Seeing Every Patient Encounter as a Research Opportunity

Robert A Harrington, MD; Rodney H Stables, DM, FRCP


August 24, 2015

This feature requires the newest version of Flash. You can download it here.

Robert A Harrington, MD: Hi, this is Bob Harrington from Stanford University here on Medscape Cardiology and Thanks for joining us today. One of the topics that's interested me greatly over the course of the past few years has been how to do a better job of integrating clinical research into clinical practice. Around the globe in cardiology we are very interested in both developing evidence that forms the basis of practice guidelines as well as collecting evidence within the context of clinical practice to better inform the guidelines. I have talked several times on this podcast with folks from Sweden, folks from the US, regarding some recent Patient-Centered Outcomes Research Institute (PCORI) projects. I often think about how we can create truly continuously learning systems, or as the American Institute of Medicine has called it, the Learning Health Care System.

One of the topics that has grabbed my attention and the broader interventional community's attention has been a fabulous trial done out of the UK called HEAT-PPCI,[1] which is a great example of how we can learn continuously within a healthcare system and utilize that not just for the local environment but to provide generalizable information for the broader community. There has been the course of some discussion around HEAT-PPCI, and there were questions raised about the single-center nature of the trial and some of the issues of consent. But during the course of all those discussions, we missed an opportunity to have the broader discussions of how this was a very good example of how to better integrate research with clinical practice.

I am delighted to have as our guest Dr Rod Stables. Rod is an interventional cardiologist at the Liverpool Heart and Chest Hospital and the director of the Cardiac Catheterization Laboratory who was the overall principal investigator of HEAT-PPCI and who has given a great deal of thought into how one thinks of the integration of research with practice. So Rod, thanks for joining us here today on Medscape Cardiology

Dr Rod Stables: My pleasure, thank you.

Dr Harrington: So Rod, let's dive right into it and talk about what interests you these days in terms of integrating clinical research with clinical practice. What I mean by that is, as a busy practitioner, why do you see this as such an important topic?

Dr Stables: Well, I think if we look at the world literature, there is a quite an established body of opinion (some might say almost an accepted wisdom) that institutions and physicians who participate in research benefit enormously through the experience, as do the patients.[2] There is good evidence that institutions that are research active will be more aware of changes in practice, more responsive to guidelines, and provide better evidence-based medicine.[3] The experience of the way that they work in this respect will stimulate, educate, and develop the staff they will recruit and maintain higher-quality staff. Their patients may receive more systematic care, more precise care, and greater attention to detail. It is possible, of course, that patients may receive more personalized, direct attention. Even in studies where patients are receiving placebo or double placebo, there is a tendency that those individuals may experience a better outcome than the institutional norm. Now, I think we'd like to feel that we could deliver that to all of our patients and all of our staff on a more or less daily basis and at the same time make a substantial contribution to the development of the evidence base. That's got to be a noble vision.

Dr Harrington: That is as articulate a view of a learning healthcare system as I've heard. Which is that we are learning continuously and that continuous learning benefits our patients first, benefits ourselves as practitioners, and benefits the broader community.

Dr Stables: Locally, we have to commit ourselves to develop the means locally, the institutional facilities that will allow us to move toward a vision where we hope all of our patients, every healthcare encounter, in some way will be making a contribution toward well-planned, well-structured research.

Dr Harrington: My former colleague at Duke, Chris Granger, who directs and has directed the Duke Cardiac Care Unit for many, many years, liked to say the same thing, Rod, which is that every patient encounter ought to be an opportunity to engage in research, because they will get better care and we'll learn something, and I think you're both right. Rod, let's talk about HEAT-PPCI first. Not so much from the bivalirudin [vs unfractionated heparin] issue. You and I talked a little bit offline that that generated a lot of furor. Let's focus on the structure and the opportunity. I find this one of the most remarkable examples of the integration of research with practice that I've seen in a number of years. Talk a little bit about how you structured the trial, why you structured it that way, and give the audience some sense of the commitment of your practitioners, which I find extraordinary.

Dr Stables: Most of your listeners will be aware that the vast majority of randomized control trials that guide our practice tend to be conducted in relatively few specialized centers and tend to recruit a highly selected population that may not necessarily be representative of the true clinical everyday norms. I wanted to have a trial (which may actually be the first trial in cardiovascular medicine if not the first trial in any branch of medicine) that would recruit every single eligible patient without fail.

We were working in the setting of primary PCI, which is a pressurized, time-critical environment. A service that operates 24 hours a day, 7 days a week, 365 days a year, and in my institution we are activated to receive a primary PCI about 1400 times every year. I collaborate with about 11 other colleagues who share the rota with me and of course a wide variety of other professional disciplines—nursing, radiographic, and support staff. We set ourselves the demanding ambition that every single patient who entered the cath lab in the context of primary-PCI activation would be randomized. People were interested in the general question but, for example, I had two colleagues who favored low-molecular-weight heparin, which wasn't even one of the substances under evaluation in the trial, but their attitude was that the trial is important, it's important to be cohesive: I won't use my normal drug because I am going to follow protocol. I was incredibly moved and proud of the effort that saw 22 months of continuous activity and nobody missed a single patient. Two in the morning, three in a row, Sunday afternoon, every patient was randomized. I think that is tremendous commitment.

Dr Harrington: I hope our listeners really focused on that. Twenty-two months, 24/7, every single patient who was eligible was enrolled into the study. For someone like me who has done clinical trials for 25+ years now, that is an extraordinary accomplishment. I did not know that piece about the low-molecular-weight heparin. That did require dedication. So talk to me, you're the service line leader there with the cardiac cath lab. One of the comments that frequently comes up is 'Well, I'll enroll my patients, but I'm not sure that my colleagues will enroll theirs.' How did you get community buy-in so that your colleagues were all willing to participate in the randomized trial? Clearly as a community, meaning, at Liverpool, you as a group had equipoise that it was ethical to do the study at Liverpool. What was it about Liverpool, the setting you've created, the environment, that your colleagues were all willing to say we're in 24/7 for the next couple of years?

Dr Stables: I am very fortunate in that I work in a wonderful hospital with great colleagues and we are incredibly cohesive and team-based in everything we do. We make joint decisions about which stock we will hold and which stock we won't hold, which trials we will support and which trials we won't support. In the military the term is called concentration of force. If an institution like ours turns its support toward one endeavor and does it in a wholehearted cohesive manner, what your EP colleagues would call synchronous, then you're going to get a far greater effect. We can bring much more force to bear if we collaborate and cooperate in this way, and that's the key.

We've got other studies to demonstrate it as well. Recently we were looking at a prospective registry study on issues of hypersensitivity in allergic reaction with hydrophilic radial sheaths, and we completed a 400-patient prospective study in 50 days. Once a unit mobilizes behind an endeavor (we do about 2900 PCIs a year), we are in a position to mobilize a big force when we work together.

Dr Harrington: Again, you are making it sound easier than we both know it is, but that is an extraordinary ability to get people to work together. Talk a little bit about, not your physician colleagues, but your radiology colleagues, your nursing colleagues, your support-staff colleagues. Same issue here—that people feel part of the whole, so they are willing to collaborate like this?

Dr Stables: We worked out from the software that more than 150 different people actually performed a randomization action. The case-record forms were completed in real time by frontline clinical staff. Things that at the time perhaps seemed mundane or unimportant, eventually, as we got into the furor after HEAT-PPCI, became incredibly important. When we were talking about the dosing anticoagulant agents and their effect, who would have thought that the body weight of the patient would have been so important, but for a weight-adjusted drug it is critical. And in the primary-PCI setting frequently you don't know the weight of the patient, so we recorded the estimated weight as you'd use in a lab, and 24 hours later we weighed the patients and we were able to shed light on the accuracy and the spread. Every aspect of the hospital collaborated—I mean, people in clinical coding (we used audit clerks to check clinical coding) picked up on the one minor bleeding event that we missed clinically in a patient that progressed to bypass grafting. It was a tremendous hospitalwide effort. It was a positive and unifying force.

Dr Harrington: I know that you're intending (because we talked offline about this) to publish more. Some of these methodologic issues will be incredibly important to share with the rest of the community so that we can all learn from the lessons at Liverpool. Let me talk about another constituency that there has also been much discussion around the trial and that is the patients. I'm on record as having been very much a supporter of the consent methodology that you used in the trial. I think it was innovative, it was completely within the realm of how we think about the ethics of clinical trials, but not all of the community agree with you and your approach. Do you want to talk a little bit about the consent process? Why you chose to do it that way, and what have you learned from that, now having the benefit of having done the study and talking with patients about their experience? What have you learned, and would you do it the same way?

Dr Stables: Yes, I would definitely do it the same way and I am proud and consistent and solid in my belief that we behaved sensibly and appropriately. But, of course, there are some caveats that we need to bear in mind. The first one is every research endeavor is clearly different and has different nuances as it affects the patients. Now in our cath lab before HEAT-PPCI initiated, both bivalirudin and heparin were available and were in regular use on the basis of physician preference and so on. A patient entering the cath lab in the context of a primary PCI could very well have either heparin or bivalirudin, and they would never have been asked or consulted, it would have just been administered on the feelings of the physician at the time. Now the trial randomized the patients between these two accepted guideline-endorsed and well-established products. As such there was nothing novel or experimental, and they were not being subjected to anything out of the normal.

Ideally the following day, but when patients were at ease and fully recovered, the trial was discussed with them. They were informed of what had happened and we sought their permission then to continue to use the data we had accumulated to date and to remain in the touch with them for the purposes of clinical follow-up. Although it is obviously a post hoc observation, it is clearly to me very positive that only three patients out of 1827 chose not to continue with the trial. Interestingly, although we didn't in any way ask the patients to explain or justify their choice, these three individuals chose to do so. They had issues unrelated to ethics and unrelated to the concept of the trial but more to do with aspects of their personal life. There were no major ethical issues there. I think that the aspect of the nature of the trial is important in accepting this kind of way forward.

The other key issue is of course this was in the context of primary PCI, and again, as you can see from the published paper, we are performance managed on the time interval from ambulance hand brake to artery open. The median was 29 minutes. There isn't an enormous amount of time for meaningful informed discussion even if the patient were at rest and compos mentis, but for a patient in pain, distress, under the influence of opiates, to discuss the ins and the outs of a randomized trial in a handful of seconds, it's not what constitutes informed consent. We don't even ask the patients to provide written consent for the primary PCI itself.

Dr Harrington: These are terrific issues that I fully agree with. You have nicely laid out what the caveats are around the type of research that this kind of design is most applicable toward, and I fully agree with the way you approached this. Although, as I have said, there are people in the community that have not agreed.

Dr Stables: Just in the interest of balance, whenever you choose a line you have to accept its limitations. I am conscious of one important limitation of the approach we chose. I think it's important we share that so that your listeners can hear both sides of the story. The counterargument that appeals to me is that while I think most reasonable observers would accept that it would be impossible for a patient to provide truly informed reasonably meaningful consent in these circumstances, I can accept that the converse may be true. Let's imagine a patient has a preformed view that they would never, ever, ever under any circumstances wish to participate in a research endeavor, in the same way that they may choose to never donate a kidney or they may hold similar strong views of a variety of issues. You could imagine a scenario that a patient, even in a primary-PCI setting, given the chance could express this preformed intention not to participate. The HEAT-PPCI approach denied those patients that opportunity. This a limitation, and we have to weigh whether or not it's such a significant limitation that it overtakes the potential advantages.

Dr Harrington: I accept that as well and I have heard people articulate that. It brings up the notion of distributive justice as an ethical principle. Oftentimes in clinical trials I think that the very selective nature of how patients enter the trials in some ways goes against the principle of distributive justice. That we are being too highly selective, and one of the things that appeals to me about the HEAT-PPCI design is that you're very, very much adhering to the principle of justice because you are including everybody, but I fully accept, as you say, that there may be people who would not want to participate. Did any of that come out, Rod, as you came back to patients a day later? Did you get people (I realize this is not a quantitative exercise but perhaps a qualitative exercise) who may have expressed discomfort, and did you have a mechanism by which you could capture that information a day later?

Dr Stables: No, we didn't really. Again the consents were captured by a team of initially three, in the height of the trial, five individuals who were trained for this purpose. We didn't database anything more than the outcome. I have already hinted the outcomes were very positive. So I don't think that it was a major problem, but again there is another side here about the nature of the trial and the nature of the setting and so on. It could be very different if we tried to extend this too much.

Dr Harrington: Rod, one of the things we talked about offline before we went live in recording was a study that you've been doing looking at the mechanism or the method of follow-up. I find this intriguing as somebody who has designed a lot of trials and particularly focused on end-point ascertainment, adjudication, etc. You've had some unique experiences in this trial to look at different methods of follow-up. You want to comment on that?

Dr Stables: Yep, certainly. Anyone who's run a trial, whether a local or multinational endeavor, will know that a disproportionate amount of resources, both fiscal and human, are expended on trying to obtain the follow-up. If it were possible to obtain similar quality with reduced effort or reduced cost this would make research much more acceptable. We were interested in exploring these issues.

So in HEAT-PPCI, we performed what might be regarded as traditional gold standard follow-up with individual personal contact with every patient during the follow-up duration. But at the same time we also downloaded the information that was available through UK routine computerized health records about hospital admissions. I'm just going to ditch the question of adjudication and clinical end-point committee because I actually do believe in that, I am very positive about that. Before you get to that stage the investigators have to know that the patient has experienced some kind of issue or event and usually the trigger is hospitalization. We start off by asking the patients have you visited and spent a night in the hospital since your discharge on such and such a date. Now what we found was that in the end the results that we published in the main paper in the Lancet were actually the product of the synthesis of both methods, because we rapidly realized that the so-called gold standard was wrong. The computerized method had errors but they were different errors. For example, a patient would look you in the eye and swear that they had never been hospitalized in the follow-up period. We would later find that the computerized records said that they had been and when we went to the institution and sought that medical record, indeed the patient had been hospitalized and an event had occurred. Traditional follow-up would have underestimated the events, and of course there were computerized records that were suggesting a medical diagnosis that turned out to be wrong and was adjudicated as such later. The magnitude of the error was approximately equal in both limbs, and so our broad simple conclusion was that the headline conclusions of a trial would probably look very similar if you simply chose the computerized method over traditional method. I think that this is a really interesting zone to be moving into.

Dr Harrington: You know a lot of us in the US are spending a lot of energy working with PCORI trying to understand how to best utilize electronic health records to do clinical studies, in particular randomized clinical trials. These same issues come up over and over, and I think as we all share information with each other on the methods it's going to be enormously valuable for all of us. Rod, the final word I'd like you to comment on is the training issues, the training of research fellows, the training of interventional cardiologists during the course of the trials. You actually probably had a couple of classes of fellows come through during the course of the trial. What do you see the role of trials like this for the trainee to be able to learn from, to be part of, and what they can take away?

Dr Stables: The general learning experience in participating in research is fantastic. It's not enough simply to do the mechanics on the floor every day, you need to back it up with education seminars and understanding. This brings us back to where we started really. The buy-in of an institution hinges on belief. The belief is based on the fact that you believe the trial is worthy, noble, and ethical, but also you believe that it's methodologically sound and that you're engaged with what you are doing and why you are doing it. We provided continuous feedback seminars, newsletters, posters. The posters were kind of location specific, so that the posters in the ambulance-receipt area were about eligibility, and the posters in the cath lab were about dosing and blood testing and [activated clotting times] ACT, and the posters in the ward were about follow-up observations and schedules and blood-test schedules. This creates the kind of learning environment I think we are both looking for.

Dr Harrington: Rod, that is a terrific description of how to engage trainees in research, and I totally agree with you there is both a didactic and experiential piece of training in research that is absolutely critical. It sort of brings us back to where we started, which is the real issue that you've highlighted in your work, which is how to integrate research into practice, and so I think we've had a terrific discussion. I want to thank you. My guest has been Rod Stables, who is an interventional cardiologist in Liverpool and a director of the cardiac cath lab there and really has been doing some very innovative work around this notion of integration of research in practice. Rod, thank you for joining us here today on Medscape Cardiology.

Dr Stables: Thank you, it has been a pleasure.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: