The Evolution of Clinical Trials in Critical Care: An Interview with Dr. John Marshall
John Marshall, MD, FRCSC, FACS, FCAHS,
Trauma Surgeon and Intensivist at St. Michael’s Hospital,
Department of Surgery, University of Toronto
Member, Institute of Medical Science
By: Hira Raheel
Clinical trials in critical care differ from trials in other fields. They focus on validating the safety and efficacy of common treatments and procedures, and recruit subjects who are typically unconscious and incapable of making decisions for themselves. As a leader in critical care research both nationally and internationally, Dr. John Marshall has nearly 30 years of experience with clinical trials. It was my pleasure to speak with Dr. Marshall about his career as a scientist in the field of clinical trials in critical care.
Could you give us a brief background about yourself? How did you get to where you are today?
I completed medical school at the University of Toronto. I was torn between a career in public health and surgery. I completed a general surgery residency in 1984 from Dalhousie University, and became interested in trauma, sepsis, and acute care during that time. Following that, I spent time at McGill University doing a research fellowship, a mixture of basic research and clinical research.
What drew you to clinical trials in this area of medicine? Why do you think clinical trials are important?
During my time at McGill, it became apparent to me that we often don’t know whether we are helping or harming our patients, and that many interventions provided with the best rationale may actually be deleterious. The only way one can determine whether something helps is through clinical trials–using the principles of probability, and understanding what happens in populations of patients.
How did you become involved with clinical trials in your area of medicine?
I was fortunate that in the late 1980’s, a group of Canadian intensivists became interested in the idea of doing collaborative clinical trials, and formed the Canadian Critical Care Trials Group (CCCTG).We grew from about 10 or 12 people who came to our first meeting in 1989, to more than 300 Canadian members. We’ve published 250 papers, 18 of which are in the New England Journal of Medicine. We’ve developed a new model of clinical research, based on investigators identifying the questions that arose through their clinical experience. It was a model heavily based in trust and collegiality.
How do you think clinical trials in Critical Care differ from those in other research areas? Do you think you have developed a better model for trials?
I think we have a model that is successful and that merits close attention, but I may be doing a disservice to colleagues who are doing equally well in other fields. I do think that the model of collaborative research, where you fly in to an airport, meet in the airport hotel, and look to see whether you can catch an earlier flight is lethal to good science. Good science requires time, collegiality, and collaboration. The CCCTG meets in Lake Louise and in rural Ontario and Quebec. We expect people to spend time together socially and get to know and trust each other enough to be able to work well together.
The other key strength of our model is that before we actually do a trial, we do a series of studies: systematic reviews, observational studies, standardized questionnaires, and focus groups. We fully understand the question we are asking and the knowledge base behind it. This makes for much more compelling grant applications and also benefits young trainees who might conduct and publish a systematic review even if they do not run the clinical trial themselves.
How have clinical trials in your field grown and developed since you first became involved?
After we published our first paper, other countries looked at us and liked the model. The Australians set up the Australian and New Zealand Intensive Care Society Clinical Trials Group in 1994. Then an American group with extensive funding from the National Institutes of Health was formed to study Acute Respiratory Distress Syndrome (ARDS). Other countries followed this trials group model. We discovered that we have some common approaches and needs. We saw that when international groups worked together, they could address important questions, and publish studies in high impact journals. So, we set up the International Forum of Acute Care Trials (InFACT) as a channel for these groups to collaborate and interact with each other.
What do you hope to gain out of international collaborations in clinical trials?
We are trying to promote collaboration amongst groups, because we’re finding that doing a 5,000 patient study is challenging in a single country, and it is both more efficient and more generalizable to have multiple countries collaborate. We are keen on trying to expand this model outside of Europe, North America, and Australia. There are now critical care clinical trials groups in China, Southeast Asia (which includes Sri Lanka, Bangladesh, Pakistan, India and Nepal), sub-Saharan Africa, North Africa, Brazil, and Latin America. Through InFACT we are trying to mentor and promote clinical research in groups from less developed nations.
From single-centre studies, to multi-centre trials, and now international research collaborations, clinical trials in critical care have come a long way, so what’s next?
We are developing a model called a platform trial, currently run in Australia, New Zealand, and Europe to study severe community acquired pneumonia. We are seeking funding for this trial in Canada. A platform trial focuses on a disease, like community acquired pneumonia, and looks at multiple interventions. These interventions can be different antibiotic combinations, varied immunomodulatory strategies, and alternate approaches to mechanical ventilation. Using Bayesian adaptive approaches, one establishes the parameters for deciding whether something does or does not work. Once an intervention is shown to be better, it becomes part of the control arm of the trial moving forward. If it is not found to be of any use, then it is dropped, to be replaced by another intervention to evaluate. In essence, the trial that can continue in perpetuity, and becomes a hybrid model of research and quality improvement. Tens of thousands of scenarios are developed in advance to create the algorithm that drives the trial. But, once you have done this, the algorithm guides subsequent patient recruitment. Platform trials blur the boundaries between clinical research and clinical care.
Do you have any general advice for young investigators and grad students who want to potentially transition from doing basic science research and move towards more clinically-focused research?
For the clinician, research is no longer a luxury. It is the mechanism through which we understand what we’re doing and whether it is helping patients. It is critical to become literate in science and to engage in that science. There’s a sense of humility in science, which transforms ignorance from “I don’t know” to “We don’t know” and activates a powerful tool to generate knowledge.