Participant “withdrawal” from a study: what it really means and how to manage it

Published on Monday, 13 January 2025

People who agree to participate in a research study have the legal right to stop taking part, or change their involvement in the research, at any time. This is typically described as participants “withdrawing” from a research study, and historically has been talked about in all-or-nothing terms: participants are either taking part, or they are not. However, in reality, things may not be that simple. A typical NPEU Clinical Trials Unit (CTU) randomised controlled trial recruits premature or sick babies and collects data both from their hospital records and via parent-completed questionnaires, and there are multiple ways in which a baby's and their parents' involvement may change during their time on the trial (see Figure 1).

Do all (or any) of these count as “withdrawal”?

Do all (or any) of these mean we will no longer receive any trial data for the baby?

What do we mean when we say “withdrawal” anyway?

While it is vital that participants in research are protected and able to reduce their involvement in a study without difficulty, it is also imperative that studies produce accurate, relevant, and meaningful results. Randomised controlled trials with a low retention rate, or a high proportion of missing data, may fail to identify a difference between treatment arms, or may produce results that are viewed as unreliable. Therefore, collecting as much of the required data as possible on the maximum number of randomised participants is important. Both recruiting site staff and participants may assume that if someone does not want to continue providing data, or doesn't want to continue receiving a trial intervention, none of their previously-provided data will be relevant or will contribute to the findings. This is not the case at all.

For example, perhaps a parent no longer wants to complete trial questionnaires about their baby, but is happy for the trial to continue collecting data from their baby's clinical notes. If they are given the option to do this – continuing involvement in the trial in a way that works better for them – it grants them autonomy and also provides some usable, relevant data for the trial. If the parents only had an all-or-nothing choice (as in Figure 2) or being entirely involved or withdrawn entirely, the trial would lose data unnecessarily.

In addition, people often have altruistic motives for taking part in research, wanting to help others experiencing their (or their child's) condition. Offering only an all-or-nothing choice reduces their ability to do this and may make it harder for them to withdraw entirely (for instance, if they feel they are “letting the trial down”, or that choosing to step away would make their earlier involvement meaningless). Allowing varying levels of involvement (as shown in Figure 3) means they can still make a contribution.

Participants (or their parents or carers in the case of a neonatal trial) may want to reduce their level of involvement in a study rather than halt it entirely, or they may even want to increase their level of involvement (for instance, if they originally declined to take part in an aspect of the trial). Ideally, a trial should be able to accommodate these types of requests.

The PeRSEVERE project (PRincipleS for handling end-of-participation EVEnts in clinical trials Research) (https://persevereprinciples.org), initiated within the UKCRC Registered Clinical Trials Unit Network, has investigated this in depth and produced a list of principles to guide everyone who is involved in managing participant changes in research. It emphasises not only the importance of participant autonomy and data completeness, but the need for triallists to consider what types of “withdrawal”, or changes in involvement, are possible and practical, and what systems are needed in order to manage these efficiently. The recently-released World Health Organisation guidance on best practice for clinical trials also highlighted the importance of clarifying with participants their wishes around changing their involvement in a research study.

Thus, at the NPEU CTU, we are moving away from the blanket term “withdrawal” and instead talking about “change of consent”. For each of our trials, we are working to delineate what that might mean, and what changes are possible. This includes adding features to our systems to allow these different changes to be recorded and correctly implemented. This work, as well as considering how to communicate key principles of withdrawal (and what we mean by it) to parents, families, recruiting site staff, and each other, is an ongoing and complex journey, but vital in order to maximise the benefit of our trials and allow participants to make informed choices.

Madeleine Hurd
Data Manager, NPEU Clinical Trials Unit