Is it ethical for doctors to ask their own patients for research consent?
It’s not uncommon for Saoirse Cameron to go back to work in the middle of the night to speak to a family about a study she’s recruiting for. Cameron is the research coordinator for the pediatric critical care unit in the Children’s Hospital at London Health Sciences Centre, where one trial currently underway requires that patients be signed up just before they receive a blood transfusion. (The study is looking at whether the age of blood received in a transfusion affects clinical outcomes. As a rule, the oldest blood in a bank is provided first; this study randomizes some patients to receive new blood.) Cameron needs to sit down with parents to explain the research—the risks, benefits, the ethics—as well as the multi-page consent form they have to sign. This can take up to half an hour, and often coincides with one of the scariest moments in a parent’s life, a moment when “any information is too much,” says Cameron. Many families can’t make a decision at that moment, though sometimes they come back the next day and say they’d like to participate. At that point, the child is no longer eligible to participate.
So when Cameron attended the Clinical Trials Ontario conference this past March and learned about an alternate way of obtaining consent, she wondered if it might be a better option for some of her team’s studies. The integrated consent model is being used by researchers at the Ottawa Hospital Research Institute in trials comparing one standard-of-care treatment to another standard-of-care treatment; that is, two treatments commonly used to address a specific ailment—different anti-nausea agents, for example—between which there is no consensus among clinicians as to which of the two is best. Such trials are considered to be of minimal risk to participants, compared, for example, to studies testing a new treatment.
The integrated consent model allows clinicians to obtain verbal consent from their own patients, and thus differs significantly from the more conventional informed consent model, which requires that someone other than a patient’s care provider obtain consent, and also that the patient provide written documentation of their agreement to participate.
Both of these aspects of informed consent have long been central to the approval of studies by research ethics boards. So how do these differences affect the ethics of the new model? Is it in fact ethical? And what do patients who’ve been recruited with the new model have to say?
What is purpose of the integrated consent model?
The model was developed largely in an effort to simplify what some consider an unnecessarily cumbersome process that often impedes enrollment into clinical trials. With traditional informed consent, though a patient’s own doctor or nurse is usually the one to broach the idea of participating, someone other than the clinician—usually a research coordinator or assistant from the study—must do the actual enrolling. “And they’re not always handy,” says Dean Fergusson, epidemiologist at OHRI and lead scientist for its Rethinking Clinical Trials (REaCT) program, which is currently using informed consent in some 11 studies. “They’re off in different parts of the hospital, or they’re busy, or it’s after hours, or they’re on vacation, and that patient may never even meet up with the coordinator. That happens a fair amount.”
In many studies, patients are given at least 24 hours to consider the request; most never follow up, says Fergusson. Patients are frequently asked to come in for further baseline testing—a pregnancy test, for example—and these can be a challenge to schedule. Fergusson describes enrollment in cancer trials as “dismal”—less than five percent across Canada, and less than three percent for breast cancer trials. This is one of the big “knocks” against clinical trials, says Fergusson—a lack of generalizability. “If you find results in the five percent who participated, can you generalize them to the other 95 percent?”
Enrollment in REaCT trials can happen at the initial step, when the clinician first speaks to the patient. Care providers are expected to read a script explaining the purpose of the study, the process of randomization, and the fact that the patient can withdraw at any time. The patient is given an information sheet covering much of the same ground which they can take home. They can also sign up right there and then, in which case the care provider documents the patient’s consent in their clinical notes, signs off electronically in the REaCT app, and randomizes the patient to one of the two treatments—all in one visit.
The REaCT trials have been running for about three years and have enrolled about 1,700 patients—well over 80 percent of those approached. In part, this difference reflects the fact that the REaCT trials are low-risk, whereas most cancer trials are of novel therapies (for which the informed consent process typically has several more steps). Fergusson estimates that the integrated model allows a five-fold reduction in costs, due to the ease of data collection and management, and to the efficiency of enrollment.
Is the integrated consent model ethical?
When Ray Saginur, chair of the Ottawa Health Science Network Research Ethics Board, which is responsible for vetting all research at the Ottawa Hospital, first considered the integrated consent model, the hardest thing to wrap his head around was the absence of a signature. “We’re very used to having things written,” he says. “In part because we know what was said, and [in part because] we know that that the person was there.” At the same time, he welcomed the “discomfiture” caused by the departure from convention. “Because I have concerns about our conventional practice,” he says.
Saginur explains that consent documents have become increasing long, complex and jargon-ridden. “They start looking inadvertently more like contracts,” he says, speculating that they’re primarily designed to try and protect the sponsors of research from liability. “Oftentimes people who are asked to sign these forms either can’t or won’t read a whole document. They just sign them. So the consent process may be subverted.” The integrated model’s script, the information sheet provided to participants, and the fact that participants can withdraw at any time (which is true about any trial), sufficiently mitigate Saginur’s concern around the absence of written consent.
But these aspects of the model may not be enough to assuage some ethics boards when it comes to consent being obtained directly by clinicians. “How do we ensure that there’s no coercion involved?” says Sharon Freitag, director of research ethics at St. Michael’s Hospital in Toronto. “How do we ensure that the patient or participant is really understanding and not just agreeing because of the trust they put in the relationship with their physician or treating surgeon?” Freitag also wonders if some people might feel unable to decline. “If this is the person who’s going to be operating on me, who’s going to be treating me going forward, how do I say no?”
Charles Weijer, a research ethicist at Western University who focuses on randomized control trials, does not see this issue as particularly problematic in low-risk trials. “When research is comparing two interventions that are routinely used in clinical practice, when research participation involves minimal risk, when data is collected from routine health data sources, I think in those circumstances and possibly others, it seems reasonable for the patient to be approached by her own physician,” he says.
Weijer is more concerned about trials where no consent is obtained at all. An example would be the Padit trial, a randomized cluster crossover trial in which patients who came through the doors of certain institutions received one prophylactic antibiotic and patients who came through the doors of other institutions received a second prophylactic antibiotic. Researchers requested a waiver of consent on the grounds that the study could be very inclusive—and thus very generalizable—but would be prohibitively expensive if it required staff to gather informed consent from each potential participant.
“I understand the motivations of researchers to want to do these important studies,” says Weijer. “But at the same time, when patients are receiving a drug in a trial that might have been different from the one they would have been prescribed in practice, I think it’s really important that their consent be obtained.” He thinks alternative models such as the integrated consent model might be used in trials like these, “as a way of doing the study in a pragmatic way and obtaining the consent of research participants.”
What are patients’ experiences with the integrated consent model?
While undergoing treatment for breast cancer in 2017, Rebecca Hollingsworth participated in three separate REaCT trials. Her oncologist, Mark Clemons, who was the lead investigator in each, told her about them during an appointment to discuss what kind of chemotherapy she would be receiving. One of the studies compared different ways of administering chemotherapy (Portacath vs PICC line); another looked at whether more frequent injections of a drug to boost immunity during chemo were necessary or not. Hollingsworth says she weighed her decision each time. “The ones he proposed for me all seemed in the name of research, and I didn’t think any one of them was necessarily going to harm me,” she says. “I was very excited about it. I have so much respect for Dr. Clemons. He’s just trying to make patient care better and better and better and better.”
Soon, Saoirse Cameron may not have to get up as much in the middle of the night. Her team is hoping to pilot the integrated consent model in some of the studies on the unit, which would mean a patient’s doctor or nurse could obtain consent for research at the same time that they’re obtaining consent for a procedure. “It’s one process, one person, a lot less burden to the families,” says Cameron. She agrees it is possible parents would be more trusting of their child’s doctor or nurse than a research coordinator they’ve never met before, and that this may result in higher enrollment. “And that’s a double-edged sword from an ethics point of view,” she says. “For sure, it’s great if we’re able to get more people into studies, but ultimately we think it’s just better practice. We think it’s more ethical to do integrated consent than informed consent in some studies.”