1 DEPARTMENT OF HEALTH AND HUMAN SERVICES SECRETARY'S ADVISORY COMMITTEE ON HUMAN RESEARCH PROTECTION MEETING TUESDAY, MARCH 30, 2004 The Advisory Committee met in the Commonwealth Ballroom in the Holiday Inn Hotel and Suites, Commonwealth Center, 625 First Street, Alexandria, Virginia, at 8:30 a.m., Ernest Prentice, Ph.D., Chairman, presiding. MEMBERS PRESENT: ERNEST D. PRENTICE, Ph.D., Chair BERNARD A. SCHWETZ, D.V.M., Ph.D., Acting Executive Secretary CATHERINE SLATINSHEK, M.A., Executive Director THOMAS L. ADAMS, CAE, Member MARK BARNES, J.D., L.L.M., Member CELIA B. FISHER, Ph.D., Member ROBERT G. HAUSER, M.D., Member NANCY L. JONES, Ph.D., Member FELIX A. KHIN-MUANG-GYI, Pharm.D., Member SUSAN KORNETSKY, M.P.H., Member MARY L. POLAN, M.D., Ph.D., Member, M.P.H. SUSAN L. WEINER, Ph.D., Member MEMBER ABSENT: E. NIGEL HARRIS, M. Phil., M.D., D.M. EX OFFICIO MEMBERS: RACHEL BRAND, U.S. Department of Justice HOWARD L. BRADLEY, Social Security Administration FRANK BRICKFIELD, U.S. Central Intelligence Agency KATHRYN LYNN CATES, U.S. Department of Veterans Affairs ROGER CORTESI, U.S. Environmental Protection Agency PATTY DECOT, U.S. Department of Defense ANITA EISENSTADT, J.D., National Science Foundation SALLY FLANZER, Agency for Healthcare Research and Quality Food and Drug Administration DAVID LePAY, M.D., Ph.D. DAVID SHORE, National Institutes of Health 2 OHRP STAFF PRESENT: MICHAEL CAROME IRENE SMITH-COLEMAN PAT EL-HINNAURY KELLEY BOOKER JULIA GOREY KARENA COOPER KRISTINA BORROR RINA HAKIMIAN SHIRLEY HICKS JUDITH BROOKS JULIE KANESHIRO GLEN DREW MELINA HILL CARLA BROWN 3 A G E N D A ADDITIONAL BUSINESS: . . . . . . . . . . . . . . .4 Ernest Prentice, Ph.D. Chairman, SACHRP LITIGATION ISSUES PRESENTATION: . . . . . . . . . .5 Haavi Morreim, Ph.D. Department of Human Values and Ethics University of Tennessee College of Medicine BREAK: . . . . . . . . . . . . . . . . . . . . . 57 HIPAA PRIVACY RULE PANEL: . . . . . . . . . . . . 58 Susan McAndrew, Office of Civil Rights, Department of Health and Human Services Susan Ehringhaus, Associate General Counsel for Regulatory Affairs, Association of American Medical Colleges Mark Rothstein, Director Institute for Bioethics, Health Policy and Law Joanne Pollak, Vice President and General Counsel, Johns Hopkins Medicine INTERNATIONAL ISSUES - REVIEW/DISCUSSION: . . . .124 Dr. Mary Lake Polan LUNCH: . . . . . . . . . . . . . . . . . . . . .132 ADVERSE EVENTS ISSUES - REVIEW/DISCUSSION: . . .132 Ernest Prentice, Ph.D. PUBLIC COMMENT: . . . . . . . . . . . . . . . . .167 FUTURE BUSINESS ISSUES: . . . . . . . . . . . . .178 Including discussion on Subpart B Conflict of Interest Decisionally Impaired Subjects ADJOURN: . . . . . . . . . . . . . . . . . . . .245 4 P R O C E E D I N G S 1 (8:40:19 a.m.) 2 DR. PRENTICE: Okay, folks. Good morning 3 everybody. We're getting ready to start again. 4 Another long interesting day. I want to briefly 5 overview the schedule for you. As you can see, we are 6 beginning with additional business. There really is 7 no additional business, except to introduce our first 8 speaker, which I'll do in a moment, but we're going to 9 have an awful lot of other business later on during 10 the day so we're not going to spend time talking about 11 that at all. 12 We will move through the sessions. We'll 13 try to stay on time. As you can see, we're going to 14 have a talk on litigation. We're going to have a 15 panel on the HIPAA Privacy Rule. We are going to 16 change the schedule slightly. Dr. Mary Lake Polan has 17 to go back to Stanford on an early afternoon flight, 18 so what I would like to do is to have her talk be 19 given between 12 and 12:30, so we're going to delay 20 lunch for a half an hour. I hope that's okay. We'll 21 have lunch between 12:30 and 1:30, and then we'll move 22 into the talk on adverse events, which I will give at 23 1:30 to 2. We'll have the public comment period 24 between 2 and 3, so if you have a comment that you 25 5 would like to address to the committee, please sign 1 up. And then 3 to 3:15 a break, 3:15 to 4:00 we'll 2 continue with some future business issues, and then 3 hopefully we'll be able to adjourn on time. 4 All right. Moving right along, it's my 5 distinct pleasure to introduce to you a friend and a 6 colleague, Dr. Haavi Morreim, and she's a Professor of 7 Bioethics in the College of Medicine at the University 8 of Tennessee. She has a very long bio here, I'm not 9 going to read it all, but for over 20 years, initially 10 at the University of Virginia School of Medicine, and 11 thereafter, at the University of Tennessee, she's done 12 an awful lot of clinical teaching and consulting in 13 medical ethics. 14 She chairs the Independent Patient 15 Advocacy Council, created to serve patients enrolled 16 in the Abicor Artificial Heart Trial, and I'm sure 17 that a lot of you have read about that particular 18 trial and the subsequent lawsuit. So Haavi has really 19 gotten heavily involved in bridging ethics and law, 20 shall we say. 21 And a little side note, Haavi is a BMW 22 race car driver, so she comes here eminently 23 qualified, and when you hear how fast she talks, you 24 will understand why she drives race cars. 25 6 DR. MORREIM: Thank you. As a matter of 1 fact, I've really got to live up to that today because 2 I'm not going to be going through all of the slides. 3 I don't know if you all have copies of my slides out 4 there or not, but I'm not going to talk about all of 5 them. There's way more than we can possibly talk 6 about in the approximately 20 minutes that I get to 7 yap before we open it up, and so what I'm going to do 8 is basically just focus on a few slides and highlight 9 some issues. 10 By way of overview, I'm going to look a 11 little bit at litigation trends but not really, 12 because you all can read an article that I sent a 13 while back. I'm going to look a little bit at some of 14 the sequelae that might come about, insurance 15 challenges, a certain special issue, the propensity 16 that we're seeing to settle without litigating 17 anything. And above all, I'm going to focus on what 18 I think are some solutions for protecting IRBs, and I 19 think that's really important. 20 For this stuff, like I said, I've got to 21 just go through retro speed. I'm not going to read 22 this stuff. It's very hard to find out what's going 23 on out there by way of litigation because a lot of 24 this stuff gets settled before it even gets filed, so 25 7 we don't really know what's going on there. For the 1 overview, I refer you to an article that is almost on 2 the ground. It should have been out by now, but it 3 will be very shortly, so you've got the site. 4 Potential sequelae, again those of you who 5 have the outline can read that. Some of the problems 6 include just getting people to serve and stay on IRBs 7 as we find more and more litigation naming IRBs. I'm 8 going to be looking a little bit at some of the other 9 players in research, obviously investigators, CROs, 10 SMOs, clinical research coordinators, other folks, but 11 my focus is going to be on the IRBs. 12 Insurance challenges - what I did here was 13 I took a look. I found a number of people who are on 14 IRBs, both independent and institutional IRBs. I 15 talked to a few investigators, CROs, SMOs, some 16 sponsors and people like that just to get an idea of 17 what was going on in the insurance industry. 18 Basic problems - some very disturbing 19 trends I want to call your attention to. One of them 20 is, particularly for investigators, their medical 21 malpractice insurance is not covering research, 22 increasingly true. Some are. In Tennessee we've got 23 State Volunteer Mutual which is a physician- 24 owned/physician-controlled malpractice insurer. They 25 8 do cover research. A lot of insurers, however, are 1 saying Mr. Doctor Investigator, this isn't medical 2 practice; therefore, it ain't medical malpractice; 3 therefore, sayonara, good luck. 4 We're also finding other kinds of recision 5 actions. We are finding, deeply troubling here, if 6 you'll take a look at this one - refusals to cover 7 pediatric trials. And now we're starting to get to 8 see refusals to cover OB trials, so that an IRB that 9 is reviewing a pediatric trial may have to try and get 10 indemnification from the sponsor, or find some other 11 means. This is a very troubling development. 12 We also are finding some problems getting 13 indemnification for, for example, independent IRBs or 14 others who want to. There are some IRBs who do not 15 like the idea of getting indemnification from the 16 sponsor because it looks potentially like getting into 17 bed. A lot of IRBs do because they don't have a lot 18 of choice in this matter, but now we are seeing 19 sponsors wanting cross-indemnification from not just 20 IRBs, but investigators as well, saying look, you all 21 pay for our problems. If litigation happens, we want 22 you to indemnify us, so that's getting increasingly 23 problematic. 24 Premiums are based on a number of things, 25 9 and if you look at this slide and the next one, we are 1 seeing increases in insurance premiums, but as we see 2 30 to 50 percent rises per year, you have to 3 understand how the premiums are calculated. For a lot 4 of independent IRBs they are based on revenues. If 5 revenues are going up, and they are for a lot of 6 independent IRBs, then of course the malpractice 7 premium is going up as a function of doing more 8 business, so it's hard to parse out the actual rises 9 in costs. Similarly, for an SMO it's based per 10 patient. If they're doing more patients then, of 11 course, they're paying more. Again, it's hard to 12 track, and I really did not have much of a chance to 13 get at the details there. We do know the premiums are 14 going up rather substantially. 15 A major challenge here, and I've gotten 16 this from more than one source, insurers do not 17 understand much about research. They think it's 18 medical practice, or they don't understand where the 19 real risks come from. They think that IRBs perform 20 experiments on people. I mean, they don't understand 21 what's going on. A number of insurers have gotten out 22 of the business all together. There are basically two 23 insurers left, so those are some of the things that we 24 have to at least be aware of. 25 10 Also to be noted is that the increasing 1 costs of insurance really are not built into the 2 overhead costs. For example, NIH grants for sponsors. 3 Where are they going to come up with the extra money 4 to pay for this? 5 In the insurance industry, there are some 6 possible responses, and some of them are already 7 underway. As the economy improves, the entire 8 insurance market is going to improve. The whole 9 insurance market has been in the tank for years. Med- 10 mal has been in the tank for years, well before 9/11, 11 and now it's been in the tank more since then. 12 We're starting to get a comeback on the 13 economy, that's going to help. Brokers are now in the 14 process of trying to educate insurers about what are 15 the risks, where are they located, how to manage them 16 more intelligently than just yikes, raise the premium. 17 And we are seeing some market-based activity going on. 18 We're seeing consortium purchasing on the part of 19 independent IRBs. We're finding insurers now offering 20 special riders to investigators to cover research, 21 things like that. So the market processes are 22 beginning to activate, and we can look forward to some 23 improvement there. 24 Some of the solutions that I'm going to 25 11 offer or recommendations as I focus on Section 5 of 1 our outline, I think that that is going to help 2 considerably. If we can limit insurance risk, and 3 make the insurance risk more predictable, it'll be a 4 lot easier for insurers to have intelligent, 5 intelligible, reasonable pricing on their product. 6 Another challenge here is the temptation 7 to settle without litigating. As your outline 8 indicates, there are a number of reasons why it is 9 very tempting to settle without litigating, including 10 the difficulties of predicting what juries are going 11 to do, and the expense of litigating. 12 Some of the problems of settling, and 13 that's mainly what's going on. Right now even as we 14 speak, the Hutch trial is underway in Seattle. Maybe 15 not exactly even as we speak. It's almost 9:00 here, 16 so it might be a little early out in the Seattle area, 17 but the Wright against the Frank Hutchinson Cancer 18 Institute, that trial is underway. The plaintiffs 19 have made their case. The defense case is now 20 underway, and that is one of the relatively few cases 21 being actually brought to trial. 22 And, of course, then if things go further 23 and somebody doesn't like the outcome, you can be 24 rather confident that appeals will be made and so 25 12 forth. On the whole though, most of these cases are 1 getting settled no matter how flimsy, and that is 2 deeply troubling to me. 3 Part of it may be that insurers who work 4 in research are not as familiar with litigation in 5 this area, partly because it is a rather new 6 phenomenon as, for example, med mal insurers for 7 ordinary medical malpractice. You find that an awful 8 lot of med mal claims don't get settled, they get 9 tossed. They basically just get þþ the plaintiff 10 attorney realizes we don't have a case here, and then 11 the thing gets flat dropped with no money changing 12 hands. It's a very common outcome in ordinary med 13 mal. 14 We don't know how common it is here in 15 research, but it seems to be less common here. 16 Insurers seem more willing at this point to just hand 17 the plaintiff's attorney some cash and make him go 18 away. We don't know. But one of the problems with 19 premature settlement is this fourth one that's 20 mentioned, the third and fourth one, is that you get 21 all these unexamined, unrebutted claims that are left 22 out there hanging as though they might be some sort of 23 standard of care. 24 I culled from some of Alan Millstein - he 25 13 is an attorney who has been very active on the 1 plaintiff's side in litigation. Many of you know that 2 person's name very well here. I culled from some of 3 his complaints some of the so-called responsibilities 4 that IRBs have, and some of them are deeply troubling. 5 The idea that the IRB is supposed to approve all 6 aspects of a clinical trial. Take a look at number 3 7 - ensure that the research complies. How can anyone 8 ensure that all of the factors of the research comply 9 with federal requirements? 10 Number 4- assisting the sponsors in 11 developing protocols. Is that the IRB's job? That's 12 being put out there as though it's some sort of 13 standard. I am troubled by this concept. Take a look 14 at the fifth one - appropriately monitoring the 15 informed consent process and the conduct of the 16 experiment. Yes, there is some opportunity for IRBs 17 to monitor. It is very unclear what kind of 18 monitoring IRBs really are supposed to do. Are they 19 supposed to be in the room every time a consent is 20 done? That's the only way that you're going to find 21 every inadequate consent. I think that's an 22 unreasonable standard to place on IRBs, and make sure 23 - take a look at the bottom one - ensure proper 24 reporting. No IRB can do that. Make sure that all 25 14 the investigators report all the adverse events they 1 are supposed to. How can an IRB do this? These 2 standards have been proffered, they're out there, they 3 have not been rebutted. I find that troubling. 4 Okay. Potential solutions. I'm going to 5 look at four. First of all, arbitration, which I 6 think is a very interesting option, and then I'm going 7 to look at three different models of importing federal 8 immunity. And I think this is a very promising kind 9 of approach. As I say, three different models, and 10 each one of them has a lot to recommend it. 11 First of all, arbitration. There is 12 actually since 1925, the Federal Arbitration Act 13 basically has said the federal government strongly 14 endorses for public policy reasons, strongly endorses 15 the idea of resolving disputes by arbitration, 16 including binding arbitration. State statutes, a lot 17 of them replicate the FAA, and courts have been very, 18 very favorable in upholding and enforcing arbitration 19 agreements; not just any old arbitration agreement, 20 because there are some constraints. The arbitration 21 agreement has to be written in certain ways that are 22 fair, and I'll outline I think what should be an 23 arbitration agreement that might go into a research 24 situation, but there are a number of advantages to 25 15 arbitration. It's simpler, rules of discovery are a 1 lot more relaxed. Time frames are smaller. The 2 expense at least in terms of the litigation process is 3 vastly smaller in most cases. You may or may not have 4 a smaller outcome, but the administrative costs are 5 vastly less than the cost of litigation in most 6 instances. And it's more expeditious. The injured 7 person does not have to wait years, and years, and 8 years before getting perhaps a jackpot, perhaps 9 nothing. 10 Some of the major themes of courtsþ 11 decisions - I went through a number of decisions 12 regarding arbitration contract, and not just in 13 healthcare, but you look at employment, you look 14 across the board, there are a number of different 15 places where you find arbitration contracts. Courts 16 emphasize - a couple of things I'll highlight here. 17 One, they emphasize this is a contract. This is a 18 contract between two parties. And number two, that 19 this is not the þ and this is the U.S. Supreme Court, 20 very important - it is not a waiver of rights. It is 21 merely a different venue for seeking to enforce one's 22 rights. 23 So that as you look at the IRBs, you know, 24 the consent forms and so forth that IRBs have to pass 25 16 on, and they say oh, well, we can't have the research 1 subject waiving rights. Arbitration is not a waiver 2 of rights. It is just a different venue for exploring 3 and vindicating one's rights. Arbitration contracts 4 are not generally overturned unless you have done 5 something that would overturn any kind of contract, 6 like it's fraud, it's undue inducement, it's seriously 7 unequal bargaining parties, strength of the respective 8 bargaining parties, unconscionability, things like 9 that. 10 So here are some recommendations for what 11 might go into an arbitration agreement that would go 12 into research. One is, it's got to be clear and 13 conspicuous, big letters, and maybe even a separate 14 sheet of paper. Very good idea to have a take-home 15 copy of something like this, of the consent form, and 16 especially the arbitration agreement. It needs to be 17 fully voluntary. In other words, it cannot be a 18 prerequisite for getting into this trial, and that 19 would be especially important for things like Phase I 20 trials for people who are very ill. A little bit 21 different for normal healthy volunteers, perhaps, who 22 are being paid for their participation, so we've got 23 a spectrum of considerations to deal with here. 24 I recommend making it fully voluntary and, 25 17 indeed, in one case, or more than one case that I 1 read, a 30-day opportunity to revoke was looked on 2 very favorably by the courts as a reason to enforce 3 this contract. Look, the guy had a take-home copy. 4 This was not in the research context, but the fellow 5 had a take-home copy. He had 30 days to change his 6 mind. The court said what more do you need? Of 7 course it's binding. 8 We also want to make sure that they are 9 fair bilaterally, that you have reasonable time frames 10 and discovery opportunities. And don't make the 11 research subject pay the cost of the arbitration. 12 That'll kill it in a heartbeat. Okay. Make sure that 13 the research subject does not have to pay cold hard 14 cash in order to vindicate his or her rights in this, 15 and make sure that it's implemented carefully. In 16 other words, a legally authorized representative, if 17 not the patient - not just any old buddy who looks 18 like might be a suitable person to sign. 19 Now federal immunity, and I'm going to 20 look at three different types. The basic concept here 21 comes from Old English law going back to the 1200s, if 22 I'm not mistaken; and that is, sovereign immunity - 23 the king can do no wrong. Governments on the whole 24 worldwide do not allow themselves to be sued as they 25 18 go about the business of governing the people, unless 1 the government has drawn some sort of exception, a 2 waiver to sovereign immunity. 3 Our government has drawn a number of 4 exceptions to sovereign immunity, and I'm going to 5 explore several of them. The idea behind sovereign 6 immunity is that as people try, and as government 7 officials try and govern and do their things, they 8 need to have some kind of realm of flexibility for 9 making the discretionary judgments that they have to 10 make without being judicially second-guessed every 11 time they turn around. The government has got to 12 work, has got to function. 13 The idea for an IRB is, the underlying 14 concept, IRBs are federally mandated, federally 15 specified in their activities, and federally 16 supervised. They are, in a sense, quasi-government 17 operators in this way, and there are a number of ways 18 in which we might flesh that out. Oh, and by the way, 19 for the longshot advantages of this - to the extent 20 that we make IRBs at least immune in a qualified way, 21 and by qualified immunity what I'm talking about here 22 is the idea that they're presumed to be immune, but 23 not completely immune. Yes, they could be liable 24 under certain circumstances. That's the qualification 25 19 of the immunity. And to the extent that we could use 1 one or more of these models, we could very 2 substantially reduce the cost of suits because with 3 immunity, what you want first off the bat is for a 4 judge, not a jury, a judge to decide, is this immune 5 from suit? Are these people immune from damages? Is 6 this the kind of thing that renders immunity? If the 7 answer is yes, you don't even go to discovery, which 8 substantially cuts down on litigation costs. 9 Typically, these immunity also have 10 unavailability of punitive damages. And in some cases 11 you also have the option of the defendant gets to 12 recover the legal costs from the plaintiff, and that's 13 a big disincentive. If Alan Millstein stood the 14 chance, or any other plaintiff's attorney stood the 15 chance of having to pay not only his own expenses but 16 the other side's expenses, that would be a significant 17 disincentive to frivolously naming an IRB or some 18 other entity in the context of research-oriented 19 litigation. So let's take a look at the first one - 20 the Healthcare Quality Improvement Act of 1986. 21 The story behind this basically is, you 22 take a look at hospitals and their peer review 23 committees, those peer review committees have a very 24 public policy-oriented job of look at their fellow 25 20 physicians and if they find, for example, the surgeon 1 has a terrible technique. He's killing off people or 2 causing injuries right and left, getting a lot of 3 infections, they might want to review the activities 4 of that surgeon or other kind of doctor, and 5 potentially either curtail, or suspend, or even 6 outright withdraw that person's privileges to practice 7 at that hospital. Peer review does a very important 8 function for quality of healthcare. 9 What was happening is, back in the early 10 80s and before, what was happening is that some of the 11 doctors who were being reviewed and having their 12 privileges either suspended, curtailed or revoked were 13 suing the peer reviewers. They're saying wait a 14 minute, you're not trying to improve healthcare 15 quality. You're just trying to get rid of business. 16 I'm going to sue you on anti-trust grounds, treble 17 damages. This chills peer review, needless to say. 18 So Congress said we need to give these people some 19 immunity to protect this very important function. And 20 so basically, if the reviewers meet these very basic 21 standards, then they are immune from liability. Not 22 immune from being sued, but immune from liability; 23 namely, they were reasonably in the belief that their 24 action was to improve the quality of healthcare or 25 21 promote þþ they reasonably tried to get the facts, 1 they reasoned carefully, gave adequate notice of 2 hearing to the person, and their conclusions were 3 reasonably based on the evidence that they had. 4 Important features - this is a rebuttable 5 presumption. The burden is on the plaintiff. Now 6 remember in typical tort litigation, the burden is 7 also on the plaintiff, but here it's especially on the 8 plaintiff. They have to prove that the peer reviewers 9 were out-and-out unreasonable, not just that they made 10 a mistake, but they were unreasonable. So even if 11 they make an error that's okay, as long as it was a 12 reasonable error, and they then are immune, very 13 important. 14 Judges also, not juries, decide this. 15 You're not leaving this to twelve good and true. 16 You're leaving this to a judge who understands the 17 concept of immunity. Also, there is the prospect of 18 the defendant who prevails collecting attorney fees, 19 which I think is a very interesting thing. And as I 20 say, a very strong potential deterrent to frivolously 21 naming the IRB just because they happen to be 22 somewhere in the picture. 23 Courts have very happily upheld the 24 Healthcare Quality Improvement Act. I have yet to 25 22 find a case, there must be one somewhere, I have yet 1 to find a case, however, where judges have said yeah, 2 these guys weren't very reasonable. Go ahead, 3 collect. I have yet to find such a case. Very strong 4 endorsement by courts here. 5 The basics with IRBs are parallel, 6 obviously. Important healthcare quality and research 7 human protection functions. Instead of due process 8 for the accused, what you're looking for is 9 reasonableness of care in gathering facts and so forth 10 to make reasonable decisions regarding protection of 11 human research subjects. And here's a very rough 12 draft on slide 27 of how you might actually parse this 13 out, and adapt the HCQIA to protect IRBs. I'm going 14 to shoot on from there. Sorry. You've got it in 15 your hands hopefully to read that and go back later. 16 The second one, the Federal Tort Claims 17 Act, which was enacted back in 1946. This is very 18 directly based on sovereign immunity, and as pointed 19 out here, the sovereign immunity, this act waives it 20 so that the federal government will allow itself to be 21 sued under circumstances where somebody else might 22 also be suable, a private person under similar 23 circumstances, but there are very important exceptions 24 that I'll get to in a second. What happens here is 25 23 that instead of suing that individual, typically it 1 would be some government worker, you sue the federal 2 government. You file an administrative claim, the 3 Attorney General says yeah or nay, that this is, 4 indeed, a person who is employed by the government 5 acting within the scope of his employment, and then 6 the exclusive remedy is against the government. 7 Judges, not juries, decide these cases - no punitive 8 damages. And think about when you translate to IRBs, 9 think about the ways in which that will limit the 10 costs for insurers, where they're not worried about 11 this wildcard out there of what kind of multi-million 12 dollar punitive damage suit are we going to have to 13 cough up the money for. That's off the table, once 14 you get into the immunity territory. 15 Exceptions especially focus on 16 discretionary functions. Although the government 17 allows itself to be sued, no can do if we're talking 18 about a government official exercising discretion in 19 ways that have to do with public policy issues, or 20 similarly social-economic, as well as political kinds 21 of policy. Think about the ways in which IRBs 22 exercise discretion, making decisions that are policy- 23 oriented decisions. 24 Okay. Interesting historic example here, 25 24 and it's not historic, it's actually contemporary - 1 back in 1992 the federal government, Congress decided 2 that it's federally funded community and migrant 3 healthcare centers needed protection from the cost of 4 malpractice insurance. It was another of our 5 malpractice insurance crises, and these federally 6 funded centers wound up spending a lot of federal 7 money paying for malpractice insurance premiums, and 8 they were able to do less and less by way of providing 9 care for the people that they were federally funded 10 and set up to care for. And so the Congress said, you 11 know, let's at least try for three years as an 12 experiment pilot project to give Federal Tort Claims 13 Act coverage to the people who provide care in these 14 community centers. 15 Well, three years later they made it 16 permanent, and actually at this point although the 17 healthcare centers can buy GAP insurance coverage, 18 they don't need to buy comprehensive healthcare, or 19 excuse me, malpractice and other kinds of liability 20 insurance. 21 What happened there, and by analog, the 22 idea of protecting IRBs via this kind of thing is that 23 once again, you have something, although IRBs are not 24 federally funded per se, they are federally mandated, 25 25 they are federally specified and federally supervised. 1 They are created to serve the public interest, and the 2 parallels are obviously very strong. 3 The advantages obviously of bringing IRBs 4 literally under the FTCA, and I think this would be a 5 very interesting analogy, you would, if you want to 6 sue an IRB, you have to sue Uncle Sam, and Uncle Sam 7 decides if you get to sue him or not. That would be 8 a fairly significant deterrent to litigation, and 9 above all, it would protect the IRB members. They 10 would no longer even possibly be individually liable 11 so long as it was covered by the FTCA. Some things 12 are not covered by the FTCA; intentional torts. And 13 as I say, there would still be a need for GAP 14 coverage, but it would be substantially less, and 15 substantially more circumscribed. 16 Okay. We then get to the last model which 17 is a little bit more complicated, and I'm still, I 18 confess, kind of sorting my way through the Section 19 1983 and Bivens stuff. And Mark Barnes and I were 20 chatting on the phone about this. I'm trying to get 21 that book you recommended, and I called the library. 22 Yeah, yeah, we have it. I went to the library, nah. 23 So at any rate, what I'm presenting here is the 24 product of several weeks of very intensive work as I 25 26 was invited to do this, and to think about litigation 1 issues, especially with reference to IRBs, and so I 2 have not been able to absolutely clear the plates on 3 some of these concepts. 4 Basically, 1983 and Bivens are sort of 5 dovetailing. Again, you begin with the presumption of 6 sovereign immunity, the king can do no wrong. You 7 cannot sue the government, except. Section 1983 8 actually was enacted in 1871. It was called the Ku 9 Klux Klan Act. It was civil rights litigation, 10 because we were finding after the Civil War, because 11 we were finding that a lot of folks who were acting 12 under the color of state, meaning not government 13 generically, but literally the 50, or at that time 14 maybe, what was it - 48 states, however many states we 15 had at the end of the Civil War, they would act under 16 color of state law and basically ignore the federal 17 protections for minority persons, such as freed slaves 18 and others. And so basically what Section 1983 wanted 19 to do was to make sure that people could enforce 20 constitutional rights, even if they are being trampled 21 on by state officials, such as police officers or 22 state troopers, whoever that might be. 23 Okay. So that's in the ambit of the 50 24 states. A Bivens action is with respect to federal 25 27 employees. You note a moment ago that I said one of 1 the exceptions for the Federal Tort Claims Act is that 2 when you have a federal agent or government official 3 who is exercising discretion, making policy decisions, 4 normally that person is exempt, and the government is 5 exempt from litigation. The exception carved out by 6 the two of these is, if this government actor trampled 7 on a constitutional right that was obviously a 8 constitutional right, and knowable as such to a 9 reasonable person at the time, then there can, indeed, 10 be liability. So it's people acting under color of 11 state or federal law performing discretionary 12 activities, and trampling constitutional rights. 13 The idea here is on the one hand, you 14 really need to protect citizens from overbearing 15 public officials. On the other hand, you don't want 16 public officials to be subjected to judicial second- 17 guessing every time they make a decision, so that's 18 the balance the Feds are trying to create with both 19 Bivens and Section 1983. 20 What we're looking at for IRBs here is 21 kind of the obverse of all that, what would they be 22 protected from if they were brought into this kind of 23 model? So we would look at very similar themes to the 24 ones that I've been drawing for, the Healthcare 25 28 Quality Improvement Act and Hospital Peer Review, and 1 similarly, the FTCA. You're looking at giving IRBs 2 qualified immunity. In this case where they would be 3 liable is only when they are literally trampling on 4 someone's constitutional rights, clearly establish 5 constitutional rights knowable to be such at the time; 6 things like due process, revocation of due process, so 7 that as long as IRBs were engaging in careful 8 procedures and so forth, they would probably be and 9 remain immune from tort litigation so long as they 10 were honoring people's constitutional rights. So 11 that's that idea. 12 And by way of summary, and then we can 13 open it up to discussion, the reasonableness of the 14 conduct across the board here is what tends to secure 15 the immunity. Judges, not juries, determine immunity 16 on this theory for the IRB. The burden would be on 17 the plaintiff to show that the IRB's action was 18 clearly unreasonable or that it trampled 19 constitutional rights, knowable as such to a 20 reasonable person. And immunity, ideally would want 21 immunity from suit. ACQIA gives immunity only from 22 damages. Ideally, it would be nice to have immunity 23 from suit, but one or the other, either way, whatever 24 you can get would be a great help, no punitive 25 29 damages. And then again a very important thing that 1 I would like to see built in is to protect potential 2 defendants by allowing them to collect attorney's fees 3 where the plaintiff's action was unreasonable in 4 naming the IRB. 5 Overall recommendation - one thing I would 6 like very much to see from the administrative end, and 7 this goes back to some of the first things that I said 8 in this brief set of remarks, is that we need to make 9 it a little bit clearer as to what our IRB's 10 responsibilities, particularly regarding ongoing 11 monitoring both of the informed consent process, and 12 of the undertaking of the research itself, especially 13 what are not their responsibilities for ongoing 14 monitoring. This is a very big hole, and IRBs are 15 falling into it, and down and down they go until we 16 get some kind of administrative clarification on that, 17 and hopefully constraint, circumscription of what kind 18 of responsibilities IRBs have there. 19 Also from the administrative level, I 20 would like to see some kind of mechanism to track 21 litigation. It would be nice. I'm not sure that we 22 easily can, but it would be kind of helpful. Also, to 23 encourage in some way a creative marketplace with 24 respect to the insurance issues. And then finally, it 25 30 would be interesting to see the legislature enact 1 qualified immunity. 2 I might mention, by the way, that with 3 respect to this third option for qualified immunity 4 that I mentioned, Section 1983/Bivens kind of thing, 5 there are some arguments, and I've been digging around 6 in this in case law, it is at least conceivable that 7 some courts might grant IRBs Section 1983/Bivens-type 8 immunity even now, because looking through some of the 9 older Supreme Court decisions here - Supreme Courts, 10 for example, in one decision 20 or 30 years ago 11 extended federal immunity to school board officials. 12 In this particular case what happened was 13 some school kids spiked the punch for a meeting of an 14 extracurricular activity, so the school board guys 15 suspended these two students, and the students sued, 16 and the U.S. Supreme Court said in previous cases we 17 have found that state troopers, we have found 18 policemen, National Guardsmen are immune from 19 liability because they are performing quasi-government 20 activities. We hereby find this extends to school 21 board people because they are engaged in quasi- 22 government kinds of activities. They are serving a 23 public policy function, and they deserve immunity. 24 Well, wouldn't it be interesting if a court said IRBs 25 31 are serving a public policy function that should not 1 be chilled, same rationale applied from case, to case, 2 to case. And it's a very familiar set of themes that 3 go across the case law here. And IRBs fit very 4 squarely within an awful lot of the case law regarding 5 all three of these kinds of federal immunity, so I'm 6 going to be quiet and turn it back over to Mr. 7 Chairman. Thank you for your attention. 8 DR. PRENTICE: Thank you, Haavi. You've 9 fully lived up to my expectations. 10 DR. MORREIM: Well, somebody said I speak 11 at 285 words a minute with gusts up to 482. 12 DR. PRENTICE: I've got a number of 13 questions, but I'm going to limit my questions to I 14 think one, with one comment. In looking at your 15 extraction from various complaints, there is one that 16 I would like to amplify on. And I don't know whether 17 you touched on that in your research. That's the 18 Richard Cuss v. Sherman Hospital case in Illinois, 19 where the appellate court basically determined that 20 the IRB had a duty to ensure that Richard Cuss, who 21 was the plaintiff, actually signed the IRB approved 22 consent form. So it wasn't good enough for the IRB to 23 simply approve a consent form, and then inform the 24 investigator he had to use that consent form. They 25 32 actually had to make sure that he, indeed, used the 1 correct consent form. That was settled out of court 2 because the insurance company mandated a settlement 3 when it was remanded back to district court. 4 The question I have concerns your comments 5 on arbitration. I've been involved in one case in 6 California where there was arbitration. It was a 7 research-related case. Arbitration was mandated as a 8 condition for participating in the clinical trial, 9 which goes counter to what you're recommending here. 10 So my first question would be, why would a subject 11 want to voluntarily agree to arbitration? The chances 12 are that if you sue, you're going to get a lot more 13 money if you don't arbitrate, so that would be one 14 question. The second question would be, does 15 arbitration offer any protection for IRBs? 16 DR. MORREIM: Your question or your 17 hypothesis that you get more money if you sue, I'm not 18 sure I would buy that from the data in terms of 19 medical malpractice, if that's an analog. Most 20 medical malpractice cases get either tossed or 21 settled, and so if one is looking at settlement, then 22 in those kinds of cases, arbitration versus 23 settlement, arbitrators are sometimes more generous to 24 plaintiffs than juries are in some cases, and they are 25 33 sometimes more generous to plaintiffs than would be a 1 judge. And I say sometimes. I think we would 2 actually have to track the data, but in terms of 3 collecting money for the plaintiff, given the number 4 of suits that either get dropped or that wind up 5 taking endless amounts of time, and remember the 6 attorney gets a third in liability litigation, 7 malpractice, research-oriented and so forth, so that 8 what the person actually winds up with in the end 9 years out, that's not so clear. Some people hit the 10 jackpot, a lot of people don't. 11 Another factor, and this is one of the big 12 advantages for arbitration, is that smaller claims can 13 actually collect. Typically, the cases that wind up 14 attracting a plaintiff's lawyer are the ones that are 15 big enough to attract the big bucks, and so a lot of 16 the people have small claims that are just not worth 17 bothering with, if you're a plaintiff's attorney 18 because it's expensive to file and pursue litigation. 19 It's very expensive, and so a lot of those cases 20 simply are never taken up. And so somebody who has a 21 small claim has a chance with arbitration, and 22 basically no chance. 23 We also have the expeditiousness of the 24 settlement, that you don't have to spend endless years 25 34 dragging this thing out. There are a number of 1 advantages for prospective plaintiffs. It becomes an 2 interesting question how much "informed consent" would 3 you have to engage in for the arbitration. There are 4 downsides to these, which is why I'm more inclined 5 toward the federal immunity approach than the 6 arbitration approach, but I think we need to put it on 7 the table and look at it carefully, because if you 8 don't get any arbitration - or excuse me - if you 9 don't get the immunity, arbitration may be a 10 reasonable alternative. 11 DR. PRENTICE: What about protection of 12 IRBs? 13 DR. MORREIM: It would seem to me that 14 perhaps you could write that in. I invite others 15 comments on that, but that can be a contract. 16 Arbitration is a contract. And perhaps, you know, 17 another possibility - we talk about indemnification, 18 I guess þþ I don't want to get too far into 19 speculation here, but there are various possibilities 20 that one might write by way of arbitration contracts 21 that would include the IRB within any kind of 22 arbitration agreement. 23 DR. PRENTICE: Okay. One final comment. 24 I would think that if you did have an arbitration 25 35 agreement, a plaintiff would be less likely to include 1 the IRB in any kind of a suit. 2 DR. MORREIM: I believe so. 3 DR. PRENTICE: Okay. 4 DR. MORREIM: Yes. 5 DR. PRENTICE: All right. It's opened up 6 to the floor now. Raise your hands so I can see you 7 if you want to ask a question. Yes. 8 DR. HAUSER: Haavi, that was terrific. 9 RPMs were maximum on that. I have a couple of 10 questions. One is, what type of mechanism should we 11 develop to track litigation, and perhaps understand 12 better the occurrences out there regarding 13 malpractice, and incidents that would give us a better 14 base of knowledge regarding this problem. Do we need 15 to put together some sort of registry or data 16 collection mechanism so that we better understand 17 exactly what we're dealing with, so that better policy 18 can be devised? 19 DR. MORREIM: I would like to see better 20 information. I really would, and I think we need some 21 creative problem-solving in this realm. We might þþ 22 I hate to mandate that IRBs that have been sued file 23 somewhere, but I would sure like to encourage some 24 sort of registry, at least an informal one, where IRBs 25 36 that have been named, or obviously other people 1 involved in research can at least identify that they 2 have been named. Here is the complaints. Complaints 3 tend to be, obviously, public documents, and so if 4 someone has been named in a suit, then your insurer 5 already knows, so it's not like you're spilling the 6 beans. And so I would hope that there would be 7 relatively few disincentives to some kind of central 8 informal registry, but I think we need better data, 9 and I would hope that at least sharing information 10 about what kinds of complaints are being brought out 11 against what kinds of entities, and what happens to 12 them. I think it would be a useful thing. I'm not 13 familiar with the mechanics of assembling such a 14 thing, but I like the idea. 15 DR. HAUSER: The second question I have is 16 around immunity. Have you thought through how you 17 would extend immunity to IRBs, and what the IRB may 18 need to do in order to qualify for that immunity? 19 DR. MORREIM: The ways in which one would 20 extend, if you were to use the Healthcare Quality 21 Improvement of 1986 Act, Congress would enact it. And 22 the wording that I offered as a very rough draft might 23 be the way in which such legislation would be enacted. 24 It would require legislation, at least on the HCQIA 25 37 approach. And the rationales would be pretty much the 1 same, and so you would simply carve it out and say 2 IRBs, because they are federally mandated, their 3 activities are federally specified, and they are 4 federally supervised are immune so long as their 5 decision-making is reasonable according to these 6 standards, so that's how you would do it there. 7 Alternatively, if you follow the FTCA, the 8 Federal Tort Claims Act approach, you would do it, and 9 again it would take Congress if you're going to do it 10 the way that they did it for the federally funded 11 community and migrant healthcare centers, but again 12 you would do it on that model if you took the FTCA 13 approach, and you would write into the legislation we 14 hereby bring IRBs under the FTCA, and they are immune 15 so long as - and that means if you want to sue an IRB, 16 you're suing Uncle Sam. And they're immune so long as 17 they follow these criteria. Obviously, intentional 18 torts are not part of the FTCA. Nobody gets immunity 19 from the feds for intentional torts like battery, 20 fraud, things like that. I don't know if any insurers 21 who are going to immunize you for that either, but 22 those are a couple of approaches that would require 23 legislation. 24 The Section 1983 thing is þþ one way you 25 38 could get there is hope to goodness that some court, 1 like the Supreme, would just declare that IRBs þþ 2 DR. HAUSER: Okay. So the legislation 3 would contain certain qualifications that the IRB had 4 to meet in order to qualify for immunity. That's the 5 substance of the þþ 6 DR. MORREIM: Yes. I mean, you look at 7 FTCA, it's not that they specified a lot of ditsy 8 details for these community health centers to qualify. 9 If they're a federally funded community health center, 10 however those are set up, then they're under. If they 11 choose þþ it was voluntary, by the way, but now it's 12 permanent for those who choose to sign up. And maybe 13 you could make it a voluntary thing. It's hard to 14 picture an IRB not wanting to be brought under that 15 kind of immunity. 16 In other words, I think one thing that's 17 worth pointing out is that for HCQIA, the peer review 18 bodies themselves didn't have to meet some sort of 19 detailed list of qualifications in order to become a 20 peer þþ if it's a hospital peer review body, as the 21 hospital sets these up, then you're under this. And 22 the only criteria are, did you engage in reasonable 23 procedures to get the facts, give reasonable notice to 24 the person you're looking at, et cetera, et cetera, 25 39 fairly broad. 1 DR. PRENTICE: Mark. 2 MR. BARNES: Thank you, Haavi, for all the 3 work that you did leading up to this, as well, because 4 I know in conversations you had with Ernie and me, you 5 were deeply involved in this for a while. 6 Just a couple of points. One is that if 7 one wanted to capture data in regard to what kind of 8 suits are going on out there, there actually is a way 9 that the department has right now to capture, perhaps 10 not all of the data but a good deal of the data, and 11 I think it's worthwhile for us to note that; and that 12 is that under the Healthcare Quality Improvement Act 13 of 1986, that also created a registry of med mal 14 payments in the National Practitioner Data Bank, or 15 NPDB, which is a national registry of any kind of 16 malpractice payments when a physician is one of the 17 co-defendants, a defendant or one of the co-defendants 18 in an action. And any resolution, including 19 settlement of those cases, must be reported to the 20 NPDB with the designated amount of settlement and what 21 the apportioning of liability is between the different 22 physicians, the hospital, and others who are named as 23 co-defendants in the suit. So the federal government 24 keeps this registry right now, and all malpractice 25 40 insurers, all hospitals, all group medical practices, 1 they're all required to report to the NPDB when there 2 is a settlement. 3 One thing that could be done is that it's 4 not in the NPDB methodology now, but it would be 5 possible for the department, which has administrative 6 authority over the NPDB, with full authority to 7 implement the NPDB and design regulations for NPDB 8 under the authorizing legislation, the HCQIA, it would 9 be possible for the department to put some checkboxes 10 or basically to design the system in such a way that 11 research-related claims are flagged, so that there 12 would be a registry of cases and of settlements, and 13 of judgments. That is something the department could 14 do right now. And it may be that we should think 15 about recommending that as a group to the department. 16 HRSA is the entity that oversees the NPDB, so that's 17 one point that I wanted to make. Did you want to 18 respond? 19 DR. MORREIM: You have noted very 20 appropriately an important distinction that we would 21 want to keep track of; namely, where you have med-mal, 22 ordinary medical malpractice, all the NPDB, the data 23 bank keeps track of is what happened in the end. And 24 in the med-mal context, we have had a certain skewing 25 41 of activity to avoid, you know þþ listen, I'll go away 1 quietly if I don't have to pay a settlement so that it 2 doesn't have to go on the data bank. 3 As you have very nicely noted, what we 4 would want to make sure and ask for by way of any 5 expansion is to keep track not just of outcomes, but 6 of just the filings themselves, and the initial 7 filings, which is not something the data bank 8 currently gathers. But you've nicely kept track of 9 the difference there, and I just want to accent that. 10 Thank you. 11 MR. BARNES: And then one other point. 12 And I want to express a point of view which is not my 13 own, but I think it's worthwhile expressing. In fact, 14 I loathe the point of view I'm about to express, but 15 I think that it's worthwhile putting on the table so 16 that you can respond to it; and that is, and we hear 17 this þþ this would be Alan Millstein's point of view, 18 not to speak for him, obviously, but I've heard him 19 say what I'm about to say in different fora. And that 20 would be look, IRBs have been negligent. They've been 21 lazy. They have been þþ investigators have been lazy 22 and negligent. They experiment on people. People get 23 hurt, and the tort system is a good way to punish 24 people who do bad things. It's a good way to punish 25 42 IRBs and investigators who don't play by the rules, 1 who don't inform people of their rights, and so why 2 not let the American tort system let it work, and it 3 will reform behavior. 4 DR. MORREIM: I agree. I loathe the þþ 5 I'll tell you, I think there is a certain theoretical 6 measure of credibility if the tort system actually 7 worked the way it's supposed to. If the tort system 8 actually ferreted out and blamed blame-worthy conduct, 9 yes, okay. And in my writings þþ and Millstein and I 10 have actually appeared on the same platform - that was 11 colorful. And it was right when he had sued and the 12 suit was in process for the Number 5 Patient for the 13 Abicor Artificial Heart Trial, with which I am deeply 14 involved, and that was really an interesting sub-plot. 15 At any rate, I think that one of the 16 biggest problems here is that the tort system isn't 17 working the way it hypothetically ought to. And as I 18 said to Millstein at the time, I have been very public 19 in being in favor of appropriate kinds of tort 20 liability across the board, but it needs to be based 21 on blame and reasonableness, and that is not where we 22 are right now. 23 Indeed, the kinds of immunity that I am 24 structuring or recommending that we structure might 25 43 more closely resemble the way the tort system ought to 1 work, where you're focusing on reasonableness. Now it 2 would be with a certain kind of accent; namely, the 3 plaintiff would have to prove unreasonableness in 4 order to prevail, as opposed to merely having to show 5 something of a softer standard than that. But I think 6 the kinds of immunity that I am proposing would more 7 closely resemble the kinds of things that the 8 hypothetical Millstein argument is trying to achieve. 9 And frankly, I think that some IRBs have been less 10 careful historically than they need to be, and I think 11 some of the both federal actions and the litigation 12 has been a wake-up call, and wake-up calls are useful. 13 The problem is that now that everybody is wide awake, 14 we are at the point where we can þþ this kind of 15 litigation can inspire very perverse activity, time 16 wasting, effort wasting activity. And I commend to 17 your attention the slide that talks about potential 18 sequelae. It's slide number - whatever the heck it is 19 - slide number 5. I think we're going to start to see 20 perverse responses that do not improve human research 21 safety. 22 DR. PRENTICE: Felix. 23 DR. GYI: Thank you for a very 24 enlightening talk. Those are þþ not being an 25 44 attorney, I find those options to be very interesting. 1 From your perspective in having gone 2 through advising subjects and looking at the wealth of 3 information they have to sign prior to entering into 4 a study, I wonder if we are placing the burden on 5 subjects by even looking at arbitration as a feasible 6 alternative, because that's one of those areas where 7 we then go back to the subject and say read some more, 8 adopt, digest information prior to participating in 9 the study. And then, by the way, forget all the 10 things you read in the informed consent form because 11 here's another document for you to digest. I wonder 12 if you might comment on that and see what the utility 13 of something like that is in this particular setting, 14 even though that may be a reasonable legal option. 15 DR. MORREIM: I think that the challenges 16 of achieving another informed consent of sorts are 17 significant, and I think that would be a significant 18 downside. I think probably the best thing to 19 recommend arbitration - again, if we can't secure 20 appropriate kinds of immunity - is that it's something 21 that could reduce some of the onus on IRBs. To the 22 extent that it could at least displace from the 23 litigation forum onto the arbitration forum, to the 24 extent that it could displace a lot of that activity 25 45 into a more confined, a more constrained, a more calm 1 headed forum, I think to that extent - and thereby 2 reducing IRBs' fears of being sued, et cetera, et 3 cetera, et cetera - to the extent that it can then 4 encourage IRBs to be more relaxed in the sense of not 5 getting into perverse responses to litigation, it 6 could foster a better atmosphere for IRBs to work, so 7 that in that indirect sense, benefit patients. 8 Also, I think it is important to remember 9 that arbitration does hold certain very significant 10 potential advantages for the research subject class 11 patient, and it's not like they're giving up something 12 major. Yes, they're giving up some things, but as the 13 courts have pointed out again and again, the person on 14 the patient side, the smaller party, can gain a great 15 deal from arbitration. 16 So many cases wind up, as I pointed out, 17 either not being brought about in the first place 18 because they're too small, or they wind up being 19 endlessly lost in litigation, or the person winds up 20 collecting very little at the end because of all the 21 court expenses that get deduced. There are a lot of 22 advantages. It is true, a very significant 23 disadvantage could be the possibility of informed 24 consent for the arbitration. 25 46 On the other hand, something that can be 1 rescinded within 30-days. It's not like you have to 2 achieve instant understanding this moment before you 3 step out of the office. If this person has a take- 4 home copy, the opportunity to ask questions and 5 rescind the agreement any time within 30 days, that 6 might be a good protector for those up front 7 information costs, if you will, for an arbitration 8 approach. 9 DR. PRENTICE: Celia. 10 DR. FISHER: Hi, Haavi. Thank you very 11 much. I think for me, I have a number of points. 12 They're points and questions. For me what I think is 13 most significant with respect to some of your 14 suggestions is the notion of how to operationalize 15 what an IRB is supposed to insure. And I think that 16 that's something that one could take some kind of 17 short-term action on, what does it mean for an IRB to 18 monitor. So I think those points are something that 19 is very important to follow through because the 20 language is vague at this time. 21 I'd also say that to collect the data 22 might be difficult, because it's my understanding that 23 institutions settle to avoid publicity. Many times 24 the settlements request that this is not public 25 47 information, so to find out about settlements may be 1 very difficult, and maybe you have some ideas about 2 that. 3 I also have a question about talking about 4 protecting an IRB versus talking about individual IRB 5 members, versus protecting the PI, so I can understand 6 the protection of individual IRB members who come 7 under an umbrella of the institution, or the 8 organization to which the IRB is constituted, but I'm 9 not sure, and maybe I'm a loathsome person, but I'm 10 not sure I totally think that þþ I totally agree with 11 the concept of immunity for an IRB, but let me go on. 12 Another question I have is if, in fact þþ 13 I know in your paper you made very important 14 distinctions between medical tort law and how it would 15 be inappropriately applied to research. However, I am 16 confused about the type of research in which 17 participants use their health insurance to pay for 18 their participation, and I'd like to know what would 19 happen in those situations. 20 I'm also not comfortable with the analogy 21 of the IRB to peer review. In my mind, peer review - 22 one of the purposes is to recommend punitive action 23 if, in fact, it is reasonable, or to say that someone 24 has practiced in an ill manner. My understanding of 25 48 an IRB is to ensure subject protection and that to the 1 extent feasible, the proposed subject protections are, 2 in fact, enacted by the PI. And once again, I put in 3 to the extent feasible. 4 I also have difficulty in thinking about 5 the IRB as an arm of the government. I think there's 6 a difference between being regulated by the federal 7 government and being an arm of the government. And I 8 think that there's certainly a chilling effect if, in 9 fact, IRBs are seen from the participant's perspective 10 as having the authority and mandate, and power of the 11 government which, in fact, I don't think was the 12 purpose because they're supposed to be local and come 13 to decisions that reflect the local community. 14 I also agree with Felix that the notion of 15 reasonableness seems to place an undue burden on a 16 participant who not only is basically getting an 17 informed consent to understand what the researcher's 18 responsibilities are, and now is being asked to sign 19 something for arbitration or whatever way you would 20 organize it against an IRB to which they know very 21 little about. 22 And my last confusion has to do with in 23 informed consent. Typically, you are given 24 information that you call the PI if you have questions 25 49 about the research, and you're given the number of 1 someone with the IRB or in Human Subjects Protections 2 to talk about your research rights; yet, the kind of 3 arbitration that you proposed immediately sets up this 4 adversary system, so who is going to be the 5 participant advocate if in fact any time they call a 6 representative of the IRB or the individual who is 7 supposed to talk to them about their rights, that 8 individual has to be protective of potential 9 arbitration that may come down the line. 10 And finally, if in fact there's immunity 11 for the IRB, the IRB has conflicting tensions to begin 12 with. They have a responsibility to their 13 institution, and one of the beauties of the IRB system 14 is that there is supposed to be a wall there so that 15 they can implement subject protections. And I'm not 16 sure, it sounds to me like a lot of what you're 17 suggesting might undermine protections for subjects. 18 DR. MORREIM: Okay. Lots of questions, 19 very good ones. I think we got lots of answers, good 20 ones. First of all, how to operationalize sort of 21 voluntary reporting and so forth; admittedly, a 22 challenge. I would refer us, I guess, at this point 23 to efforts by JACAHO and others to collect information 24 about adverse events in hospitals, you know, patient 25 50 safety movement needs to be a model here. Not all 1 together successful yet, but we are trying to get 2 reports of adverse incidents, you know, sentinel 3 events, things like that from within the patient 4 safety movement. We might be able to pick up some 5 tips from them on how to encourage reporting. 6 I'm thinking that it's likely that 7 voluntary reporting is only going to be þþ is going to 8 be it rather than any kind of mandate. But again 9 note, when somebody has been served with process, it's 10 now a done deal, honey. You ain't going to give away 11 anything by saying we just got served, so that 12 reporting that we've now been served is not nearly as 13 sensitive as reporting we had a boo-boo and nobody 14 knows about it yet. And that's the level we're trying 15 to get reports on for patient safety, so this would be 16 at a lesser level, it seems to me. 17 Secondly, looking at protecting IRBs, very 18 important, versus IRB members and PIs and so forth. 19 Very important to note, and you did not trample this 20 distinction but I want to highlight it. It is one 21 thing to protect IRBs. All the immunities that I'm 22 talking about do not touch Pis, they do not touch 23 sponsors. I am not talking about making drug 24 companies immune from suit, or investigators or 25 51 anybody like that. I am strictly talking about IRBs. 1 And when we talk - and I'll jump to one of 2 your later questions - when we talk about a 3 potentially adversarial setting when somebody phones 4 an IRB to make a complaint; darling, if you're about 5 ready to get sued, that's also adversarial, so the 6 possibility of any kind of adverse action by a 7 prospective plaintiff against the IRB sets up the 8 adversariality that is going to be true, whether it 9 would be resolved in an arbitral forum or in a 10 litigation forum. Either way, if people aren't 11 getting along in the same sandbox, we're going to have 12 some sand flying in each other's eyes potentially. So 13 in other words, it doesn't make the problem worse to 14 talk about arbitration, as opposed to litigation. 15 You've still got the challenge. 16 With respect to the third thing that I 17 wrote down, for patients who are using healthcare 18 insurance to participate in research, I may have to 19 come back to that. I'm not sure I understood your 20 question exactly. You're noting that peer review in 21 hospitals is a potentially punitive sanctioning thing; 22 whereas, IRBs are very different. Yes. That's not 23 where the analogy is. The analogy is they are both 24 engaged in a particular kind of process to safeguard 25 52 healthcare, or in this case, safeguard the well-being 1 of research subjects, patients, research subjects. 2 What activities they do in order to 3 protect research subjects, protect patients are a very 4 different focus, absolutely right. And in one case, 5 you're getting rid of bad apples. In the other case, 6 you are sort of taking a more positive thing, trying 7 to make sure that certain protections are in place in 8 the first place, so that where you see disanalogies, 9 I agree - definitely disanalagous, but the important 10 things are where the analogies do lie; namely, a 11 federally mandated public service/public policy 12 protected function that is going to be substantially 13 chilled, and perhaps significantly misdirected, and 14 perversely misdirected when the individuals, and yes, 15 the bodies as a whole stand to be named in litigation 16 if the litigation is not meritorious. 17 I will say that you rightly distinguish 18 between individual members versus IRBs on a whole, but 19 I can also say that in some of private conversations 20 with various IRB people, including heads of 21 independent IRBs, I know of some who would just get 22 out of the business if they got sued. Or much more 23 significantly, although some individuals might just 24 say I've had it if we get sued, and that would be a 25 53 more personal decision. 1 On the other hand, if it becomes 2 impossible to get insurance - I mean, the independent 3 IRBs serve a very significant function for those who 4 do research outside of the academic health center 5 context. And to the extent that certain kinds of IRBs 6 might find it impossible to exist in a litigious 7 environment, that could be a significant downside for 8 the overall protection of human research subjects. It 9 could add to the complexities under which IRBs 10 function. 11 Furthermore, seeing IRBs as an arm of 12 government - I will say, it's an interesting thought. 13 I'm not sure that this kind of immunity would 14 necessarily prompt patients in their minds to say ah- 15 hah, the IRB is an arm of government. I think a lot 16 of people don't really know much about the IRB. 17 You're quite right in the first place, even though a 18 plaintiff's attorney might later suggest to sue them. 19 They know about it perhaps a little bit through the 20 consent form and a phone number and so forth, but IRBs 21 play such a different role than investigators. And 22 their role in the potential harms - and I think here 23 is kind of a bottom line in my somewhat lengthy 24 response to your many interesting and thoughtful 25 54 questions. 1 I think that the IRBs play such a 2 different role, and as you pointed out, that role 3 needs badly to be further clarified. The injuries 4 that happen to subjects are usually not going to be 5 from inadequate IRBs. They might be from a very good 6 protocol not being carried out very well, so that we 7 need to parse out where are the true injuries most 8 likely to come from. 9 IRBs are a broader umbrella function, and 10 their responsibilities - and yes, their susceptibility 11 to whatever kind of punitive response needs to be kept 12 where the IRBs' responsibilities are; namely, in this 13 broader umbrella area. 14 DR. PRENTICE: Okay. We have three 15 individuals on the list who want to ask questions. 16 We've got a choice here. We can go with the first 17 person on the list, which is Susan, or two of you 18 could yield to the third or the second. And the 19 people on the list are Susan, Mary Lake and Tom, but 20 only one of you can ask a question. We're already 21 three minutes into the break, so what would you like 22 to do? Who has the most important question to ask 23 Haavi? 24 DR. MORREIM: Give each person 30 seconds 25 55 and a give me a minute and a half. 1 DR. PRENTICE: All right. Each person 20 2 seconds, you have a minute and 15 seconds. And then 3 we'll cut our break short. Susan, succinct. 4 MS. KORNETSKY: My question is about 5 arbitration. You had mentioned that this is not 6 viewed as a waiver of rights, and I would just like 7 you to expand upon that. And then also, if HHS - I 8 mean, I want to put you on the spot whether you would 9 look at that the same way. And just a comment; my 10 concern would be that we're going to have to start 11 asking research subjects to bring attorneys to the 12 informed consent session. Between the informed 13 consent, the HIPAA, and now this I would be very, very 14 concerned. 15 DR. MORREIM: Oh, no. I mean all three. 16 DR. PRENTICE: Oh, all three? 17 DR. MORREIM: Yes. 18 DR. PRENTICE: Okay. All right. Mary 19 Lake. 20 DR. POLAN: A really rapid and practical 21 question. As a physician, the disincentives to serve 22 on IRBs are enormous. And if you add the fact that 23 you don't get paid, you have enormous pressures for 24 financial activity in your practice. And then to that 25 56 you add the potential for getting sued layered upon 1 malpractice claims. It's a difficult situation. 2 If you were to prioritize your three 3 levels of immunity in terms of likelihood of seeing 4 one of those eventuate in a practical solution within 5 the next 12 months, what would your priority be, and 6 why? And perhaps the general counsel, somebody else 7 like Mark or the general counsel could weigh-in on if 8 you really wanted to do this and make this happen, how 9 would you do it? 10 DR. PRENTICE: Tom. 11 MR. ADAMS: The thoughts that you have 12 shared with us centered on issues that would require 13 legislation for the most part. The OHRP has as part 14 of its existing mandate the ability to educate the 15 public. Would you see a potential federal role in 16 contacting insurance entities and the broader public 17 to begin to talk about this problem, and basically 18 jaw-bone, if you will, for meaningful change given the 19 fact that historically medical malpractice legislation 20 at the federal level has not fared well? 21 DR. MORREIM: Okay. Quick answers one, 22 two, three. I don't see a whole lot of enthusiasm for 23 arbitration. I just throw it out there, darlings. 24 Okay. Second, disincentives to þþ okay. It's a toss- 25 57 up between HCQIA and FTCA. I'm thinking HCQIA is a 1 lovely model because of granting attorney's fees, no 2 punitive damages, a lot of protection, and it's very, 3 very parallel. It would be an easy thing to sift that 4 out. FTCA, though, has such a precedent with the 5 community health centers. I think that that could be 6 readily adaptable, so I have a hard time choosing. 7 Finally, the federal role for educating 8 insurers, et cetera - my guess is that probably the 9 better educators are going to be the brokers, people 10 who are actually carving the deals in the marketplace. 11 Government might, indeed, have some role. Just help 12 them understand where the risks reasonably are. If 13 somebody can please educate certain plaintiffs' 14 attorneys out there so they understand a little more 15 about research, that would be really nice, but I don't 16 think Uncle Sam is going to be very effective in that 17 respect either. Thank you. 18 DR. PRENTICE: Okay. Thank you, Haavi. 19 That was fascinating. We could listen to you all day. 20 We're going to take a little over 10 minutes. Please 21 try to get back on time because we now have the HIPAA 22 Privacy Rule Panel. 23 (Whereupon, the proceedings in the above- 24 entitled matter went off the record at 9:50:38 a.m. 25 58 and went back on the record at 10:04:26 a.m.) 1 DR. PRENTICE: Okay, folks. If everyone 2 could take their seats. Okay. Now we're going to 3 move into the HIPAA Privacy Rule Panel. I think 4 everybody recognizes the fact that this is an 5 extremely important topic to IRBs, as well as clinical 6 investigators, so it's a delight to have four 7 distinguished panelists who are going to present to us 8 today. What I'm going to do is introduce them 9 individually in sequence, and then at the end of the 10 fourth panelist, I would like all four experts to go 11 up to the table in front of the room where they can 12 respond to questions, so you would hold your questions 13 until after all four presentations are complete. 14 The first panelist is Susan McAndrew, and 15 I'm going to abbreviate her bio in order to be a 16 little bit more efficient and not read everything 17 here, out of time considerations. Susan's a Senior 18 Health Information Privacy Policy Specialist in the 19 Office of Civil Rights at HHS. Ms. McAndrew came to 20 HHS in May, 2000 to lend her expertise to the final 21 privacy regulation, so it's really an honor to have 22 you here, Ms. McAndrew, to talk about the HIPAA 23 Privacy Rule and some of the issues that we are 24 dealing with. 25 59 MS. McANDREW: I want to thank you so much 1 for inviting me here today to talk about the Privacy 2 Rule. I am from the Office for Civil Rights, which is 3 in the Office of the Secretary, that was given 4 responsibility for enforcing and implementing the 5 Privacy Rule. 6 The Privacy Rule stems from legislation in 7 1996 called the Health Insurance Affordability and 8 Accountability Act, hence HIPAA. The regulations were 9 issued first in December of 2000, and then modified 10 substantially in August of 2002. And entities were to 11 be in compliance with the Privacy Protections as of 12 April of last year. 13 The Office for Civil Rights does two 14 things. One, we work with entities to provide 15 technical assistance and to help them come into 16 compliance with the rule. And we also investigate 17 complaints about potential violations of the rule. 18 And to date, we have probably received over 5,000 19 complaints, and the number continues to grow, but we 20 are happy to say that we've been able to resolve about 21 45 percent of those complaints satisfactorily. 22 I wanted to use the time today just to 23 give a very brief overview of how the Privacy Rule 24 works, particularly in connection with research 25 60 endeavors. I'm assuming that a lot of this is known to 1 you all, and so I want to move rather rapidly through 2 it and allow time to talk a little bit about some of 3 the issues that have been resolved to date, and some 4 of the issues that remain. And then the other 5 speakers will þþ the other panelists will have 6 additional perspectives to offer in terms of the 7 issues that the research community is facing in trying 8 to absorb the privacy requirements, and interacting 9 with entities that are now subject to privacy 10 requirements. 11 Basically, the rule operates to protect 12 privacy in two major ways; one, it gives the 13 individual certain rights with respect to their 14 information - that is, they now have a federal right 15 to access their information, their medical records and 16 billing accounts. They have the right to amend those 17 records if they are wrong. They have the right to ask 18 for an accounting from covered entities about how 19 their information has been used. 20 The rule also protects the privacy of 21 identifiable health information by limiting how a 22 covered entity can use and disclose that information. 23 Use is usually a term that we use to describe how an 24 entity will employ the information within itself, a 25 61 disclosure is how it would give the information, or 1 share the information with someone outside of the 2 covered entity. So by controlling how a covered 3 entity uses and discloses information, we can keep 4 that information private. 5 The rule sets out þþ does these controls 6 in a way of providing permissions. We say how it can 7 be used. Clearly, we allow the information to be used 8 for treatment, payment, and healthcare operation 9 purposes. These are the core functions of treatment 10 providers and the insurance industry. We allow þþ in 11 addition to those primary purposes we do recognize in 12 the rule a more limited set of important public 13 purposes, where information can be disclosed or used 14 without the individual's permission. If, in fact, the 15 information is to be used in a way that is not within 16 the scope of one of those permissions, then the only 17 way that information can be used for that purpose is 18 with the individual's written authorization. 19 Research is an area where we have allowed 20 under certain conditions for the information to be 21 shared with others without the individual's written 22 authorization. Otherwise, not that much different 23 from when you would conceive of the need for an 24 informed consent. Other research purposes would go 25 62 forward with a written authorization, and I want to 1 talk about those circumstances just briefly. 2 But before we get into that, the research 3 that is þþ just to step back for a second. The 4 research that's affected by the rule includes both 5 records research, research just on information, as 6 well as research involving human subjects in a 7 clinical trial setting is the most common. But the 8 rule also effects research in a variety of contexts. 9 It's not all research. It's not just research that's 10 subject to the Common Rule. It's not just research 11 that's subject to the FDA protections. It's any 12 research endeavor, and so we have tried to make the 13 provisions of the rule applicable in a generalized 14 way, and not specific to any other environment in 15 which research operates. 16 The counter of that is that the Privacy 17 Rule does not override, does not change the Common 18 Rule, does not change the FDA protections that are in 19 their regulations. And so an entity that is subject 20 to the Common Rule or is subject to FDA's protections, 21 and is also subject to privacy needs to blend all of 22 these obligations in order to continue to operate. 23 So what does it mean to have research with 24 patient permission? Under the Common Rule, this would 25 63 be research that's gone through IRB approval and there 1 is an informed consent. For the Privacy Rule, we 2 parallel that function by requiring that an 3 authorization valid under the Privacy Rule needs to be 4 signed by the research participant. 5 We have made attempts in the regulations, 6 primarily in August, to streamline the authorization 7 process so that it is more likely that the 8 authorization that's needed for privacy and the 9 informed consent that would be needed for the same 10 trial can be merged into a single document. In some 11 cases, the preference may be to have two separate 12 documents. But to make the merger of these two 13 documents easier, we did allow research authorizations 14 and exemption from the usual requirement of the 15 typical privacy authorization; and that is, it doesn't 16 need an expiration date. It doesn't have to be time 17 limited, or even event limited. And we also allow an 18 IRB in the process of merging the documents to waive 19 some of the authorization requirements in order for 20 the two to work better together. 21 We realized that there was going to be 22 circumstances in the typical research setting where 23 informed consent was not needed, and typically þþ the 24 Common Rule has criteria under which the need for the 25 64 informed consent can be waived. In the Privacy Rule, 1 we have tried to adapt those criteria to apply in a 2 setting where you would also be thinking about having 3 to waive the privacy authorization so that the same 4 research can go forward without either an informed 5 consent or a privacy authorization. 6 And there are also a number of activities 7 that were either outside of the Common Rule, or where 8 IRB approval wasn't necessary. And we have tried to 9 provide for some of those activities to also go 10 forward under the Privacy Rule without authorization; 11 including research on decedents, activities that are 12 simply preparatory to research, as well as creating in 13 August of 2002 something we call a Limited Data Set, 14 which allows information that's not totally de- 15 identified by our privacy standards, but do lack any 16 personal identifiers to be used for research purposes 17 without needing individual's authorization for that 18 information to flow for research purposes. 19 So just briefly what these þþ how these 20 four options work for when you can do research under 21 the Privacy Rule without needing the individual's 22 authorization - the first one is, when you have an IRB 23 that has waived the need for the authorization. And 24 that's based on three criteria that are set forth in 25 65 the rule, that the use of the PHI for the research 1 purpose represents only a minimal risk to the privacy 2 of the individual. That the research could not 3 practicably be conducted without waiver of the 4 authorization, and that the research could not be 5 practicably conducted without access to the 6 identifiable health information. 7 A little more on minimal risk. We had a 8 lot of discussion internally about how to þþ in a 9 privacy context what this means. This is similar to 10 our criteria that the IRB uses in terms of waiving the 11 informed consent in the normal research context, but 12 what does minimal risk mean when you're only looking 13 at the privacy aspects, and so we have further 14 identified that. It essentially means that there's a 15 plan in place that will adequately protect the 16 identifiers. There's a plan in place in the research 17 that will remove the identifiers as soon as 18 practicable consistent with the research purpose. 19 There may be times when, in fact, the identifiers need 20 to stay with the data, and that's allowed. And 21 there's a need for written assurances that the 22 information will not be reused or disclosed in other 23 than a research or similar context. 24 So if those conditions are in place, then 25 66 it would be permitted þþ describe circumstances where 1 the risk to the individual's privacy would be 2 considered minimal enough to allow their identifiable 3 information to flow without their written 4 authorization for that research purposes. You would 5 also need then to meet the second and third prongs of 6 the waiver criteria to get the waiver. 7 Just briefly on the preparatory for 8 research. This was essentially a way to allow 9 researchers access to information within a covered 10 entity so that they could prepare their hypothesis and 11 prepare the material necessary to present the research 12 proposal to the IRB for the IRB's review. This was 13 not the research itself, but just certain activities 14 that are preparatory to that research. And that the 15 condition for the access to the information is that 16 the identifiable information is not removed from the 17 covered entity, so it's used only for this limited 18 testing purpose. 19 We attempted to merge the fact that the 20 Common Rule does not cover information of decedents, 21 but the Privacy Rule does, and so we have allowed for 22 research that uses decedent information to go forward 23 without authorization. 24 And finally, the limited data set was 25 67 created in an attempt to track research need for 1 information which is largely unidentifiable, but still 2 contains certain elements. Primarily, the 3 conversations we were having mostly with the research 4 community was the need for birth dates and geographic 5 identifiers to come as part of these data sets. 6 With that degree of specificity within the 7 data, the information is still potentially 8 identifiable, so we didn't want to take it totally 9 outside of the protections of the Privacy Rule, but 10 did allow for research purposes for those kinds of 11 indirect identifiers to remain in the data sets. The 12 balance there was struck in terms of requiring a data 13 use agreement to govern how that information is 14 controlled once it is given over to the researcher. 15 There were a number of other requirements. 16 One of the rights I had mentioned earlier was the 17 right for the individual to know how their information 18 has been used or disclosed, actually disclosed. Here 19 for research, there were a couple of changes that we 20 made in August of 2002, to make it easier for a 21 covered entity that does a lot of þþ participates in 22 a lot of research, to simplify how they account to the 23