By Akshay Patel, Senior Statistician at Exploristics
What is an estimand?
To facilitate clarity on precisely what a clinical trial intends to investigate, international trial regulatory guidelines (ICH E9(R1)) call for trialists to clearly define the treatment effects being investigated by using estimands. Following the ICH guidance, estimands are being recognised as a critical component of achieving patient-centricity in quality healthcare, which allows clear evaluation of treatments on patients’ lives. This will ensure trials provide the best evidence to define patient care.
An estimand is a precise statement of a research question and is the thing we want to estimate. Defining estimands is more than saying simply that we want to estimate the effect of a new treatment. Estimands also define who we count as having been successfully treated (what if people discontinue treatment?), and what population are we describing (if some people die before their outcomes are measured, we may find we are drawing conclusions on healthier people, who were able to survive the whole trial).
Why should we involve patients in discussions about estimands?
A key pain point for developers of new medicines is the recruitment and retention of clinical trial participants. This can be addressed to some extent by involving patients earlier in discussions around the planning of clinical trials. By bringing them in to the trial design process, patients feel that their input into clinical research is valued, and so are more likely to stay engaged in the clinical trials process.
Now with increasing attention being placed on using estimands in trial design, there is an opportunity to get patients involved in this critical part of trial design. This would be beneficial to many stakeholders. Not only could this encourage patient participation in clinical research, but it would also provide these patients with the direct benefit of evaluating treatments more clearly enabling them to make more informed decisions about their treatment choices. This in turn supports improve patient-centred outcomes and the evaluation of which treatments should be funded by healthcare systems and payers.
How can we support discussions on estimands?
While involving patients in defining estimands would benefit multiple stakeholders, what evidence is there that they would engage with this and how can it be facilitated in practice? As part of an NIHR Fellowship at Queen Mary University of London, I had the opportunity to work on a patient tool specifically designed to support the communication of estimands to public partners (i.e., patients and/or members of the public who are involved with or provide input to the research team) . During this project, the team:
- co-developed a practical estimands explainer tool with public partners that describes what an estimands is and what impact it may have in trial results interpretation.
- explored public partners’ perspectives on the value of discussing estimands with them when designing a trial.
Developing a patient tool to explain estimands
Public partner interaction with the tool starts with being taken through an analogy of buying a new car and explains why individuals need to ask clear and specific questions about a car to understand whether it will perform as desired for them, as different individuals have diverse needs (Figure 1.). This example was used to illustrate how when testing a new treatment, all stakeholders need to be clear about the precise question being asked, so that the answer is meaningful and useful. The tool then takes users through further examples which demonstrate why this is important.