Cadmore media player playing audio Alex John London, PhD, discusses equipoise in research.
Audio Player
Accessibility
Using forms mode, you can utilize keyboard shortcuts to make usability more efficient.
Space bar will play and pause the audio. Pressing escape will close all displays and pressing it again will jump to the end of the player.
You can press A to open the Keyboard Shortcut Dialog which also provides an audio description of the player and all of its features.
Transcript
Result Count
of
Search
Segments
Result Count
of
Search
Name:
Alex John London, PhD, discusses equipoise in research.
Description:
Alex John London, PhD, discusses equipoise in research.
Thumbnail URL:
https://cadmoremediastorage.blob.core.windows.net/65aaa389-459d-4940-8d10-1041f1f64d7a/thumbnails/65aaa389-459d-4940-8d10-1041f1f64d7a.jpg?sv=2019-02-02&sr=c&sig=LAd%2Bua9VNJEe8RBNg%2BgRBaTAbYT5Hx3qVt1cF8zGE3I%3D&st=2024-12-30T17%3A44%3A29Z&se=2024-12-30T21%3A49%3A29Z&sp=r
Duration:
T00H21M47S
Embed URL:
https://stream.cadmore.media/player/65aaa389-459d-4940-8d10-1041f1f64d7a
Content URL:
https://cadmoreoriginalmedia.blob.core.windows.net/65aaa389-459d-4940-8d10-1041f1f64d7a/19042419.mp3?sv=2019-02-02&sr=c&sig=3D5NglH53hcbQ4BTwJG5TXHPtCCeQTxTgt2fC0WTSUA%3D&st=2024-12-30T17%3A44%3A29Z&se=2024-12-30T19%3A49%3A29Z&sp=r
Upload Date:
2022-02-28T00:00:00.0000000
Transcript:
Language: EN.
Segment:0 .
[ Music ] >> Hello, and welcome to JAMAevidence, our podcast series based on core issues and evidence-based medicine. Today, we will be discussing the topic of equipoise in research. I'm your host, Dr. Roger Lewis, Statistical Editor for JAMA and Co-Editor of the JAMA Guide to Statistics and Methods series. I'm here with Dr. Alex John London, who is the Clara L. West Professor of Ethics and Philosophy at Carnegie Mellon University where he directs the Center for Ethics and Policy.
His book -- For the Common Good Philosophical Foundations of Research Ethics, published in 2021 from Oxford University Press. It's a pleasure to have Dr. London with us today. >> It's a pleasure to be here. Thanks very much. >> Thank you. So I'd like to start with just a first question, since equipoise is something that many of our readers have heard about, but actually may not be very clear on its exact definition. So let's start with, what is equipoise?
>> So equipoise is a state of conflict or uncertainty about the relative clinical merits of a set of interventions. In the piece in JAMA, the set of interventions was a video laryngoscope or a Macintosh direct laryngoscope and the relative merits involved a comparison of successful first-pass intubation for patients in the ICU. >> So equipoise is usually brought up as a consideration in designing randomized clinical trials or other trials in which we allocate treatments to patients.
So how is equipoise related, if it is, to the ethical acceptability of randomized assignment of treatments? >> Clinicians have important ethical obligations to patients and so the idea is that when equipoise obtains, it's considered to be ethically permissible for clinicians to offer study participation to those patients. So what that means is, when there's uncertainty or conflict about the relative merits of the set of interventions that are included in the trial, that it's permissible for clinicians -- ethically permissible -- for them to offer participation in that study to their patients, even if that means that the treatment that the patient receives isn't going to be the one that the clinician directly recommends.
The treatment will be allocated through a process like randomization. >> So when you use the term "conflict," what do you mean by conflict? >> Yeah, so there are two ways in which equipoise can obtain. One way is if people are just uncertain -- if experts are uncertain about which from a set of interventions is going to be best for patients. That can be the case, but it can also be the case that clinicians can have strict preferences about interventions.
That is to say, one group might be convinced that a particular approach or strategy or intervention is superior to the alternatives. And then you can have a different group of clinicians that also believes one intervention is preferable to the others. It's just that, that group disagrees. And so that would be a case where there's conflict in the expert community about how to treat or intervene with patients for the same condition. >> So would it be fair to say that, in a setting in which individual clinicians have strong preferences for one treatment or another, that as long as different clinicians have different preferences, then a state of equipoise exists?
>> That's right. That's the idea. So you could have equipoise if the experts -- the group of experts -- say we don't have enough information to tell us which of these set of possible interventions is superior so we want to run a clinical trial. And during COVID-19, for example, you saw this. There were a lot of therapeutic or prophylactic candidates that were supported by preliminary evidence or some hypotheses, and so then the clinical community said there's a lot of uncertainty and so it's ethically permissible to offer those interventions -- to allocate those interventions to patients using a random process inside the confines of a well-designed clinical trial in order to ascertain which, if any, of these interventions work and which ones don't.
In other circumstances where you have some experts who feel very confident that they've seen a lot of evidence that one intervention is superior, that if the evidence that exists isn't sufficient to persuade other fully-informed experts of the superiority of that intervention, then it can be -- also be permissible to run a trial where the goal of the trial is to generate the information that's necessary to see which of those groups of experts is correct. >> So sometimes when we're discussing interventions that are being considered for incorporation in a randomized clinical trial, a clinician with strong feelings will believe or state that it would be unethical for them to enroll a patient in the trial if they personally would not recommend both of the arms in the trial.
How would you respond to that? >> So I think that that is a misconception and I think it's one of the unfortunate consequences of the name "equipoise." And also of the long history of debate about the concept. I mean, so the notion of -- that the term equipoise evokes this image of a knife's edge where someone is equally poised between, you know, two interventions, about which one is better. And the way I like to think about this is, if a clinician believes that one intervention is superior and other clinicians disagree, is it consistent with acceptable medical practice that some patients would see clinicians from this first group, let's say, and that other patients would see clinicians from this other group?
If it is, that is to say, if one of those groups of clinicians isn't practicing substandard or inferior or unethical medicine, then it's got to be permissible to allow patients, to ask patients, if they would be willing to be allocated at random to those same set of interventions, even though experts within those different groups may feel that patients who are allocated to the alternatives that are favored by their colleagues are getting substandard or inferior care.
And the truth is that there are interventions that have support of prior evidence and some strong theory and then, in actual practice, it turns out that they're inferior to alternatives. So we don't want to foreclose randomized, controlled clinical trials that will settle those kinds of disputes because settling those kinds of disputes is an important source of social value. We'd like to eliminate variation in practice that isn't supported by evidence that tells us that that variation is caused by our ability to titrate care to the specifics of individual patients and improve outcomes for them.
That kind of variation is great, but variation that isn't supported that way, that's just random variation, we want to get rid of. >> So I think this is a critical point. So I just want to make sure that I understand it correctly. So what you're saying is that, even if no individual physician feels tremendous uncertainty in what they would recommend for an individual patient, as long as physicians in general recommend different therapies, that the principle of equipoise would hold and it may be ethical to randomly assign those therapies within the context of a well-designed clinical trial.
Is that correct? >> Yes. That's right. Know -- you could imagine a circumstance in which each member of the expert community has a strongly-held preference in favor of one intervention. And as long as it's not in favor of the same intervention, right, as long as there's disagreement about which intervention is best, then the value of a clinical trial would be to put to test the views of these different experts in order to find out which of that set of experts is correct.
>> Thank you. There was a particular phrase from your article I'd like to give you a chance to explain. And it's -- you stated, "the principle of equipoise reconciles two potentially conflicting ethical imperatives, to ensure that research involving human participants generates scientifically sound and clinically relevant information, while demonstrating proper respect and concern for the rights and interests of the study participants." Can you expand on this? >> Sure. So two pillars of ethically acceptable research with humans are that, first, such research has to generate valuable information.
And so the principle of equipoise gives us a way of trying to make that requirement operational. So if we don't know what the optimal way to intervene, to treat, or prevent a disease is, and a study is going to shed information -- generate information that we need in order to answer that question, then that's a form of social value or value that the study could generate. Similarly, if there's conflicting opinion among experts, so there's a diversity of practice for treating a particular condition or for trying to prevent a particular condition and the study would address that disagreement, it would resolve or narrow that disagreement, then that's another source of important social value.
So when equipoise obtains, it's a way of helping us to establish that, yup, this is a study that's worth doing. The second pillar of ethically acceptable research is that not only does the research have to be scientifically sound, but it has to respect the rights and welfare of study participants. And in this, the aspect of that that's relevant to equipoise is that we want to make sure that no participant is knowingly provided with care that is substandard. And so we can operationalize that here in terms of -- we don't want any patient to be allocated to care that not even a reasonable minority of experts would recommend for them.
And so, if equipoise obtains, then it should be the case that, no matter what you're allocated to in the study, it's consistent with the recommendation of at least a reasonable minority of expert clinicians who are fully informed. That is to say they're familiar with the literature. They know what we're doing. >> What are common misconceptions regarding equipoise that you've run into among clinical investigators? >> Yeah. So one of those is something that we've already talked about. The most common misconception about equipoise is that it refers to the belief state of the individual clinician.
That is to say that it requires the individual clinician to be uncertain about the relative merits of the set of interventions for treating or preventing a particular condition. That's a mistake, though, because, if that were the case then it would foreclose studies in which there's disagreement or conflict in the expert community about what we ought to do. And that would be odd, then, to say, well, it's okay for there to be diversity in actual practice in what people do, but it's not okay for us to run a study in which we would randomize people to those very same interventions in order to generate the information that we need in order to see whether those different practices are on par or whether one of them is inferior to the others.
And so from, if you think about it, preventing that kind of trial makes nobody better off. It doesn't protect patients because patients get access to those same interventions out in clinical practice if both of those interventions are available in clinical practice. But it deprives the scientific community and patients and policy-makers and other stakeholders of the information that they need in order to optimize patient care. So, even though it's very common to think that equipoise refers to the belief state of the individual clinician, I think that that's an error.
>> So one of the important initial steps in a patient's participation in a clinical trial is participating in a voluntary and informed discussion, during which they can have their questions answered prior to giving consent for participation. So how should equipoise be communicated in an informed consent document or during an informed consent discussion with a patient considering enrolling in a clinical trial? >> Yeah. So an important goal in the informed consent process is to let participants know why a trial is being conducted, what makes the study important, so that they have a sense of what their participation is contributing to.
That, alongside of, you know, patients have to understand what procedures will be performed on them, what the risks of study participation are, and so on. And so explaining the nature of the uncertainty or the disagreement among experts that a trial is designed to resolve or reduce, it's a good way to help participants understand how their participation is going to contribute to science and to the advancement of medical practice. And it's a good way -- it's a good scaffolding, then, to explain to participants what the procedures are that will be performed on them and how, if at all -- you know, how do those procedures differ from what would be performed on them in ordinary clinical practice.
And are there elements of the research that are included specifically for the research. So things that would not be done in clinical practice but that are necessary to advance the research, for example, study-related blood draws or tests that are required to generate data that are needed for the study, but that wouldn't be performed if the participant were not to participate in the study. So you don't have to use the term equipoise, but communicating the information that term equipoise refers to can be a good way to help patients understand what's going to happen to them, and generate buy-in to the idea that participating in this study is worthwhile.
>> So, obviously, one of the purposes of a randomized clinical trial or clinical research in general is to create evidence that shifts medical opinion towards acceptance of a single or a set of preferred treatment strategies. So how do you know when enough evidence has accumulated so that equipoise no longer holds, implying that it would no longer be ethical to randomize patients between new treatments? >> Yeah. That's a great question, and there are a couple of different parts to it. So one of those is, sort of, a narrower statistics question about -- at what point would we think that the evidence we generated is sufficient for us to say we resolved the conflict or uncertainty?
A second part of that, though, is, are we making sure that the endpoints that we're looking at in our trial -- so this is, sort of, before we get to the technical statistics part -- when we're designing the study, are we designing our study around endpoints that are meaningful to the people who are going to be the consumers of the research? And so I say that, not to be crass in terms of the consumers of the research, but just the idea that, look, the consumers of a phase three study are going to be practitioners and patients.
The consumers of a phase one or a phase two study may be the researchers that are carrying out subsequent research in the trajectory of trying to evaluate the merits of a particular strategy or whatever. But it's important to recognize that these studies are a tool. They're a tool to generate information that the consumer down the line, the next researcher or the clinician, needs in order to resolve an important uncertainty. And so if we're talking about a phase three study, then it's important that we say -- that we be confident that the endpoints that we've chosen, you know, and the information that we're generating, goes to the subject of the disagreement or the nature of the uncertainties.
If we're uncertain about what the best means of prophylaxis is, that we design our trial to generate information that we can agree ahead of time, yes, answering that question is likely to change practice. If there's a disagreement among clinicians, it's important to say, what is the nature of that disagreement? So that if we generate information about that, how likely is it, then, that in light of that information, open-minded scientists who disagree will see that information and say, ah, yes, I have to move, shift my belief in favor of this intervention rather than that intervention.
I think that's the important first step and then the second step is making sure that we power our study or design our study so that it generates enough information that it's going to actually persuade people. Because if we generate information that leaves room for doubt, then we might complete the study but not have disturbed equipoise. And I think you see that in COVID-19 when there were rafts of very small studies that were underpowered, that were basically -- they were designed to detect a silver bullet.
And part of the problem with those studies is, yes, if they found a penicillin-like effect then they would disturb equipoise. But if they don't find a penicillin effect, it's not necessarily the case then that they eliminate those interventions that were being studied from consideration because your prior going into research is that, well, it's unlikely that anything has an effect that big. And so the fact that we didn't find it is not really unusual and it could be the case that there's a more meaningful, smaller effect there. And that's why, you know, part of what was objectionable about some of those studies is they weren't looking for meaningful effects with adequate sample size and power or other operating characteristics, so that negative results are also informative.
That is to say that people go, yup, this isn't something that we should continue to look at. Let's move on. >> So in follow-up to your last set of points, what would you say about research that, by its very design, is unlikely to have a meaningful chance of disturbing equipoise? >> Yeah. That research, I would say, is morally problematic because it isn't just that it's an important condition or a necessary condition that research begin in equipoise. Right? That is an important condition, but the point of research is to generate the information that's going to move the beliefs of clinicians or researchers or policy-makers.
And so if you look at a study beforehand and you say the results of this study are unlikely to do that, then it's not clear that it meets that first pillar of ethically acceptable research, namely that it has sufficient value that it justifies us moving forward. >> Your point about studies that are unlikely to be able to disturb equipoise as being morally problematic is extremely well taken. But it does raise a question about learn phase studies in general and, by that, I mean studies that are intended to move the science further but not to directly affect clinicians' beliefs or preferences regarding the treatments that are being randomized.
>> Yeah. That's a great point. So I think it's important to emphasize that the point about studies being designed to disturb equipoise applies most directly to sort of pivotal phase three studies. When you're looking at early phase studies, you have to evaluate those studies in a larger context of a trajectory of research, that the point of those studies is to explore particular facets of a dose or an indication or the things like that, and it must be the case, though, that they're doing that in a way that contributes to a hand off, like a epistemological or scientific bucket brigade where, you know, they're pouring information into the bucket of the next stage where the idea is all these stages are going to build up to studies that have the potential to disturb equipoise.
So it can be the case that earlier phase studies can be objectionable if, for instance, they use endpoints that subsequent researchers or clinicians wouldn't really regard as relevant to the underlying scientific questions. It's slightly different there. It's not about disturbing this uncertainty or disagreement in the expert community, but it's similar in the sense that the question would then be - was there adequate value in this study handed off to what I earlier called the consumer of that information to really move the science forward?
And I think equipoise is about ultimately we want to move science forward not to satisfy individual curiosity. We want to move science forward to improve our ability to prevent or treat or diagnose disease. >> This is Roger Lewis, and I'd like to thank our guest today, Dr. Alex John London. More about this topic is available in the JAMA Guide to Statistics and Methods series which is published in JAMA and on our website, JAMAevidence.com. For more audio, visit us at JAMAnetworkaudio.com. You can listen and follow wherever you get your podcasts.
[ Music ]
Segment:0 .
[ Music ] >> Hello, and welcome to JAMAevidence, our podcast series based on core issues and evidence-based medicine. Today, we will be discussing the topic of equipoise in research. I'm your host, Dr. Roger Lewis, Statistical Editor for JAMA and Co-Editor of the JAMA Guide to Statistics and Methods series. I'm here with Dr. Alex John London, who is the Clara L. West Professor of Ethics and Philosophy at Carnegie Mellon University where he directs the Center for Ethics and Policy.
His book -- For the Common Good Philosophical Foundations of Research Ethics, published in 2021 from Oxford University Press. It's a pleasure to have Dr. London with us today. >> It's a pleasure to be here. Thanks very much. >> Thank you. So I'd like to start with just a first question, since equipoise is something that many of our readers have heard about, but actually may not be very clear on its exact definition. So let's start with, what is equipoise?
>> So equipoise is a state of conflict or uncertainty about the relative clinical merits of a set of interventions. In the piece in JAMA, the set of interventions was a video laryngoscope or a Macintosh direct laryngoscope and the relative merits involved a comparison of successful first-pass intubation for patients in the ICU. >> So equipoise is usually brought up as a consideration in designing randomized clinical trials or other trials in which we allocate treatments to patients.
So how is equipoise related, if it is, to the ethical acceptability of randomized assignment of treatments? >> Clinicians have important ethical obligations to patients and so the idea is that when equipoise obtains, it's considered to be ethically permissible for clinicians to offer study participation to those patients. So what that means is, when there's uncertainty or conflict about the relative merits of the set of interventions that are included in the trial, that it's permissible for clinicians -- ethically permissible -- for them to offer participation in that study to their patients, even if that means that the treatment that the patient receives isn't going to be the one that the clinician directly recommends.
The treatment will be allocated through a process like randomization. >> So when you use the term "conflict," what do you mean by conflict? >> Yeah, so there are two ways in which equipoise can obtain. One way is if people are just uncertain -- if experts are uncertain about which from a set of interventions is going to be best for patients. That can be the case, but it can also be the case that clinicians can have strict preferences about interventions.
That is to say, one group might be convinced that a particular approach or strategy or intervention is superior to the alternatives. And then you can have a different group of clinicians that also believes one intervention is preferable to the others. It's just that, that group disagrees. And so that would be a case where there's conflict in the expert community about how to treat or intervene with patients for the same condition. >> So would it be fair to say that, in a setting in which individual clinicians have strong preferences for one treatment or another, that as long as different clinicians have different preferences, then a state of equipoise exists?
>> That's right. That's the idea. So you could have equipoise if the experts -- the group of experts -- say we don't have enough information to tell us which of these set of possible interventions is superior so we want to run a clinical trial. And during COVID-19, for example, you saw this. There were a lot of therapeutic or prophylactic candidates that were supported by preliminary evidence or some hypotheses, and so then the clinical community said there's a lot of uncertainty and so it's ethically permissible to offer those interventions -- to allocate those interventions to patients using a random process inside the confines of a well-designed clinical trial in order to ascertain which, if any, of these interventions work and which ones don't.
In other circumstances where you have some experts who feel very confident that they've seen a lot of evidence that one intervention is superior, that if the evidence that exists isn't sufficient to persuade other fully-informed experts of the superiority of that intervention, then it can be -- also be permissible to run a trial where the goal of the trial is to generate the information that's necessary to see which of those groups of experts is correct. >> So sometimes when we're discussing interventions that are being considered for incorporation in a randomized clinical trial, a clinician with strong feelings will believe or state that it would be unethical for them to enroll a patient in the trial if they personally would not recommend both of the arms in the trial.
How would you respond to that? >> So I think that that is a misconception and I think it's one of the unfortunate consequences of the name "equipoise." And also of the long history of debate about the concept. I mean, so the notion of -- that the term equipoise evokes this image of a knife's edge where someone is equally poised between, you know, two interventions, about which one is better. And the way I like to think about this is, if a clinician believes that one intervention is superior and other clinicians disagree, is it consistent with acceptable medical practice that some patients would see clinicians from this first group, let's say, and that other patients would see clinicians from this other group?
If it is, that is to say, if one of those groups of clinicians isn't practicing substandard or inferior or unethical medicine, then it's got to be permissible to allow patients, to ask patients, if they would be willing to be allocated at random to those same set of interventions, even though experts within those different groups may feel that patients who are allocated to the alternatives that are favored by their colleagues are getting substandard or inferior care.
And the truth is that there are interventions that have support of prior evidence and some strong theory and then, in actual practice, it turns out that they're inferior to alternatives. So we don't want to foreclose randomized, controlled clinical trials that will settle those kinds of disputes because settling those kinds of disputes is an important source of social value. We'd like to eliminate variation in practice that isn't supported by evidence that tells us that that variation is caused by our ability to titrate care to the specifics of individual patients and improve outcomes for them.
That kind of variation is great, but variation that isn't supported that way, that's just random variation, we want to get rid of. >> So I think this is a critical point. So I just want to make sure that I understand it correctly. So what you're saying is that, even if no individual physician feels tremendous uncertainty in what they would recommend for an individual patient, as long as physicians in general recommend different therapies, that the principle of equipoise would hold and it may be ethical to randomly assign those therapies within the context of a well-designed clinical trial.
Is that correct? >> Yes. That's right. Know -- you could imagine a circumstance in which each member of the expert community has a strongly-held preference in favor of one intervention. And as long as it's not in favor of the same intervention, right, as long as there's disagreement about which intervention is best, then the value of a clinical trial would be to put to test the views of these different experts in order to find out which of that set of experts is correct.
>> Thank you. There was a particular phrase from your article I'd like to give you a chance to explain. And it's -- you stated, "the principle of equipoise reconciles two potentially conflicting ethical imperatives, to ensure that research involving human participants generates scientifically sound and clinically relevant information, while demonstrating proper respect and concern for the rights and interests of the study participants." Can you expand on this? >> Sure. So two pillars of ethically acceptable research with humans are that, first, such research has to generate valuable information.
And so the principle of equipoise gives us a way of trying to make that requirement operational. So if we don't know what the optimal way to intervene, to treat, or prevent a disease is, and a study is going to shed information -- generate information that we need in order to answer that question, then that's a form of social value or value that the study could generate. Similarly, if there's conflicting opinion among experts, so there's a diversity of practice for treating a particular condition or for trying to prevent a particular condition and the study would address that disagreement, it would resolve or narrow that disagreement, then that's another source of important social value.
So when equipoise obtains, it's a way of helping us to establish that, yup, this is a study that's worth doing. The second pillar of ethically acceptable research is that not only does the research have to be scientifically sound, but it has to respect the rights and welfare of study participants. And in this, the aspect of that that's relevant to equipoise is that we want to make sure that no participant is knowingly provided with care that is substandard. And so we can operationalize that here in terms of -- we don't want any patient to be allocated to care that not even a reasonable minority of experts would recommend for them.
And so, if equipoise obtains, then it should be the case that, no matter what you're allocated to in the study, it's consistent with the recommendation of at least a reasonable minority of expert clinicians who are fully informed. That is to say they're familiar with the literature. They know what we're doing. >> What are common misconceptions regarding equipoise that you've run into among clinical investigators? >> Yeah. So one of those is something that we've already talked about. The most common misconception about equipoise is that it refers to the belief state of the individual clinician.
That is to say that it requires the individual clinician to be uncertain about the relative merits of the set of interventions for treating or preventing a particular condition. That's a mistake, though, because, if that were the case then it would foreclose studies in which there's disagreement or conflict in the expert community about what we ought to do. And that would be odd, then, to say, well, it's okay for there to be diversity in actual practice in what people do, but it's not okay for us to run a study in which we would randomize people to those very same interventions in order to generate the information that we need in order to see whether those different practices are on par or whether one of them is inferior to the others.
And so from, if you think about it, preventing that kind of trial makes nobody better off. It doesn't protect patients because patients get access to those same interventions out in clinical practice if both of those interventions are available in clinical practice. But it deprives the scientific community and patients and policy-makers and other stakeholders of the information that they need in order to optimize patient care. So, even though it's very common to think that equipoise refers to the belief state of the individual clinician, I think that that's an error.
>> So one of the important initial steps in a patient's participation in a clinical trial is participating in a voluntary and informed discussion, during which they can have their questions answered prior to giving consent for participation. So how should equipoise be communicated in an informed consent document or during an informed consent discussion with a patient considering enrolling in a clinical trial? >> Yeah. So an important goal in the informed consent process is to let participants know why a trial is being conducted, what makes the study important, so that they have a sense of what their participation is contributing to.
That, alongside of, you know, patients have to understand what procedures will be performed on them, what the risks of study participation are, and so on. And so explaining the nature of the uncertainty or the disagreement among experts that a trial is designed to resolve or reduce, it's a good way to help participants understand how their participation is going to contribute to science and to the advancement of medical practice. And it's a good way -- it's a good scaffolding, then, to explain to participants what the procedures are that will be performed on them and how, if at all -- you know, how do those procedures differ from what would be performed on them in ordinary clinical practice.
And are there elements of the research that are included specifically for the research. So things that would not be done in clinical practice but that are necessary to advance the research, for example, study-related blood draws or tests that are required to generate data that are needed for the study, but that wouldn't be performed if the participant were not to participate in the study. So you don't have to use the term equipoise, but communicating the information that term equipoise refers to can be a good way to help patients understand what's going to happen to them, and generate buy-in to the idea that participating in this study is worthwhile.
>> So, obviously, one of the purposes of a randomized clinical trial or clinical research in general is to create evidence that shifts medical opinion towards acceptance of a single or a set of preferred treatment strategies. So how do you know when enough evidence has accumulated so that equipoise no longer holds, implying that it would no longer be ethical to randomize patients between new treatments? >> Yeah. That's a great question, and there are a couple of different parts to it. So one of those is, sort of, a narrower statistics question about -- at what point would we think that the evidence we generated is sufficient for us to say we resolved the conflict or uncertainty?
A second part of that, though, is, are we making sure that the endpoints that we're looking at in our trial -- so this is, sort of, before we get to the technical statistics part -- when we're designing the study, are we designing our study around endpoints that are meaningful to the people who are going to be the consumers of the research? And so I say that, not to be crass in terms of the consumers of the research, but just the idea that, look, the consumers of a phase three study are going to be practitioners and patients.
The consumers of a phase one or a phase two study may be the researchers that are carrying out subsequent research in the trajectory of trying to evaluate the merits of a particular strategy or whatever. But it's important to recognize that these studies are a tool. They're a tool to generate information that the consumer down the line, the next researcher or the clinician, needs in order to resolve an important uncertainty. And so if we're talking about a phase three study, then it's important that we say -- that we be confident that the endpoints that we've chosen, you know, and the information that we're generating, goes to the subject of the disagreement or the nature of the uncertainties.
If we're uncertain about what the best means of prophylaxis is, that we design our trial to generate information that we can agree ahead of time, yes, answering that question is likely to change practice. If there's a disagreement among clinicians, it's important to say, what is the nature of that disagreement? So that if we generate information about that, how likely is it, then, that in light of that information, open-minded scientists who disagree will see that information and say, ah, yes, I have to move, shift my belief in favor of this intervention rather than that intervention.
I think that's the important first step and then the second step is making sure that we power our study or design our study so that it generates enough information that it's going to actually persuade people. Because if we generate information that leaves room for doubt, then we might complete the study but not have disturbed equipoise. And I think you see that in COVID-19 when there were rafts of very small studies that were underpowered, that were basically -- they were designed to detect a silver bullet.
And part of the problem with those studies is, yes, if they found a penicillin-like effect then they would disturb equipoise. But if they don't find a penicillin effect, it's not necessarily the case then that they eliminate those interventions that were being studied from consideration because your prior going into research is that, well, it's unlikely that anything has an effect that big. And so the fact that we didn't find it is not really unusual and it could be the case that there's a more meaningful, smaller effect there. And that's why, you know, part of what was objectionable about some of those studies is they weren't looking for meaningful effects with adequate sample size and power or other operating characteristics, so that negative results are also informative.
That is to say that people go, yup, this isn't something that we should continue to look at. Let's move on. >> So in follow-up to your last set of points, what would you say about research that, by its very design, is unlikely to have a meaningful chance of disturbing equipoise? >> Yeah. That research, I would say, is morally problematic because it isn't just that it's an important condition or a necessary condition that research begin in equipoise. Right? That is an important condition, but the point of research is to generate the information that's going to move the beliefs of clinicians or researchers or policy-makers.
And so if you look at a study beforehand and you say the results of this study are unlikely to do that, then it's not clear that it meets that first pillar of ethically acceptable research, namely that it has sufficient value that it justifies us moving forward. >> Your point about studies that are unlikely to be able to disturb equipoise as being morally problematic is extremely well taken. But it does raise a question about learn phase studies in general and, by that, I mean studies that are intended to move the science further but not to directly affect clinicians' beliefs or preferences regarding the treatments that are being randomized.
>> Yeah. That's a great point. So I think it's important to emphasize that the point about studies being designed to disturb equipoise applies most directly to sort of pivotal phase three studies. When you're looking at early phase studies, you have to evaluate those studies in a larger context of a trajectory of research, that the point of those studies is to explore particular facets of a dose or an indication or the things like that, and it must be the case, though, that they're doing that in a way that contributes to a hand off, like a epistemological or scientific bucket brigade where, you know, they're pouring information into the bucket of the next stage where the idea is all these stages are going to build up to studies that have the potential to disturb equipoise.
So it can be the case that earlier phase studies can be objectionable if, for instance, they use endpoints that subsequent researchers or clinicians wouldn't really regard as relevant to the underlying scientific questions. It's slightly different there. It's not about disturbing this uncertainty or disagreement in the expert community, but it's similar in the sense that the question would then be - was there adequate value in this study handed off to what I earlier called the consumer of that information to really move the science forward?
And I think equipoise is about ultimately we want to move science forward not to satisfy individual curiosity. We want to move science forward to improve our ability to prevent or treat or diagnose disease. >> This is Roger Lewis, and I'd like to thank our guest today, Dr. Alex John London. More about this topic is available in the JAMA Guide to Statistics and Methods series which is published in JAMA and on our website, JAMAevidence.com. For more audio, visit us at JAMAnetworkaudio.com. You can listen and follow wherever you get your podcasts.
[ Music ]
Form Title Download
Media
Transcript
Form Title Bookmark
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Form Title Share
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Form Title Accessibility
The cadmore media player is a professional video and audio player experience designed to deliver media as well as relevant transcript, segmentation, and metadata. On the screen is a traditional html audio player with buttons such as play, pause, forward, captioning, language and play speed selection and more. Adjacent to the audio player are generally tabs, one of which shows a scrolling transcript of the audio and the other shows any segmentation of the media, sometimes referred to as chapters. At the top of the audio player is a menu bar containing an Explore button and Tools button. The Explore button allows you to dig deeper into the audio by opening up an explore dialog which contains an about tab which contains rich metadata about the media, a segments tab which contains further information about the audio, and a related tab which contains links to other media related to this media. The tools button opens a menu with options for sharing the media on social media, creating bookmarking links, creating embed codes for your website, generating citations, and more.
Keyboard Shortcuts
Keyboard shortcuts only work while you are in forms mode. By entering forms mode, you will be able to quickly access portions of the media player with single key commands.
Spacebar: Play/Pause Toggle
E: Explore Menu,
O: Tools Menu
T: Transcript Tab,
S: Segments Tab
F: Forward 15,
R: Rewind 15
L: Languages,
P: Playspeed
M: Mute Toggle,
V: Volume Bar
N: Video Only Toggle,
B: Full Screen Toggle
C: Captions Toggle
Escape: Closes dialogs and menus
Z: Next Transcript Chapter
X: Previous Transcript Chapter
Q: Next Transcript Paragraph
W: Previous Transcript Paragraph
Form Title Embed Media
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Embed Size
-
You Size - Responsive
-
We Size - Responsive
-
Picture Overlay Popup
-
400 x 225
-
512 x 288
-
400 x 700
-
768 x 1024
-
1024 x 512
-
1280 x 608
Form Title Cite
No citation provided.
Info
Segments
Related
Thumbnails
Accessibility
Using forms mode, you can utilize keyboard shortcuts to make usability more efficient.
Space bar will play and pause the media.
You can press A to open the Accessibility Dialog.
This describes many other keyboard shortcuts for media controls.
Exit Clip Mode
The video is currently set-up to show just a clip from 0:00 to video end. You have requested to see a portion of the media outside of the clip.
Click OK to switch to the full media or click if you would like to stay inside the clip. You can always go back to the clip with Tools and Clip Mode