Equipoise is about balancing risk and benefit. So in a clinical trial where we’re testing treatments, we need to know the treatments that go into the trial need to be no better or no worse than the other treatment, because obviously if a doctor is prescribing treatment, if he knows that one particular treatment is better than the other, he has a duty to prescribe that treatment. So in a trial, it has to be seen to be equal and fair. So in any trial, there might be two or more arms, and one of those might be a control or a placebo.
So equipoise is the principle that there’s genuine uncertainty between any of those arms, that any of those arms are neither better nor worse than the others. So in the course of my work here in the Marie Curie Research Centre at Cardiff University, we’ve embedded interviews with more than 200 clinical trial participants about different things, but an emerging theme, spontaneously, is often around concepts of equipoise and issues relating to equipoise. And I’m going to give a few examples here. One of the issues with equipoise is something called therapeutic misconception. This can happen because patients, participants, or even doctors don’t believe that there is equipoise. That they have a feeling that one of the treatments is better than the others.
So for instance, in our work, when we’ve interviewed participants, sometimes the patients have been very disappointed when they’ve been randomised to what they perceive is the non-interventional arm of the trial, not the treatment arm, so they become very disappointed. And certainly in the context of advanced cancer patients or end-of-life patients, this can actually be quite harmful to participants and cause them immense distress. We also need to make sure that there’s fair comparisons. So sometimes even if it’s a clinical trial of drug efficacy and the treatment, the trial outcomes show that the drug is– each side of the drug– is effective as each other, sometimes there’s a difference in the way the drug is delivered.
So one of trials that we looked at was looking at a trial of bisphosphonates for breast cancer, which help with bone pain. But for one arm of the trial, the treatment was delivered by IV at clinic and for the other arm of the trial, the treatment was delivered by tablet. So the participants could take the tablets in their own home. So although the drug proved to be non-inferior– The drugs were the same. The drug effect was the same– there were differences in how the treatment was received by patients, because of their idiosyncratic preferences. Some of the patients prefer to be treated at home, and others prefer the safety and support of the clinic environment.
So we need to take that into account as well. Another issue is new technology and preferences towards treatment. So particularly in trials of surgical interventions, there may be an untested surgical procedure. New technology comes again, often it’s something jazzy like robotic treatment, robotic arm that’s doing surgery, but they can’t say that it’s better or worse than the others, because it hasn’t been tested. So therefore you have a situation where you’re testing a standard treatment against something that’s perceived as high-tech, novel, and there is an innate preference for that model of treatment.
So here we find that these trials are very difficult to recruit participants in to, because of the participants, because they are intuitively drawn to the new technology, but also because, as one of the researchers here has coined it, there are subliminal messages given at sites that are trying to recruit patients, because the doctors don’t believe in this clinical equipoise either. They think that the new technology is better. Another issue towards clinical equipoise is risk perceptions. So equipoise means that the general medical community has neither preference towards one arm or another of a trial. It’s the general medical community. The lead researcher may have a preference, but that’s OK. What counts is the general medical community.
So we ran into problems with a trial that we were looking at, which was a drug to prevent blood clots in cancer. And this drug was untested beyond six months of treatment, so there was no licence to prescribe after six months. So it needed to be tested. So what we found was that the clinicians themselves, different groups of clinicians, had different ideas, different perceptions of risk. So the oncologists were not keen to recruit to the study, in case the patients or participants were randomised to the wrong arm, because they were worried about clots recurring. However, the haematologists, on the other hand, were worried to randomise patients to the trial, because they were worried about risk of bleed.
And each of these things, there was an equipoise, but their personal equipoise and clinical experience meant that the trial failed to recruit.