12/03/2014 - Session III Presentation: John Lantos

Session moderator and second-presenter introducer

Then we'll move on to our next topic which is ethical considerations for communicating risks of standard of care research. We have two presenters and our first presenter is John Lantos who is professor of pediatrics at the University of Missouri at Kansas City and director of the Children's Mercy Hospital Bioethics Center.

Dr. John Lantos

Thank you so much and thanks for having me. It's really been a great conference so far. I especially liked the opening where we said we weren't going to talk about support and then showed a video with a 20-minute presentation about support. It was brilliantly inconsistent and thus set the stage for the discussions that followed. One of the heartwarming things to me about this whole controversy about consent for standard of care research is to learn just how much lawyers and bioethicists esteem and trust physicians. We should note a watershed moment in bioethics. Clinicians used to be seen as paternalistic, as haughty, as playing God, as enemies of patients. No more. Now the physician, the personal physician is the icon of virtue. She cares about me, about only me. Her loyalty is absolute. Her judgment impeccable. Her communication skills are astute, empathic and intuitive. She helps me know what I want because she knows what I want. She helps me choose to choose what will always be the best thing for me and what we choose together will be the choice with the lowest risk and the highest benefit individualized just for me. The researcher

Dr. John Lantos

like a fallen angel is noble in a very different sort of way. Where she is humble and altruistic, he is brash, full of hubris. and fundamentally oblivious to my interests, values, and preferences. For him, I am fundamentally meat. I am an assay. If he could replace me with a rat, he gladly would. My interests are of no interests to him. He would attempt to force his choices upon me unless he is carefully overseen at every stage of his diabolically utilitarian project. And if his recommendations differ in any way from what my saintly physician would recommend when she steps down from the water upon which she walks,

Dr. John Lantos

that recommendation will inevitably be accompanied by risks that I would not face if I just say no. And here is the worst thing. The clinician is also a chameleon. and can transform herself into a researcher subtly and almost imperceptibly. It's like Roald Dahl's book, The Witches. Have you ever read that? Where

Dr. John Lantos

women who form the Society for the Protection of Children, when they go into meeting in their closed room, peel off the rubber masks that are their faces and turn into witches who, instead of protecting children, like to turn them into rats. A cartoon here? Maybe, but not so far from the truth of the way we currently think about research and clinical care. This view of the dichotomy between the clinician and the researcher is the basis for our current bizarre system in which we know that informed consent for clinical care is terrible. We know that there are all sorts of idiosyncratic, unjustifiable, and dangerous practice variations. We know that patients are harmed as a result. We know that costs are increased as a result. And we know that it's been going on forever. And we don't really care. It's also the basis for the legal responses to the differences in consent for research and therapy. As a parable, think of what happened with diethylstilbestrol, DES.

Dr. John Lantos

You may remember that that was a drug that was widely used in the 1940s and 50s, supposedly to prevent stillbirth and premature birth. It was widely used. It was rarely studied. Finally, some researchers did a prospective randomized control trial and showed that it was useless to prevent premature birth or stillbirth, and much later, much later, was found to cause vaginal adenocarcinoma in the offspring, in the babies who were born to women who took it during pregnancy. So who got sued when this was discovered that the women had the vaginal adenocarcinomas? Not the doctors who were prescribing it willy-nilly without studying it. The researchers who failed to get informed consent warning of what must have been seen as a reasonably foreseeable risk of vaginal adenocarcinoma in the offspring 20 years later. If data mattered, we'd have to admit that we have pretty robust data now showing that the well-designed clinical study is as safe as clinical care. Patients in well-designed clinical studies have outcomes that are comparable, not better, but not worse, than similar patients who are not in studies.

Dr. John Lantos

And we know that in addition then to being safe, they also generate valuable information. There have now been three meta-analyses. Fernandez in the Canadian Medical Journal just last month Gross in public plus medicine and Vist did a Cochrane meta-analysis analyzing over 200 studies in which there was a comparable group not in the study. And they showed in the meta-analysis the results are pretty similar whether you're randomized, not randomized in the study, not in the study, looking at similar outcomes, looking at different outcomes. So why then are we so lax in our oversight of clinicians and so draconian in our suspicion of researchers? The answer, I think, comes from the history of research ethics. And the key figure, the person who I think was the first to propose this dichotomy was Otto Gutentag. He was a German physician who emigrated to the US in the 30s, taught at UCSF for many years, and in 1953 wrote a great essay published in Science in which he talks about two aspects of the physician-patient relationship. In one, the physician's goal is solely to care for the patient. In this situation, Gutentag writes, one human being is in distress, in need, crying out for help. Another fellow human being is concerned and wants to assist him. The cry for help and the desire to render it precipitate the relationship. In the other type of relationship, Gutentag writes, the physician is quote, the one who performs experiments of no immediate value to the person under observation, experimentation is the basis on which the two meet. The original bond between them and Gutentag imagines the gulf between these two roles, the clinician and the scientist, to be quite wide and to present a significant challenge to the basic concepts of the original doctor-patient relationship. And this theme of the dual loyalties of the clinician and the investigator has come to be what I would call the central dogma of research ethics. It's been repeated like a mantra by scholars including Larry Churchill, Sam Hellman, Franklin Miller, Jerry Menikoff, Lynn Lance repeated it yesterday, Ruth Macklin, and others. And as an example, OHRP's critique of the Voldemort study said, Quote, ultimately, ultimately, key word, ultimately, the issues come down to a fundamental difference between the obligations of clinicians and those of researchers. Doctors are required to do what they view as being best for their individual patients. Researchers do not have that same obligation, unquote. But here's the key point. If you go back and read Gutenthal, this is not what he had in mind. He's quite explicit about it. So is Henry Beecher. Gutentag was explicit in stating that the idea of this split, of this divide, was only relevant in studies, in his words, that did not have as a goal the direct benefit of the patient. He stresses that when the goal is to benefit the patient, a very different and more complex set of issues arises. Henry Beecher makes the same distinction. He notes the distinction between experimentation that's not for the patient's benefit and experimentation that is designed to benefit the patient and writes, quote, to treat the ill is to experiment. Experimentation is a common necessity in fitting the remedy to the disease. The patient gives his permission in the act of going to the physician for relief and the usual doctor-patient relationship. In other words, the conflict of interest and goals that arises when research is not for the benefit of the patient either disappears or is much less concern in studies where the goal is benefit for the patient. Jay Katz, who I think probably has written more insightfully about informed consent than anybody ever and is more widely ignored than anybody who's written about informed consent than anybody ever. wrote in 1969, quote, the oft-made distinction that the physician is primarily concerned with the patient qua patient and the investigator with the research is not a useful one in the majority of medical situations, for most investigations occur in the context of clinical research combined with professional care. Katz goes on, the art and the science of medicine are intricately interwoven. Research and therapy, pursuit of knowledge and treatment are not separate but intertwined. The emphasis on informed consent for research raises troubling questions about what all patients should be told about all interventions. And he concludes that codes that regulate researchers and their consent process only are what he calls, quote, pious exercises in futility. And his key phrase is this, since they aspire to ideals and are divorced from the reality of human interaction, they invite judicious or injudicious neglect, which is just what happens with current approaches to informed consent for research. Here's a story from our place. Our IRB was very concerned. not because of the support study, we weren't involved in that, but another neonatal research network study called the Top Transfusion of Premature Baby Studies. And the IRB said, oh, you know, we're going over your informed consent form. So I asked the PI there, did you change the consent form? And he said, oh, yeah. It says death in every paragraph. I mean, it's just, you know, you enroll your baby in this, death, death, death. I said, so are people not signing up? He said, nobody reads those things.

Dr. John Lantos

And this is the most important problem with regulations that require things that, in Katz's words, are divorced from the realities of human interaction. It makes ethics and informed consent a meaningless ritual. So maybe we could move the discussion forward by identifying that the real problem is not being the difference between research, whatever that is, research on standard of care, whatever that is, standard care, whatever that is, or even quality improvement, whatever that is. The real problem is informed consent and shared decision making for any intervention in healthcare when there is uncertainty and choice. So let me finish with a brief proposal for what maybe a draft guide, I think a draft guidance ought to look like and what patients need to be told. It picks up on the integrated consent model that Kim and Miller published in the New England Journal earlier this year.

Dr. John Lantos

I think what we need to do is identify situations where there is significant practice variation, where different doctors recommend different treatments. There are many. Could be treatment for macular degeneration that John Tyson was talking about yesterday, or screening for colon cancer that Ruth Macklin was talking about, or aspirin for prevention of heart attacks, or whatever. And in these situations, if there's a proposal to do a study, then all patients who are facing that choice should get a short form that has five key components. Component one, your current situation. You have a disease. Here's the natural history. Here's the prognosis. That's what you're facing. Number two, there are options. What are they? What treatments are out there and what do we know about them? Number three, why doctors disagree. And in this I would say people need two of the most common risks and one most serious risk, two of the most common benefits and one most important benefit. People can't hold more than four ideas in their head at the same time anyway. So narrow it down to two or three. And then here's the key proposal that I think would incorporate some of what we were hurting from the patient representatives yesterday. Say to people, you now have a choice. We're doing a randomized trial. You can enter the randomized trial and your decision will be decided by the gumball machine. Or you can choose one of the treatments or the other. And number five, whatever you choose, whether it's randomization or one treatment or another, give us permission to collect outcome data. And then we'd have data. for all these situations on patients who are eligible, patients who got randomized, and patients who made their decision based on personal choice. It would allow people to

Dr. John Lantos

exercise their autonomy. It would give them all the relevant information. And if OHRP and risk management said you had to staple the 20-page consent form on after it, bless their hearts, I have no problem with that. Thank you.

Session moderator and second-presenter introducer

Any questions, comments?

Audience member 1

John, I have a quick question. Actually, first of all, thank you. And thank you for your five-point proposal. Actually, my question about the proposal is simply, as you're envisioning that, how long do you think it would take to have that conversation? for the average person involved in facing one

Dr. John Lantos

of these issues? That would be something better for you guys to try to study. But I would picture five to 10 minutes. Most of my data to answer such questions comes from cocktail parties, where I find I can explain most randomized control studies to people who have had a few drinks in five or ten minutes. That helps too.