Good morning. Thank you very much for this kind invitation. My name is Dr. Humberto Martinez-Cordero. I'm a hematologist at Military Hospital in Columbia and also at National Cancer Institute working in malignant hematology and bone marrow transplantations programs. The objectives of this talk speak about pragmatism from a philosophic perspective. What is counterfactually? We are going to speak about also real or evidence, classification of trials and pragmatic trials, details and characteristics of pragmatic trials and tools to evaluate them, and the behavior in the world about this kind of trial. So I want to speak first about what is pragmatism. And the pragmatism rejects the existence of absolute truths. Ideas are provisional and subject to change in light of future research. So this comes from traditional schools of philosophy, along with idealism, realism and existentialism. But for pragmatists, truth and goodness must be measured according to how successful they are. In other words, pragmatism is based on utility, which is the most important word, utility being the basis of all of meaning. So trials, as we can know, are optimized to demonstrate efficacy, sometimes overestimating benefits and underestimating harms. They are designed to prove effectiveness and efficiency taken into account the risks. So we're going to speak about what is the counterfactually and the counterfactually scenario moving to reality. So as we can see here in this slide, the shade of gray represents the exposure status and the X indicates the presence of outcome, the disease. So people with X, they have the outcome. The blue dots in this slide represent a foreign factor to the association under the study. If the foreign factor is distributed independent of the exposure intervention, we can say that we can interchangeability. So we can say that the control group is a good replacement for the counterfactual scenario. And we can see here the outcome and the exposure and the foreign factor in this slide. The blue dots represent a foreign factor to the association under the study. But in this slide, you can see the foreign factor is not independent of the exposure intervention. So we can see here the interchangeability is lost. So the contrast group is not a good substitute for the counterfactual scenario. That is why we need randomized control groups. If we have a foreign factor that is not independent of the exposure intervention, we can have systematic errors in our cities. But we need to speak about this situation in real-world evidence because in real-world evidence, we are not to control all the time this kind of exposures. So in real-world evidence, we have RCTs, randomized control trials, probably the different kind of studies from real-world evidence. The real-world evidence, we can have data coming from observational basis obtained generated during the routine clinical practice. And this data can come from electronic health records, claims and billing activities, product and disease registries, patient-generated data, including in-home use, and data gathered from other sources that can inform on health status, such as mobile devices. But on the other hand, as we can say before, RCTs are different kind of studies, and when we can control everything inside of the trial. What are the design of real-world evidence? Our simple trials, the pragmatic trials that we are going to speak now, observational studies, prospective and retrospective, but they are not limited to them. We can have another other kind of studies being part of real-world evidence. So why is it important to have real-world evidence? To monitor post-market safety and adversity events and make regulatory decisions. They are frequently called phase four studies. FDA and EMA make decisions about real-world evidence in this specific regard. The healthcare community is using this data to support coverage decisions. NICE, for instance, they are using real-world evidence to approve some drugs, and they are important to develop guidelines and decision support tools for using clinical practice, as NICE did in the past, or as NICE or another regulatory agencies are doing now. And the medical product developers are using real-world data and evidence to support clinical trials, for instance, to design large simpler trials or pragmatic trials and observational studies to generate innovative new treatment approach. So these kind of studies are a very attractive option for countries like countries being part of Latin America. What is the course classification of trials? We have, in one hand, explanatory trials, and the other hand, we have pragmatic trials. The explanatory trials confirm physiological or clinical hypothesis. So this is the most important thing to have in mind. And pragmatic trials, the most important thing is they inform a clinical or policy decision by providing evidence for adoption of the intervention in real evidence. So we are trying to have information about they are working in our clinical practice. So the most important thing is no study is perfect being part of pragmatic or explanatory. They frequently are mixed of them. So the explanatory trials, the main characteristics of this kind of trials, they have high internal validity because they have a very well-controlled scenario, a smaller sample size, sophisticated design, controlled environment, and mostly they are phase two or three. So the most important thing here is RCTs are frequently explanatory trials. So in the other hand, we have pragmatic trials. They have highest external validity because they have frequently designs in real-world practice. They have frequently large sample size. The design is simple. We can have this kind of trials in diverse settings. And mostly, as we can say, as we said before, they are mostly phase four trials. So no one is exclusive. Sometimes we have this kind of trials being part of both. So to illustrate a very pragmatic trials with this tool, one, we are going to speak about these characteristics of pragmatic trials. One, there are not inclusion or exclusion criteria. We are going to speak about one example. Two, practitioners are not constricted by guidance on how to apply the experimental treatment. So they are very flexible. The experimental intervention is applied by all practitioners, thus covering the full spectrum of clinical settings. We can have, for instance, this kind of trials in emergency department or in wars. The best alternative treatments are used for comparison with no restrictions. So sometimes we are comparing, for instance, this intervention with another real-world practice treatment. The comparative treatment is applied by all practitioners. As we said before, not formal follow-up sections. Sometimes the outcome is collected by the clinical history. The primary outcome is a clinical meaningful one that does not require extensive training to assess. It is important to mention that it's not necessary to have good or complete training in making or designing trials. So all practitioners can do this. There are no plans to improve or alter compliance for the experimental or comparative treatment. So we can compare our treatment with the best clinical practice. Not a special strategy to motivate practitioners' adherence to the trial protocol, which is frequently done in our cities. And analysis includes all participants in an intention to treat patients. It is important to mention this. This is the PRECIS II tool. We can say that it can be divided in recruitment of investigators and participants. This is a very simple thing. The intervention is delivering within the trial. The nature of follow-up and the nature of determination and analysis of outcomes. We can use this to assess if one kind of trial is pragmatic or explanatory trial. For instance, this trial, the CRASH trial. The CRASH trial was developed to give corticosteroid treatment for participants with significant head injury. So this was an individually randomized placebo-controlled trials. But this kind of trial was developed in an emergency department scenario. So no consent was required for this specific situation. And this could explain why they have local variations. So, but the most important thing that explain why this is a pragmatic trial is because you can randomize patients. It doesn't matter if they are, you know, allocated by computer or something like that. So this is a very, very simple design. The pragmatic outcome is the very pragmatic outcome is all cause of mortality. You know that mortality is a very, or it's a very important outcome for us. And the most, one of the most important things is more than 10,000 participants in 239 hospitals were included in 49 countries. So we can find a very, very important population to study. So this trial demonstrated that compared with placebo, the risk of death from all causes within two weeks was higher in the group allocated to corticosteroids. And the relative increase in deaths due to corticosteroids did not differ by injury severity or time since injury. You can think about this design is one of the most important trials designed to prove that corticosteroids are not beneficial for patients with head injury. So corticosteroids are not used for patients anymore with this kind of evidence. And it's not possible to have another better evidence, another trial to prove this. So this is the best tool to do that. And to speak about challenge to design this kind of trials, we need to speak about consent. As you can see in the crash trial, the concept was not necessary because it was performing emergency department so sometimes the consent is not necessary. But you need to think about the consent if you are thinking about design one study in your field, for instance, in this specific regard in gynecology. Investigators or investigators outside academic centers, we can do that. As I mentioned before, it's not necessary to have a full training in trials to develop this kind of trials. It's not necessary to have experienced investigators, but it is important to have one of two coordinating this kind of trials. The heterogeneity is important to be reckoned with because sometimes we need to adjust some characteristics of the population. Sometimes we need to give incentives to investigators, but it's not necessary. Sometimes we need self-reported outcomes with tools, for instance, patients reporting neuropathy of treatments or diarrhea or some specific adverse reaction to treatments. And as you can see here in this slide, only 111 trials are registered as pragmatic trials in clinicaltrials.com. So speaking about intervention, blinded interventions are not fully pragmatic. We can have non-drug interventions as scales, evaluating scales, or you can have this kind of tools evaluating adverse reactions or different things from drug interventions. It is necessary that the intervention, if done, should be delivered as a normal means. You need to have evidence of what you are doing all the time in your clinical practice. It is important to have adherence, delivery of protocol and safety. You need to monitor this. And training in the intervention, even with constraints of the environment where the trial is conducted. It's important to have a training of what we are going to do. Follow-up. This is, in my concept, the best thing of pragmatic trials. Because you can have an abstract collection of the data. Clinical records. We are doing all the time clinical records of what's happening with every single patient. So we can have the data from this clinical history. This is an opportunity to unify clinical history. For instance, to have the diagnosis, the date of diagnosis, the cardinal symptoms at the start, when the patient started the treatment, other interventions. Registered with the patient's phenotypes. It is important to registry and follow up this. And the most important thing in this kind of trial is low-cost opportunities. Because you are giving this kind of treatments that you are giving for those patients in normal clinical practice. Outcomes. Sometimes in RCT, the outcomes are important for pharma. And this pragmatic trials, the outcomes are important to patients. And sometimes the outcomes are very hard outcomes, like mortality, for instance, or relapsed in cancer matter. Safety endpoints and established interventions you can have as an outcome. Lack of effectiveness of intervention. We can have simplistic and minimized trial procedures. They are easy to collect data because, for instance, mortality is an easy way to have the one kind of outcomes. Right? And we can register long-term safety data as well. We can see here in this slide how the pragmatic trials are increasing. So this figure shows articles per year cataloged in Medline that have the word pragmatic or naturalistic trials. So you can see here how important they are in the last 30 years. You can see here in red the countries that have more pragmatic trials. So the States, part of Canada, Europe, they have an important fraction of pragmatic trials in their trials. So to finish, this is our take-home message to understand what is pragmatism. And pragmatism is important because the utility, the counterfactual concept, it is important because in one hand we have RCTs with a very controlled counterfactual scenario. It's not necessary to have this kind of counterfactually controlled in pragmatic trials because it's a situation that we are working all every single day to understand what is explanatory versus pragmatic trials. Explanatory, sometimes we are making decisions in clinics only with explanatory trials, but it's important to take into account pragmatic trials. And we need to have our own data to have decisions, to make decisions. Real-world evidence, we can say, now what is real-world evidence? Post-marketing is the best moment to perform it. So in the moment when the drug is approved, we need to perform or design post-marketing trials to know how is the behavior of this drug in the clinical practice. What is the impact in public health? Cheaper intervention could be more pragmatic than explanatory. It is important, more resources are necessary to perform explanatory trial. So our pragmatic trials are cheaper than explanatory trial. So one important thing is while RCTs can have an important internal validity, in our pragmatic trials, we can have an important external validity. And no laissez-faire approach, high quality trials are required. So when we are talking about is not necessarily counterfactual scenario, and that doesn't mean that we need to be simplistic as such in designing pragmatic trials. We need to have important and very well-designed pragmatic trials as well. So with this, I am finishing my talk. I am up for questions. This is my institution now. You can see here the military hospital in Colombia. And this is the place when I had my training in lymphoma and myeloma field. Thank you very much for listening this talk.

pragmatism

counterfactual scenario

real-world evidence

explanatory trials

pragmatic trials

high-quality trials