NIH CLINICAL CENTER GRAND ROUNDS
Episode 2009-012
Time: 58:00
Recorded April 1, 2009
ETHICS ROUNDS
Looking for Clinical Findings in Research Subjects: When is it a Good Idea?
Susan M. Wolf, JD, McKnight Presidential Professor of Law, Medicine and Public Policy
Faegre and Benson Professor of Law, University of Minnesota Law School
Professor of Medicine, University of Minnesota Medical School
ANNOUNCER: Discussing Outstanding Science of the Past, Present and Future – this is NIH Clinical Center Grand Rounds.
(Music establishes, goes under VO)
ANNOUNCER: Greetings and welcome to NIH Clinical Center Grand Rounds. We bring you a special "Ethics Grand Rounds" on this edition. The topic, "Looking for Clinical Findings in Research Subjects: When is it a Good Idea?" Our speaker is Susan M. Wolf, McKnight Presidential Professor of Law, Medicine and Public Policy and Faegre and Benson Professor of Law, University of Minnesota Law School, also Professor of Medicine, University of Minnesota Medical School. We take you to the Lippsett Ampitheater at the National Institutes of Health Clinical Center in Bethesda, Maryland, Dr. David Wendler, head of the Unit on Vulnerable Populations in the Department of Bioethics at the NIH Clinical Center, will introduce today's topic.
WENDLER: So today's Ethics Grand Rounds began 10 years ago with the publication of what I think is one of the most interesting studies that I read coming out of this place in a while. There are lots of important research that comes out of the Clinical Center and every once in a while it's not only important but it's interesting. This was about 10 years ago. Some people in radiology realized that we get a lot of what we consider healthy, normal volunteers who come to the Clinical Center and a lot of those people undergo MRI scans. There had been some debate about what a normal MRI looks like, how many abnormalities there are in people that don't have signs or symptoms and they realized they have a great resource for addressing that question. So at the point they started that study, they did an analysis of the scans of 1,000 people who had been enrolled in research here as normal volunteers and undergone an MRI scan and they evaluated all 1,000 of the MRIs to look for abnormalities and the quiz question for today is how many brain tumors did they find? They identified in 1,000 people who had come here and undergone MRI, as normal volunteers for research purposes. Which that is the fascinating part. For me you go to a cocktail party and you say, we all think we are normal here and it looks like there is a rate of maybe 3,000 of us and 3 have brain tumors if the data is right. It also led to a worry of one way of characterizing that is maybe when we are screening normal volunteers, we are missing clinically significant findings and as a result of that, a policy was put in place to address that concern, which you'll hear about in a minute. And that policy led to a consult that Betty will tell us about and she is an investigator at NIAID in Baltimore as part of intramural NIH and she has been doing MRI research for about 12 years. Welcome, Betty Jo.
BETTY JO: Can everybody hear me? Yes. So, I’m concerned whether this is really a good idea. Our research, we do fMRI research in this particular study and it's addressing issues with smokers. Smoking is a major health problem. A lot of mortality attributed to it. Most individuals who quit smoking relapse very quickly. It's very hard to kick the habit and the processes that perpetuate smoking in smokers is not that well understood. We are trying to understand that a little bit better. So the study will involve functional magnetic resonance imaging in smokers and in controls. And these people are 18-55 years old. They are all screened with a physical, including a thorough neurological exam and if they have any neurological findings or complaints related to neurological findings, they are excluded from the study. So they will all go under fMRI scanning and they'll do a couple of 50's we'll look at the activation and compare that between the smokers and the nonsmokers and the whole business that this might lead to a better understanding of why smoking is so hard to kick and might even lead to some novel approaches to treatment.
So, as part of our scans, we need to get an anatomical scan to register the functional scans on so we can tell what parts of the brain we are dealing with. They sometimes reveal something that might be an abnormality in the subject's brains. And there is now quite a bit of literature on incidental findings in research scans ranging from sinusitis to tumors. And a fair number of these are judged to be clinically significant.
Now the definition of that as far as I can read is basically somebody looked at it and thought, that looks scary and we ought to look at it closer. Very few studies have follow-up even to say whether the finding was confirmed and there is virtually no long-term follow-up on how these folks do with having been given this information.
So, now that said, with our scans, we are very likely to miss some of these incidental findings or miss -- if we don't find them they can't be incidental, but we may miss things for a couple of reasons. The scans we do for our anatomies are not optimized to pick up clinical problems and the researchers are not trained to look at these scans for clinically significant findings. So, because of that, NIH has a policy in place to reduce the chances that a clinically significant finding might be missed and that policy states that anyone who is participating in a study has to have a clinical scan read by a radiologist once a year. So our concern was whether this was really a good idea.
So these scans are not otherwise required for the research. MRI is not validated as a screening tool and healthy folks and while the initial impetus for this was finding stuff early is good, this is good for people, it might end up being a violation of the ethics principle and it is a violation of, don't do anything more than you need for the research. We don't need these for the research. The current policy doesn't allow participants decline this clinical scan and I’m concerned that's a violation of the respect for persons principle since this is really not needed for the study itself. So, there is a lot of concerns about screening tests in general and this has been in the news quite a bit lately with the recent results on the psa screening thing. Our gut instinct is fighting disease early is good. Turns out that's not always the case. Sometimes it just prolongs treatment periods and we still die at the same time point but you're sick for longer.
Some positive findings -- some that are not destined to cause clinical problems for the person being treated carries a lot of risk and significant expense and other things. There is a lot of concern in general about screening tests. And in particular, about MRI as a screening, a number of investigators have raised concerns that it's probably not a good idea. The first study raises or follows through what happens when you find something incidentally and what happens is people tend to get alarmed wants something done about it whether it makes sense from the clinical data or not. That's a problem. And Judy published a little piece on comparing this -- or questioning the wisdom of looking for these findings and pointing out a number of incidents in medicine where people kind of flock to doing something that turns out to not be the best course of action.
So, there are certainly possible benefits like I said, early detection might or would lead to early intervention and it might be beneficial but searching the literature, don't find any evidence that early intervention improves outcome for CMS pathology and particularly for intercranial aneurysms. Some data suggests it might be harmful in some situations that people have unnecessary procedures that carry significant risk. Possible harms, you can have false positives that may lead to unnecessary tests and this is something I have seen in our practice. We had a 13-year-old kid who ended up getting a head CT scan and then didn't verify the findings of the so this kid had radiation exposure she didn't need. That's not good.
So since no one suggests MRI as a screening tool, the clinicians are left with little guidance as to what to do. Anxiety over brain lesions may lead to risky interventions. There is published information on this. This is a more common practice in Japan. They do brain checkups where they will do a neurological and MRI scan and come up with quite a few things. There is a published series of 60 asymptomatics followed for 43 years and 40 did not grow in 3 years. So the recommendation by these Japanese investigators is to do nothing but follow them when they are found.
Patients, despite knowing there is risk with surgery, want these things out. So people are undergoing very risky, very invasive procedures because they are scared something is bad in their brain and they want it out. There is an editorial with a similar vain, a neuroscientist investigator who participated in his own study and was found to have an aneurysm and he puts it in the paper. What did I do? If you look at the data, that may not have been the breast thing to have done. And then also, you're sentencing these folks to a lifetime of follow-up which carries a lot of risks and anxiety and costs and you may importantly have significantly impact on their insure ability. Try going to an insurance company and saying I need insurance but I know there is an aneurysm in my brain but I don't think it's going to do anything. They won't be insurable in the private market at all. That's a major risk. Given the questions about MRIs and screening tools, we were questioning whether a clinical quality MRI scan should be done on the participants in our study. And if they are performed, should they be done as part of a research study to determine their impacts? Is this a good screening tool? If we did it that way, we would remove the perception that this is useful as a screening test.
So if people had to sign a consent form saying, we don't finish it's useful. We will follow-up any findings and see if it is, it would put the results in a different light for the participants. And then in general, what if anything should be told to the participants in the consent form about incidental findings? This is a broader question in general but it's particularly poignant when dealing with findings in the brain. And then finally should the participants be able to decline this clinical scan we don't need for our research purposes?
WENDLER: Anybody have any questions for Betty Jo about this case or the study? Paul?
QUESTION: If you, on your initial screening find a neurologic finding -- what would you do?
BETTY JO: We would refer those folks on to a neurologist or whatever. >> so if they find something in their study, they'll refer them
QUESTION: So given the current state of all of this, what you tell your patients now about the scan?
BETTY JO: Currently what we tell our participants is that this scan is not a clinical quality scan and it will not routinely be read by a radiologist. If they have concerned about their health, they should talk their physician. We tell them if something jumps out as abnormal, we do have a relationship with the radiologist and we will show that research scan to the radiologist and see if they are concerned about it. And we don't feel like we should leave the participants with the burden of, this might be bad. So if the radiologist says, it needs a ct scan for follow-up or he should have a clinical MRI. We get that on our dime so we can resolve it as much as possible before referring them on to whatever treatment they need.
WENDLER: Any other questions? All right. Thank you, Betty Jo.
INTRODUCTION OF NEXT SPEAKER BY Dr. Paul Plotz, chief of the Arthritis and Rheumatism Branch at the National Institute of Arthritis and Musculoskelatal and Skin Diseases at the NIH.
PLOTZ: it's a pleasure to introduce professor Suzanne Wolf today. Suzanne is a native Washingtonian and a fourth generation member of her profession. In her senior year at Princeton in 1975, she wrote her undergraduate thesis on the responsible use of racial and ethnic categories in biomedical research. And she has returned to this subject frequently over the following decades. Princeton gave her the Isidore Brown Thesis Award and graduated her summa cum laude. Her law degree is from Yale and she did graduate work at Harvard. She clerked for Judge Leonard Sand in the U.S. Court for the Seventh District of New York and practiced law at a distinguished law firm. She taught at NYU Law School and was a fellow at Harvard and a National Endowment for the Humanities Fellow and an associate at the Hastings Center where, by the way, I’m told she played all the infield positions on the team. She has been at the University of Minnesota for the past 16 years where she is the McKnight Professor of Law, Medicine and Public Policy, the Professor of Law and the plain ole' Professor of Medicine, and Director of the Joint Degree Program in the Law, Health and Life Sciences and a member of the University Center for Bioethics and chair of the Consortium on Law and Values and Health Environment and the Life Sciences.
She has published -- eat your heart out people -- repeatedly in the New England Journal, Science, Nature Genetics, JAMA. She is a lawyer. In neurology, and a wide variety of law ethics journals and often published books and conference proceedings. She lectured at all the top universities throughout the country and appeared on top television programs. And has made the final 4 in each of the last 25 years. And I have been told she is favored to win this coming weekend. But most impressive to me about professor wolf is what she writes about. She writes about big, solid, important and difficult topics and cases. Nancy Cruzan, Terri Schiavo, Jamahl Burk, genetic testing for fatal inherited diseases and the implications for insurance, euthanasia and physician assisted suicide and feminism in relation to many ethical, medical and legal topics, and for the today, she has written managing incidental findings in human medical research.. It's for the work and interest that she has been invited here today. Professor Wolf. [applause]
WOLF: Thank you for that too generous and frighteningly thorough introduction. I’m delighted to be here and I think Betty Jo brought to you an enormous complex and subtle case. I’m not going to pull my punches. I’m going to argue this NIH policy is a mistake and we can have a robust debate.
Just to give you my disclosure, okay. There are two sign posts here that operated the key context. In 2005 I was part of a trans-NIH/Stanford two-day workshop off campus. Judy Ellis, who is now at the University of British Columbia took the lead on this. I helped among others with the planning. You see the list of institutes that were involved. That focused on detection and disclosure of incidental find negligence neuroimaging research and I’m going to talk about the conclusions because that group was skeptical about this NIH policy. Of routine clinical scans for all research secretaries. The other sign post -- research subjects -- a two-year project funded by NHGRI that I led as PI and this is the long list of people involved in it. The gold are my co-investigators and the big symposium was published last summer. And I’m going to use that as another source.
Now, I think I should stop for a second. What do I mean by an incidental finding? You may want to debate that as well. And that was a source of tremendous confusion in the literature. This is the definition that we came up with in our consensus paper, which took two years to negotiate. A finding concerning an individual research participant -- so here we are talking about individual findings, not aggregate findings. There is particularly in the genetics literature a whole separate conversation about returned of aggregate findings to the pool of research participants. These are individual. Now findings that have potential health or reproductive importance. Our study looked at this in genetic and genomic research contrasting neuroimaging where the conversation on is was far more advance than in genetics and genomics. This is why in HDR I was interested this. But they tolerated our performing the comparison. They were pleased that we would perform that comparison. So we could fight about, or reproductive.
Betty Jo's case doesn't raise that but that's part of the IF debate. And it's discovered in the course of research but is beyond the aims of the study. This contrast with some other definitions floating around are unexpected. You are going to see one of our punch lines was, stop unexpecting it! Expect it!
We know enough about incidents and so on to expect. Now contrast research findings. Particularly in genetics and genomics and I’m part of that debate too, about whether to return individual research findings. Findings from the focal variables in meeting the aims of that study. But we are talking about the incidental stuff here. I thought I’d throw up a picture of some incidental findings from -- my pronunciation was just corrected on that -- from the New England Journal of Medicine 2007. This is much more recent. 2000 asymptomatic mean or median age, 63.3 years and I’ll give you some of their findings. These were -- I already mentioned the foci of our project, the contrast between neuroimaging -- imaging verses genetics.
There is too complex. I’m only asking you to focus really on this stuff. Here is the data. Mean age 63.3 years. 7.2% asymptomatic brain infarction. 1.3% cerebral aneurysms. 1.6% brain tumors and 3% other. But, you should also notice that outside of this in the literature, there is a distressingly large range seated. Incidents of incidental findings and we should be call logging what kinds. Now historically there is no consensus on handling is. I think our NHGRI-funded study was the first funded to get a grip ethically, legally, on what should be done with incidental findings.
But part what have we did was look at the web, publicly available, consent forms, guidance documents, a whole protocol I won't describe here unless you want to talk about it. I just threw up certain form language which said, we're not going to tell you anything. We are not going to disclose them. This is a different policy than you just heard where they were saying to people, if we do see something, we are going to get it read by a neuroradiologist and we are going to talk to you about it. But, I think consensus now is really beginning to emerge and that is what I want to talk about.
What are the key questions on handling incidental find something first of all, do researchers, including PhD PIs have an obligation to hunt? This is the question that Betty Jo put before us and the NIH policy forces us to confront. Not just react if we see one but actually hunt and under the NIH policy, get clinical stance and get clinical radiologists and neuroradiologists to hunt. What should researchers do once they spot an incidental finding? Should they seek a consult when you don't have this design of compelled consult? What information then should be offered back to the research participant or to the participant's physician? And there is a kind of cite in the literature although the advise is -- advice is this belongs to the research participant although there are plenty of instances where you want to involve the physician and you have to worry about minors and guardians and those with diminished capacity. What should consent forms say in advance about how this should be handled. What should IRBs require and what should funders require and how should these findings be handled in reanalysis of archived data?
We are looking at this right now. This is a very big question. Now, the workshop that I eluded to from 2005 identified 5 possible approaches. And it's the fifth, this bottom one in essence that we are here to debate. That is not the one that our study recommended. I’m going try to hold my hand study -- steady which was up here. Essentially they do nothing or if researchers tell the research participant without seeking expert clinical review, number 3 researchers spotting incidental findings should seek expert review to say, do you see it too? We know this is just a research scan not a clinical grade scan that is optimized -- I’m back in the neurocontext here -- for diagnosis. And are you of the view that this is of sufficient potential importance that we should be telling the research participant? That is the option that we ended up recommending in our project. 4. Have all research scans reviewed by a clinical expert. And number 5, routinely acquire both research and clinical grade scans. Which is what we are really debating here today.
Now the workshop majority at the bottom of the slide felt that options one and two were ethically inferior to the others. That simply doing nothing or never consulting an expert was a mistake ethically. But that among the 3 there was not a clear recommendation, though there was tremendous skepticism raised about 5.
So the workshop recommended to give you a feel, researchers -- incidental findings come up all the time this. Is part of research. Researchers are responsible for managing them ethically. Let's get those ostrich heads out of the sand. Anticipate, set up a pathway. If you need to have access to a neuroradiologist, you need to set all that up in advance. Some of these findings rarely are urgent and you need to be able to deal with them quickly. You'll see in the third bullet, subjects or their surrogates are first in line to get this information which should happen in the timely fashion. This is a rejection of the idea that the primary care doc gets it first. Incidental findings should be addressed specifically in consent forms with disclosure of the anticipated pathway and IRBs should require this.
What did the workshop say on the policy that Betty Jo brings to us to debate? We saw no ethical requirement to acquire additional screening or clinical scans beyond those required for research. And here in the white there is a notation that this policy already was in existence. But the group was concerned about -- at the gold language in the bottom, unknown incidents of true positive, clinically significant incidental findings in asymptomatic individuals. And went on to say that the data that were available raised the possibility of a high rate of false positives and incidental finds, which is what Betty Jo was talking about too. Our own recommendations were in accord.
If you look at the third bullet, researchers do not have a duty to hunt for incidental findings. They have other duties to manage and anticipate incidental findings and establish a pathway, to deal with how incidental finding will be defined and handled in reanalysis of archived data. Here is the language saying no obligation to affirmatively search for incidental findings. The goal of research is to seek generalizable knowledge not to provide health information to individuals. Researchers are not obligated to perform extra MRI scans or modify their scans, their research scans to provide clinical information. And this is a very rich domain of ethical discussion about the difference between the ethics of research and the ethics of clinical care.
I’ll anticipate a question out of the gate. Of course there are instances where the researcher is also the clinician and we can talk about those dual-role models but that is seen in the ethics literature as a combination of two very different ethical roles. This gives you a feeling of the pathway we recommended, which was option 3 in the workshop that, in other words, the researcher tells the participant ahead of time about the potential for is, including probabilities, planned pathway, a feeling for what is going on here, and asks the participant whether he or she wants to be told of incidental findings. This is important in genetic research where there is a very well recognized right not to know. If the research thyme team identifies a suspected incidental finding it goes to the PI. It may not be the pi that recognize that is but the buck stops there. And the PI who may need a consult is ultimately responsible for managing the process of determining if this is there. If there is an incidental finding there and if it has potential health or reproductive importance.
Now this is all in the domain not of diagnosis that is terribly important to recognize. But simply to say, is there something we are worried about here? Does somebody who is a neuroradiologist or looks at scans for a living, agree there is something we are worried about here? And something that we are sufficiently worried about that we ought to disclose it to the research subject and set off a cascade that Betty Jo eluded to. And we went further to say, we would like to offer guidance on what kinds of incidental findings should be offered, maybe offered or should not be offered. We recognize 3 categories which are in gold at the bottom. And we pegged the obligation, the discretion and discouragement of offering to, what is the likely consequence for the research participant, him or herself?
So this is a subject centered or participant-centered assessment. Is disclosure likely to offer -- likely is all you can do with an incidental finding. You don't have a firm diagnosis. To offer strong benefit, yes, disclose. You think you might be able to save a life, disclose. Possible net benefit? Discretionary. Unlikely benefit, do not disclose. We can talk about all that. This is too much but just don't look at the genetic stuff unless you want to. Strong net benefit, we pegged to likely to be life-threatening or likely to be great but there is something you can do about it. Possible net benefit where there is discretion to disclose, non-fatalities that were fatal you would be in the first category. It is not something you talk about, but the research participant is likely to deem that information important. We are at the beginning stages of understanding how participants think about these things. Unlikely net benefit. Do not disclose. Information revealing condition, not likely it to be a serious health or reproductive importance or whose health or reproductive importance cannot be ascertained. Much to worry about here but remember, the genetics and genomics was part of what we were worrying about here.
Here is the policy which Betty Jo was kind enough to unearth for me. As an outsider, I couldn't access this myself, you were right. The key language is in gold. I just gave you-all of the kind of logistical technical stuff leading up to it so you can see how little there is here. All there is, it seems, and I love to be corrected, is a sentence. Clinical screening scans must be obtained at least once a year on all subjects or as specified in the IRB protocol. I’m told that the OR stuff is a reference to, it might be more than yearly rather than yes -- less than yearly. So that is it. There you go.
One sentence. Problems. Where is the specification, the standardization and the evidence-based justification for this approach? I’m unaware of it right now. Should you collect this additional clinical scan in the same scanning session? Or are you going to have people come back? You're talking about burden either way. Degree of burden is significant. Why yearly? Where did that come from? What is the justification for that timeframe? As Betty Jo properly raised, can the participant refuse this clinical scan and what are the consequences for their continued inclusion in the research study? And it's the clinical scan unreadable, if they move or something, what does that do to their continued participation in the research? How rapidly should these clinical scans be read? If it's 7-14 days, we know that in a rare case is going to be too slow for some urgent incidental findings. Can the researchers themselves, if they spot something, trigger a faster clinical read? In that case you have two sources of potential findings. Which ones should be reported?
The literature is deep into this question. There is no guidance in this NIH policy. That upped to each research team to come up with their own answer? That seems very undisciplined, unstandardized and unjustified. Are you going to report all anatomical variations and deviations from normal even if no clinical significance to research subjects? What about incidental findings of minimal clinical significance, like sinusitis? To whom will they be reported? How will they be recorded? If there is going to be a read of a scan, I guess we are talking about two separate records. But all of that needs to be handled. It does have HIPPA implications and it does have privacy implications. And finally, is some evidence basis for this being assembled? Are data being collected on incidents, false positives, cost, consequences, to provide an evidence base?
The big ethical question here are whether the burdens of this practice are justified by the benefits. And Betty Jo already eluded to a lot of the burdens. They include -- I think, very understandable confusion on the part of the participant between -- so am I in research or being clinically evaluated? This is much discussed in the research ethics literature under the banner of a misconception which I thought we were trying to dispel. This is a real open invitation to coming back with a vengeance. Of course there is a burden of more time in the scanner or separate scanning session and the risk, which for some research populations is going to be higher as indicated here.
Generating findings leading to anxiety, burdens of work up are bad.. That can happen even if you have a true finding and you're doing benefit. You may also do that harm. But you definitely also may find it where you're not going to confer a best and it's just burden. False positives and costs and staff time including. Betty Jo eluded to the costs of clinical follow-up. Is NIH going to pay for clinical follow-up on all of these find something that's quite a bill. What about on the benefit side? The benefits of screening asymptomatic populations by brain MRI?
There is an interesting publication from Mayo Clinic Proceedings 2008, and among their conclusions is that brain MRI scans should not be recommended for screening healthy populations. Benefit that justifies has been demonstrated as of yet. Brain MRI screening of asymptomatic patients is an in affective screening program that would produce many in sequential finding and a low rate of clinically relevant findings. So, that is a disparagement of this as a screening tool.
Is this a good action approach to incidental findings? I’m going to argue no. First of all, what is the incremental grain in true beneficial positives? Compared to workshop alternative 3, which is what we said we were doing or 4, a routine clinical read of all research scans simply for this. Do we spot incidental findings? I don't see any articulation. I don't know of a literature that would justify this policy as opposed to policies 3 or 4. And further, how does this practice of routine clinical scans comport with common rule requirements that researchers minimize the risks of research? This is adding risks by adding this clinical protocol on top and assure that risks are reasonable in relation to anticipated benefits if any. If you pull from here and you just look at the statistics coming out on the incidents of incidental findings as opposed to what we know about clinical beneficial finds, I think there is a common rule problem here. And the workshop suggested it itself that these requirements militated against extra scans and the NIH policy.
Now I want to wind up -- I eluded to the ethics literature contrasting the role of researcher with the role of clinician. I think the current consensus in incidental findings literature is that researchers do not have a duty to hunt or suddenly become clinicians but they have eye duty to handle responsibly if. Partly because of the notion that researchers don't have the same duties, they are generating generalizable knowledge. However, there is to me a very persuasive literature that researchers cannot ignore clinically significant incidental findings. That is a hard obligation. And I would point to, happy to discuss more, Belskin Richardson’s work on the duty of ancillary care. They argue for a duty of ancillary care. They are pioneers in doing that. Based on research participants entrustment, partial entrustment. Not entrustment as big as when that patient goes from patient to clinician but partial entrustment of their body and health information. Elicit et al. Participated in that. Dr. Miller: presumptive entitlement to information. And I have talked both in the ethics and law sphere of duties to warn and last clear chance. I would argue researchers are not obligated to provide clinical scans, however, NIH need not assert that it's fulfilling an obligation here.
Let's be precise. It could argue instead that it is electing a larger role. That as an institution, it is determined to establish a clinical care relationship with all research participants and conduct routine clinical scans. However, if that is the position that NIH is taking, I would argue participants must consent to this separately. They must be able to decline this clinical scan, which has its own I think mental burdens. Explicit policy is needed and I don't see it present, to guide researchers and the clinicians conducting the clinical scans to assure appropriate evidence based practices.
And lastly, this is the punch line. I don't see an evidence base currently for this policy. Routine MRI screening conducted by NIH, I would think, has to be sound clinical practice based on evidence that establishes benefits justifying its burdens. Okay. I think I have said enough. 6 is the punch line. And here are my acknowledgments. Thank you. [applause]
QUESTION: Thank you for the presentation. I have a question for you. You give a good argument why the choice number 5, which is more or less the current policy, isn't necessarily the best choice. I have some issues with the choice that was recommended, number 3, it seems to be taking a middle ground that relies in the sane diplomate that if somebody is in my study, and I seem to be pretty decent at looking at MRIs I might find it. And if a way it might be better to go with -- although it's a little bit scary to think of that option number 1 where Betty Jo and you both gave very good reasons why these patients would otherwise not have an MRI, they might well be better off not having one because you find the findings and why not upfront tell them, if you want to have an MRI clinically, otherwise we are not even going to look for clinical stuff or maybe do number 4 where if you're going to look for stuff, have an expert rather than relying on the general pediatrician looking. I think I gave enough there.
BETTY JO: I actually agree with you and I argued that at the workshop you eluded to. I don't -- I think you can make an argument for not taking any action on any of these. Now clinically, that ends up being hard to do when you see something there. But, I don't think that the data is there to suggest you really are doing something good for these folks. And the case I eluded to is it happened under our policy currently and we ended up giving a ct scan to a 13-year-old who didn't need it. If we were operating under, this is absolutely not a clinical assessment. If you're concerned over your health, go to your dr., that wouldn't happen.
WOLF: This is a problem. There is no cost free problem free way through this. If you go with option one, the reason there has been ethical arguments pounded, which I find persuasive that one is a mistake, is because you are going to see things that you go, oh, my goodness. The signal case that brought this problem to us was an adolescent in fMRI cognitive research and there it was. And it is hard as can be to shut your mouth when this kid is sitting there and you see something and you think, I might save a life. And so, I think in that kind of circumstance, it makes sense to mount a policy that would allow you to act on that impulse. Now, saying that, you're still going to trigger some false positives. You're going to trigger some burdens. But I think not only is that a laudable impulse but I think truly research participants expect that you will do that. Not that that should be the end of the debate but I think it's true. It's in data from their expectations. So we are trying to weave a course here through these problems. I think there is an argument for number 4. 4 -- I mean extra-mural researchers get very worried they don't have the budget. They don't have the budget to do 3 much less a budget to do 4. And you need to have neuroradiologists who are willing to do it. One of the things that Betty Jo will recall we discovered in our workshop was that some radiologists really balked and said, what? You want me to look at a research grade scan and say what? I’m not even willing to do it. But then there were a large number who disagreed who said, I’m already doing it. For my research colleagues. That's also another issue. I think honestly, there were some liability concerns that were floating around in there. So I think you can argue for number 4. I think 1 is a tough sell.
QUESTION: As a medical scientist, I have been participating as a guinea pig in a numerous studies for the last 40 years and I expect a good study and not only for the clinical part but as a volunteer, I’d like to know more about myself. So, how do you grade this sub clinical study in your presentation, Betty Jo? A fewer number of -- reduction -- or some other issues? How do you differentiate?
BETTY JO: The anatomies that we get for research purposes are one millimeter slices and the clinical ones are 3 millimeters. So, the clinical ones distinguish between different tissue types much better. So, for our purposes, we need the finer grain anatomy but that doesn't do the best job of distinguishing tissue types.
QUESTION: I have been part of submitting an MRI study of the affect of aging on brain and they found a little cyst. So the issue is we must have the good quality with in expensive procedures. So if a good study is done, it provides useful information that should be available to the participant. Thank you.
QUESTION: I was wondering if any could speak to the impacts these recommendations have had on the policy and whether or not it's going to be changed.
WOLF: That may be why we are here. I’m not sure. There has been a very large impact in the world out there. I’m not privy to whether there is reconsideration of this policy afoot at NIH. I would hope because the policy, that one sentence, predates all the work that I laid out. And the empirical work. So I think there is a lot of grist for the mill now and it really does bear consideration.
WENDLER: I can say as a result of this consult, this issue came to the Clinical Centerethics committee who is now in the process of discussing it. So there will be more to come. That is much too early to say.
QUESTION: We do pay normal healthy volunteers for their time and inconvenience when they enroll in a study. So how much your patients get paid for these functional MRI studies?
BETTY JO: It depends on how long the study is and how in-depth it is anywhere from about 80 bucks to a couple of hundred.
QUESTION: The reason is, if you're giving them a compensation that is kind of disproportionate to the inconvenience, 2000 dollars, then you can say you are doing it as a for-pay research and you're not obliged. You're paying them for something that you're not obliged to disclose anything to them.
BETTY JO: I don't know if that would remove the obligation.
QUESTION: that's my question. I’m raising the question for the -- well that raise -- where you just pay somebody and do whatever you feel like doing with them and then not inform them of anything? You are kind of -- walking on that pin line.
WOLF: You can't buy your way out of the common rule. So, no. Even though you're paying -- in fact, payment raises more issues. It doesn't decrease the number of issues as you suggested yourself.
QUESTION: Thank you for a very interesting discussion. Regarding the issues of cost that may be associated with further clinical evaluation after an incidental finding, I wonder if you couldn't look at it as it paralleled to what is done widely as I understand at least at academic institutions where if an adverse event occurs in the context of an investigation al study, subjects are told upfront in the consent form that immediate emergency medical care will be provided so they will be taken care of in terms of a medical need. However, they will be responsible for the cost of additional diagnostic and therapeutic procedures either individually or through their private insurances. Could something like this be included in the consent form regarding incidental findings?
BETTY JO: That's essentially what we include. We just -- our population is generally under served medically. So we didn't feel that it was appropriate to leave folks with terrible ambiguity about whether this is a tumor when a 500 dollar scan could resolve the issue one way or the other. It's a show-stopper for most of our participants. We have taken the position if a further study is needed to clarify, we'll go that far and then refer them on their own. But that's what is in our consent now.
WOLF: And that is the dominant recommendation. But, I’m not sure my sense is that here at NIH, they may be paying for more. There may be a sense that the clinical obligation extends beyond diagnosing. There is a confusion in the literature which you're calling out, I’m glad, between adverse events and incidental findings. And they are not the same thing.. Because incidental findings are not caused the phenomenon by the research. They are discovered by the research. So it's important to keep those two things in mind. This business about uninsured populations is a very big problem because do they have the capacity to follow-up anything you discover? And if you're doing research, let's say not here at NIH, and your policy is not to pay for all the follow-up but you're going to load it on them, what are they expected to do? And so then does the researcher have an obligation to set up a pathway for these uninsured people? That is really unsolved.
QUESTION: I’m going ask a question for another day and another session. But first I have been on the side of clinical MRIs and I thank you because you started to nudge me a little bit away from that. But I think I want to remark on the fact from my experience here, clinical -- patients who come to research in universities or at NIH, do expect a certain level of care. And separating the research from the clinician, it shows good but it's really very, very difficult for a lot of studies where a lot of incidental findings come out. And where a finding may or may not relate to the treatment of an unknown disease. So I think there is a whole universe of dealing with incidental findings that are not MRIs of the brain and each research group finds a way to do that and get either outside care or appropriate inside care and I have the diploma of, should we include brain MRI in that big universe or do they require the kind of delivery you have just given us?
WOLF: The biggie is CT colonography research where they are imaging from the base of the lungs to the pubis and the incidents last published competed 80% of extra colonic findings. Jay cats famously among others, has written about, can you combine these roles? You're right. I agree with you that research subjects, participants, show up often with confusion. You're the one in the white coat. Aren't you taking care of me? But that is an enormous set of problems. You're right for another day. Right now, I don't think we are funded, set up or equipped to have the CT colonography researcher now take clinical responsibility from here to here.
WENDLER: Thank you for a terrific session.
[applause]
(Music fades in, under VO)
ANNOUNCER: We were pleased to bring you a special "Ethics Grand Rounds", recorded April 1, 2009. The topic, "Looking for Clinical Findings in Research Subjects: When is it a Good Idea?" Our speaker was Susan M. Wolf, McKnight Presidential Professor of Law, Medicine and Public Policy and Faegre and Benson Professor of Law, University of Minnesota Law School, also Professor of Medicine, University of Minnesota Medical School. You can see a closed-captioned videocast of this lecture by logging onto http://videocast.nih.gov -- click the "Past Events" link, or by clicking the "View Videocast" link for today's podcast at the Grand Rounds podcast page at www.cc.nih.gov/podcast/grandroundpodcasts.html. The NIH CLINICAL CENTER GRAND ROUNDS podcast is a presentation of the NIH Clinical Center, Office of Communications, Patient Recruitment and Public Liaison. For more information about clinical research going on every day at the NIH Clinical Center, log on to http://clinicalcenter.nih.gov. From America’s Clinical Research Hospital, this has been NIH CLINICAL CENTER GRAND ROUNDS. In Bethesda, Maryland, I’m Bill Schmalfeldt at the National Institutes of Health, an agency of the United States Department of Health and Human Services.
