Scenarios: PeRSEVERE in practice

What is it?

We have put together a few example scenarios, showing how the PeRSEVERE principles can help manage these sorts of situations, or avoid them altogether.


Who is it for?

Anyone interested to learn more about the practical applications and impact of the PeRSEVERE principles.


More details

Below, you will find a series of hypothetical scenarios and some suggestions for how they should be managed or avoided by research teams.



Scenario #1

You are preparing to share a summary of the study results with the study’s participants. Some of the participants are recorded as having ‘withdrawn consent’, with differing implications for their ongoing participation – some withdrew from further clinical data collection, some withdrew from completion of further posted questionnaires, and some withdrew from all further researcher contact (and some a combination of these).

  • Who can the researchers contact?
  • How can the Persevere principles help prepare for this?
Read More

Who can you contact?

  • It seems reasonable to at least contact participants who are still involved in some way, i.e. those who are still participating in the study in some active way (and who have not withdrawn consent for further contact).
  • Consider how to make results available in ways that mean early-stopping participants can still find them easily, for example online and via routes that a general audience might be likely to look (e.g. via relevant charities).

  • Take advice from research sites – are any of these participants still in regular routine follow-up and are there opportunities to ask if they want the results?


How can the Persevere principles help prepare for this?

  • O2: the extent of participation change is participant’s choice. Knowing this, we should find out how participants want their involvement in a study to change, including to find out if they might want the study results or any other contact about the study after they stop taking part.

  • O7: stopping participation early does not affect participants’ right to receive information later on, if they want it. Just because they stop taking part in some or all of a study, we should not assume they do not want to find out the study results.

  • D2: clear instructions in the study protocol should remind research staff to check participants’ preferences for receiving the study results at the time participants stop or reduce their participation. Protocol instructions might also cover the approach to take at the time the study results are ready.

  • D3: potential participants should get clear information about stopping their participation, including that stopping participation will not mean they cannot have the study results, and that they should expect to be asked their preferences for further contact.

  • M1: researchers should collect good enough data about participation changes, including about participants’ preferences for receiving study results when they stop or reduce their involvement in a study. With good quality data, there should be less uncertainty at the time the study results are ready.



Scenario #2

A participant has previously decided to stop study intervention and study-related visits, but continues to see the same doctor as part of routine care. The doctor (who is the main study investigator at the site) knows that the final scheduled study questionnaire is particularly important for the study’s objectives.

The doctor has a discussion with the participant (instigated by the participant) at a routine healthcare visit about the overall study progress. This leads to the participant subsequently deciding to complete the questionnaire during the visit.

Some on the study team have suggested that the participant should formally reconsent to the trial by completing another consent form, before the questionnaire can be used in the study. However, it seems this will be difficult to do at short notice as the participant lives quite far from the site.

  • Can the questionnaire data be used for the trial?
  • How can the Persevere principles help prepare for this?
Read More


Can the questionnaire data be used?

  • It may not be right to assume the questionnaire data cannot be used – the participant might well have done it willingly, so discarding the data might not be fair to them.

  • Look at what the participant said before about how they wanted their participation to change. How does this new data provision fit with that?

  • Study researchers need to be reassured that the participant was not pressured or coerced into completing the questionnaire. It might be relevant to consider whether there is a big risk of coercion in this context, or how high the ‘stakes’ are for the different people involved. Might it be possible for the participant to confirm their wishes to someone other than the study doctor?

  • Questionnaire completion is not without risk for the participant, but is relatively low-risk compared to some other study activities. It is often considered in-scope for ‘proportionate approaches’ to consent. Might we consider the participant completing this questionnaire to imply their consent?

  • Completing a new consent form will be burdensome for the participant, especially if it needs to be done in person. There is unlikely to be an ethically approved consent form just for this purpose. So it may not be feasible to do this – is it fair to the participant to discard the information they provided, just because of this?

  • Excluding the data from the study will have a negative impact on the study results. Is that justified in this case?


How can the Persevere principles help prepare for this?

  • O2: the extent of participation change is the participant’s choice. This means participants can change their mind about stopping participation, as long as it still helps meet the study objectives.

  • O3: decisions about participation should be informed by understanding of the value of collecting the study data. This means a participant could make an informed, freely-given choice to provide one last questionnaire, with the knowledge that it could be helpful to the study.

  • O7: it is OK to keep a participant informed about the study even after they’ve stopped taking part. The participant might just be interested to hear how the study is going, even if they have no intention of making any more contributions to it.

  • D1: plan to allow reduced participation (e.g. completion only of final questionnaire) if participants might be interested to keep making a contribution. The study team in this case might have built the possibility into the protocol, so that if a participant is thinking of stopping participation, they could be offered the chance to stop all the questionnaires apart from the last one, if they want.

  • D5: encourage dialogue about participation to help discuss this sort of case. This might make it easier to discuss and agree for the participant to complete fewer questionnaires, rather than stopping them all.

  • D6: site staff should get training on how to handle these sorts of cases. Research staff are keen to do the right thing, and it can be challenging to serve the needs of individual participants and studies at the same time. Training might increase staff confidence in managing the initial participation change, and the situation where a participant wants to re-start making a contribution to the study.

  • M1: ensure good record keeping about participation changes. This includes a clear record of how the participant wants their participation to change at different times in the study. It also includes keeping a record of the discussions (for example in the participant’s medical notes), particularly at the time of that final questionnaire.



Scenario #3

A study has publicised its main study results at a conference, and the research team is currently working on the final publication for a suitable scientific journal.

In doing some final routine data checks, it has been discovered that a participant is recorded as having withdrawn their consent for all further data collection, a few months before they died.

The date of the participant’s death is relevant to the study results and has been sent (in error) by the study site to the research team. This issue has only just come to light and so the participant’s death has been included in the primary analysis that was already reported at the conference, and that is intended for inclusion in the main study publication.

  • Should the information about the participant’s death still be used in the main analysis, or should the analysis be redone?
  • What should happen to the report already published at the conference?
Read More

Should information about participant’s death be used in the main analysis, or should the analysis be redone? What should happen to the report already published at the conference?

This is a particularly difficult scenario, and the specific details would need to be reviewed carefully. The following are some of the possible considerations.

  • The analysis results are already in the public domain. Would this issue mean they should be retracted and removed from the public record?
  • The analysis was done using most up-to-date information available at the time, in good faith – this might be a reason to retain the record. The results in the public domain will be summary results, so the individual’s details won’t be visible to anyone.
  • The research team might have limited resources to redo the analysis, but this may not be a strong argument if it is agreed that redoing it would be the right thing to do.
  • It might be that removing the data from the analysis would not change the overall study results. Again, this may not be a strong argument if it is agreed that there are ethical reasons to redo the analysis.
  • Consider checking the records about the participant’s ‘withdrawal’ – what did they specifically say they wanted to do?
  • It is possible that the participant was (very) unwell around time they said they wanted their participation to stop. How might this impact the decision about the right approach to take?
  • Should the research team consult an independent ethics committee for their view?


How can the Persevere principles help prepare for this?

  • O5: data collection continues until participant says they want it to stop. Did the participant definitely say they wanted this? If not, are the right conditions met for the researchers to assume the prior consent still applies to that aspect of participation, implying an ethical case for the data to still be used?

  • D2: the protocol should include instructions to help ensure participants’ wishes are understood and recorded wherever possible. This should reduce the ambiguity, particularly in cases like this where it is no longer possible to ask the participant what they want to do.

  • D3: clear information for potential participants about what will happen if participation stops/reduces might put researchers on ‘firmer ground’ for their decisions about further data collection.

  • D6: training and support for research staff who interact with participants can help ensure participants’ wishes are understood and clearly recorded, even in difficult scenarios such as where participants’ health may be declining fast.

  • M1: better record keeping around participation changes helps to make sure participants’ wishes are recorded clearly. It is also important to have robust processes for carrying out participants’ wishes, for example stopping all further data collection if that’s what a participant has said they want.

  • M2: monitoring processes can check that participants’ wishes have been carried out correctly. In this case, they might have picked up the discrepancy earlier between the participant’s recorded wish for no further data collection and the receipt of further data.


Scenario #4

A study involves asking for participants’ consent for their biological samples to be collected in a ‘central’ lab for the purposes of collecting some study outcome data.

The samples will be in the form of paraffin blocks, which, after receipt, the central lab will section into slides then perform staining. They will then review each one and report back to the study team with the relevant outcome data. This will be done in ‘real time’ during the study, rather than in a batch at the end.

Occasionally, participants withdraw their consent for this activity to take place. In some cases, study sites do not inform the study team in a timely manner, so the study team is delayed in notifying the central lab of the withdrawals.

  • What should happen to samples that have already been sent to the lab when the withdrawal notification is received by the study team?
  • What should happen to the data has already been reported back to the study team when a withdrawal notification is received?
  • And what should be done in cases where a participant stops taking part in the study but their wishes about sample use are not clear?
Read More

In line with PeRSEVERE principle O6, data already collected for a study at the time a participant stops taking part should be retained and used in the study analysis. The same idea is not usually applied to how informed consent is applied to the use of biological samples (and the PeRSEVERE principles do not cover this as it is not in scope).

This means we might take a different approach to how samples are used, and what data is collected. The suggestions below set out a possible approach to take to deal with this potential complexity. In cases where data might be retained after a participant says they no longer want their biological samples to be used, this should only be done if it’s fair, as explained in principle O6. This includes being clear with participants about the approach before they agree to take part in the first place.


What should happen:

  • To samples already sent to the lab?
    • If they have not already been analysed, it is fair to not analyse them. Instead, they should be handled in line with the participant’s wishes (for example, they might need to be securely destroyed).
    • If they have already been analysed, the most appropriate action may depend on when the participant said they no longer wanted their samples used – if the sample had already been analysed and therefore study data had been generated before that date then perhaps the generated data can be used?

  • To data already reported to the study team?
    • The data probably can be used unless the date the participant said they no longer wanted their samples used was before the samples were actually analysed at the lab.

  • Where participants’ wishes are not clear?
    • Find out if it may be possible to seek clarification from the participant in a sensitive way.
    • See what participants were told previously about what would happen in this case. It might be appropriate to assume that prior consent is still in place, but only if certain conditions are met (see PeRSEVERE principle O5).
    • If participants have not been given clear information then the best approach may be a cautious one – i.e. do not use the samples.


How can the Persevere principles help prepare for this?

  • O5: this principle is about data collection continuing until a participant says they want it to stop. However, the same ideas can be applied to other situations (such as biological sample storage) and can guide the right approach where participants’ wishes about aspects of participation are not known.

  • D2: the study protocol should include instructions about this sort of participation change and the implications of different participant wishes, expressed at different times.

  • D3: give participants good information about what will happen to their samples if they stop taking part in the study – including what their choices are.

  • M1: good quality data collected about participants’ wishes, reported in a timely manner, gives everyone clarity about what participants want to do.

  • M2: monitoring may help ensure there is clear record-keeping and suitable processes at study sites and at the lab around participation changes.


Scenario #5

A patient has recently considered taking part in a study and completed the study consent form, but then decided that the proposed study treatment was not what they wanted due to fear of potential side-effects.

This led them to initially change their mind and withdraw their consent, just after randomisation but before starting any treatment.

However, the research nurse has now contacted the study team to say that the patient has changed their mind again and would now like to take part and receive the allocated treatment.

  • Can the patient do this?
Read More

Can the patient re-join the trial?

  • Yes, as long as:
    • The study team are satisfied that it is their freely-given and informed decision, based on the information from the site staff. They might check the record of discussions with the participant from that time (for example, in the medical notes) as part of monitoring activity for some extra reassurance.
    • The participant remains eligible to take part in the study.
    • The design/rules of the study allow it (aside from the eligibility question).


How can the Persevere principles help prepare for this?

  • O2: the extent of the participation change is the participant’s choice. This doesn’t mean the participant has limitless choices about how their participation change, but as long as the conditions above are met and the participant’s decision is informed and freely-given, there shouldn’t be a problem with them changing their mind and deciding to re-start the study.

  • D5: encourage dialogue about participation – this allows openness to discuss different (including unusual) choices.

  • D6: training and support can help research staff manage unusual situations such as this one.

Scenario #6

A participant in a study about mental health interventions stops responding to messages about their next intervention sessions. The study protocol does not say what to do in this situation.

The research site team concludes that they cannot carry on trying to deliver the intervention to the participant as they cannot reach them. They complete the required form to indicate that the study intervention is stopping, and send that to the study coordinating centre.

Automated text messages continue to be sent from the study coordinating centre to the participant requesting that they complete study questionnaires in line with the follow-up schedule.

Several months later, during routine communication, the research site team tell the study coordinating centre that the participant has ‘fully withdrawn’ from the study, although they have not had any further contact with the participant.

  • Was the coordinating centre team right to continue requesting data from the participant?
  • Has the research site team taken the right action in reporting a ‘full withdrawal’?
  • Would it be appropriate to make any further contact with the participant about the study?
  • Might any other corrective or preventative actions be needed?

Read More
  • Was the coordinating centre team right to continue requesting data from the participant?
    • The only information they had was that the participant was stopping study intervention, so it seems reasonable that they continued to request data from the participant.
  • Has the research site team taken the right action in reporting a ‘full withdrawal’?
    • Based on the information here, it is unclear why they are reporting a ‘full withdrawal’. They have lost contact with the participant, but they do not know that the participant wants to stop taking part. That is one possible conclusion, but would be an assumption and may not be correct.
  • Would it be appropriate to make any further contact with the participant about the study?
    • Given the participant’s wishes are not known, it could be reasonable to attempt further contact with the participant. However, if they have not responded to several contact attempts then it might be best to stop trying, or stop trying for now. The context of the study (mental health) may be relevant to the decision about this.
  • Might any other corrective or preventative actions be needed?
    • As the text messages are automated, the study team should ensure these are switched off if it is agreed that the study has lost contact with the participant.
    • There may be a need to discuss the issue with the research site to get a shared understanding around the difference between loss of contact with a participant, and a participant expressing a wish to stop taking part.
    • It would be good for the protocol to be amended to address this issue, providing instructions and guidance for any future, similar cases.
  • How can the Persevere principles help prepare for this?
    • O2: it is primarily for participants to decide how their level of involvement should change. Adherence to this idea should avoid any members of the study team making assumptions about what the participant wants to do.
    • O4: it is important for everyone involved in studies to understand the different implications of losing contact with a participant and a participant expressing a wish to stop taking part.
    • O5: data collection should continue until a participant says they want it to stop. In this case, without such a statement from the participant, the study team can be confident in their justification to request further study data.
    • D2: the protocol should contain clear instructions about how to handle loss of contact between participants and research staff.
    • D4: participants should get clear information about what will happen if they lose contact with study staff, so that they know what to expect.
    • D5: the consent and recruitment processes should be carried out in ways that encourage participants to say if they are having difficulties taking part. We don’t know that this was not done in this case, but it is an important idea that can help avoid this sort of issue.
    • D6: good training and support for all involved can help them take the most appropriate action, with confidence.
    • M1: good quality data about participation changes would have led to the study coordinating centre team being aware earlier of the nature of the participation change.

Scenario #7

A study participant returns a postal questionnaire with most questions left blank. They write a note at the end, saying that they want to withdraw from the study. They say they feel the questions are not relevant to them, and that they wish they had not joined the study.

The questionnaire is returned directly to the study coordinating centre. The comments are not properly escalated by the person who deals with the form, so two additional questionnaires are sent to the participant before the study team become fully aware of the issue. One of the questionnaires is returned, again with comments from the participant about wanting to stop taking part.

In this study, participants are asked to give consent for data about them to be collected directly from their medical notes a while after they join. The study team establish that data has been collected from this participant’s medical notes between the time of the first returned questionnaire and the team’s full awareness of the issue.

  • Should the study team use the data they collected from the medical notes?
  • The study team would like to contact the participant to apologise. Can they do this, if the participant has said they want to withdraw from the study?

Read More

  • Should the study team use the data they collected from the medical notes?
    • It is problematic that the participant’s request to stop taking part in the study was not picked up in a timely manner.
    • The participant’s request to stop taking part was quite general, just saying that they wanted to ‘withdraw’ from the study. In theory, they might be contacted to try to ascertain exactly which aspects of the study they want to stop. However, this could be difficult to do sensitively in this case, where it seems likely the participant is unhappy or upset out their experience in the study.
    • Therefore, it might be best to not contact the participant about this and take a conservative approach, assuming their wish to withdraw means stopping all possible aspects of the study. This would mean it is best not to use or retain the data collected from the participant’s medical notes.
  • The study team would like to contact the participant to apologise. Can they do this, if the participant has said they want to withdraw from the study?
    • Although it might be reasonable to conclude that the participant does not want any further contact with the study, it could be the right thing to do to contact the participant to acknowledge the error that was made. This would need to be done sensitively, but the fact of the participant having stopped taking part in the study does not rule out the contact.
  • How can the Persevere principles help prepare for this?
    • M1 is about collection of informative data about participation changes. In this case, the ‘data’ available to the study coordinating centre was just a handwritten comment, but they should have processes in place to be able to identify such issues.
    • O2 says that it is for participants to decide how their participation changes. One implication of this is that study teams should make it clear to participants what their options are, and how they can communicate them. Perhaps the study documentation in this case could have made clearer what the ‘expected’ routes would be for the participant to notify the study team that they wanted to stop taking part.
    • O5 says that data collection can continue until a participant says they want it to stop. It might have been possible for that to happen in this scenario, if the study team had been clear about the approach in advance. It might have been helpful to notify the participant of this intention to continue collecting data (including why that would be useful to the study) soon after the participant stopped taking part. Principle O5 does make allowances for cases where further data collection cannot be justified, like this one.