps we do need to have some better centralized system. This carries over into the area of pharmaceutical and device research also where protocols may go to one IRB at one institution and be turned down but that information is never centralized to be looked at by other institutions that may be reviewing the same protocols. It is a somewhat similar situation of the common protocol and being treated not in your case identically but it could be treated quite differently. DR. FLEISCHMAN: I think it is a very good analogy to the 1970's sickle cell anemia story but generally I think IRBs do feel the responsibility to protect the subjects in a cohort that they have responsibility for so that, you know, in -- I know a little bit about neonatology -- you cannot do 12 studies in the same preemie, you know. Not even two or three because you will bleed them dry. But the -- less than articulate. And IRBs are very sensitive to that and usually divisions are sensitive to that but the question really has to do with this unique circumstance where IRBs doing their job but still not have done that. I think Adil's point about the special panels is an interesting one and I look forward to that conversation tomorrow morning and in that context I will disagree with Adil about the need for such panels in most circumstances. In this context I thought it very important that the IRBs actually do their job and then have another layer, rather than say they could not do their job, and that they would then have a second level of review, which might result in coordination. Might result in collegial agreement to collaborate on work. The problem that Felice raises -- there is a social science group who know one another extremely well. One that studies that post-traumatic stress disorder and pathological bereavement and another that studies responses to disaster and trauma, and they really do know each other extremely well and they go to the meetings and they talk to each other. But there are a host of other people who nibble around the edges of that science who are also interested in doing research and who, you know, have very fine protocols that could be ethically performed. So it isn't even a matter of, you know, just the collaborating groupings. It is a whole host of other people. The School of Public Health at Columbia, the New York Academy of Medicine, the Department of Health, the CDC, its internal people came up with large numbers of potential, very fine research protocols. DR. SHAMOO: Alan, special standing panel almost is similar to what you are talking about. Is the local IRBs are not devoid of responsibility. They refer it to the SSP when it is unique circumstances where they, themselves, feel -- needs another layer and then the special standing panel after deliberation a few times, they then give up that responsibility when it is no longer necessary back to the local IRB. So it is not a national IRB for the purpose of forever that they do it that way. So it is very similar to what you demanded. CHAIRPERSON MARSHALL: I have -- did you want to respond, Alan? DR. FLEISCHMAN: We can talk about it tomorrow. CHAIRPERSON MARSHALL: All right. I have Susan, Greg and then Felice. Greg? DR. KOSKI: Well, this may come as a surprise to some of you but there are those among the research community who would find the suggestion of yet another panel for a review to be yet another impediment. In this case, depending upon, you know, in the absence of any mechanism within the regulatory structure for imposing such an additional level would be cited as, you know, illegal and so on. So that, you know, what I am basically trying to get to is the question about, you know, if -- I mean, clearly we are now in a somewhat different world than what we were in prior to September 11th and we do not know what the future holds. But clearly there may be man-made disasters of the type of terror that we have seen. There may, of course, be other kinds of disasters along the way. The question is, is there a need on an ongoing basis such as NBAC has identified to begin to work within government and the scientific community to develop appropriate guidelines, procedures, policies for identifying when such additional oversight process might be desirable or required or whatever? I am seeking advice here on what to do because, indeed, anything that is viewed as an impediment to research can be damaging also. So achieving the proper balance there is a very important one and I am just pleading for some guidance. DR. SHAMOO: Can I respond to Greg? CHAIRPERSON MARSHALL: Sure, Adil. DR. SHAMOO: I think the SSP is not the only unique God send method of dealing with these issues. For example, OHRP or NOHRO within NBAC, if there is a new legislation, but I am saying barring that, let's say, even OHRP within its own advisory committee can serve that purpose. I do not know if that is allowed within your own authority. I do not see anything wrong within that authority to be the oversight for such kind of unique and compelling circumstances that there will be another layer where at least OHRP will look at it and try to coordinate rather than having, you know, ten research protocols -- I am just using your example -- taking blood from the preemie and dry them to death, you know. DR. KOSKI: I will just quickly say that our general counsel is not here but my suspicion would be that we would be told that that is outside the regulatory authority of OHRP in its present condition. CHAIRPERSON MARSHALL: Alan? DR. FLEISCHMAN: Greg, we share your concern and, in fact, that is why we walked slowly in thinking about this because as we read the regulatory authority, we would think you do not have it. We also have great respect for the social science community and the need for research. So I do think we ought to spend some time, and it may take us three months, it may take us six months, it may take us a year, but to sort out a careful series of recommendations about this issue. I do think we need to weigh each of these very carefully and to figure out who ought to have what authority, if any, in what circumstances because I do think it is important. Just to conclude, it has been alleged and I do not have primary data on this, but it is alleged that there were some people who thought they might just be able to create a law in the State of New York which would say you may not do research with certain sets of people because it would be illegal without permission of whoever. I would think the research community would respond more negatively to that approach than to some kind of academically developed collegial, thoughtful, sensitive approach, which is why we raised the question for analysis. CHAIRPERSON MARSHALL: I have Greg responding and then I have Felice, Mark and Elliot. DR. KOSKI: I want to make sure that we include in the discussion not just the social and behavioral sciences here because, you know, there are also issues, for instance, that we can easily envision, whether it is vaccine research or others that are going to cross into that, you know, biomedical end. I always speak up for whichever side has not been either attacked or-- [Laughter.] DR. KOSKI: -- just so we get that out there. We are maintaining balance. DR. LEVINE: I wanted to sort of clarify two dimensions. My referring to those research traditions and groups was not to say that those were privileged bodies. It was really to say we might have some lessons learned with respect to how that kind of coordination works under rapidly changing circumstances so that we can better understand how to ferret out ways of doing it among persons who have been giving it some thought. But, I think, in particular, the difficulty of adding multiple tiers -- part of what I was trying to think about was whether there might be an alternative mechanism because here we have a situation where, indeed, there are an accelerated set of events that -- where research has -- runs risks but also has benefits to individuals in the host community. And also, of course, involves complex issues of tension and problems for those doing the research. So we may need to not as much think about multiple layers but are there some ways of having a human subjects review that is -- that accelerates the pace in ways that does so still with the highest standards of ethical practice. CHAIRPERSON MARSHALL: Greg has a remark on point and then Mark. DR. KOSKI: One thing I would want to urge is that we now look beyond the possibility of doing something that is non-regulatory here outside of government. One of the big pushes that we have been trying to advance is to actually have the scientific community and the public take their appropriate responsibility for basically doing themselves what they ought to be doing. And it seems to me that there even could be an opportunity here where organizations such as CASA, which would cover all of the social and behavioral sciences, or the AAU and AAMC, could develop, for instance, a mechanism that would be advisory to their members and so on that might be a helpful step forward that could be done without having to go through a regulatory process. MR. BARNES: Greg, I think that your general counsel is not here and I am not your general counsel but I will observe the following, which is -- it really builds on what you just said. Although your office does not really have the authority to mandate a kind of community IRB and that one would have to go through it in this disaster or in the Buffalo Creek disaster or anything that might be similar. What you do have the authority to do is to say to that IRB in Palo Alto and that IRB in Cambridge and the IRB in D.C. and the IRB in Trenton, all of whom have authorized these studies of this population affected by this disaster, you do have the authority to say to them, "You know what, you really were not doing an adequate job of protecting human subjects because you were, in fact, approving this research in a population about whom you never asked the question are they being studied to death, and so you did not do your job of oversight." And if you did that and I think it is within your office's authority to put out guidance about that, then what that would do -- it would not be that you would be requiring them to adopt any particular model but you would drive them towards a cooperative agreement among themselves so that with the municipality, with the community, with the whatever political subdivision or division there would be, then they would understand from the beginning that they would have a responsibility to coordinate their research as part of their own obligations to approve their own research. DR. KOSKI: Mark, that certainly is true. However, I would also say that there is great concern about regulatory agencies using the guidance process to create de facto regulation without going through an appropriate rules making process. So one needs to be very careful about navigating in that particular area. CHAIRPERSON MARSHALL: Elliot? DR. DORFF: What I was going to say is more or less what Greg just said in terms of the use of professional organizations to create their own ethic of these things. That was the first point that I was going to say and I think that that is an unutilized sort of body. And what Felice mentioned before in terms of getting information from researchers that have done this in other kinds of disasters, I think would be very appropriate. The other thing that I wanted to mention was that we need to -- I mean, we need to try to figure out -- in a disaster sort of situation, by the very nature of them, you have not had advance planning so the -- so getting people to coordinate themselves, their efforts, in a way that does not bleed the subjects to death, is harder because you have not figured this out ahead of time. And so, I mean, I think that one of the things that would need -- that we would want, some of these, you know, professional organizations to think about is the specific definition of what a disaster is for these purposes and then, secondly, what mechanisms within their own organizations should come into play immediately in order to make sure that the subjects are not harassed. CHAIRPERSON MARSHALL: Sandy? DR. CHODOSH: My approach is somewhere in between, I think, and that is that somehow this can be planned ahead of time. The idea of central registry of such studies. Not to do another evaluation of each study but simply to assess that you are not impacting on the same population for too many of -- it does not have to be the same kind of study. Just too many studies. That a central registry, it seems to me, would be helpful and I think it is within the scope of what you could possibly do without getting into trouble with your legal people even though it would be best if we could do it through professional organizations. The trouble is that, you know, that there are too many professional organizations to line up. We need something that it seems to me has some centrality to it to begin with to accomplish that and without making judgments about did the IRB do the right job but rather is the population that is going to be affected going to be unduly affected by too many studies. DR. KOSKI: A good point, Sandy. I think from the government's side, any time our office collects information in a systematic way for more than nine people, it is considered a survey and it is subject to clearance through OMB and the Paperwork Reduction Act and so on. So that again there are constraints that are placed there. I think we need to make sure we do not lose sight of the fact that the investigators also need to take a major role and responsibility and to perhaps quell their own enthusiasm that could result in the overuse or exploitation of a vulnerable and limited group of subjects. So that again it is possible to have the IRBs and the investigators sensitized to that. To the extent that we are moving into a new era, one of the things that might be very helpful would be actually have our, you know, professional organizations that now have shown a renewed willingness to exercise their responsibilities in these areas to actually take the lead and then offer, make recommendations perhaps to the government on how we could work together to achieve the appropriate goals that we want because, as I said, we may in some instances want to facilitate and speed, expedite a process in a responsible manner. Whereas, at the same time we also might want to have additional protections afforded there. So if we could get the community to actually come together and say this, you know, government is what we need to help meet our goals, too, that would be, I think, the best of all possible worlds. But I guess I am enough of an idealist to believe that could actually happen. CHAIRPERSON MARSHALL: Susan, and then I am going to give Jonathan the floor. MS. KORNETSKY: I think one of the real problems here is, you know, a registry or whatever only serves some purpose. I think a real problem that needs to be addressed is then who decides. When the numbers or the types of research gets up there, who is the policeman to say this one go ahead, this one stop, this one wait? And I think that is the huge challenge. Within our own institutions when we have a highly desirable population, we usually ask someone or some group to take that responsibility but that is where I really see the challenges here. Who is going to do that? DR. ZULLO: Susan Zullo from OHRP but I am not speaking with my OHRP hat. I respectfully hear everyone's comments and I actually have a friend that was involved with the New York City accident. And I think it is a fine line of trying to balance human subjects protection and not regulate what people can do because I know she is very -- looking for some way to participate in a study or something so that she feels like she is doing something and I think we have to be careful. Although she is vulnerable, I think we need to weigh -- I think she is perfectly -- I know she is very stressed from all of this but I think she is perfectly capable of making an informed choice at this point and I think we need to be careful not to over regulate. CHAIRPERSON MARSHALL: I just want to give Jonathan ample time and then open the floor for full discussion. DR. SUSSMAN: Don Sussman, not speaking for the Department of Transportation. I want to go back to something Greg said. Sometimes you have to be able to expedite. Local organizations, police organizations, even federal organizations are often not comfortable with immediate objective evaluations of disasters and there has to be a way to keep this information sometimes, I do not want to use the word "suppressed" but being held down. So sometimes it works the other way. CHAIRPERSON MARSHALL: And I think that is a perfect segue to Jonathan who is going to sort of enlarge our framework for discussion. DR. MORENO: I want to thank Mary Faith, Kate and Alan, also, for letting me squeeze into the program at the last minute. As some of you know, I have had a long-standing interest since my work for the Radiation Advisory Committee and the use of human subjects, the National Security Experiments, and I published a book about it, which I think there might still be a couple copies of. And I never thought that the role of sort of ethics historian would be so contemporary. In the last couple of weeks I have had occasion to think about that work again and it struck me in preparing for this meeting that there are some things that NHRPAC really needs to think about. I would say as important as everything else is that is going on at this meeting, there is probably nothing more important in the short-term at least than some of the issues that have to do with the human subjects research environment in light of September 11th. And so immodestly I have listed them and I am going to run through particularly my number one on my list in some detail because I think it is quite pressing. I do this, by the way, in the spirit of the original idea of NHRPAC which was that we would provide the new OHRP with resources and counsel that in the course of their busy lives, however talented Greg and his staff are, they don't really have time often to step back and think about. Though I have a hunch that Greg has thought about this stuff already, as his colleagues have. I wanted to make sure that we got it on the table here and I am sure this is not the end of the conversation, though I do have an action item for us at the end of this. So I want to talk about four items very briefly. First of all, the status of the Radiation Advisory Committee recommendations on classified research. The position of OHRP and the department on these recommendations. And I have a couple of really information items that I would like OHRP perhaps to help us with, namely information about the anticipated protocol for the smallpox vaccine trials that have been announced that may start as soon as January. And what I think of as a de facto Phase IV trial of widespread prophylactic use of certain antibiotics for anthrax exposure, Cipro being the obvious but not the only example. These were the recommendations that the advisory committee offered the Clinton Administration in 1995 with respect to classified research. And as you probably know, the advisory committee had a special responsibility to look back at classified research, and along with the Executive Order that established the Radiation Advisory Committee there was a parallel order in January of '94 to put on a fast track for declassification any records of radiation experiments involving human subjects or apparently involving human subjects. That was information that might not have been available. Some of that information also shed light on government policies, often secret government policies, on biological and chemical experiments as well. These were the recommendations of the advisory committee: First of all -- none of them, I think, are going to be particularly surprising -- to obtain informed consent of all human subjects in classified research, which in many cases will mean that the subjects themselves will have to have the appropriate security clearance. To inform the subjects of the identity of the sponsoring agency. Some subjects may not wish to participate in research that is conducted by one agency or another. Inform subjects that the project involves classified research if that has not otherwise been disclosed. Establish permanent records. This became a major problem for the Radiation Advisory Committee in trying to look back to see, as the President asked, if any Americans were put at risk by classified research during the Cold War. The records often were not available. And, finally, that the research be approved by "an independent panel of nongovernment experts and citizen representatives, all with the necessary security clearances." Now this was the Clinton Administration's response in 1996 in a document called "Building Public Trust." The Administration agrees with the first four recommendations, the ones I showed you. Notice that it didn't necessarily agree to the last recommendation, the panel. The President is issuing a memorandum directing federal agencies to jointly propose modifications to the federal policy for the protection of human subjects, the Common Rule, as it applies to classified research in order to implement these changes. Further, subjects will be informed of the sponsoring agency except in limit minimal risk cases. In all secret studies, researchers will obtain informed consent, disclose that the project involves classified research, and keep permanent records. The response went further than that as a matter of fact. It said that the federal agencies will jointly propose first amending the Common Rule to require that IRBs for secret projects include a nongovernmental member. This was their way around the special panel effort. Establishing an appeals process so that any -- this is very important I think -- that any member of a review board who believes that a project should not go forward can appeal the board's decision to the head of the agency and, if necessary, the assistant to the President for science and technology. And, three, requiring the sponsoring agency to keep permanent records and so forth, and to declassify these records as soon as possible. Finally, the Administration is taking two additional steps. They said the President is directing the heads of the federal agencies to disclose annually the number of secret human research projects undertaken by the agency and the number of human subjects participating in each project. So here are my questions in light of my own ignorance of any action that has been taken with respect to these recommendations: First, a question for the Administration, directed through OHRP or our new Assistant Secretary for Health. What is the status of the modifications of the Common Rule? Where are they in the agencies? I spoke to somebody at OSTP last year who told me that they were in process among the signatory agencies. I think now is the time to move that process to a rapid conclusion. What is the status of the special review process? Has that been rejected or will they accept that particular recommendation for a non-agency member of the IRB for classified research with the appropriate security clearance? What is the status of disclosure of the number of classified experiments and of subjects of those experiments? I have not heard any numbers at all in the last five years but it does not mean that they are not out there. I just may not be aware of them. So I am going to resolve, if I can get a second, that this advisory committee forward these questions to the appropriate officials in the Department of Health and Human Services and the Administration. And if we are told that they can't give us answers then I think we should ask why we can't get answers. And if they will not tell us that, then I think at least they should articulate a policy with respect to classified human experiments given our less than happy history. I am not, by the way, of the view that classified human experiments may not take place. I think they ha to take place under appropriate circumstances and I think they can. I took a lot of heat for this position from my friends on the left when I published that book but I believe I have been vindicated by recent events so I am not opposed to classified human experiments but I think they can be done in an appropriate way. Now I have three other concerns and the second of which -- the first of which of these last three involves another resolution that I would like to put on the floor. Namely that the advisory committee should resolve that the ACHRE recommendations be adopted by DHHS. We are not an advisory committee to all the signatories of the Common Rule but we are, as I understand it, an advisory committee within DHHS and it does not seem to me to be inappropriate for us to make such a recommendation to our own agency. This will have two effects. It will both require that any DHHS and other agency research done with DHHS be in compliance with those four rules. It will also establish a benchmark for other federal agencies to come into compliance. I also would like to know if we can be kept informed, so far as OHRP is aware of these plans and protocols, for the anticipated diluted smallpox vaccine trials. And another point of information, and this again in a very positive vein, the use of antibiotics -- the widespread use of antibiotics due to the current anthrax scare amounts to a large Phase IV trial. There should be some kind of look back at what is happening with respect to the people who are taking these powerful medications. And a question, I think, is whether, first of all, any agency plans to do that or perhaps if we should encourage the IOM or some other appropriate quasi-governmental entity to do that. And, also, of course, we would have an interest in how that is being done. I am finished. Thank you. CHAIRPERSON MARSHALL: Thank you very much, Jonathan. You have put some concrete recommendations before the committee. So, committee members, let me ask for some concrete responses. Adil? DR. SHAMOO: I just have a clarification. I heard Dr. Fauci saying the smallpox vaccine dilution experiment -- which is, by the way, already going on in my own institution because I saw signs all over the campus encouraging medical students to be patriotic and volunteer for it. [Laughter.] DR. SHAMOO: It is true. And they came to classes, et cetera, and the funding -- obviously there is an emergency here and it is very appropriate. I heard him on television saying that this is a dilution experiment. Please correct me, the FDA people here. That it does not require all over again -- he did not say the word IRB but my understanding was it does not require an IRB approval or another IND or anything because it is the same drug. It is just dilution. That is what he said. I want somebody from FDA to clarify what he said because that will have an impact on our discussion. DR. MORENO: I did not hear that so I am happy to have that information. I would like to hear more about that. MS. KORNETSKY: Jonathan, I would like to just ask for your opinion on number three and four. Why -- your rationale for why you think NHRPAC should be kept informed on these two particular trials. I can foresee many other different situations or trials and I am just trying to understand your rationale for that. I am a little uncomfortable with that. DR. MORENO: I could only give you a general answer, Susan. I think that we are in a -- I am sympathetic to your point. Yet I think we are at the same time perhaps undergoing a qualitative change in the atmosphere with respect to certain kinds of public health oriented research in an emergency. And I think we might -- NHRPAC may itself be able to learn something from the way in which these studies are being done that may be helpful to our deliberations in the future. I am not sure we need to revisit -- I take your point -- every single trial along these lines but these two potential agents are such enormous public interest -- and we are the National Human Research Protection Advisory Committee -- that I don't think it is inappropriate for us at least to stay informed about how they are being conducted. MS. KORNETSKY: I understand the public health implications of this and the seriousness of this. I just really do not know if that is within our charge to be sort of -- I mean, to be kept informed. What does that mean? What if there is a difference of opinion? I mean, are we looking at the ethics of this? Are we looking at whether this is appropriate to subject to the public to it? What if there is a difference of opinion here? I do not know what our authority is over that and I just think it is a slippery slope for this committee to be going down. CHAIRPERSON MARSHALL: Greg? DR. KOSKI: On point four I understand your point saying that it is very much like a Phase IV open label study but until, in fact, it is proposed as a research study it is not a research study. It is medical care. It is being done to provide antibiotics for treatment or prophylaxis in infection. So that, you know, if it is going to be done as a study it should be proposed as such and be appropriately reviewed and all there. But I do not think we can make it a de facto Phase IV trial by calling it one. DR. SHAMOO: How about the anthrax dilution study? Smallpox I mean. Smallpox. I apologize. Smallpox dilution study. So they could increase the number of potential people taking it from 15.4 to up to 60-70 million. MS. MEYERS: It was never really officially taken out of what they call the Orange Book so it is an approved -- the vaccine is still an approved drug on the books, I believe. [Simultaneous conversation.] CHAIRPERSON MARSHALL: Let's get back to a little bit of order here. I will recognize you if you put your hand up. Alan? DR. FLEISCHMAN: My understanding -- this is a side comment and then my real comment. My understanding about the smallpox vaccine trial, the diluted trial, is it is a research study, it would require an IRB, it has such already, but because we are really talking about just an approved drug at a different dose then it doesn't require going through an entire new IND process at the FDA. It is a dose -- it is a study of a dose of a drug. DR. KOSKI: Alan, I am sorry. But may I correct him on this, please, Mary Faith? No, if you are going to basically be changing the labeling of the smallpox vaccine, it would require a new IND even though it is an approved drug. So it would. DR. SHAMOO: And they did not go through that. CHAIRPERSON MARSHALL: Alan has another comment. DR. FLEISCHMAN: Let me suggest a little bit of a softening of Jonathan's language in his two resolutions. I think they are very important issues that have been raised. I also think we are in a very sensitive moment in the history of the United States and in the history of the Department of Health and Human Services with a brand new Assistant Secretary for Health. And I would -- I have not got the language but I would work on the language of asking for information because of our concerns. And that we would be quite willing to help, to work with the people who are now involved in these issues who were not part of the prior Administration when those recommendations came out, to think these through based on the present issues. So, I mean, I would hasten for us to craft, as only Mary Faith can, language that would be collaborative, supportive and helpful and at the same time in our desire to understand where, if any, these processes are going. CHAIRPERSON MARSHALL: Thank you, Alan. DR. KOSKI: One note of caution. For instance, in our discussion of the smallpox vaccine dilution studies, we are in a very sensitive time here so that I do not know whether or not a new IND was filed or whether that would even be required since it is entirely reasonable to cross reference an existing or previously approved IND. I think we -- the plea here is that -- and this is not to target you, Adil, at all but unless we have firsthand accurate knowledge about any of these issues, we should not say anything at all because, you know, there are things that people can pick up on very easily, and I know you meant no offense but I just do not know the details. And it is just a plea for caution that when we have facts, we should stick to the facts, and when we don't have facts then we should acknowledge that appropriately. DR. SHAMOO: I agree with you fully, Greg, and that is why I asked information from the FDA liaison when I spoke. CHAIRPERSON MARSHALL: Thank you. Our FDA liaison is not here. Yes/no? She left. Yes, she left. That is what I thought. So let us proceed. Sandy? I would like to get a sense from the committee of a response to Alan's thoughtful suggestion. DR. CHODOSH: My thinking about this is that the smallpox trial is not a lot different than the study of the families and the impact on the families. This is a fallout from a tragedy, in essence, for which we are now going into a heightened level of research to accomplish some things that need to get accomplished. So that it sort of fits in again with how do you handle these kinds of circumstances? It is somewhat different but basically it has some of the same elements and maybe there needs to be a system. For instance, if they were supposed to get an IND and didn't get it but it is really important to get the work done, maybe there needs to be a mechanism to go to emergency kind of situations, which would be addressed in the same way as studying the families in multiple ways. CHAIRPERSON MARSHALL: Greg, and then Adil. DR. KOSKI: I think it is important to recognize that under the Common Rule each of the agencies has -- the Secretary or the agency head has the authority to weigh the provisions of the Common Rule when it is justifiable or whatever, for appropriate reasons. And that is intended to provide the kind of flexibility that is necessary to respond to emergency or other challenging situations but to make sure that it is done appropriately and that is something that has preserved. So I will just throw in at this point that in the recommendations of our staff, we, in fact, have included a discussion along these lines on some of these issues for the meeting of the HHRS meeting next week where some of these issues will be discussed there also. CHAIRPERSON MARSHALL: Mark? MR. BARNES: Let me try to offer a resolution perhaps. Can I just ask, though, was there a first resolution on the slide that preceded this? If you could, please. [Pause.] MR. BARNES: The status of the modifications. Let me offer this as a resolution. Would you stay over there? You can forward from here. Okay. No, you do not have to stay over there. Then you can sit down. Okay. Good. I would like to offer the following resolution: Basically that we do recommend as a committee that there be a -- that we receive information about the status of the modifications to the Common Rule. [Pause.] DR. __________: Could I use this time to ask a question maybe of Greg or somebody else? But what happened to the Presidential memo or order dictating, I believe, Section 125 of the Common Rule? It has just disappeared into bureau-space or something like that and I was wondering where it is. DR. KOSKI: The Department of Health and Human Services issued a report as they were required to do in response to the ACHRE committee. Reference has been made to that report and we will distribute copies of that to the members. The Presidential directive to, you know, basically deal with the confidential research did not disappear into some sort of cyber space. It disappeared into the Human Subjects Research Subcommittee apparently some years ago into the various agencies where they have been working, albeit without great speed, toward appropriate revisions, consideration and adoption of those provisions. Current events may provide greater impetus to look at that and move that process forward. Again, as I said, that is one of the issues that will undoubtedly arise in the discussion at the meeting next week. To my knowledge, there are very few -- I think there are only four of the agencies within the federal government that have acknowledged doing classified research and so that it is a relatively limited area and it does not apply to many of the other agencies but we will just have to again sort of work this through. As to the status of the other recommendations, you know, I do not know what those are simply because there has been a lot to do since we got here and we have been taking on other issues. But again I think that I would like to thank Jonathan and all for raising the issues because it is clearly something that needs to be given a higher priority. MR. BARNES: I think I can phrase the resolution without looking at the overhead. Let's try this: That we as a committee -- the resolution is that we as an advisory committee write a letter to whatever the appropriate official would be, whether it is to Greg or whether it is to the ASH or to Secretary Thompson, basically endorsing the recommendations of the ACHRE committee, reinforcing them, inquiring politely as to the status of what President Clinton had promised to be done. That is the first part of the resolution. The second would be that we ask the staff here, not demand, but that we ask that the staff here go back and answer these questions for us, which are legitimate questions that NHRPAC would want answered, which is what is the legal mechanism by which the diluted smallpox vaccine trial is being done. And the second would be has there been thought about the status of whether the prophylactic use of antibiotics would really be a Phase IV trial or whether it would -- because it is generalizable knowledge or is it instead regarded by CDC and the other entities, the public health authorities who are recommending it, instead as being simply a kind of off-labeled use of an approved drug. That is really the question. So the first is in the form of a resolution. I mean it is all a resolution. But the second is really a request from NHRPAC to the staff that we receive the information. CHAIRPERSON MARSHALL: Alan? DR. FLEISCHMAN: I hesitate to use the language of "endorse" the ACHRE recommendations. We are aware of them. We think they are important and we would like to know the status of what is going on with them. I think if we are going to -- at least I, for one, if we are going to do the process of endorsing those recommendations, we need a little bit more -- [Laughter.] DR. FLEISCHMAN: Well, we have heard a lot about them today and I have actually -- I have somewhere in my files those giant big blue things. But, no, honestly, I hesitate to use the language of "endorsement" but I think we ought to say we think they are important and we would like to know more about where they stand. DR. MORENO: I think our task is simpler than that. We can simply note that we are aware of the fact that the previous administration adopted certain policies in light of the ACHRE recommendations. Not all of them. And simply inquire where in the diplomatic spirit to which you have all persuaded me in the last half hour -- simply inquire where they are. And if we want to note their particular poignancy under current circumstances. CHAIRPERSON MARSHALL: All right. Adil, is this on point to this conversation? DR. SHAMOO: Yes. I have a friendly amendment really. We should neither use "resolution" or "command" or "inquire" or "resolution." I think they are strong words and within the current circumstances our's should be a letter to the Assistant Secretary maybe of Health, the new one. That we avail ourselves, due to the current circumstances, these are the current issues. We are concerned about them. We avail our services in those areas if you choose to do so. This is very soft language to enter into a dialogue with the Assistant Secretary and then later on we could decide what to do next. CHAIRPERSON MARSHALL: We do not have an Assistant Secretary yet. We have someone who has been nominated so we might want to bear that in mind in thinking how we want to proceed. Felice? DR. LEVINE: It seems as if there is a strong consensus/unanimity around your writing a letter or being authorized to write a letter. I don't think we have to state formal resolutions to seek further briefing and information on the status of the ACHRE recommendations and other activities surrounding 3 and 4. Especially in light of recent events, NHRPAC would like to be briefed and to be as helpful as possible. CHAIRPERSON MARSHALL: All right. Thank you. I want to come back to that when we close and just -- but it seems as though we do have some consensus on that. I want to recognize Susan and then open the floor. I saw Vera. Those of you who are ex officio or public members, please head up to the microphones and we would like to hear from you. Susan? MS. KORNETSKY: I just really want to make it clear. I mean, I feel a lot more comfortable in I think the way Alan had phrased it. I think sort of being informed about these things is fine. I just would not -- be careful of the wording that -- I am just not so sure what our action will be as if we are informed and we do not like something. And so, you know, that is -- I think we have to be very, very careful about sort of putting within the letter, well, why are we asking for this. Just to be informed, so what. DR. SIEGEL: One of the things I was thinking might help is I asked Alan why are we even getting into this conversation in the first place. And for this particular advisory committee it would be because there is a concern given to the catastrophic situation that we find ourselves in that human subjects protection may start to veer away from what we all believe is where it should be. And so maybe, Susan's point, that is the preamble about why we would even discuss this at this table. MS. KORNETSKY: I think that is very, very important. CHAIRPERSON MARSHALL: So let me just say procedurally that -- and I want to recognize Vera -- that we will draft something and send it out to the committee. Nothing will go directly to anyone without crafting and wordsmithing, which you all do very well. Vera? DR. HASSNER: I want to interject something I am very troubled by, which is the earlier part of the discussion was the disaster as a research arena. Now we have had in New York -- you know, initially when those plane loads of social workers and psychologists came, I think it was 22 very significant psychologists signed a petition, "Hands off. Don't intervene because intervention has been found sometimes to do more harm than good." So this has to do with in other words all those protocols running loose. That part which we are not sure is for the benefit of the victims, there is a lot of activity and wish to do research. On the other hand, the vaccines, their usefulness and what it does, the entire nation wants to know, and you are backing off. So I have a problem. That is a real life issue which the sooner we all know, the more confident the public will be and the lower the anxiety level. On the other hand, the victims, particularly if you talk about intervening with children, there were all sorts of wild, wild myths about 10,000 orphans. It is all nonsense. It is a myth created by professionals rather than from the actual disaster victims. So I would suggest, yes, a great deal of caution, as Alan, as you really articulated. One has to be very, very laid back even though this is an opportunity because not doing more harm to those who have already been harmed. And another thing is that there are studies about trauma victims and what is found is that the best remedy is, in fact, normalcy within the family unit and not professional intervention. There is lots of material on that which we have in our database but I am not the expert on that. So the medical aspect here about vaccines for these two looming, you know, real fears that we all have, I would think that whether it has been set up as research or not as an IND does not matter. The outcomes, the information has to be available and you should, I would think, want to be part of that to show your own value. CHAIRPERSON MARSHALL: Thank you, Vera. Have I missed anyone? [No response.] CHAIRPERSON MARSHALL: No. Any other comments or observations about the topic at hand and, if not, we may move on to wrap-up. [No response.] WRAP-UP CHAIRPERSON MARSHALL: Move on to wrap-up. All right. I just want to note that we are eight minutes ahead of schedule. So what I would like to do first is to go back to Elliot and Felice for the third party language. And then I am going to turn the floor over to Kate, who has some housekeeping and procedural things for us that we need to go over and also things that we want to make sure we do before our ex officio and public members head out the door. So, Elliot and Felice? DR. DORFF: The proposal that we have is that on page 9 of the document that Mary Kay gave us, we would in number one -- the only changes are in the section, "Parties to Research Projects." The second and third sections remain the same under this proposal. Under number one we would add (D) to number one, which is "the sponsors of the research." Just something since we are listing the research -- you know, the parties to it. DR. PELIAS: If applicable. DR. DORFF: If applicable. Don't all research projects have sponsors, no? DR. PELIAS: Not funding. DR. DORFF: All right. Whatever. If applicable, fine. Whatever. And then in place of number two what we would say is this: "Third parties become human subjects if they voluntarily enter a research protocol. Otherwise they are not classified as human subjects. Even so, IRBs are charged with protecting the confidentiality of information about third parties." And then we would leave it to the IRB to determine -- I have been and you cannot go by me because I have learned an awful lot this afternoon and I am still kind of feeling my way. The idea -- no, all kidding aside. The idea behind doing this was to say that -- I mean, if we are talking about the examples that were given were gang research in which you have a member of a gang and the researcher is getting various members of the gang to talk about the leader. So you are talking about gang leadership or something like that. You are not going to get -- for any really significant research on that, you are not going to get, you know, informed consent kinds of things. Virginia was here from the Food and Drug Administration, I think, and she was talking about -- from National Institute for Child Health and Human Development. DR. KOSKI: Actually, no. She is from the NIH Social and Behavioral Sciences. DR. DORFF: Okay. Anyway, she was talking about alcohol studies and the alcohol abuse studies, I think, right. And so the thing is that we -- on the one hand, we want to make sure that those kinds of studies can go forward but on the other hand the concern, it seemed to me at any rate, of those that wanted to classify these people as human subjects all together, third parties as human subjects all together, was what would that do to the confidentiality of the information that is obtained. And so hence the third sentence in what I just said. I really do think that this is a kind of compromise, that this is meeting the concerns of both groups. So I would replace just number two. Number two would now be, okay, "Third parties become human subjects if they voluntarily enter a research protocol. Otherwise they are not classified as human subjects. Even so, IRBs are charged with protecting the confidentiality of information about third parties." CHAIRPERSON MARSHALL: Although procedurally I am going to say if we don't resolve this within a couple of minutes, we are just going to go back to the drawing board either later this evening or tomorrow at some point. Adil? DR. SHAMOO: Yes. I was not involved -- I was supposed to be -- with the earlier remarks so I don't know anything about it. Our alternative, some of us -- probably Sandy and I at least, Sandy and I are on the same side now. You guys watch it. [Laughter.] DR. SHAMOO: We said the IRB may determine that third parties without any specificity, and then the rest of that sentence there. And that is generic and there is no specificity to it. Okay. "The IRB may determine that third parties who are identified by human subjects become human subjects themselves. The requirement for informed consent or a waiver of informed consent from third parties is a question for IRB judgment." The second sentence could be made in better English. Even I, as a foreigner, know that. [Laughter.] DR. SHAMOO: But the same meaning. The same meaning. DR. CHODOSH: The original amendment did not include the statement that they would not be called human subjects and I think that that is not what was intended. It was intended that it was the IRB that could determine that third parties, indeed, were human subjects. CHAIRPERSON MARSHALL: Correct. Mary Kay? DR. CHODOSH: Positive rather than negative. CHAIRPERSON MARSHALL: Right. Mary Kay, did I see your hand? DR. PELIAS: Yes. CHAIRPERSON MARSHALL: Thank you. DR. PELIAS: I would like to suggest that we take the paragraph that has just been worked over and read it in concert with paragraph 2 on the next page where we talk about what the IRB is charged with evaluating, and I think if we work over the ideas in paragraph 2 on page 9 and paragraph 2 on page 10, we can put these things together so that the ideas will come out appropriately. CHAIRPERSON MARSHALL: You are talking about -- I just want to clarify. You are talking about number 2 on page 10? DR. SHAMOO: Yes. DR. PELIAS: I am talking about 2, what we called 2(A) on page 9 and paragraph 2 on page 10. CHAIRPERSON MARSHALL: Correct. DR. PELIAS: Okay. CHAIRPERSON MARSHALL: All right. So would it be acceptable to the committee if we then ask if ten minutes from now when we close if Mary Kay, Elliot, Mark, Adil, Felice, whomever would like to participate, draft something for us to take a look at tomorrow. I think this will go quickly. Or at some point before -- DR. DORFF: That is fine. DR. PELIAS: That is fine. DR. LEVINE: A threshold issue. I do not think this is a ten minute crafting issue. I think we are back -- and we debated it quite substantially and there is -- we could act as if the 6:5 split does not exist but there is not yet language that yet, I think, meets something that could be crafted in 10 minutes. I mean, maybe we should meet tomorrow morning for an hour before we get together. I think it remains a complex issue because, you know, I think Elliot thought and Mary Kay an I thought we had it actually but we obviously did not. So if the determination of human subjects -- whether subjects when they are making reference to others, whether the determination of whether those are themselves human subjects rests with IRBs, and I think that is a very complex issue and one we have not arrived at. And that seems to be what Sandy -- well, that is one position and that was the split in the Third Parties Group and I thought it sounded like there is still a split. CHAIRPERSON MARSHALL: Greg? DR. KOSKI: Well, I don't know. Maybe I dosed through the last part of the discussion but it just seems to me that we do, in fact, have something that is agreeable to everyone. Let me just try this and see what you say. Taking page 9, paragraph 1, leave it as it is, except in 1(C) strike the words "from human subjects" so that (C) reads "third parties about whom researchers obtain information but who have no direct interaction with researchers." Okay. That defines a third party without making a determination as to whether or not they are a human subject under the regulations. Then strike in its entirety the first paragraph 2(A) or whatever it is, the first part of that, and adopt a separate point 2 that says, "Third parties may become human subjects; the requirement for informed consent or waiver of informed consent from third parties is a question for IRB judgment." Words thereby. You may want to fine tune that but basically it gets to the point that the actions for protection of the third parties to the research are something that would be determined by the IRB. Okay. So we may need to craft the wording there a little bit. Then you take that and you put it in context of both the next sections on confidentiality of research information and then the roles and responsibilities of the IRB, and you have something that I think incorporates pretty much everything but does not go to the point of saying that third parties are not human subjects, which is something that we cannot agree on. So if we do that we get to, I think, something that covers everything. CHAIRPERSON MARSHALL: Felice? DR. LEVINE: I am, frankly, hearing this reg as exactly what the OHRP position was in the VCU situation and what I am hearing is third parties -- that if subjects are making reference to others, those could constitute human subjects and that is exactly what the tension and challenge is. And it is -- at least I would see it as certainly not a friendly amendment in trying to craft how we understand when subjects of study are making reference to others. I think that is the challenge. DR. KOSKI: Let me just be very explicit that this is in no way a statement of OHRP policy. I am only offering this in the context of what I heard from the group prior to breaking for this session. So that I am only doing this in the context of this meeting, reflecting only what I have heard here without regard for any of the other issues outside of this committee. CHAIRPERSON MARSHALL: Susan and then Elliot? MS. KORNETSKY: Greg, you can correct me if I am wrong. I think with VCU the problem was that OHRP made a statement that they were, the third parties were human subjects. I think what we are trying to do here is to give some room for IRBs to think about it, to discuss it, to come up with an arrangement one way or another, and to consider issues of informed consent. But I can't speak for what OHRP said. But certainly within the IRB community, I think that is what we heard that was troubling. CHAIRPERSON MARSHALL: Elliot, Mary Kay, Abbey? DR. DORFF: As I understand it, the issues are two. One is the confidentiality on which we are all agreed that steps have to be taken in order to protect the confidentiality of people -- of third parties. The other issue is whether you have to get informed consent from them, right, and if you classify them as human subjects then you do and if you don't, you don't. Is that right? [Chorus of no.] DR. DORFF: No? [Chorus of no.] DR. DORFF: You do or you get a waiver. All right. So the question is whether you are going to require researchers to get waivers for this and the question is do you want to put that kind of a burden on both the researchers and on the IRB or do you want to say that as a matter of course they are not and that you do not have to get this kind of waiver? CHAIRPERSON MARSHALL: Mary Kay? DR. PELIAS: I think that we are resolving this issue very well. If we come in with language that Greg has suggested saying third parties may become human subjects. I think that is an absolutely critical phrase. They may become. They are not defined as human subjects. They may become human subjects. And then we look at the circumstances and we have the IRB look at the circumstances as well. CHAIRPERSON MARSHALL: That was my understanding of what we had agreed to. I just want to be clear on that. That is my understanding earlier that we had agreed to that, yes, as a committee. Yes. DR. PELIAS: Okay. CHAIRPERSON MARSHALL: Okay. Thank you, Mary Kay. I am going to make a procedural remark in just a minute but I want to acknowledge Abbey. If there is someone in the audience who I saw a hand, please step up to the microphone. Abbey? MS. MEYERS: Well, I think if you say that they may become human subject, it should be qualified. We need to say that they may become research subjects if and when they become the goal, but not goal is not the word I'm looking for, the goal of the research or they become -- they are directly interviewed by the researcher. DR. SHAMOO: We can't do that. MS. MEYERS: Excuse me. CHAIRPERSON MARSHALL: Okay. DR. PELIAS: I would refer you to the second paragraph on page 10 that says that the IRB will take all this into account and make a judgment. CHAIRPERSON MARSHALL: Denyse? DR. THORNLEY-BROWN: I am just a little unclear now. You say third parties may become human subjects but it is not -- what is a human subject? I hate to do this but what is a human subject that a third party may become? I mean, when? If you use 2, you have still got your identifiable private information of human subject or subset 2, identifiable private information so that is potentially a third party, is it not? [Simultaneous conversation.] DR. THORNLEY-BROWN: Okay. Why not? [Laughter.] DR. THORNLEY-BROWN: I just see some internal contradictions. CHAIRPERSON MARSHALL: I think we are trying to keep the door open in terms of allowing flexibility at the local level for the IRB. That is fair. Okay. DR. SHELTON: I gather there is some language -- Jim Shelton, USAID. It does strike me that IRBs need to have some basis upon which they would make that determination and I gather there's a little bit of guidance on the second page but it doesn't strike me as particularly all that helpful and instructive as to when IRBs are supposed to do that. I mean, to me, your point is in some ways important that somehow under the Common Rule either you're a human subject in which you get protection of or you're not, in which case you get nothing, which is sort of a problem here because there ought to be a way to get some protection without people being necessarily a human subject. MR. BARNES: I think that what we had said before this latest round of discussion on this issue was that there is a test, which is that the IRB needs to look at the intent of the research, what the research is trying to study, whether the research is really going after the index subject that were the ostensible subjects themselves, or really intent. And the impact is to go after information beyond those subjects, about people beyond those subjects, in which case it would be more likely that those so-called third parties would themselves become research subjects, thus triggering a requirement either for informed consent or waiver of informed consent. I also think that it is not -- it is absolutely not true that third parties under the position that is -- the consensus position that we are trying to articulate -- have no protection. They have multiple protections. These are confidential records. The IRB is required to assess and guarantee the confidentiality and the privacy of research records. State law guarantees it. Medical professional conduct regulations guarantee it. HIPPA guarantees it when it goes into effect on April 4th of 2003. So it is not true they have no protection. MS. GOTTFRIED: I just want to make a recommendation that revises what Greg said as sort of an over arching principle and then say that we should then go to further explication, which has not been done yet, that incorporates what Mark just said with respect to intent to the research index subjects, impact, et cetera. And this may work, I don't know, but whether or not third parties may become human subjects, which would require either informed consent or waiver of the informed consent requirement, is a question for IRB judgment and determination. And then go into whatever the more specifics are going to be. CHAIRPERSON MARSHALL: Good. Okay. So I think we are close. So, Kate and others, let me ask -- Mary Kay? DR. PELIAS: I will add that we had considered the idea of explication and had not eliminated it all together but rather when we put these things -- these suggested guidelines together this time, we were just outlining them for purposes of discussion with anticipation of adding some explication later on. CHAIRPERSON MARSHALL: Okay. So here is what I am going to take the prerogative of the chair and do, is to recommend that a small group of those that I named previously and Kate get together either after we adjourn today or tomorrow morning. And I can tell you that it will not take from 9:00 to 9:30 for me to give updates on our financial relationship document or the letter to the FDA and the OHRP. That will probably take somewhere around 60 seconds. MS. GOTTFRIED: 62. CHAIRPERSON MARSHALL: 62. So let us say that tomorrow morning between 9:05 and 9:30 we will revisit the draft that this group pulls together for us and come to a conclusion on it. So thank you very much for -- DR. LEVINE: Can I make a request? CHAIRPERSON MARSHALL: Yes. DR. LEVINE: I don't care how early we do it tomorrow but I have a doctor's -- CHAIRPERSON MARSHALL: You can't do it now. DR. LEVINE: I have a doctor appointment. I think I need to be in the group. CHAIRPERSON MARSHALL: Yes, you absolutely do. That is just what I was going to say. MR. BARNES: We wouldn't dare leave you out. [Laughter.] DR. LEVINE: I think it is perhaps the most important issue relating to the social sciences. All of our work involves social interaction and I think that is most important in terms of our understanding of human subjects protection and I think some of that in points in the conversation got lost. CHAIRPERSON MARSHALL: Thank you. Okay. So we begin at 8:30 tomorrow morning. Small group, what is your pleasure in terms of when you would like to get together? [Simultaneous conversation.] CHAIRPERSON MARSHALL: 7:30 breakfast. It gives you an hour, Adil. An hour is long enough. So let me turn the floor over to Kate because she has a couple of announcements. MS. GOTTFRIED: Just briefly, the next meeting, so everyone who is here today but may not be here tomorrow, is scheduled for January 28th and 29th, which is a Monday and a Tuesday. And then I want to bring the committee's attention to a date that we had already identified in July, which is the 29th and the 30th, which I believe is a Monday and Tuesday. And I want you to, if you would, check your schedules to let me know whether or not you could meet the week of the 22nd, the week before, 22-23 or 24-25 or 25-26, somewhere in that week, or alternatively 7/31-8/1 or 8/2. So that what I'm really requesting is that last week in July, shift the date from Monday-Tuesday to Wednesday, either Wednesday-Thursday/Thursday-Friday or the week prior, any two days the week prior. And if you can give me input by the end of tomorrow that would be great. If not, we'll have to do it through e-mail. CHAIRPERSON MARSHALL: So we are meeting in the lobby at 6:15, I believe, those who are going to dinner. MS. GOTTFRIED: I believe that Adil, Sandy and myself have cars so anyone going who wants a ride since I don't think it's that walk-able can hook up with us there in the lobby. CHAIRPERSON MARSHALL: I want to thank all of you all. I would like to adjourn. People are leaving. So thank all of you committee members, ex officios, public members, and I hope to see you tomorrow morning. Thanks for all of your hard work today. [Whereupon, at 5:15 p.m., the proceedings were adjourned, to be resumed at 8:30 a.m., on Wednesday, October 31, 2001.] CERTIFICATE OF OFFICIAL REPORTER This is to certify that the attached proceedings before the NATIONAL HUMAN RESEARCH PROTECTIONS ADVISORY COMMITTEE held Tuesday, October 30, 2001, were held as herein appears, and that this is the original verbatim transcript thereof, and is a full correct transcription of the proceedings. ________________________ Gerald T. Brooks, Sr. Official Reporter ?? GB/ct GB/ct PAGES INK COURT AND CONFERENCE REPORTERS 752 West Kingsway Road Middle River, MD 21220 Ph: (410) 574-6420 Fax: (410) 574-6443 PAGES INK COURT AND CONFERENCE REPORTERS 752 West Kingsway Road Middle River, MD 21220 Ph: (410) 574-6420 Fax: (410) 574-6443 PAGES INK COURT AND CONFERENCE REPORTERS 752 West Kingsway Road Middle River, MD 21220 Ph: (410) 574-6420 Fax: (410) 574-6443