
Attaching the term “therapeutic” to research may create a “therapeutic misconception,” in which participants construe the protocol design as therapy directed at the individual rather than as research designed to generate generalizable knowledge”
“Principles of Biomedical Ethics” Seventh Edition 2013- Tom L Beauchamp and James F Childress
The ethical principle of equipoise is particularly important in randomised controlled clinical trials. Research is carried out because there is uncertainty about a treatment, or procedure, or to find something out; to convey otherwise is not just professionally but morally wrong. The term relates to our ability for “equal balance” in our opinions and in RCTs it is used to convey the sense that we don’t know (or believe we know) which arm is therapeutically superior at this point in time.
This doesn’t mean of course that clinicians don’t have hunches and biases. Equipoise is a state we put ourselves into. This has got much easier for me over the years. I have no problems these days with putting myself into equipoise as I have learned that we really don’t know the answers until we have put the hypothesis to the test and there have been many “surprise” results in my 13 years as a research nurse.
The research design process undergoes rigorous ethical and scientific testing, and this includes bias checks.

In his blog post on the HRA Website , Jonathan Montgomery states: “Carefully designed research reduces bias, checks whether apparently exciting results are actually just luck, and helps us to understand why things have worked in the past, so that we can reliably predict when they might work in the future.”
Sometimes I do come across colleagues who may struggle with the idea of equipoise in clinical research. This is quite natural. We are used to evidence based practice and the idea that a patient may enter a research study which may not benefit them at all is alien to all we have ever been taught.
The therapeutic misconception arises when clinician or patient erroneously believes that an individual will benefit therapeutically from their participation in a study. This can particularly occur in RCTs if they believe that one arm is superior to another.
This is different to accepting other potential benefits of participating in research – an example perhaps being increased surveillance through follow-up which patients might find reassuring even though they are not therapeutic activities in themselves. It is also valid for a patient to have a sense of general hope linked to participation in a study and we should not dismiss this. Hope, in itself, is not evidence of a therapeutic misconception.
So, why is the therapeutic misconception a problem? Well, in essence, it can bias the informed consent process right from the start because the foundation on which is is being carried out is unsound. Jennifer McCormick, in her article “How Should a Research Ethicist Combat False Beliefs and Therapeutic Misconception Risk in Biomedical Research?” suggests that:
Reiterating that research goals are based on research questions—not the participant’s condition—and that benefits to her are not expected can help minimize therapeutic misconception and the false beliefs it can generate.”
Dr Jennifer McCormick
When we approach a patient to offer research, it is important we are aware of our biases. Many of us will be hyper-alert to this need already, especially clinicians who are very experienced in delivering research.
But it is also essential that we are aware of the biases patients may have and understand that it is not unreasonable for them to have the starting point of assuming that anything we offer them will be treatment and will benefit them. After all, they have come to us most likely as someone seeking help. Of course research may help them and it may end up benefitting them. But it may not. And this is why we need to ensure that we genuinely are able to be in a state of equipoise and be aware of the concept of the therapeutic misconception in order to safely and correctly enter patients into research trials.